Giant Arachnoid Granulations Mimicking Pathology

VOLUME 27 - No. 3 - JUNE 2014 CENTAURO S.r.l., BOLOGNA Bimestrale - Poste Italiane s.p.a. - Sped. in a.p. - D.L. 353/2003 (conv. in L. 27/02/2004 n°46) art. 1, comma 1, DCB/BO Euro 30,00 ISSN 1971-4009 Official Journal of: AINR - Associazione Italiana di Neuroradiologia and: The Neuroradiologists of Alpe-Adria ANRS - Albanian Neuroradiological Society PANRS - Pan Arab NeuroRadiology Society Radiological Society of Saudi Arabia, Division of Neuroradiology Egyptian Society of Neuroradiology ISNR - Indian Society of Neuroradiology Indonesian Society of Neuroradiology Neuroradiology Section of the Radiology Society of Iran Israeli Society of Neuroradiology College of Radiology Malaysia Neuroradiology Section - Pakistan Psychiatry Research Center Section of Neuroradiology - Polish Radiological Society The Neuroradiologists of Romania Section of Neuroradiology of Serbia and Montenegro SILAN - Sociedad Ibero Latino Americana de Neurorradiologia Neuroradiology Section of Singapore Radiological Society Slovenian Society of Neuroradiology The Neuroradiological Society of Taiwan TSNR - Turkish Society of Neuroradiology d istrib ute d b y Via Ne rvia no , 31 20020 La ina te (Mi) te l. +39 02 93305.1 fa x.+39 02 93305400 www.a b me d ic a .it Official Journal of: AINR - Associazione Italiana di Neuroradiologia and The Neuroradiologists of Alpe-Adria ANRS - Albanian Neuroradiological Society PANRS - Pan Arab NeuroRadiology Society Radiological Society of Saudi Arabia, Division of Neuroradiology Egyptian Society of Neuroradiology ISNR - Indian Society of Neuroradiology Indonesian Society of Neuroradiology Neuroradiology Section of the Radiology Society of Iran Israeli Society of Neuroradiology College of Radiology Malaysia Neuroradiology Section - Pakistan Psychiatry Research Center Section of Neuroradiology - Polish Radiological Society The Neuroradiologists of Romania Section of Neuroradiology of Serbia and Montenegro SILAN - Sociedad Ibero Latino Americana de Neurorradiologia Neuroradiology Section of Singapore Radiological Society Slovenian Society of Neuroradiology The Neuroradiological Society of Taiwan TSNR - Turkish Society of Neuroradiology Index Generalia Brain Magnetic Resonance Imaging: Perception and Expectations of Neurologists, Neurosurgeons and Psychiatrists 261 Paulo Branco, Margarida Ayres-Basto, Pedro Portugal, Isabel Ramos, Daniela Seixas 293 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT 299 Wieslaw L. Nowinski, Ryszard S. Gomolka, Guoyu Qian, Varsha Gupta, Natalie L. Ullman, Daniel F. Hanley Assessment of Cerebrospinal Fluid Flow 268 Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early Clinical Experiences Kayoko Abe, Yuko Ono, Hiroko Yoneyama, Yu Nishina, Yasuo Aihara, Yoshikazu Okada, Shuji Sakai The Etiology of Ring Lesions on Diffusion-Weighted Imaging 280 Lentiform Fork Sign: a Magnetic Resonance Finding in a Case of Acute Metabolic Acidosis 288 Pasquale F. Finelli, Ethan B. Foxman Daniela Grasso, Carmela Borreggine, Francesco Perfetto, Vincenzo Bertozzi, Marina Trivisano, Luigi Maria Specchio, Gianpaolo Grilli, Luca Macarini Innervation of the Cerebral Dura Mater Xianli Lv, Zhongxue Wu, Youxiang Li Giant Arachnoid Granulations Mimicking 316 Pathology. A Report of Three Cases Bart De Keyzer, Sven Bamps, Frank Van Calenbergh, Philippe Demaerel, Guido Wilms Progressive Multifocal Leukoencephalopathy: a Rare Cause of Cerebellar Edema and Atypical Mass Effect. A Case Report 322 Varicella Zoster CNS Vascular Complications. A Report of Four Cases and Literature Review 327 Chris Ojeda, Rachid Assina, Maureen Barry, Ada Baisre, Chirag Gandhi Francisco Chiang, Theeraphol Panyaping, Gustavo Tedesqui, Daniel Sossa, Claudia Costa Leite, Mauricio Castillo Cover: Ferdinand Hodler (1853-1918) Landscape on Lake Geneva, 1906. oil on canvas, 59,8 x 84,5 cm. © Frank Kovalchek - http://www.flickr.com Histoplasmosis Brain Abscesses in an Immunocompetent Adult. A Case Report and Literature Review 334 Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis 339 Ana Ines Andrade, Maren Donato, Carlos Previgliano, Mardjohan Hardjasudarma Atchayaram Nalini, Aaron de Souza, Jitender Saini, Kandavel Thennarasu Aneurysms Effect of Electromagnetic Radiation on the Coils Used in Aneurysm Embolization 350 Xianli Lv, Zhongxue Wu, Youxiang Li Letter to the Editor CT Angiography Source-Images and CT Perfusion: Are They Complementary Tools for Ischemic Stroke Evaluation? Response to: Reliability of CT Perfusion in the Evaluation of Ischaemic Penumbra Nicola Morelli, Eugenia Rota, Paolo Immovilli, Ilaria Iafelice, Emanuele Michieletti, Donata Guidetti, John Morelli Response to Letter to the Editor 368 "CT Angiography Source-Images and CT Perfusion: Are They Complementary Tools for Ischemic Stroke Evaluation?" José Eduardo Alves, Ângelo Carneiro, João Xavier Letter to the Editor 369 Response to Letter to the Editor 370 Erratum Brain targets: can you believe your own eyes? 371 Laurent Spelle, Thomas Liebig Early Endovascular Treatment 356 of Aneurysmal Subarachnoid Hemorrhage Complicated by Neurogenic Pulmonary Edema and Takotsubo-Like Cardiomyopathy Andrea Manto, Angela De Gennaro, Gaetana Manzo, Antonietta Serino, Gaetano Quaranta, Claudia Cancella 365 Robert Hurst Stefanoni G, Tironi M, Tremolizzo L, Fusco ML, DiFrancesco JC, Patassini M, Ferrarese C, Appollonio I Spine An Unusual Case of Isolated Hypoglossal 361 Nerve Palsy Secondary to Osteophytic Projection from the Atlanto-Occipital Joint Satya Narayana Patro, Carlos Torres, Roy Riascos AINR News In memoriam Dr Mario Savoiardo 372 Books Information and Congresses Instructions for Authors 375 260 381 Indexed in: National Library of Medicine’s MEDLINE database (http://www.ncbi.nlm.nih.gov/pubmed/ - Search: neuroradiol j) EMBASE - Scopus (http://www.scopus.com) Google Scholar (http://scholar.google.com) ISSN 1971-4009 Take Control. Capture More. 48% Less Delivery Force Easy to Deliver Easy to Place Easy to See • 48% less delivery force required as demonstrated in bench testing* • Distal FlexCell Design including softer distal segment,† shorter landing zone‡ and bright radiopaque markers • Full-length radiopacity†† for precise placement, visible integration and interactive retrieval See for Yourself: TrevoXP.com All photographs taken by Stryker Neurovascular. Bench test results may not necessarily be indicative of clinical performance. Testing completed by Stryker Neurovascular. Data on file and available upon request. * Bench testing included Trevo XP ProVue, 4x20mm (n=38) and Solitaire FR, 4x20mm (n=5). † Bench testing included Trevo™ XP ProVue, 4x20mm (n=57) and Solitaire FR, 4x20mm and 4x15mm (n=8). ‡ Compared to Trevo™ ProVue Retriever. †† Bench model photo. Solitaire is a trademark of Tyco Healthcare Group LP composed of Covidien, Inc. INDICATIONS FOR USE The Trevo Retriever is intended to restore blood flow in the neurovasculature by removing thrombus in patients experiencing ischemic stroke within 8 hours of symptom onset. Patients who are ineligible for intravenous tissue plasminogen activator (IV t-PA) or who fail IV t-PA therapy are candidates for treatment. THIS DOCUMENT IS INTENDED SOLELY FOR THE USE OF HEALTHCARE PROFESSIONALS. A physician must always rely on his or her own professional clinical judgment when deciding whether to use a particular product when treating a particular patient. Stryker does not dispense medical advice and recommends that physicians be trained in the use of any particular product before using it in a procedure. The information presented is intended to demonstrate the breadth of Stryker product offerings. A physician must always refer to the package insert, product label and/or instructions for use before using any Stryker product. Products may not be available in all markets because product availability is subject to the regulatory and/or medical practices in individual markets. Please contact your Stryker representative if you have questions about the availability of Stryker products in your area. The Stryker products listed above are CE marked according to the Medical Device Directive 93/42/EEC. Copyright © 2014 Stryker NV00007584.AA Stroke: Our Only Focus. Our Ongoing Promise. CORSO ITINERANTE 2014 Ospedale Bellaria (Bologna) 23-24 Ottobre 2014 AINR - CONSIGLIO DIRETTIVO Massimo Gallucci, Presidente Alberto Beltramello, Vice Presidente Fabio Triulzi, Past President Ferdinando Caranci, Segretario Carla Uggetti, Tesoriere Consiglieri: Marcello Bartolo, Massimo Caulo, Francesco Causin, Elisa F.M. Ciceri, Francesco Di Paola, Maria Ruggiero Coordinatori di Sezione: Alessandro Bozzao, Neuroradiologia Funzionale Salvatore Mangiafico, Neuroradiologia Interventistica Bruno Bernardi, Neuroradiologia Pediatrica -----------------------------------------------------------------------------------------Cari Amici, Sono molto grato al CD dell’AINR che mi ha incaricato di organizzare il Corso Itinerante, che si terrà a Bologna il 23 e 24 Ottobre 2014. Una data prossima alla conclusione della mia attività di Direttore del Servizio di Neuroradiologia, fondato nel 1960 dal Prof. Giovanni Ruggiero all’Ospedale Bellaria, oggi IRCCS delle Scienze Neurologiche di Bologna. Il prossimo 1 novembre 2014 infatti andrò in pensione, dopo 47 anni di vita ospedaliera, e questo sarà l’ultimo mio impegno culturale ufficiale. Mi offre l’occasione di presentare tutti i miei collaboratori, una équipe della quale sono molto fiero . Il programma del Corso è stato preparato cercando di essere esaustivi dal punto di visto scientifico, volendo però trasmettere anche quella che è la nostra visione della Neuroradiologia: una Disciplina medica completa, a tutto tondo, dall’anatomia alla funzione, dal metabolismo all’interventistica. Un forte benvenuto a tutti coloro che si iscriveranno al Corso. Ricordo che l’iscrizione è gratuita. La sede sarà presso l’Ospedale Bellaria. Il numero massimo di partecipanti è di 90 persone. Per ogni informazione e per l’iscrizione rivolgetevi direttamente a me. A presto Prof. Marco Leonardi Cattedra di Neuroradiologia - Università di Bologna IRCCS delle Scienze Neurologiche Ospedale Bellaria, Via Altura 3 - 40139 Bologna Tel.: +39 348871453 - e-mail: marco.leonardi@unibo.it marco.leonardi@centauro.it Orario 23 ottobre 2014 - Moderatore: Mino Andreula 10:00 Evoluzione della neuroradiologia - Marco Leonardi 10:30 Anatomia cerebrale e nervi cranici - Massimo Gallucci IL NEURORADIOLOGO CLINICO L’importanza della correlazione clinico-radiologica Moderatori: Andreula, Righini, Scotti 11:00 I reperti occasionali - Hodman Ahmed Sheikh Maye 11:30 L’approccio al paziente pediatrico - Monica Maffei 12:00 Patologia della sostanza bianca - Stella Battaglia 12:30 Idrocefalo - Fiorina Bartiromo, Anna Federica Marliani 13:00 Epilessia e ricerca della focalità Anna Federica Marliani, Fiorina Bartiromo 13:30 – 14:30 - PAUSA PRANZO (PARCO) Moderatori: Falini, Salvolini, Triulzi 14:30 Il conflitto neurovascolare - Luisa Raffi IL NEURORADIOLOGO RADIODIAGNOSTA Conoscenza delle tecniche 15:00 Angio RM senza e con mdc – Francesco Toni 15:30 Angio TC Dinamica - Luigi Cirillo 16:00 Tecniche di diffusione e perfusione - Raffaele Agati 16:30 Ruolo, tecniche e modalità degli studi di attivazione: Il coma - Daniela Cevolani 17:00 Presentazione di casi clinici Orario 24 ottobre 2014 - Moderatore: Marco Leonardi 10:00 La qualità in neuroradiologia - Patrizia Cenni IL NEURORADIOLOGO CHIRURGO L’indicazione al trattamento … Moderatori: Briganti, Longo, Sirabella 10:30 Traumi cranio-encefalici - Carlotta Barbara 11:00 Grading e guida alla biopsia nei tumori Antonalle Bacci 11:30 Diagnosi della regione sellare Monica Messia, Antonella Bacci 12:00 Le malformazioni vascolari cerebrali - Luigi Cirillo 12:30 La patologia steno-occlusiva vascolare – Luigi Simonetti 13:00 Il rachide degenerato - Luca Albini Riccioli 13:30 – 14:30 - PAUSA PRANZO (PARCO) IL NEURORADIOLOGO CHIRURGO … e la terapia Moderatori: Beltramello, Carollo, Cirillo 14:30 Il trattamento: Fistole durali e MAV - Massimo Dall’Olio 15:00 Il trattamento: Aneurismi - Ciro Princiotta 15:30 Il trattamento: ictus ischemico - Luigi Simonetti 16:00 Il trattamento: ernia al disco - Fabio De Santis 16:30 Compilazione questionario Conclusione e Saluto Marco Leonardi The Neuroradiology Journal 27: 261-267, 2014 - doi: 10.15274/NRJ-2014-10051 www.centauro.it Brain Magnetic Resonance Imaging: Perception and Expectations of Neurologists, Neurosurgeons and Psychiatrists PAULO BRANCO1, MARGARIDA AYRES-BASTO2, PEDRO PORTUGAL1, ISABEL RAMOS3, DANIELA SEIXAS1,4 Department of Imaging, Centro Hospitalar de Vila Nova de Gaia/Espinho; Vila Nova de Gaia, Portugal Department of Neuroradiology, 3 Department of Radiology, Centro Hospitalar São João; Porto, Portugal 4 Department of Experimental Biology, Faculty of Medicine, University of Porto; Porto, Portugal 1 2 Key words: ethics, interdisciplinary communication, magnetic resonance imaging, neurology, neurosurgery, psychiatry, radiology SUMMARY – Magnetic resonance imaging (MRI) has rapidly become an essential diagnostic tool in modern medicine. Understanding the objectives, perception and expectations of the different medical specialties towards MRI is therefore important to improve the quality of the examinations. Our aim was to better comprehend the reasons and expectations of neurologists, neurosurgeons and psychiatrists when requesting brain MRI scans for their patients, and also to perceive the degree of confidence of these specialists in the images and respective reports. Sixty-three specialists were recruited from two tertiary hospitals and answered a tailored questionnaire. Neurosurgeons were more concerned with the images themselves; neurologists lacked confidence in both MRI images and reports, and one third of the psychiatrists only read the report and were the most confident of the specialties in MRI findings. These results possibly reflect the idiosyncrasies of each of these medical specialties. This knowledge, driven by efficient communication between neuroradiologists and neurosurgeons, neurologists and psychiatrists, may contribute to improve the quality of MRI examinations and consequently patient care and management of health resources. Introduction Since its development in the 1980s, magnetic resonance imaging (MRI) has revolutionised medical practice and today has a crucial role both in diagnosis and research 1. There are various medical specialties that routinely request MRI studies of the brain for patient care, mostly neurology, neurosurgery and psychiatry. Due to the intrinsic differences among these medical specialties, their perception of several aspects of a conventional MRI study may differ accordingly. Neurosurgeons typically use MRI for surgical planning and patient follow-up and relatively less for making a diagnosis. Their practice relies heavily on image guidance technology 1. In contrast, neurologists use MRI mainly for diagnostic purposes. Neurologists do not seem to always depend on a neuroradiological report in order to make clinical decisions 2. Psychiatrists are not looking for “organic” proof to establish diagnosis, because most psychiatric diseases do not produce lesions visible on routine anatomical MRI 3. In addition, differentiation of various psychiatric neuropathologies is rather poor based on MRI findings because few “organic” indicators found in such diseases are frequently shared across a broad range of psychiatric conditions 3. Neuroradiologists should understand the specificities associated with medical practices involving imaging in order to improve the quality of their service and/or shape the clinicians’ perception of the quality of the service. According to Wallis and McCoubrie, the radiological report seems to be the main method of communication between radiologists and clinicians 4. Hence, neuroradiological reports can be potentially optimised for a specific specialty to im261 Brain Magnetic Resonance Imaging: Perception and Expectations of Neurologists, Neurosurgeons and Psychiatrists prove communication with clinicians and have a greater impact on diagnosis, disease monitoring and treatment 5. Studies addressing this issue are scarce, although it has been shown that many malpractice cases may derive from poor communication 4. Radiological reports that do not address the clinician’s question, use unfamiliar terms and abbreviations, or include statements out of clinical context stand as the main complaints 6,7. Furthermore, only 38% of clinicians read the entire report whereas 43% only read the summary if the report has more than one page 7. It seems that miscommunication plays an important role in the relation of radiologists and the other specialties. The purpose of this study was to characterise the perception and expectations of neurosurgeons, neurologists and psychiatrists regarding conventional MRI of the brain. Materials and Methods Hospital Centres Neurologists, neurosurgeons and psychiatrists were recruited from two medical centres, Centro Hospitalar São João (CHSJ) and Centro Hospitalar de Vila Nova de Gaia/Espinho (CHVNG), Portugal. CHSJ and CHVNG are both public tertiary referral hospital centres of two neighbouring cities in the northern region of Portugal. At the time of the study, CHSJ had 1124 beds and an influence area of 3,000,000 inhabitants 8 and CHVNG had 558 beds and an influence area of 700,000 inhabitants 9. These centres had all the medical specialties, including neurology, neurosurgery, psychiatry and neuroradiology. CHSJ had seven neuroradiologists and an MRI Unit with two MRI scanners (1.5T and 3T) and CHVNG had four neuro- radiologists and one MRI scanner (1.5T). Annual records from the year of the present data showed a total of 3293 brain MRIs performed in CHSJ, most of which (60.8%) were requested by neurologists (646), neurosurgeons (1329) and psychiatrists (26). In CHVNG, 1205 brain MRIs were performed, most of which (58.6%) were also requested by these three specialties (430, 268 and eight respectively). Participants Sixty-three clinicians from the two hospitals participated in the study: 20 neurologists, 15 neurosurgeons and 28 psychiatrists. Table 1 describes the subject sample. Ethical approval was not obtained because this study is a questionnaire study (descriptive research) and participants are not patients and are able to consent. Informants consented to take part by filling in the questionnaire and handing it to the researcher. Prior to filling in the questionnaire, the informants received information on the study, what participating entailed, told that anonymity was assured and that they had the right to withdraw at any time for any reason. Instrument Each participant filled in a tailored questionnaire (Figure 1) addressing the objectives, expectations and use of brain MRI in the subject’s professional practice that included open multiple-choice questions and visual horizontal bars (100 mm). Subjects were instructed to select the best given choice for all questions, and if necessary to use the open field or multiple choices. For visual bars, subjects were instructed to mark an “X” on the scale, proportional to the certainty of the answer. Visual bars were manu- Table 1 Characteristics of the subject sample. CHSJ CHVNG TOTAL 37 (10.2) 40.5 (10.4) 38.4 (10.4) Male n=20 n=15 n=35 Female n=19 n=9 n=28 Neurologists n=13 n=7 n=20 Neurosurgeons n=10 n=5 n=15 Psychiatrists n=16 n=12 n=28 Mean age in years (SD) Gender CHSJ – Centro Hospitalar São João CHVNG – Centro Hospitalar de Vila Nova de Gaia/Espinho 262 Paulo Branco www.centauro.it The Neuroradiology Journal 27: 261-267, 2014 - doi: 10.15274/NRJ-2014-10051 1. Age _______________ 2. Gender • Female • Male 3. • • • • Պ Պ Medical Speciality Neurology Պ Neurosurgery Պ Psychiatry Պ Other Պ 4. Specialist or resident? Specialist Պ Resident Պ 5. In your daily hospital practice, what is the most common reason for requesting a brain magnetic resonance imaging examination? • To confirm a diagnostic hypothesis Պ • To exclude another diagnosis Պ • To follow-up an already known pathology Պ • To make sure I do not miss anything Պ • Other(s) _________________________________________________ 6. • • • • • • In your daily hospital practice, what do you usually expect from a magnetic resonance imaging examination? To confirm my diagnostic hypothesis Պ To suggest another diagnosis Պ To present a negative result Պ To present a nonspecific result Պ To present an unexpected result Պ Other(s) _________________________________________________ 7. What is your level of confidence in the quality of images and protocols of brain magnetic resonance imaging examinations? No confidence Maximum confidence 8. What is your level of confidence in the quality of the reports of brain magnetic resonance imaging examinations? No confidence 9. • • • • Maximum confidence How do you usually evaluate the result of a brain magnetic resonance imaging examination? I only read the report Պ I read both the report and the images Պ I only read the images Պ Other(s) _________________________________________________ Thank you for your collaboration! Figure 1 Questionnaire. Tailored questionnaire applied to the sample of neurologists, neurosurgeons and psychiatrists (translated into English for comprehension). ally measured at the centre point of the cross in millimetres. Data were anonymised to ensure privacy protection. way ANOVA with post-hoc analysis using the Tukey test were conducted to compare groups. Statistical Analyses Results Data were processed using SPSS 21.0. Descriptive statistics were used to evaluate the reason for requesting an MRI examination, the expectations regarding the MRI and the assessment of the examination result. One- Regarding the most common reason for requesting a brain MRI examination (Table 2), 95% of the neurologists and 93% of the neurosurgeons answered “to confirm a diagnostic hypothesis”. By contrast, only 39% of the psy263 Brain Magnetic Resonance Imaging: Perception and Expectations of Neurologists, Neurosurgeons and Psychiatrists chiatrists wanted to “to confirm a diagnostic hypothesis”, preferring instead “to exclude another diagnosis” (57%) with brain MRI. Fiftythree per cent of the neurologists also wanted “to exclude another diagnosis”, while neurosurgeons rarely did so (20%). Forty-two per cent of Paulo Branco the neurologists chose additionally “to follow-up an already known pathology”, unlike the neurosurgeons (27%) and psychiatrists (7%). Finally, 16% of the neurologists selected “to make sure I do not miss anything”, compared to only 7% of neurosurgeons and 4% of psychiatrists. Table 2 Most common reasons for neurologists, neurosurgeons and psychiatrists requesting a brain magnetic resonance imaging scan. CHSJ CHVNG TOTAL Question Neurologists NeuroPsychiasurgeons trists Neurologists NeuroPsychiasurgeons trists Neurologists NeuroPsychiasurgeons trists “To confirm a diagnostic hypothesis” 100% 90% 31% 83% 100% 50% 95% 93% 39% “To exclude another diagnosis” 54% 10% 63% 50% 40% 50% 53% 20% 57% “To follow-up an already known pathology” 46% 10% 0% 33% 60% 17% 42% 27% 7% “To make sure I do not miss anything” 23% 0% 6% 0% 20% 0% 16% 7% 4% Other(s) 0% 0% 0% 0% 20% 0% 0% 7% 0% CHSJ – Centro Hospitalar São João CHVNG – Centro Hospitalar de Vila Nova de Gaia/Espinho Table 3 Expectations of neurologists, neurosurgeons and psychiatrists on the result of brain magnetic resonance imaging examinations. CHSJ CHVNG TOTAL Question Neurologists NeuroPsychiasurgeons trists Neurologists NeuroPsychiasurgeons trists Neurologists NeuroPsychiasurgeons trists “To confirm my diagnostic hypothesis” 85% 100% 63% 100% 100% 92% 90% 100% 75% “To suggest another diagnosis” 69% 0% 25% 14% 40% 17% 50% 13% 21% “To present a negative result” 0% 0% 13% 0% 20% 17% 0% 7% 14% “To present a nonspecific result” 8% 0% 0% 0% 20% 0% 5% 7% 0% “To present an unexpected result” 8% 0% 0% 0% 20% 0% 5% 7% 0% Other(s) 8% 0% 6% 0% 0% 0% 5% 0% 3% CHSJ – Centro Hospitalar São João CHVNG – Centro Hospitalar de Vila Nova de Gaia/Espinho 264 www.centauro.it The Neuroradiology Journal 27: 261-267, 2014 - doi: 10.15274/NRJ-2014-10051 Figure 2 Confidence in brain magnetic resonance images, protocols and examination reports. Graphic representations across specialties – Neurology, Neurosurgery and Psychiatry – illustrating confidence in brain magnetic resonance images, protocols and reports. With respect to the expectations on brain MRI examinations (Table 3), all specialties expected to confirm their diagnostic hypothesis (100% of the neurosurgeons, 90% of the neurologists and 75% of the psychiatrists). Fifty per cent of the neurologists, 21% of the psychiatrists and 13% of the neurosurgeons also expected that brain MRI suggested another diagnosis. The only specialties that anticipated brain MRI “to present a negative result” were psychiatry (14%) and neurosurgery (7%). Furthermore, only 5% of the neurologists and 7% of the neurosurgeons selected “to present an unexpected result”, and “to present a nonspecific result”. None of the psychiatrists selected these last two options. When questioned about their assessment of the brain scan results, most of the neurologists (90%) and neurosurgeons (93%) read both the report and the images. Thirty-nine per cent of the psychiatrists and 11% of the neurologists only read the report but not the MRI images. Interestingly, no neurosurgeons reported reading only the report, but 7% declared they only read the MRI images. None of the other spe- cialties, neurology or psychiatry, reported analysing only the images. Mean confidence in brain MRI images and protocols on a 100-point visual bar (Figure 2) was for psychiatrists 81.2 (SE 2.6) and for neurosurgeons 74.0 (SE 4.5). Neurologists were the least confident on brain MRI images and protocols (mean rating of 69.9, SE 4.5). Considering confidence in the quality of examination reports (Figure 2), mean confidence on a 100-point visual bar was 76.7 (SE 3.3) for psychiatrists, 70.7 (SE 5.6) for neurosurgeons and 60.5 (SE 5.7) for neurologists. There were differences between specialties (F(2, 60) = 3.382, p < .05). Tukey post-hoc showed differences between neurologists and psychiatrists (p < .05). Discussion We found conceptual and practical differences regarding the use, perception and expectations of brain MRI among neurologists, neurosurgeons and psychiatrists. Neurosurgeons’ 265 Brain Magnetic Resonance Imaging: Perception and Expectations of Neurologists, Neurosurgeons and Psychiatrists most common reason to ask for a brain MRI was to confirm a diagnostic hypothesis, and their main expectation of the examination result was to confirm their initial hypothesis. This was probably because surgeons usually investigate potentially resectable lesions and very often before asking for a brain MRI, particularly in a tertiary hospital, they already know what type of pathology to expect, either because the patient has another image examination and/or was referred by another medical specialty. In addition, due to the interventionist nature of neurosurgery, none of the neurosurgeons only read the report; they need to plan biopsies and surgeries and hence to visualise the pathology and brain anatomy. Neurosurgeons revealed likewise a fairly high confidence in the quality of the images and report, which may be interpreted as how relevant they both are to the neurosurgical diagnosis and intervention. Moreover, because neurosurgical pathology is usually readily visible in MRI scans, there is less space for lack of confidence in images, and error or subjectivity by the neuroradiologist. Since visualising brain images seems to be fundamental for the neurosurgeons’ work, it is important that neuroradiologists assist them in correctly reading the examinations. Structural MRI is more complex than computed tomography, and this may not be readily apparent for the untrained reader. Psychiatrists also relied on brain MRI scans to confirm their diagnostic hypothesis, but by excluding another diagnosis, most likely due to the “non-organic” nature of the pathology of this specialty. Consequently, out of the three specialties, psychiatry showed the most confidence in the MR images, protocols and reports. One third of the psychiatrists only read the MRI report, hinting either little interest in the images because their pathology is “invisible” or their limited familiarity with this type of examination, at odds with the low number of examinations requested by this speciality. Neuroradiologists should care that psychiatrists are able to read not only the reports but also the images, important for a critical opinion on the quality of the examinations. New challenges await psychiatrists in the near future, with the foreseen growing use of functional magnetic resonance imaging (fMRI), other unconventional MRI techniques and advanced image post-processing in the diagnosis of neurodegenerative and psychiatric diseases 10,11. Neurologists showed the least confidence among the three specialties in MRI images and 266 Paulo Branco reports. This can be explained again by the nature of the pathology investigated by neurologists. Neurological diseases and their differential diagnosis are both numerous and include many rare diseases. Furthermore, neurological examination frequently narrows down the diagnostic list and, most of the time, permits the localisation of the pathology in the central nervous system. On the other hand, diagnosis in neurology can be achieved with other types of examinations, not just brain MRI. In fact, only some of these pathologies are visible in brain scans, while many others give only subtle or subjective imaging signs of their existence, making the work of the neuroradiologist more difficult and less objective. Likewise, clinicians sometimes do not perceive how important detailed clinical information is for the neuroradiologist, both in prescribing the adequate MRI sequences and in evaluating imaging signs. Understanding the particularities of the different medical practices involving brain imaging may help neuroradiologists improve the quality of their service. Since the examination report is the most important means of communication with other specialties, it should reflect these particularities. Besides the normal effort placed on the clarity and the quality of its content, neuroradiologists should not forget to address the clinician’s question in the report, most importantly, as seen before, in the case of the more elusive neurological diseases. The report should also include a reflection on differential diagnosis and the neuroradiologists’ uncertainties (when they exist), instead of just affirming or excluding pathology. This could help neurologists increase their confidence in the quality of the neuroradiology work. Moreover, other communication strategies should be implemented/reinforced, especially in a time of crisis in Europe when the financial aspects of radiology are of most concern. Formal and informal discussion of cases and multidisciplinary meetings and research are of the utmost importance for good communication with clinicians and for better patient care. Additionally, neuroradiology subspecialisation may also be desirable in large hospital centres, for example in fields like paediatric neuroradiology, dementia, movement disorders, and epilepsy, among others. Good communication between clinicians and radiologists has been shown to have a beneficial diagnostic and therapeutic impact 5. Some of the great challenges of medical imaging today are to measure accurately the quality of www.centauro.it The Neuroradiology Journal 27: 261-267, 2014 - doi: 10.15274/NRJ-2014-10051 examinations and the performance of neuroradiologists 12, and to combat examination overutilization 13. The limitations of our study concern the type of health institutions investigated and the restricted geographical area. It would be important to replicate this study in other centres and countries in order to validate and generalise our results. Conclusions This work sheds light on the motivations of various specialties in requesting brain MRI examinations. We showed that the most common reasons, perception and expectations seem to depend on the particular specialty requesting the scan. Our results illustrate the relevance of understanding these factors to improve imaging services and communication among the different specialties dedicated to the study of the central nervous system. Acknowledgments We thank the following for their collaboration: Professor Rui Vaz and the Department of Neurosurgery, Professor Carolina Garrett and the Department of Neurology, and Dr Roma Torres and the Department of Psychiatry of Centro Hospitalar São João, and Dr António Jorge and the Department of Neurology, Dr Marques Baptista and the Department of Neurosurgery, and Dr Jorge Bouça and the Department of Psychiatry of Centro Hospitalar de Vila Nova de Gaia/Espinho, Portugal. References 1 Kettenbach J, Wong T, Kacher D, et al. Computerbased imaging and interventional MRI: applications for neurosurgery. Comput Med Imaging Graph. 1999; 23 (5): 245-258. doi: 10.1016/S0895-6111(99)00022-1. 2 McCarron MO, Sands C, McCarron P. Quality assurance of neuroradiology in a District General Hospital. Q J Med. 2006; 99 (3): 171-175. doi: 10.1093/qjmed/hcl012. 3 Linden DE, Fallgatter AJ. Neuroimaging in psychiatry: from bench to bedside. Front Hum Neurosci. 2009; 3: 49. doi: 10.3389/neuro.09.049.2009. 4 Wallis A, McCoubrie P. The radiology report – Are we getting the message across? Clin Radiol. 2011; 66 (11): 1015-1022. doi: 10.1016/j.crad.2011.05.013. 5 Dalla Palma L, Stacul F, Meduri S, et al. Relationships between radiologists and clinicians: results from three surveys. Clin Radiol. 2000; 55 (8): 602-605. doi: 10.1053/crad.2000.0495. 6 Grieve FM, Plumb AA, Khan SH. Radiology reporting: a general practitioner’s perspective. Br J Radiol. 2010; 83 (985): 17-22. doi: 10.1259/bjr/16360063. 7 Clinger JN, Hunter BT, Hillman JB. Radiology reporting: attitudes of referring physicians. Radiology. 1988; 169 (3): 825-826. 8 Centro Hospitalar São João. Instituição. 2013. Available at: http://www.chsj.pt/PageGen.aspx?WMCM_PaginaId=27542. Accessed 2 October 2013. 9 Centro Hospitalar de Vila Nova de Gaia/Espinho. Caracterização. 2013. Available at: http://www.chvng. pt/assets/html/chvnge_caracterizacao.html. Accessed 2 October 2013. 10 Malhi GS, Lagopoulos J. Making sense of neuroimaging in psychiatry. Acta Psychiatr Scand. 2008; 117 (2): 100-117. 11 Bandettini PA. Twenty years of functional MRI: the science and the stories. Neuroimage. 2012; 62 (2): 575588. doi: 10.1016/j.neuroimage.2012.04.026. 12 Anumula N, Sanelli PC. National initiatives for measuring quality performance for the practicing neuroradiologist. Neuroimaging Clin N Am. 2012; 22 (3): 457466. doi: 10.1016/j.nic.2012.04.010. 13 Yousem DM. Combating overutilization: radiology benefits managers versus order entry decision support. Neuroimaging Clin N Am. 2012; 22 (3): 497-509. doi: 10.1016/j.nic.2012.05.013. Daniela Seixas, MD, PhD Dept. of Experimental Biology Faculty of Medicine University of Porto Alameda Professor Hernâni Monteiro 4200-319 Porto Portugal Tel.: +351 225513654 E-mail: dseixas@med.up.pt 267 The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 www.centauro.it Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early Clinical Experiences KAYOKO ABE1, YUKO ONO1, HIROKO YONEYAMA1, YU NISHINA1, YASUO AIHARA2, YOSHIKAZU OKADA2, SHUJI SAKAI1 1 Department of Diagnostic Imaging and Nuclear Medicine, 2 Department of Neurosurgery, Tokyo Women’s Medical University; Shinjyuku-ku, Tokyo, Japan Key words: cerebrospinal fluid, hydrocephalus, MRI SUMMARY – CSF imaging using the time-spatial labeling inversion pulse (time-SLIP) technique at 3T magnetic resonance imaging (MRI) was performed to assess cerebrospinal fluid (CSF) dynamics. The study population comprised 15 healthy volunteers and five patients with MR findings showing expansive dilation of the third and lateral ventricles suggesting aqueductal stenosis (AS). Signal intensity changes were evaluated in the tag-labeled CSF, untagged brain parenchyma, and untagged CSF of healthy volunteers by changing of black-blood time-inversion pulse (BBTI). CSF flow from the aqueduct to the third ventricle, the aqueduct to the fourth ventricle, and the foramen of Monro to the lateral ventricle was clearly rendered in all healthy volunteers with suitable BBTI. The travel distance of CSF flow as demonstrated by the time-SLIP technique was compared with the distance between the aqueduct and the fourth ventricle. The distance between the foramen of Monro and the lateral ventricle was used to calculate the CSF flow/distance ratio (CD ratio). The CD ratio at each level was significantly reduced in patients suspected to have AS compared to healthy volunteers. CSF flow was not identified at the aqueductal level in most of the patients. Two patients underwent time-SLIP assessments before and after endoscopic third ventriculostomies (ETVs). CSF flow at the ETV site was confirmed in each patient. With the time-SLIP technique, CSF imaging is sensitive enough to detect kinetic changes in CSF flow due to AS and ETV. Introduction Aqueductal stenosis (AS) is one of the most common causes of non-communicating hydrocephalus and can be divided into congenital and acquired types based on aspects of the clinical presentation, such as brain tumor pressure or cicatricial changes due to inflammation or trauma 1. The onset of symptoms due to aqueductal stenosis varies depending on severity. The usual symptoms and signs of increased intracranial pressure may also be present, including headache, vomiting, and eventually deterioration of the level of consciousness leading to coma and death 2. Symptomatic patients with AS require surgical treatment, such as 268 shunt placement, endoscopic third ventriculostomy (ETV), endoscopic aqueductoplasty or stent placement 3-5. The diagnosis of AS is best made by magnetic resonance imaging (MRI), which can be used to characterize tissue morphology in a non-invasive fashion 6. However, it may be difficult to diagnose AS in some cases, because the aqueduct is a tiny structure, and the condition’s cause varies. Therefore, it is also important to assess cerebrospinal fluid (CSF) dynamics. MRI can be used to assess CSF flow. Evidence of a flow-related signal void and phase-contrast images demonstrating the absence of CSF flow at the level of the aqueduct support a diagnosis of AS 7-9. Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... Kayoko Abe CSF imaging with the time-SLIP technique is based on the arterial spin-labeling technique 10-12 . After inverting the overall signals with a nonselective inversion-recovery (IR) pulse, a selective IR pulse (tag) is set for CSF in arbitrary cross-sections and directions for observation and then movement of CSF are captured at each black-blood time-inversion pulse (BBTI). Therefore, without using a contrast medium, CSF within this range is labeled as an endog- enous tracer. Although background signals can be recovered by altering BBTI, and the observation period is limited to five to eight seconds, it is possible to render the gradual changes in time-axis-labeled CSF data by altering the BBTI. This method is non-invasive, repeatable, and unaffected by heart rate. Because the time-SLIP technique can easily be used to image arbitrary cross-sections, CSF dynamics can be observed not only in the ventricle but Table 1 Clinical information and MR findings by time-SLIP technique in the control group. No Age Gender CSF flow from Aq. to 3rd. Detection (over MI) CSF flow from Aq to 4th. CSF flow from Monro to lat. Travel distance (mm) CD ratio (%) Travel distance (mm) CD ratio (%) 1 8 F Positive 23.6 62.4 11.1 78.8 2 10 M Positive 21.8 62.7 12.8 86.6 3 12 F Positive 23.7 58.9 13.5 83.1 4 19 M Positive 26.1 58.7 16.6 74.4 5 19 M Positive 27.2 58.0 14.3 77.8 6 23 F Positive 22.9 50.2 9.3 60.2 7 23 F Positive 22.8 52.6 14.4 68.8 8 24 F Positive 24.7 55.5 17.1 90.7 9 25 M Positive 20.8 46.0 15.1 89.2 10 27 M Positive 23.2 54.8 13 57.2 11 30 M Positive 25.3 64.6 13.9 89.7 12 33 M Positive 24.3 55.3 12.0 63.7 13 36 M Positive 22.2 59.8 10.3 57.5 14 38 M Positive 16.6 42.1 11.2 56.6 15 39 F Positive 17.2 38.4 9.7 61.6 range 8-39 16.6-26.1 38.2-64.6 9.3-17.1 57.5-90.7 average±SD 24.4±9.8 22.8±2.9 54.7±7.4 13.0±2.4 73.1±13.0 Aq: aqueduct, 3rd: third ventricle, 4th: fourth ventricle, Monro: foramen of Monro, lat: lateral ventricles, MI: massa intermedia, CD ratio: CSF flow/distance ratio. 269 Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... A B C D Kayoko Abe E Figure 1 Selective inversion-recovery (IR) pulse (tag) set to observe cerebrospinal fluid (CSF) flow using the time-spatial labeling inversion pulse (time-SLIP) technique. (A) To observe CSF flow from the aqueduct to the third ventricle, the superior margin of the tag is set at the border zone between the aqueduct and the third ventricle (→). (B) To observe CSF flow from the aqueduct to the fourth ventricle, the inferior margin of the tag is set at the border zone between the aqueduct and the fourth ventricle (→). (C) To observe CSF flow from the third ventricle to the lateral ventricle, the superior margin of the tag is set at the level of the foramen of Monro (→). (D) To observe CSF flow from the third ventricle to the prepontine cistern through the ventriculostomy site after endoscopic third ventriculostomy (ETV), the inferior margin of the tag is set at the ventriculostomy site (→). (E) To observe CSF flow from the prepontine cistern to the third ventricle through the ventriculostomy site after ETV, the superior margin of the tag is aligned with the ventriculostomy site (→). also at the surgical site. For these reasons, CSF imaging using the time-SLIP technique is useful for capturing changes in CSF dynamics. The purpose of this study is to evaluate CSF flow by CSF imaging using the time-SLIP technique at the level of the aqueduct, the foramen of Monro, and the ETV site in healthy volunteers and patients with hydrocephalic dilation of the third and lateral ventricles, suggesting AS. Materials and Methods Subjects Institutional review board approval and informed consent were obtained for the study. The control group consisted of 15 healthy vol270 unteers with no history of brain disease, no clinical symptoms and no morphological abnormalities detected by MRI (mean age, 24.4 years [range, 8-39 years]; males, nine; females, six) (Table 1). The hydrocephalic group consisted of five patients suspected to have AS based on MR findings (mean age, 36.8 years [range, 7-68 years]; males, three; females, two) (Table 2). Four patients presented with headaches, and the remaining patient presented with gait disturbance. MRI with 3D-T2WI showed marked dilation of the lateral and third ventricles and the foramen of Monro, no dilation of the fourth ventricle, absence of the aqueductal flow void, and a membranous septum or tapering in the aqueduct in all patients. ETV was performed in two patients whose symptoms were progres- Case Gender 1 M Age 24 History np Clinical Symptoms Severe headache Treatment ETV Dizziness, gait disturbance 2 F 7 np Recent headache after head injury 39 np Mild headache Evans ETV Membranous septum Membranous septum Third Lateral Index (%) Normal Large Marked expasile dilation 45.8 None Narrow M 46 np Mild headache No operation Membranous septum 36.7 Slightly Narrow 33,1 None Narrow 38,1 None Narrow 40,8 None Narrow Normal Large M 68 Tuberculous meningitis Progressive gait disturbance In 3year-old 271 np: not particular, PVH: periventricular hyperintensity. No operation Membranous septum Prestenotic dilatation Marked expasile dilation Rt > lt, tumor in rt. trigone Normal Large marked expasile dilation no laterality Normal Large No prestenotic dilatation 5 Space No laterality Prestenotic dilatation 4 Subarachnoid Fourth Prestenotic dilatation No operation Pvh marked expasile dilation no laterality Normal Large marked expasile dilation no laterality The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 F Membranous septum Ventricles Prestenotic dilatation Tumor in right lateral ventricle 3 Aqueduct www.centauro.it Table 2 Summary of clinical information and MR findings in five patients. Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... Kayoko Abe ł Figure 2 The ROIs are set to detect changes in signal intensity according to the black-blood time-inversion pulse (BBTI). CSF labeled with a selective IR pulse (A), untagged CSF (B), and untagged brain parenchyma (C). 1 2 Ń Figure 3 CSF flow as imaged using the time-SLIP technique is evaluated by calculating the following: 1) the CSF flow/distance (CD) ratio of the distance between the lower portion of the aqueduct and the obex of the fourth ventricle versus the maximum travel distance of the rendered CSF flow. 2) The CD ratio of the height from the foramen of Monro to the superior wall of the lateral ventricle versus the maximum travel distance of the rendered CSF flow. ł Figure 4 In the control group, ROIs are set in the labeled CSF (A), untagged CSF (B), and untagged brain parenchyma (C). Signal changes in each ROI are graphed according to the BBTI. 272 www.centauro.it The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 sive. MRI was performed using the time-SLIP technique before and after the ETV in both cases. MR Imaging All MR imaging examinations were performed using a 3T MR scanner (Titan; Toshiba, Tokyo, Japan), with an Atlas SPEEDER head coil. Morphological MR data were acquired by axial T1-weighted imaging (IR technique, TR: 2650 ms, TE: 40 ms, TI: 1100 ms, field of view [FOV]: 22 × 22 cm, matrix: 224 × 272, slice thickness: 5 mm, slice gap: 1 mm, flip angle: 90°, axial section) and 3D-T2-weighted imaging (T2WI) (fast advanced spin-echo [FASE] 3D_multi planar voxel, TR: 7121 ms, TE: 352 ms, FOV: 24 × 24 cm, slice thickness: 1.6 mm, matrix: 192 × 192, flip angle: 90°, reconstructed axial, coronal, and sagittal sections). Time-SLIP technique The time-SLIP parameters for CSF imaging were as follows. FASE, pulse-wave-gated preparation scan: TR: 12000 ms, TE: 80 ms, echo train spacing: 5 ms, FOV (sagittal section/coronal section): 23 × 26/26.3 × 30.7 cm, matrix (sagittal section/coronal section): 224 × 256/192 × 224, BBTI (control group/hydrocephalic group): 1000-5800/2000-4800 ms, BBTI increases at intervals of 200 ms, labeled pulse width (control group/hydrocephalic group): 40/10-30 mm, acquisition time: about 4 min]. To observe CSF dynamics from the aqueduct to the third ventricle, from the aqueduct to the fourth ventricle and from the foramen of Monro to the lateral ventricles, T2WIs were used as registration images, and a selective IR pulse (tag) was set in the border zones from the sagittal section in the former two regions and from the coronal section in the latter region. For the two patients who underwent ETV, T2WI sagittal sections were tagged as registration images to observe CSF flow at the site of the ventriculostomy (Figure 1). The visualization of CSF flow was assessed at each site in the control and hydrocephalic groups. CSF flow from the aqueduct to the fourth ventricle was evaluated using the CD ratio, which was the distance between the lower portion of the aqueduct and the obex of the fourth ventricle versus the maximum travel distance of the rendered CSF flow. CSF flow from the foramen of Monro to the lateral ventricle was evaluated using the CD ratio, which was the height from the foramen of Monro to the superior wall of the lateral ventricle versus the maximum travel distance of the rendered CSF flow (Figure 3). The CD ratio from the aqueduct to the third ventricle was not calculated because of markedly turbulent CSF flow in the third ventricle. A one-way analysis of variance was used to evaluate these ratios as calculated among radiologists, and Student’s t-test was used to evaluate the differences in these CD ratios between the control and hydrocephalic groups. Statistical analysis was performed using Microsoft Excel 2010, and statistical significance was defined as p < 0.05. All images were evaluated independently by three radiologists. Results Signal Intensity Signal intensity by BBTI changed according to a fixed pattern in the tag-labeled CSF, the untagged cerebral parenchyma, and the untagged CSF in the control group. Signal intensity in the tag-labeled CSF remained almost fixed despite changes in BBTI. The signal intensity of untagged CSF decreased until a null point (BBTI of 2600-2800 ms) and then recovered. Signal intensity of the untagged cerebral parenchyma increased gradually to BBTI of <2400 ms and achieved fixed intensity at BBTI of >2400 ms. The signal intensity of the untagged CSF and cerebral parenchyma were equal when BBTI ranged from 1800–2000 ms and from 4200-4400 ms (Figure 4). Evaluation Methods CSF Flow Rendering and CD ratio To calculate optimal BBTI, ROIs were set in tag-labeled CSF, untagged cerebral parenchyma, and untagged CSF on the CSF images from the aqueduct to the fourth ventricle in the control group. Signal intensity within the ROIs was measured and graphed at each BBTI (Figure 2). The CD ratio was calculated using the distance between the lower portion of the aqueduct and the obex of the fourth ventricle versus the maximum travel distance of the rendered CSF flow, and the height from the foramen of Monro to the superior wall of the lateral ventricle versus the maximum travel distance of the 273 Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... A B Kayoko Abe C Figure 5 Images obtained for a 33-year-old man in the control group. CSF flow is rendered clearly using the time-SLIP technique from the aqueduct to the third ventricle (→) (A), from the aqueduct to the fourth ventricle (B), and from the third ventricle to the lateral ventricle (→) (C). A B C D E F Figure 6 Images for a 24-year-old man in the hydrocephalic group. A) A membranous septum and prestenotic dilation in the aqueduct is observed on T2WI of sagittal sections (→). Images obtained using the time-SLIP technique show the following findings. Before ETV, no CSF flow from the aqueduct to the third ventricle is observed (→) (B) and CSF flow is clearly observed from the aqueduct to the fourth ventricle (→) (C). D) After ETV, CSF flow from the aqueduct to the fourth ventricle has disappeared (→). Turbulent and pulsatile CSF flow is observed from the prepontine cistern to the third ventricle (→) (E) and from the third ventricle to the prepontine cistern through the ventriculostomy site (→) (F). 274 www.centauro.it The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 Table 3 Results of CSF flow by the time-SLIP technique in two patients who were confirmed ETV. CSF flow from Aq to 4th. CSF flow from Aq. to 3rd Case Travel distance (mm) Detection Pre-ETV Post-ETV PreETV PostETV CSF flow from Monro to lat. Travel distance (mm) CD ratio (%) PreETV PostETV PreETV CD ratio (%) PostETV PreETV PostETV CSF flow at ETV site Detection 1 Negative Negative 15.7 0 45.3 0 6.3 26.0 13.2 57.9 Positive 2 Negative Negative 0 0 0 0 0 15.5 0 43.4 Positive Table 4 Results of CSF flow by the time-SLIP technique in three patients who were not comfirmed ETV. CSF flow from Aq. to 3rd Case CSF flow from Aq to 4th CSF flow from Monro to lat. Detection Travel distance (mm) CD ratio (%) Travel distance (mm) CD ratio (%) 3 Negative 0 0 19.6 59.5 4 Positive (not reach MI) 13.1 33.6 0 0 5 Negative 0 0 0 0 rendered CSF flow. The CD ratio of CSF flow from the aqueduct to the third ventricle was not calculated because CSF flow was turbulent and not linear within the third ventricle. These figures were then assessed by three radiologists who had interpreted the images. The CD ratios were analyzed statistically using analysis-of-variance techniques, and the criterion for statistical significance was set at the level of 0.05. Because no significant differences in interpretation were found, average values were used to indicate the CD ratios. tio at this level ranged from 38.2 to 64.6% with an average of 54.7%. Reflux and pulsatile CSF flow that formed a coil at the tip was observed from the third ventricle to the lateral ventricles through the foramen of Monro. The signal void of pulsatile CSF flow was also observed in the third ventricle. The CD ratio was calculated as the height from the foramen of Monro to the superior wall of the lateral ventricle versus the maximum travel distance of the rendered CSF flow. The CD ratio at this level ranged from 57.5% to 90.7% with an average of 73.1%. The control group CSF flow was clearly identified at each level in all healthy volunteers (Table l, Figure 5). The upward pulsatile CSF flow of turbulent eddies was observed from the aqueduct to the third ventricle and over the posterior margin of the massa intermedia in the third ventricle. Downward pulsatile CSF flow that formed a coil at the tip was observed from the aqueduct to the fourth ventricle. The CD ratio was calculated based on the distance between the lower portion of the aqueduct and the obex of the fourth ventricle versus the maximum travel distance of the rendered CSF flow. The CD ra- The hydrocephalic group ETV was performed in two patients, No. 1 and No. 2 (Table 3). In Case No. 1, AS was confirmed during ETV; downward pulsatile CSF flow from the aqueduct to the fourth ventricle and reflux pulsatile CSF flow from the foramen of Monro to the lateral ventricle was observed. However, upward pulsatile CSF flow from the aqueduct to the third ventricle was not observed before ETV. After ETV, the downward pulsatile CSF flow from the aqueduct to the fourth ventricle had disappeared. Increased reflux pulsatile CSF flow from the 275 Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... A Kayoko Abe B Figure 7 A 7-year-old girl suspected of AS and known to have a tumor in the trigone of the right lateral ventricle. Although CSF imaging by the time-SLIP technique shows no CSF flow from the third ventricle to the lateral ventricle before ETV (→) (A), CSF flow is clearly observed after ETV (→) (B). A B Figure 8 The CD ratios of the rendered CSF flow are significantly higher in the control group than in the hydrocephalic group at the level of the fourth (A) and lateral ventricles (B) (*p < 0.001). foramen of Monro to the lateral ventricle was observed, and the CD ratio at the level of the lateral ventricle had increased to 59.5% from 19.6%. In Case No. 2, aqueductal obstruction was confirmed during ETV. No CSF flow was observed at any of the levels examined before ETV. However reflux pulsatile CSF flow from the foramen of Monro was observed, and the CD ratio increased to 43.4% after ETV. Toand-fro CSF flow was clearly observed between the prepontine cistern and the third ventricle 276 through the ventriculostomy site in both patients. In one of the other three patients (Case No. 3), no CSF flow was observed at the level of the aqueduct, but reflux pulsatile CSF flow from the foramen of Monro to the lateral ventricle was observed. The CD ratio at that level was 59.5%. In the next patient (Case No. 4), upward and downward pulsatile CSF flow were observed at the level of the aqueduct, and the CD ratio in the fourth ventricle was 33.6%, but no CSF flow from the foramen of Monro www.centauro.it The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 to the lateral ventricle was observed. In the last patient (Case No. 5), no CSF flow was observed at any of the levels examined (Table 4). The CD ratios of the rendered CSF flow at the levels of the fourth and lateral ventricles were significantly higher in the control group than in the hydrocephalic group (p < 0.001) (Figure 8). Discussion The time-SLIP technique for CSF imaging uses labeled CSF with a selective IR pulse (tag) as an endogenous tracer. It is thus possible to render CSF dynamic changes without contrast medium 10-12. Therefore this is a noninvasive and reproducible technique. When using the time-SLIP technique, the BBTI setting is essential to render a clear CSF flow over a suitable period. First, background signals are inhibited with non-selective IR pulses; labeled CSF is then set with a selective IR pulse (tag). The untagged CSF signals altered by the BBTI decrease gradually and then increase after reaching a null point. When the untagged CSF signals reach a null point, the labeled CSF and the untagged CSF exhibit marked contrast. In the present study, CSF flow from the aqueduct to the third ventricle, from the aqueduct to the fourth ventricle, and from the foramen of Monro to the lateral ventricle was rendered clearly for all healthy volunteers. Altering signal intensity by BBTI had similar effects in the tagged CSF, the untagged CSF, and the untagged brain parenchyma, respectively, of healthy volunteers. The untagged CSF signals reached a null point at 2600-2800 ms, and the contrast images of untagged brain parenchyma and CSF along with labeled CSF were acquired at approximately 1800-4400 ms. Setting BBTI within this range ensured the continuous visualization of CSF dynamics. The observation period, limited to five to eight seconds, is another advantage of this technique. Other types of CSF imaging do not allow for the visualization of CSF dynamics with the same time resolution. The pathway leading from CSF production to absorption is widely known as the bulk-flow theory 13-15. In recent years, the pulsatile flow theory has been reported 12,16,17. In the bulkflow theory, CSF, which is produced by the choroid plexus in the ventricles, flows unidirectionally from the lateral ventricles through the foramen of Monro, then through the third ventricle and the aqueduct into the fourth ventricle and through the foramen of Luschka and Magendie into the subarachnoid space, to then be absorbed by arachnoid granulations. The movement of CSF flow depends on the relationship between active production and passive absorption. According to the pulsatile flow theory, the reported movement of CSF depends on the expansion and contraction of cerebral arteries during the cardiac cycle as well as changes in venous pressure throughout the respiratory cycle. In this study, the rendered CSF flow at each level in the healthy volunteers was pulsatile with turbulent eddies, and traveled over the massa intermedia in the third ventricle. The CD ratio in the lateral ventricles was >50%. The rendered CSF flow was not slow and represented bidirectional movement between the ventricles. These findings suggest that CSF imaging with the time-SLIP technique can be easily used to visualize pulsatile CSF flow. The diagnosis of AS primarily involves morphological evaluation by MRI. The evaluation of CSF dynamics by MRI is also useful in the case of AS and an open ETV site 7,9,18. In one patient observed to have AS during ETV surgery, CSF flow from the aqueduct to the fourth ventricle was found, regardless of the lack of CSF flow from the aqueduct to the third ventricle. The CD ratio at the level of the fourth ventricle was similar in healthy volunteers and the patient. After ETV, CSF flow in the fourth ventricle disappeared, and no CSF flow was observed at the level of the aqueduct. In the other patients who underwent ETV, no CSF flow was observed at the level of the aqueduct, and aqueductal obstruction was confirmed during ETV. In both patients, turbulent and pulsatile CSF flow was observed to pass through the ETV site between the third ventricle and the prepontine cistern. These findings suggest that this technique is sufficiently sensitive to render CSF dynamics at the level of the aqueduct and ETV site. CSF flow from the foramen of Monro to the lateral ventricle was low or undetected before ETV, and then increased after ETV. Yamada et al. 12 performed CSF imaging using the time-SLIP technique and showed a decrease in CSF flow, which disappeared from the foramen of Monro to the lateral ventricle in a patient with hydrocephalus. In this study, the CD ratio for the distance from the foramen of Monro to the lateral ventricle was calculated as >50% in all the healthy volunteers. Changes in the rendered CSF flow and CD ratio at this level after ETV suggest that CSF imaging with 277 Assessment of Cerebrospinal Fluid Flow Patterns Using the Time-Spatial Labeling Inversion Pulse Technique with 3T MRI: Early... the time-SLIP technique could be used to capture the return of CSF dynamics to normal after ETV, which could help in predicting patient prognosis. In one of the three patients who did not undergo ETV (Case No. 4), CSF flow was observed at the aqueductal level, but was not observed in the lateral ventricle. The CD ratio for the area from the aqueduct to the fourth ventricle was similar to that observed among healthy volunteers, though the rendered CSF flow was weak and did not reach the posterior rim of the massa intermedia. In this case, T2WI did not show any evidence of deformity or prestenotic dilation of the aqueduct. These findings suggest that the degree of AS was relatively minor but nonetheless altered CSF dynamics in cases of hydrocephalus. CSF flow at the aqueduct was not observed in two patients (Case No. 3 and 5). CSF flow from the foramen of Monro to the lateral ventricles was visualized in one patient, who had a CD ratio >50% (Case No. 3). In the other patient (Case No. 5), no CSF flow from the foramen of Monro to the lateral ventricles was detected. The lack of CSF flow at the level of the aqueduct, and clinical symptoms by increased intracranial pressure support the existence of additional flow paths 19,20. The contrasting characteristics of CSF flow in the lateral ventricles of these patients suggest that time-SLIP CSF imaging can be used to characterize the pathophysiology of CSF dynamics. The time-SLIP technique is expected to visualize these flow paths by allowing researchers to label CSF at arbitrary cross-sections, angles, and ranges and to 278 Kayoko Abe predict the state of hydrocephalus. One limitation of CSF imaging with use of the timeSLIP technique is that an evaluation method has not yet been established. However, as determined in the present investigation, imaging the form and movement of turbulent CSF flow may allow for the calculation of CSF flow speed in limited spaces, such as the aqueduct, but measurement in larger areas remains difficult. Furthermore, even though the time-SLIP technique allows for a longer observation period than conventional imaging methods, the observation time is only a few seconds, and the patient must be placed in a supine position. These factors limit the analysis of CSF dynamics using the time-SLIP technique. Conclusions The time-SLIP technique for CSF imaging is a non-invasive, rapid and repeatable method with which to observe CSF dynamics. It may be helpful in predicting the severity of AS and the therapeutic effect of ETV and is also expected to aid in the treatment of other conditions affected by the CSF dynamics associated with hydrocephalus. Acknowledgment We are grateful to Seiko Shimizu, Maiko Shinohara and Takao Yamamoto for technical assistance. We thank the technicians in the MR unit for their help with image acquisition. www.centauro.it The Neuroradiology Journal 27: 268-279, 2014 - doi: 10.15274/NRJ-2014-10045 References 1 Little JR, Houser OW, MacCarty CS. Clinical manifestations of aqueductal stenosis in adults. J Neurosurg. 1975; 43 (5): 546-552. doi: 10.3171/jns.1975.43.5.0546. 2 Villani R, Tomei G, Gaini SM, et al., Long-term outcome in aqueductal stenosis. Childs Nerv Syst. 1995; 11 (3): 180-185. doi: 10.1007/BF00570262. 3 da Silva LR, Cavalheiro S, Zymberg ST. Endoscopic aqueductoplasty in the treatment of aqueductal stenosis. Childs Nerv Syst. 2007; 23 (11): 1263-1268. doi: 10.1007/s00381-007-0393-7. 4 Kelly PJ. Stereotactic third ventriculostomy in patients with nontumoral adolescent/adult onset aqueductal stenosis and symptomatic hydrocephalus. J Neurosurg. 1991; 75 (6): 865-873. doi: 10.3171/ jns.1991.75.6.0865. 5 Mori H, Koike T, Fujimoto T, et al. Endoscopic stent placement for treatment of secondary bilateral occlusion of the Monro foramina following endoscopic third ventriculostomy in a patient with aqueductal stenosis. Case report. J Neurosurg. 2007; 107 (2): 416-420. doi: 10.3171/JNS-07/08/0416. 6 Novetsky GJ, Berlin L. Aqueductal stenosis: demonstration by MR imaging. J Comput Assist Tomogr. 1984; 8 (6): 1170-1171. doi:10.1097/00004728-198412000-00026. 7 Laitt RD, Mallucci CL, Jaspan T, et al. Constructive interference in steady-state 3D Fourier-transform MRI in the management of hydrocephalus and third ventriculostomy. Neuroradiology. 1999; 41 (2): 117123. doi: 10.1007/s002340050715. 8 Wu HM, Yousem DM, Chung HW, et al. Influence of imaging parameters on high-intensity cerebrospinal fluid artifacts in fast-FLAIR MR imaging. Am J Neuroradiol. 2002; 23 (3): 393-399. 9 Stoquart-El Sankari S, Lehmann P, Gondry-Jouet C, et al., Phase-contrast MR imaging support for the diagnosis of aqueductal stenosis. Am J Neuroradiol. 2009; 30 (1): 209-214. doi: 10.3174/ajnr.A1308. 10 Miyazaki M, Sugiura S, Tateishi F, et al., Noncontrast-enhanced MR angiography using 3D ECG-synchronized half-Fourier fast spin echo. J Magn Reson Imaging. 2000; 12 (5): 776-783. doi: 10.1002/1522-2586(200011)12:5<776::AID-JMRI17 >3.0.CO;2-X. 11 Nishimura DG, Macovski A, Pauly JM. Considerations of magnetic resonance angiography by selective inversion recovery. Magn Reson Med. 1988; 7 (4): 472-484. doi: 10.1002/mrm.1910070410. 12 Yamada, S., et al., Visualization of cerebrospinal fluid movement with spin labeling at MR imaging: preliminary results in normal and pathophysiologic conditions. Radiology, 2008; 249: 644-652. doi: 10.1148/radiol.2492071985. 13 Di Chiro G. Movement of the cerebrospinal fluid in human beings. Nature. 1964; 204: 290-291. doi: 10.1038/204290a0. 14 Di Chiro G. Observations on the circulation of the cerebrospinal fluid. Acta Radiol Diagn (Stockh). 1966; 5: 988-1002. 15 Oreskoviþ D, Klarica M. The formation of cerebrospinal fluid: nearly a hundred years of interpretations and misinterpretations. Brain Res Rev. 2010; 64 (2): 241-262. doi: 10.1016/j.brainresrev.2010.04.006. 16 Njemanze PC, Beck OJ. MR-gated intracranial CSF dynamics: evaluation of CSF pulsatile flow. Am J Neuroradiol. 1989; 10 (1): 77-80. 17 Linninger AA, Tsakiris C, Zhu DC, et al. Pulsatile cerebrospinal fluid dynamics in the human brain. IEEE Trans Biomed Eng. 2005; 52 (4): 557-565. doi: 10.1109/ TBME.2005.844021. 18 Fukuhara T, Vorster SJ, Ruggieri P, et al., Third ventriculostomy patency: comparison of findings at cine phase-contrast MR imaging and at direct exploration. Am J Neuroradiol. 1999; 20 (8): 1560-1566. 19 Killer HE, Jaggi GP, Flammer J, et al. Cerebrospinal fluid dynamics between the intracranial and the subarachnoid space of the optic nerve. Is it always bidirectional? Brain. 2007; 130 (Pt 2): 514-520. doi: 10.1093/brain/awl324. 20 Koh L, Zakharov A, Johnston M. Integration of the subarachnoid space and lymphatics: is it time to embrace a new concept of cerebrospinal fluid absorption? Cerebrospinal Fluid Res. 2005; 2: 6. doi: 10.1186/17438454-2-6. Kayoko Abe, MD Department of Diagnostic Imaging and Nuclear Medicine Tokyo Women’s Medical University 8-1 Kawada-cho Shinjyuku-ku, Tokyo, 162-8666 Japan Tel.: +81-3-3353-8111 Fax: +81-3-5269-7355 E-mail: abe.kayoko@twmu.ac.jp 279 The Neuroradiology Journal 27: 280-287, 2014 - doi: 10.15274/NRJ-2014-10036 www.centauro.it The Etiology of Ring Lesions on Diffusion-Weighted Imaging PASQUALE F. FINELLI1, ETHAN B. FOXMAN2 1 Department of Neurology, 2 Department of Radiology, Hartford Hospital and University of Connecticut School of Medicine; Hartford, CT, USA Key words: diffusion-weighted imaging, MRI, ring lesion, immunocompromised, demyelinating disease SUMMARY – This study describes a series of cases and reviews the literature on cases of ring lesion on diffusion-weighted imaging to better appreciate the spectrum of disease associated with this neuroimaging finding. We retrospectively reviewed the MR studies of 15 patients with ring pattern lesions on diffusionweighted imaging from an inpatient Neurology service of a tertiary care center seen over a ten-year period, and reviewed cases in the literature. Thirty-one cases, including 15 new patients, comprise the study group. Immunocompromised patients accounted for 38% of patients with ring lesions on diffusion-weighted imaging with cerebral aspergillosis in five patients, progressive multifocal leukoencephalopathy in three, primary CNS lymphoma in two, cerebral toxoplasmosis in one, and resolving cerebral hematoma in one. In the immunocompetent group demyelinating lesions including multiple sclerosis, acute disseminated encephalomyelitis, Balo’s concentric sclerosis and acute necrotizing encephalitis, were seen in 11 patients, vascular etiology in four and neoplastic in three patients, two primary and one metastatic and pyogenic brain abscess in one. Ring lesions on diffusion-weighted imaging are associated with a spectrum of disease not previously considered. Immunocompromised patients accounted for almost one-half while demyelinating conditions in the immunocompetent patients were most common overall. Introduction Methods Diffusion-weighted imaging (DWI) is routinely used in the diagnosis of acute cerebral infarction and pyogenic brain abscesses 1 and in these settings the DWI changes are nearly always uniform in appearance throughout the region of abnormality. By contrast, when DWI abnormalities are most pronounced along the lesion periphery and therefore present as ring lesions on DWI sequences, the underlying pathology may differ. We present a series of 15 cases and review the literature on DWI ring pattern lesions, demonstrating that this MR imaging feature is associated with a broad differential diagnosis, most commonly demyelinating disease. The case selection reflects one author’s (PFF) personal experience over a ten-year period at the same institution and includes 15 new and four previously reported patients 2-5. The literature review employed PubMed.gov computer search using “restricted-diffusion ring”, and “DWI ring”, eliminating references where the ring was due to enhancement. Additional citations were selected from references within the initial search results. Image acquisition. MR imaging was performed utilizing a 1.5 T system (Sigma HDx; GE Medical Systems, Milwaukee, WI, USA) equipped with echo-planar gradients. DWI sequences were acquired using a single-shot, echo-planar technique with sampling of the en- 280 Pasquale F. Finelli The Etiology of Ring Lesions on Diffusion-Weighted Imaging A B C D E F G H I J K L Figure 1 Case 1. A) Glioblastoma multiforme showing a ring pattern with increased signal on DWI. B) Corresponding decreased signal on ADC (arrowheads). Case 2. C) Intracerebral hemorrhage showing a ring pattern with increased signal on DWI. D) Increased signal on ADC (arrowheads). Case 3. E) Cerebral infarction showing a ring pattern with increased signal on DWI. F) Corresponding decreased signal on ADC (arrowheads). Case 4. G) Progressive multifocal leukoencephalopathy showing a ring pattern with increased signal on DWI. H) Corresponding decreased signal on ADC (arrowheads). Case 5. I) Cerebral embolism showing a ring pattern with increased signal on DWI. J) Corresponding decreased signal on ADC (arrowheads). Case 6. K) Glioblastoma multiforme showing a ring pattern with increased signal on DWI. L) Corresponding decreased signal on ADC (arrowheads). tire diffusion tensor. Diffusion measurements in three orthogonal directions with a b value of 1000 s/mm2 were obtained with TR/TE/NEX of 6000/98/2, a FOV of 24 × 24 cm and a matrix of 128 × 128 pixels, a section thickness of 5 mm with a 2 mm gap, and 22 axial sections. Nonisotropic DWIs and ADC maps were generated automatically by the scanner. Case Vignettes Case 1. A 36-year-old-man with mental retardation presented with new onset lethargy, left-sided weakness and right gaze preference. Neurologic examination was non-focal with mental status changes consistent with mental retardation. MR imaging was performed (Figure 1A,B). Brain biopsy was performed and showed glioblastoma multiforme grade IV. Case 2. A 46-year-old-man with a history of cirrhosis and HIV infection for 20 years treated with highly active antiretroviral therapy presented following three generalized seizures. Neurologic examination showed normal mental status with decreased fine motor movements on the left and bilateral asterixis. MR imaging (Figure 1C,D) was consistent with T2 shine through from prior lobar hemorrhage. Case 3. A 77-year-old-man was admitted after sudden onset of dizziness and a fall. Neurologic examination showed normal mental 281 The Etiology of Ring Lesions on Diffusion-Weighted Imaging Pasquale F. Finelli A B C D E F G H I J K L Figure 2 Case 7. A) Cerebral embolism showing a ring pattern with increased signal on DWI. B) Corresponding decreased signal on ADC (arrowheads). Case 8. C) Cerebral metastasis showing a ring pattern with increased signal on DWI. D) Corresponding decreased signal on ADC (arrowheads). Case 9. E) Multiple sclerosis showing a ring pattern with increased signal on DWI. F) Corresponding decreased signal on ADC (arrowheads). Case 10. G) Multiple sclerosis showing a ring pattern with increased signal on DWI. H) Corresponding decreased signal on ADC (arrowheads). Case 11. I) Tumefactive multiple sclerosis showing a ring pattern with increased signal on DWI. J) Corresponding decreased signal on ADC (arrowheads). Case 12. K) Acute disseminated encephalomyelitis showing a ring pattern with increased signal on DWI. L) Corresponding decreased signal on ADC (arrowheads). status and absence of light touch and pinprick sensation and mild weakness on the left. MR imaging (Figure 1E,F) was consistent with hemorrhagic infarction. Case 4. A 54-year-old-man with AIDS was found by the police confused and wandering the streets. Neurologic examination noted the patient to be alert and answering all questions with “yes”. MR imaging was performed (Figure 1G,H). Subsequent PCR for JC virus was positive in the CSF. Case 5. An 82 year-old-man with coronary artery disease presented with a two-week history of falling. MR imaging (Figure 1I,J) was consistent with cerebral infarction. A year later he 282 was readmitted with a new left middle cerebral artery territory infarction on MR. Case 6. An 81 year-old-man with a twomonth history of dragging his left leg was admitted following a fall. MR imaging was performed (Figure 1K,L). A repeat MR after six weeks showed an increase in lesion size and he expired two and a half months following diagnosis of a primary brain neoplasm. Case 7. A 72-year-old man developed sudden right-sided weakness and sensory loss and was treated with intravenous thrombolytic therapy. MR imaging was performed (Figure 2A,B). Recurrent cerebral infarction in multiple vascular territories suggested an embolic etiology. Case 8. A 57-year-old man presented with www.centauro.it The Neuroradiology Journal 27: 280-287, 2014 - doi: 10.15274/NRJ-2014-10036 A B C D E F Figure 3 Case 13. A) Pyogenic abscess showing a ring pattern with increased signal on DWI. B) Corresponding decreased signal on ADC (arrowheads). Case 14. C) Demyelinating lesion of tumefactive multiple sclerosis showing a ring pattern with increased signal on DWI. D) Corresponding decreased signal on ADC (arrowheads). Case 15. E) Toxoplasmosis showing a ring pattern with increased signal on DWI. F) Corresponding decreased signal on ADC (arrowheads). 283 The Etiology of Ring Lesions on Diffusion-Weighted Imaging headache, nausea and vomiting, double vision and generalized weakness. MR imaging was performed (Figure 2C,D). CT of chest, abdomen and pelvis was suggestive of metastatic disease and biopsy of a chest wall mass was diagnostic for renal cell carcinoma. Case 9. A 25-year-old woman with known multiple sclerosis presented with a one-month history of difficulty with vision and gait. MR imaging was performed (Figure 2E,F). Case 10. A 31-year-old woman awoke with slurred speech, left facial droop and left arm weakness. MR imaging showed changes consistent with multiple sclerosis (Figure 2G,H). Case 11. A 33-year-old man presented with left arm clumsiness and weakness. MR imaging was performed (Figure 2I,J). Symptoms and neuroimaging abnormalities resolved over the next year and he is followed with a diagnosis of tumefactive multiple sclerosis. Case 12. A 49-year-old man presented with headache, left-sided numbness and diplopia. MR showed a ring lesion in the pons (Figure 2K,L). After six months he returned to baseline with marked resolution of MR imaging changes consistent with acute disseminated encephalomyelitis. Case 13. A 35-year-old man with non-insulin dependent diabetes and obesity presented with episodes of shaking of the right lower extremity followed by dragging the right leg. MR was performed (Figure 3A,B). MR spectroscopy (MRS) showed elevated lipid and lactate peaks and MR perfusion showed no increased blood volume of the lesion wall. The patient was treated with a broad-spectrum antibiotic for four days. Repeat MR showed increasing lesion size and a stereotactic brain biopsy was performed. Seven milliliters of purulent material were aspirated and methenamine silver stain showed clusters of bacteria identified as grampositive cocci consistent with streptococci. Case 14. A 20 year-old women was found with right-sided weakness and difficulty speaking. Cerebrospinal fluid was normal except for myelin basic protein of 6.4 mcg/L (n=0-4.0) and the presence of five oligoclonal bands. MR imaging was performed (Figure 3C,D). A right frontal stereotactic brain biopsy demonstrated demyelination. The patient met the diagnostic criteria for multiple sclerosis. Case 15. A 31-year-old woman with AIDS presented with a one-week history of worsening headache followed by left hemiparesis on the day of admission. MR imaging was performed (Figure 3E,F). Treatment for toxoplas284 Pasquale F. Finelli mosis with pyrimethamine and sulfanpirazone resulted in clinical improvement and subsequent neuroimaging after three weeks showed marked lesion resolution. Discussion To date, the differential diagnosis of ring lesions on DWI has only been considered in the context of single case reports 2,4,6. We present a series of 15 cases (Table 1) and review the literature on this uncommon neuroimaging finding, and define a broader spectrum of disease that includes cerebral aspergillosis, Balo’s concentric sclerosis (BCS), acute necrotizing encephalitis (ANE), multiple sclerosis (MS), acute disseminated encephalomyelitis (ADEM), progressive multifocal leukoencephalopathy (PML), toxoplasmosis, primary and metastatic brain tumors, cerebral infarction and pyogenic brain abscess (Table 2). Immunocompromised patients accounted for some 39% of patients with ring lesions on DWI, so that knowledge of the patient’s immune status figures prominently in determining the etiology. In the immunocompromised group cerebral aspergillosis was most common and diagnosed in five cases where lesions were multiple and small. Additionally, there were three cases of PML where the ring lesions were single and tended to be large (> 5 cm) and not well-defined, and two cases of primary CNS lymphoma (PCNSL) where lesions were single and small (<1 cm) and welldefined. Lymphoma, although here reported in immunocompromised patients, may also occur in immunocompetent patients. In addition, a patient with HIV infection and thrombocytopenia presented following a seizure with a resolving intracerebral hemorrhage (case 2) that on DWI mimicked restricted diffusion due to T2 weighted shine through. A DWI ring pattern was noted with toxoplasmosis in one patient with AIDS (case 15). In immunocompetent patients demyelinating disease accounted for the majority of DWI ring lesions in our series (cases 9-12 and 14) and in the literature 4,7-11. ANE was reported in three cases in the literature, all with bilateral thalamic lesions while three cases of BCS involving the white matter were described. In the cases of BCS the lesions were multiple in one and single in two cases and enhancement was seen in all three cases albeit minimal in two 9,10. Our cases 9 and 10 fulfilled McDonald criteria for definite MS and manifested multiple non-enhancing lesions on www.centauro.it The Neuroradiology Journal 27: 280-287, 2014 - doi: 10.15274/NRJ-2014-10036 Table 1 Current series of ring lesions on DWI. Patient Disease Immune status / Risk Factors MR enhancement 1 Glioblastoma multiforme Immunocompetent (+) 2 Lobar hemorrhage Immunocompromised (HIV)/ thrombocytopenia (–) 3 Hemorrhagic infarction Immunocompetent/IvtPA (–) 4 PML Immunocompromised (AIDS) (–) 5 Cerebral infarction Immunocompetent/ coronary artery disease, hypertension, diabetes (–) 6 Glioblastoma multiforme Immunocompetent (+) 7 Cerebral infarction Immunocompetent/HTN, hyperlipidemia (NA) 8 Cerebral metastasis Immunocompetent/metastatic renal cell carcinoma (+) 9 MS Immunocompetent (–) 10 MS Immunocompetent (–) 11 Tumefactive MS Immunocompetent (+) 12 ADEM Immunocompetent (–) 13 Pyogenic abscess Immunocompetent/diabetes, hyperlipidemia 14 Tumefactive MS Immunocompetent (+) 15 Toxoplasmosis Immunocompromised (AIDS) (+) PML= progressive multifocal leukoencephalopathy; IVtPA= intravenous tissue plasminogen activator; AIDS= acquired immunodeficiency syndrome; HIV= human immunodeficiency virus; NA= not available; ADEM= acute disseminated encephalomyelitis; MS= multiple sclerosis. Table 2 Etiology of DWI ring lesion (16 cases from the literature and 15 new cases). Infection Cerebral aspergillosis [2,16,17] 5* PML 3* [17,18] , (case 4) Pyogenic abscess (case 13) Toxoplasmosis (case 15) 1 1* Demyelination ANE [7,8,10] 3 Balo’s concentric sclerosis [4,9,11] 3 MS (cases 9, 10, 11 and 14) 4 ADEM (case 12) 1 Neoplasm CNS lymphoma [3,5] 2* Glioblastoma multiforme (cases 1 and 6) 2 CNS metastasis (case 8) 1 Vascular Primary ICH (case 2) 1* Hemorrhagic infarction (case 3) 1 Embolic infarction (case 5) 1 Ischemic infarction 2 [14] , (case 7) *= immunocompromised; [ ]= citations from literature; ( )= present case series. 285 The Etiology of Ring Lesions on Diffusion-Weighted Imaging MR imaging variably involving the corpus callosum, cerebral hemisphere, brainstem and cervical spinal cord. Three patients in the demyelinating category (cases 11, 12 and 14) suffered monophasic events with case 11 manifesting an enhancing symptomatic right frontal lobe lesion with several additional smaller T2 weighted lesions with five oligoclonal bands in the CSF, with case 12 showing a single nonenhancing pontine lesion with a mild CSF pleocytosis and case 14 bifrontal and left hemispheric white matter lesions on MR and five oligoclonal bands in the CSF. Follow-up of these three cases has been for four years, six months and two months respectively with a resolution of neurologic deficits and no new symptoms in cases 11 and 12 and markedly improved MR on serial imaging. Case 14, the most recently seen patient, showed demyelination on brain biopsy and is currently in a rehabilitation facility. Of these three cases only case 14 met the diagnostic criteria for MS, while the other two (cases 11 and 12) are best classified as probable MS and ADEM respectively 12. The findings of case 12 accord with results of a recent study of 97 patients with rhombencephalitis where demyelinating disease was the largest group with a known etiology 13. Further, in the immunocompetent patients with vascular risk factors lesions were single and included one case of hemorrhagic infarction and two cases of embolic cerebral infarction (Cases 3, 5 and 7), and one from the literature with Susac disease 14. Additionally, in the immunocompetent group we describe a DWI ring lesion in three patients with brain neoplasm, two primary and one metastatic. The primary tumors were glioblastoma multiforme, one confirmed by biopsy (case 5) and the metastasis was renal cell carcinoma. 286 Pasquale F. Finelli Lastly, restricted diffusion is an important feature with pyogenic brain abscess, however the ring pattern seen in our case 13 has not to our knowledge been described previously. The etiology of ring lesions on DWI differs from lesions with homogeneous restricted diffusion and a differential diagnosis can be formulated considering the patient’s immune status, age, time of disease onset and clinical course, history of malignancy, presence of vascular risk factors and MR imaging characteristics. In addition to lesion size and number, a variant of the DWI ring pattern, namely a “C-shaped” lesion typical of tumefactive demyelination may also be seen 15. Although combined DWI and MRS are reported to be valuable in differentiating pyogenic abscess and cystic necrotic tumor, these imaging techniques have limitations considering the overlap of findings 20, and brain biopsy may be required to distinguish between these two disease entities (case 13). In summary, ring lesions on DWI occur disproportionately in immunocompromised patients, and then mainly in association with cerebral aspergillosis where the lesions are multiple and small, and with PML and CNS lymphoma where the lesions are single, but larger and ill-defined with PML. A case of toxoplasmosis also occurred in the immunocompromised group. In the immunocompetent group and overall, demyelinating conditions were most common occurring with almost twice the frequency of other etiologies. In the context of 31 cases, 19 of whom are from one institution [15 cases from present series and four from prior reports 3-6], we consider a broader etiologic spectrum of disease than previously associated with ring lesions on DWI. www.centauro.it The Neuroradiology Journal 27: 280-287, 2014 - doi: 10.15274/NRJ-2014-10036 Reference 1 Schaefer PW, Grant E, Gonzalez G. Diffusion-weighted MR imaging of the brain. Radiology. 2000; 217 (2); 331345. doi: 10.1148/radiology.217.2.r00nv24331. 2 Finelli PF, Gleeson E, Ciesielski T, et al. Diagnostic role of target lesion on diffusion-weighted imaging. A case of cerebral aspergillosis and review of the literature. Neurologist. 2010; 16 (6): 364-367. doi: 10.1097/ NRL.0b013e3181b47001. 3 Finelli PF. Primary CNS lymphoma in myasthenic on long-term azathioprine. J Neurooncol. 2005; 74 (1): 9192. doi: 10.1007/s11060-004-5676-1. 4 Finelli PF, Uphoff DF. Ring Lesion on DWI. Arch Neurol. 2011; 68 (11): 1474-1475. doi: 10.1001/archneurol.2011.612. 5 Finelli PF, Naik K, DiGiuseppe JA, et al. Primary lymphoma of CNS, mycophenolate mofetil and lupus. Lupus. 2006; 15 (12): 886-888. doi: 10.1177/0961203306071431. 6 Finelli PF, DiMario FJ Jr. Diagnostic approach in patients with symmetric imaging lesions of deep gray nuclei. Neurologist. 2003; 9 (5): 250-261. doi: 10.1097/01. nrl.0000087718.55597.6a. 7 Wang HS. Concentric thalamic change on MR of acute necrotizing encephalopathy of childhood. Neuroradiol. 2003; 45 (9): 661-662. doi: 10.1007/s00234-003-1033-x. 8 Albayram S, Bilgi Z, Seleuk H, et al. Diffusion-weighted MR imaging findings of acute necrotizing encephalopathy. Am J Neuroradiol. 2004; 25 (5): 792-797. 9 Kavanagh EC, Heran MKS, Fenton DM, et al. Diffusion-weighted imaging findings in Balo’s concentric sclerosis. Br J Radiol. 2006; 79 (943): e28-e31. doi: 10.1259/bjr/36636301. 10 Kato H, Hasegawa H, Iijima M, et al. Brain magnetic resonance imaging of an adult case of acute necrotizing encephalopathy. J Neurol. 2007; 254 (8): 1135-1137. doi: 10.1007/s00415-006-0476-5. 11 Mowry EM, Woo JH, Ances BM. Balo’s concentric sclerosis presenting as a stroke-like syndrome. Nat Clin Pract Neurol. 2007; 3 (6): 349-354. doi: 10.1038/ncpneuro0522. 12 de Seze J, Debouverie M, Zephir H, et al. Acute fulminant demyelinating disease. Arch Neurol. 2007; 64 (10); 1426-1432. doi: 10.1001/archneur.64.10.1426. 13 Moragas M, Martinez-Yelamos S, Majos C, et al. Rhombencephalitis: A series of 97 patients. Medicine (Baltimore). 2011; 90 (4): 256-261. doi: 10.1097/ MD.0b013e318224b5af. 14 Grinspan ZM, Willey JZ, Tullman MJ, et al. Clinical Reasoning: A 28-year-old pregnant women with encephalopathy. Neurol. 2009; 73: e74-e79. doi: 10.1212/ WNL.0b013e3181bbfeb3. 15 Raj G, Kulshrestha D, Chauhan A, et al. Multiple tumefactive demyelinating lesions demonstrated on 3TMRI – a case report. J Med Med Sci. 2014; 5 (2): 4144. doi: 10.14303/jmms.2013.412. 16 Gaviani P, Schwartz RB, Hedley-Whyte ET, et al. Diffusion-weighted imaging of fungal cerebral infection. Am J Neuroradiol. 2005; 26 (5): 1115-1121. 17 Charlot M, Pialat J.-B, Obadia N, et al. Diffusionweighted imaging in brain aspergillosis. Eur J Neurol. 2007; 14 (8): 912-916. doi: 10.1111/j.14681331.2007.01874.x. 18 Shah R, Bag AK, Chapman PR, et al. Imaging manifestations of progressive multifocal leukoencephalopathy. Clin Radiol. 2010; 65 (6): 431-439. doi: 10.1016/j. crad.2010.03.001. 19 Mader J, Herrlinger U, Klose U, et al. Progressive multifocal leukoencephalopathy: analysis of lesion development with diffusion-weighted MRI. Neuroradiology. 2003; 45 (10): 717-721. doi: 10.1007/s00234-0030966-4. 20 Lai PH, Ho JT, Chen WL, et al. Brain abscess and necrotic brain tumor: discrimination with proton MR spectroscopy and diffusion-weighted imaging. Am J Neuroradiol. 2002; 23 (8): 1369-1377. Pasquale F. Finelli, MD Hartford Hospital 80 Seymour Street Hartford, CT 06102-5037,USA Tel.: (860) 545-3621 Fax: (860) 545-5003 E-mail: pfinell@harthosp.org 287 The Neuroradiology Journal 27: 288-292, 2014 - doi: 10.15274/NRJ-2014-10041 www.centauro.it Lentiform Fork Sign: a Magnetic Resonance Finding in a Case of Acute Metabolic Acidosis DANIELA GRASSO1, CARMELA BORREGGINE2, FRANCESCO PERFETTO3, VINCENZO BERTOZZI3, MARINA TRIVISANO4, LUIGI MARIA SPECCHIO4, GIANPAOLO GRILLI3, LUCA MACARINI2 1 Department of Imaging Diagnostics, 2 Specialisation School in Radiodiagnostics, 4 Specialisation School in Neurology, University of Foggia; Foggia, Italy 3 Radiodiagnostic Unit, “Ospedali Riuniti” University Hospital; Foggia, Italy Key words: diffusion-weighted imaging, lentiform fork sign, metabolic acidosis, MRI, putaminal necrosis SUMMARY – We report a 33 year-old woman addicted to chronic unspecified solvents abuse with stupor, respiratory disorders, tetraplegia and severe metabolic acidosis. On admission an unenhanced cranial CT scan showed symmetrical hypodensities of both lentiform nuclei. MR imaging performed 12 hours after stupor demonstrates bilateral putaminal hemorrhagic necrosis, bilateral external capsule, corona radiata and deep cerebellar hyperintensities with right cingulate cortex involvement. DWI reflected bilateral putaminal hyperintensities with restricted water diffusion as to citotoxic edema and development of vasogenic edema in the external capsule recalling a fork. On day twenty, after specific treatments MRI demonstrated a bilateral putaminal marginal enhancement. Bilateral putaminal necrosis is a characteristic but non-specific radiological finding of methanol poisoning. Lentiform Fork sign is a rare MRI finding reported in literature in 22 patients with various conditions characterized by metabolic acidosis. Vasogenic edema may be due to the differences in metabolic vulnerability between neurons and astrocytes. We postulate that metabolic acidosis could have an important role to generate this sign. Introduction The basal ganglia are particularly vulnerable to a wide range of toxins and metabolic disturbances 1. In the acute setting, bilateral basal ganglia abnormalities on MRI could be non-specific and the differential diagnosis should include hypoglycaemia, cerebral hypoxic encephalopathy, carbon monoxide poisoning, encephalitis, osmotic myelinolysis, toxin exposure, vascular causes and drug abuse 2. The present study describes a rare neuroradiological finding called the lentiform fork sign in a patient with acute metabolic acidosis, bilateral putaminal necrosis, cortical and deep cerebellar hyperintensities. Uremic encephalopathy, diabetes mellitus, methanol and ethylene glycol intoxications, acidopathies such as propionic acidaemia (PA) and pyruvate dehydrogenase deficiency were commonly associated with metabolic acidosis. Literature reviews 3-11 demonstrated that these conditions were also 288 related to a lentiform fork appearance. Methanol is a clear colourless toxic liquid, commonly found in many commercial products of daily use like solvents, antifreeze, varnish and cleaning fluids. Accidental or suicidal oral ingestion causes visual disturbance and central nervous system (CNS) depression leading to respiratory failure and severe metabolic acidosis. Bilateral putaminal necrosis and subcortical white matter lesions were the most frequent neuroimaging findings in patients with methanol intoxication 12. We describe a comatose 33-year-old woman admitted to the emergency department with mild mydriasis and bilaterally unreactive pupils. Case Report A 33-year-old woman with fever, vomiting, ocular deviation, dyspnoea and deterioration of consciousness was referred to our emergency Daniela Grasso Lentiform Fork Sign: a Magnetic Resonance Finding in a Case of Acute Metabolic Acidosis B D ° ° ³ ³ C ° A ³ ³ ³ ³ ³ G ³ F ³ E Figure 1 A) Unenhanced CT scan indicates symmetrical hypodensities of both lentiform nuclei (arrowheads). B-D) MRI FLAIR images reveal bilateral hyperintensities involving the lentiform nuclei (short white arrow), right cingulum cortex (long white arrow) and cerebellum (arrowhead) with a brightly hyperintense rim surrounding both putamina resembling a fork. E) The T2 sequence shows bilateral putaminal hyperintensities with the lentiform fork (arrowheads). F) T1 MRI shows hypointensity over the basal ganglia bilaterally (arrowheads). G) Axial T2 Fast Field Echo (FFE) demonstrates hypointensities on both basal ganglia (arrowheads). ° ° ° B ° A Figure 2 A) DWI shows diffusion restriction in both basal ganglia, in particular the putamen and globus pallidus. B) ADC map shows low signal intensities in the lentiform nucleus bilaterally and high signal intensities of both the forks. 289 Lentiform Fork Sign: a Magnetic Resonance Finding in a Case of Acute Metabolic Acidosis C ° ° ³ ³ B ° ° A Daniela Grasso Figure 3 A) Brain CT scan on day 10 shows a soft reduction of bilateral putaminal hypodensities (arrowheads). B,C) Follow-up brain MRI (20 days later) reveals a reduction of lentiform hyperintensities (short white arrows) and post-contrast T1 sequences demonstrate bilateral putaminal lesions with marginal enhancement (long white arrows). department. On admission neurological examination revealed quadriplegia, stupor and bilateral Babinski sign. No extrapyramidal signs, such as resting tremor or rigidity and no asterixis or myoclonus were detected. Glasgow coma scale (GCS) score was 7 at that time. Electrocardiography and chest radiogram were normal. Arterial blood gas analysis showed systemic metabolic acidosis (pH:7.03; PCO2 11 mmHg; P02 122; HCO3 2,9 mmol/L; BE -25 mmol/L) with high anion gap and high osmolar gap. Laboratory evaluation revealed elevated liver enzyme levels (ALT 143 U/L (2-40); AST 340 U/L (2-40)), serum creatinine 2.07 mg/dl (0.5-1.2) and hyperkalaemia 6,8 mmol/L (3.55.3). The glucose level was 190 mg/dl (50-110) and the ammonia level was 61 mol/l. Urinalysis demonstrated elevated ketone and protein levels without blood cells. No drug abuse was reported but in the last six months the patient had been addicted to taking diluted alcoholic solutions (disinfectants). After admission to the intensive care unit, the patient was treated with intubation, plasma expanders and sodium bicarbonate to correct acidosis. An initial unenhanced CT scan revealed symmetrical hypodensities of both lentiform nuclei (Figure 1A). Twelve hours after admission, brain MRI (Figure 1B-E) revealed bilateral T2 and FLAIR hyperintensities of lentiform nuclei, cerebellum and right cingulum cortex with a brightly hyperintense rim surrounding both 290 putamina that looked like a fork. T1-weighted MRI (Figure 1F) showed hypointensity over the basal ganglia bilaterally. Axial T2 fast field echo (FFE) demonstrated hypointensities on both basal ganglia (Figure 1G). Diffusionweighted imaging (DWI) showed high signal intensities in the corresponding areas (Figure 2A). The apparent diffusion coefficient (ADC) map showed low signal intensities in the putamen and globus pallidus bilaterally and high signal intensities of both the forks (Figure 2B). On day 10 a brain CT scan showed a soft reduction of bilateral putaminal hypodensities (Figure 3A). Follow-up brain MRI (20 days later) revealed reduction of the lentiform and cerebellar lesions and post-contrast T1 sequences demonstrated bilateral putaminal hyperintensities with marginal enhancement (Figure 3B,C). Discussion We describe a case of acute metabolic acidosis related to aspecific solvent abuse that showed unusual bilateral basal ganglia involvement. Conventional MRI demonstrated bilateral putaminal necrosis with a haemorrhagic component in association with another sign recently called lentiform fork sign. Kumar et al. 3 described the constitutive elements of the lentiform fork: 1) the lateral arm, formed by the oedematous external capsule and extending from the anterior end of the putamen www.centauro.it The Neuroradiology Journal 27: 288-292, 2014 - doi: 10.15274/NRJ-2014-10041 to the stem; 2) the stem, created by merging oedematous external and internal capsules at the infero-posterior end of the putamen; 3) the medial arm extended from the stem anteriorly up to one third of the medial edge where it split into two slightly less T2/FLAIR hyperintense branches engulfing the globus pallidus. These two branches are constituted by the oedematous medullary laminae, which divide the lentiform nucleus into three masses (the putamen, globus pallidus interna and externa). DWI showed bilateral putaminal hyperintensities with water restriction and low signal intensities of both the lentiform forks. The apparent diffusion coefficient (ADC) map showed low signal intensities in the putamen and globus pallidus and high signal intensities of the forks bilaterally. Diffusion imaging had the ability to discriminate between cytotoxic and vasogenic oedema showing the different water mobility in the brain regions involved. Bilateral putaminal necrosis had a cytotoxic nature caused by ischaemia or an irreversible damage. The lentiform fork demonstrated increased water mobility caused by vasogenic oedema. Metabolic acidosis was proposed as a trigger in the pathogenesis of this pattern of vasogenic oedema and is essential for generating the lentiform fork sign in several clinical settings. In our patient, vasogenic oedema had partially resolved on follow-up images after intubation and sodium bicarbonate infusion to correct metabolic acidosis. The pathogenetic basis of this characteristic vasogenic oedema may be attributable to the differences in metabolic vulnerability between neurons (that composed basal ganglia) and astrocytes (which formed the surrounding white matter in association with axons and oligodendrocytes). Through changes in vascular reactivity, metabolic acidosis may disrupt the blood–brain barrier leading to vasogenic oedema and later cytotoxic oedema, in relation to the severity of acidosis. Contrastenhanced MRI confirmed bilateral blood-brain barrier disruption on the putamina. We assumed that putaminal necrosis was associated with alcoholic intoxication such as methanol poisoning, but this finding is not specific and was also seen in a variety of conditions including Wilson and Leigh disease, Kearns-Sayre syndrome and various other neurodegenerative disorders 14-18. The mechanism underlying se- lective putaminal necrosis is unknown. It has been postulated that the necrosis results from decreased blood flow through the basal veins of Rosenthal secondary to hypotension 19 or from a direct toxic effect of formic acid that shows a higher accumulation in the putamina than in other parts of the brain 20. On the other hand, there was varying sensitivity of striatal neurons to toxic metabolites of methanol 21. Formic acid is responsible for metabolic acidosis which results from the inhibition of cytochrome oxidase, a mitochondrial enzyme required for oxidative phosphorylation. Its inhibition leads to anoxia and, furthermore, to cellular oedema and cell death. It may be difficult to diagnose methanol poisoning if no history of ingestion can be obtained. Therefore methanol poisoning should be considered in every patient with a metabolic acidosis of unknown origin 22. Haemorrhage, confined to the putamina or in the subcortical white matter, has been reported in up to 14% of patients with methanol poisoning 13,23 and has been suggested to indicate a poor prognosis 19,24. Few studies have used DWI in methanol poisoning 19,26. DWI demonstrated restricted diffusion on both putamina such as cytotoxic oedema that caused blood-brain barrier damage. Furthermore, the putamen’s high metabolic demand and its location in the boundary zone of vascular perfusion make it particularly susceptible to vascular damage. We believe that several factors, including cerebral microvascular anatomy, direct methanol metabolite toxicity and severe metabolic acidosis produce this characteristic distribution of pathologic findings. Conclusions Bilateral putaminal necrosis is a characteristic but non-specific radiological finding of methanol poisoning. It is probably a consequence of direct formic acid toxicity on the basal ganglia. The lentiform fork sign is a rare MRI finding reported in the literature in 22 patients with various conditions characterized by metabolic acidosis. Vasogenic oedema may be due to the differences in metabolic vulnerability between neurons and astrocytes. Metabolic acidosis could have an important role in generating this sign. 291 Lentiform Fork Sign: a Magnetic Resonance Finding in a Case of Acute Metabolic Acidosis Daniela Grasso References 1 Albin RL. Basal ganglia neurotoxins. Neurol Clin. 2000; 18 (3): 665-680. doi: 10.1016/S0733-8619(05)70217-6. 2 Beltz EE, Mullins ME. Radiological reasoning: hyperintensity of the basal ganglia and cortex on FLAIR and diffusion-weighted imaging. Am J Roentgenol. 2010; 195 (3 Suppl): S1-8. (Quiz S9-11). doi: 10.2214/AJR.07.7089. 3 Kumar G, Goyal MK. Lentiform fork sign: a unique MRI picture. Is metabolic acidosis responsible? Clin Neurol Neurosurg. 2010; 112 (9): 805-812. doi: 10.1016/j. clineuro.2010.06.006. 4 Wang HC, Cheng SJ. The syndrome of acute bilateral basal ganglia lesions in diabetic uremic patients. J Neurol. 2003; 250 (8): 948-955. doi: 10.1007/s00415-0031122-0. 5 Yoon CH, Seok JI, Lee DK, et al. Bilateral basal ganglia and unilateral cortical involvement in a diabetic uremic patient. Clin Neurol Neurosurg. 2009; 111 (5): 477-479. doi: 10.1016/j.clineuro.2009.01.007. 6 Lee EJ, Park JH, Ihn Y, et al. Acute bilateral basal ganglia lesions in diabetic uraemia: diffusion-weighted MRI. Neuroradiology. 2007; 49 (12): 1009-1013. doi: 10.1007/s00234-007-0299-9. 7 Kim TK, Seo SI, Kim JH, et al. Diffusion-weighted magnetic resonance imaging in the syndrome of acute bilateral basal ganglia lesions in diabetic uremia. Mov Disord. 2006; 21 (8): 1267-1270. doi: 10.1002/mds.20932. 8 Li JY, Yong TY, Sebben R, et al. Bilateral basal ganglia lesions in patients with end-stage diabetic nephropathy. Nephrology (Carlton). 2008; 13 (1): 68-72. doi: 10.1111/j.1440-1797.2007.00838.x. 9 Sheu YL, Cheng SJ, Chen YM, et al. The syndrome of bilateral basal ganglia lesions in diabetic uremic patients presenting with a relapsing and remitting course: a case report. Acta Neurol Taiwan. 2007; 16 (4): 226-230. 10 Johnson JA, Le KL, Palacios E. Propionic acidemia: case report and review of neurologic sequelae. Pediatr Neurol. 2009; 40 (4): 317-320. doi: 10.1016/j.pediatrneurol.2008.10.021. 11 Moore MM, Kanekar SG, Dhamija R. Ethylene glycol toxicity: chemistry, pathogenesis, and imaging. Radiol Case Rep. 2008; 3 (1): 1-5. doi: 10.2484/rcr.v3i1.122. 12 Arora V, Nijjar IB, Multani AS, et al. MRI findings in methanol intoxication: a report of two cases. Br J Radiol. 2007; 80 (958): e243-246. doi: 10.1259/bjr/40137535. 13 Glazer M, Dross P. Necrosis of the putamen caused by methanol intoxication: MR findings. Am J Roentgenol. 1993; 160 (5): 1105-1106. doi: 10.2214/ajr.160.5.8470586. 14 Hsu HH, Chen CY, Chen FH, et al. Optic atrophy and cerebral infarcts caused by methanol intoxication: MRI. Neuroradiology. 1997; 39 (3): 192-194. doi: 10.1007/ s002340050391. 15 Medina L, Chi TL, Devivo DC, et al. MR findings in patients with subacute necrotizing encephalomyelopathy (Leigh syndrome): correlation and biochemical defect. Am J Neuroradiol. 1990; 11 (2): 379-384. 16 Lawler GA, Pennock JM, Steiner RE, et al. Nuclear magnetic resonance imaging in Wilson disease. J Comput Assist Tomogr. 1983; 7 (1): 1-8. doi: 10.1097/00004728198302000-00001. 17 Seidenwurm D, Novotny E, Marshall W, et al. MR and CT in cytoplasmically inherited striatal degeneration. Am J Neuroradiol. 1986; 7 (4): 629-632. 18 Starosta-Rubinstein S, Young AB, Kluin K, et al. Clinical assessment of 31 patients with Wilson’s disease. Correlations with structural changes on magnetic resonance imaging. Arch Neurol. 1987; 44 (4): 365-370. doi: 10.1001/archneur.1987.00520160007005. 19 Feaney MB, Anthony DC, Frosch MP, et al. August 2000: Two cases with necrosis and hemorrhage in the putamen and white matter. Brain Pathol. 2001; 11 (1): 121-122. 292 20 Pelletier J, Habib MH, Khalil R, et al. Putaminal necrosis after methanol intoxication. J Neurol Neurosurg Psychiatry. 1992 ;55 (3): 234-235. doi: 10.1136/ jnnp.55.3.234. 21 LeWitt PA, Martin SD. Dystonia and hypokinesis with putaminal necrosis after methanol intoxication. Clin Neuropharmacol. 1988; 11 (2): 161-167. doi: 10.1097/00002826-198804000-00007. 22 Hovda KE, Hunderi OH, Rudberg N, et al. Anion and osmolal gaps in the diagnosis of methanol poisoning: clinical study in 28 patients. Intensive Care Med. 2004; 30 (9): 1842-1846. doi: 10.1007/s00134-004-2373-7. 23 Phang PT, Passerini L, Mielke B, et al. Brain hemorrhage associated with methanol poisoning. Crit Care Med. 1988; 16 (2): 137-140. doi:10.1097/00003246198802000-00008. 24 Gaul HP, Wallace CJ, Auer RN, et al. MR findings in methanol intoxication. Am J Neuroradiol. 1995; 16 (9): 1783-1786. 25 Aquilonius SM, Bergström K, Enoksson P, et al. Cerebral computed tomography in methanol intoxication. J Comput Assist Tomogr. 1980; 4 (4): 425-428. doi: 10.1097/00004728-198008000-00001. 26 Server A, Hovda KE, Nakstad PH, et al. Conventional and diffusion-weighted MRI in the evaluation of methanol poisoning. Acta Radiol. 2003; 44 (6): 691-695. doi: 10.1080/02841850312331287779. 27 Deniz S, Oppenheim C, Lehéricy S, et al. Diffusionweighted magnetic resonance imaging in a case of methanol intoxication. Neurotoxicology. 2000; 21 (3): 405-408. Daniela Grasso, MD Dipartimento di Diagnostica per Immagini Università di Foggia Viale L. Pinto 1 70100 Foggia Italy E-mail: daniela.grasso@hotmail.it The Neuroradiology Journal 27: 293-298, 2014 - doi: 10.15274/NRJ-2014-10052 www.centauro.it Innervation of the Cerebral Dura Mater XIANLI LV, ZHONGXUE WU, YOUXIANG LI Beijing Neurosurgical Institute and Beijing Tiantan Hospital, Capital Medical University; Beijing, China Key words: cerebral dura mater, sensory receptors, nerve fibers, nociception SUMMARY – The trigemino-cardiac reflex during Onyx embolization for dural arteriovenous fistula may be caused by mechanical or chemical stimulus to the terminals of the unencapsulated Ruffini-like receptors stemming from A-axons in the dural connective tissue at sites of dural arteries and sinuses. Slow A (ADž) and fast A (Aǃ) neurons may play a role in the stimulus afferent pathway due to their higher mechanosensitivity and chemosensitivity. These afferent pathway nerves are cholinergic innervations of the dura mater, which also contains vasoactive neuropeptides such as calcitonin gene-related peptide, substance P, and neurokinin A. Stimulation of meningeal sensory Àbres can evoke cerebral vasodilation through the peripheral release of neuropeptides, which play a role in headache pathogenesis. These myelinated A-Àbers terminate in the deep part (laminae IIIV) of the spinal dorsal horn. Its efferent pathway has been defined as the acetylcholinergic vagus nerve. The A11 nucleus, located in the posterior hypothalamus, providing the only known source of descending dopaminergic innervation for the spinal grey matter, can inhibit the neurons in the spinal dorsal horn. Background During endovascular Onyx embolization for intracranial dural arteriovenous fistula, a trigemino-cardiac reflex has been observed 1,2. This reflex has been described as the sudden development of cardiac arrhythmia as far as cardiac arrest, arterial hypotension, apnea and gastric hypermobility during stimulation of any of the sensory branches of the trigeminal nerve 1,2. Its efferent pathway has been defined as the acetylcholinergic vagus nerve. However, the receptor, afferent nerve, central projections and transmitters of this reflex were not clear. This study reviews the literature with a focus on the cerebral dura mater innervations. Afferent nerve fibers of the cerebral dura mater The cerebral dura mater is richly innervated by afferent nerve fibers, most of which originate in the ipsilateral trigeminal ganglion, and by sympathetic fibers predominantly arising from the ipsilateral superior cervical ganglion 3,4. For the supratentorial part the main nerve supply stems from all three branches of the trigeminal nerve. The dorsal rami of the first three cervical nerves, the ventral rami of the first two cervical nerves, the hypoglossal nerve and recurrent branches of the vagus nerve that follow the posterior meningeal artery provide innervation to the posterior cranial fossa dura mater 5. In addition, a comparatively sparse parasympathetic innervation was described predominantly arising from the ipsilateral sphenopalatine ganglion 3,4 . Meningeal sensory neurons in the dura initially course alongside the arteries en route to their territory of innervation, but the individual nerve Àbers typically exit the main bundle and travel some distance away from the artery before reaching their main territory of arborization 6, and the majority of nerve endings are not in close proximity to arteries 7. Davidson et al. 8 investigated changes in the innervation of the human dura with age in 27 individuals aged between 31 weeks of gestation and 60 years of postnatal life. Using immunocytochemistry with antibodies to neurofilament, they found that the density of innervation increased between 31 and 40 weeks of gestation, peaking at term and decreasing in the subse293 Innervation of the Cerebral Dura Mater quent three months, remaining low until the sixth decade. Transmitters Several studies have described neuropeptide immunoreactive nerve fibers in the dura mater 9 . Meningeal nerve fibers immunoreactive for substance P (SP), neurokinin A, and calcitonin gene-related peptide (CGRP) are thought to belong to the afferent (sensory) system, and nerve fibers immunopositive for neuropeptide Y(NPY) are most likely of sympathetic origin while those immunoreactive for vasoactive intestinal peptide (VIP) are of parasympathetic origin. Light and electron microscopic studies of the rat dura mater have shown that peptidergic (sensory) nerve fibers form a dense network both around blood vessels and in nonvascular regions. The cholinergic innervation of the dura mater in the guinea pig, mouse and rat has been investigated. The acetylcholinesterase (AChE)-containing nerve fibers were in close relationship with the main meningeal blood vessels and also appear in the meningeal tissue proper. A sparse to moderate supply of nerve fibers immunoreactive for NPY, VIP, SP, and CGRP was demonstrated in the walls of human middle meningeal arteries 10-12. Substance P immunoreactive fibers occur in the wall of the superior sagittal sinus and the meningeal artery 13,14 . An immunocytochemical electron microscopy study found different unmyelinated axons in one Schwann cell may contain nerve fibers with VIP, Substance P and NPY immunoreactivity. This study also observed a colocalization of NPY and norepinephrine in some nerve fibers. Trigeminal sensory nerve Àbers from the dura mater have been found to contain vasoactive neuropeptides such as calcitonin gene-related peptide, substance P, and neurokinin A 15 . At least a third of these Àbers are Aǃ Àbers 16 . Findings were similar for both CGRP and SP: the supra and infratentorial dura, and convexity and skull base dura had a rich plexus of nerve bundles and a nerve-fiber network 17,18. A rich array of mast cells was noted in the supra and infratentorial calvarial dura and they contain a substantial proportion of total brain histamine. Fewer mast cells occurred in the skull base dura than in the calvaria dura. Mast cells were also identified in the falx cerebri and tentorium cerebelli, but were absent in the dura over the sinus. Activated mast cells 294 Xianli Lv could potentially release mediators that in turn activate meningeal nociceptors. A study on the effect of various doses of acetylcholine (100, 200, and 500 mg/kg) on the mast cells of the albino rat dura mater concluded that acetylcholine has an activating effect on monoamine secretion 19, SP and CGRP-containing neurons that can in turn activate dura mast cells. Vasoactive mediators and cytokines released from mast cells then increase vascular permeability. Mast cell vasodilatory molecules such as histamine nitric oxide NO, tumor necrosis factor TNF and VIP could participate in the vasodilatory phase of migraine, which is associated with the throbbing pain 20. Interactions between autonomic and nociceptive fibers The autonomic and sensory nerves form a dense network accompanying blood vessels. Interactions between autonomic and nociceptive fibers have been investigated by measuring release of CGRP and prostaglandin E2 (PGE2) from the dura mater in vitro. The parasympathomimetic agent carbachol did not change basal release of CGRP or PGE2, whereas it diminished release induced by a mixture of inflammatory mediators. Norepinephrine did not change induced release of CGRP or PGE2, or basal release of CGRP. However, basal release of PGE2 was enhanced by norepinephrine, and this enhancement was reduced by serotonin through 5-HT(1D) receptors. It was found that sympathetic transmitters may control nociceptor sensitivity via increased basal PGE2 levels, a possible mechanism to facilitate headache generation. Parasympathetic transmitters may reduce enhanced nociceptor activity 19,21. CGRP-immunoreactive nerve fibers and trigeminal ganglion cells innervating the meninges are much more abundant than SPimmunoreactive afferents. Co-localization of SP and CGRP was shown in a small portion of trigeminal ganglion cells. Subarachnoid hemorrhage in the rat substantially reduced the SP-immunoreactive nerve fibers of the dura mater, but left the CGRP-immunoreactive innervation unchanged. Taken together, the CGRP-containing trigeminal innervation of the intracranial structures seems to play an important role in meningeal nociception. Meningeal sensory fibers can be stimulated to release neuropeptides from their peripheral endings in the meninges, where they can www.centauro.it The Neuroradiology Journal 27: 293-298, 2014 - doi: 10.15274/NRJ-2014-10052 evoke components of neurogenic inflammation, including dural plasma extravasation, as well as dural and pial vasodilation. SP acting at the neurokinin-1 receptor level appears to be responsible for extravasation, whereas CGRP mediates neurogenic vasodilation. Neurogenic inflammation has been hypothesized to play a role in headache pathogenesis. Stimulation of meningeal sensory Àbers can evoke cerebral vasodilation through the peripheral release of neuropeptides 22. Types of stimuli of dural primary afferent neurons To identify the types of stimuli capable of activating meningeal sensory fibers, electrophysiologic recording studies in animals have examined the response properties of primary afferent neurons innervating the cranial dura mater 7. These studies obtained single-unit recordings from the trigeminal ganglion or the nasociliary nerve. Dural afferents were typically identified by their responses to electrical stimulation of dural sites on or around the dural venous sinuses, namely, superior sagittal or transverse sinuses, or the middle meningeal artery (MMA). Based on their response latencies, the majority of the neurons could be classified as C or ADž, although a substantial number of Aǃ fibers were also present, in agreement with anatomic observations. Many of the neurons exhibit mechanosensitive receptive fields on the dura from which they can be activated by punctuate probing, stroking or traction. Some neurons also respond to thermal stimuli, in either the warming or cooling direction. In addition, neurons were excited by chemical stimuli, such as hypertonic saline applied either topically to the dura or by intravascular infusion into the superior sagittal sinus (SSS). Neurons could also be activated by dural application of a number of algesic agents, including potassium chloride, capsaicin, buffer solutions of low or high osmolarity, or a mixture of inflammatory mediators (bradykinin, PGE2, serotonin, histamine). Besides activating the neurons, the inflammatory mediators produced a sensitization, or enhancement, of responses to mechanical stimuli. The polymodal response properties of the dural afferent neurons, and their excitatory and sensitizing responses to algesic and inflammatory agents support the idea that some of these neurons serve a nociceptive function, and that they might be activated under pathologic conditions such as increased intracranial pressure or meningitis. Unencapsulated Ruffini-like receptors of A and C axons These dural nerves contain myelinated (Aaxons) and unmyelinated (C-axons) nerve fibers. Myelinated and unmyelinated nerve fibers may traverse midline structures and terminate opposite to their origin 4. After losing the myelin sheath the unmyelinated branches of the A-fiber mix up with the C-fibers. They can be traced within the densely packed collagenous fiber bundles they innervate. The unmyelinated branches of the A-fibers have a larger diameter than the C-fibers. At several sites, axonal protrusions exhibit a receptor matrix, mitochondria and vesicles. Large areas of the axon branches are not covered by Schwann cell cytoplasm. This arrangement resembles the ultrastructural relationship between nerve terminals and collagenous fibers in unencapsulated Ruffini-like receptors. Within the dura mater this receptor type is mainly distributed at locations where the superior cerebral veins empty into the sinus, and at the confluences of sinuses. The C-axons which follow the unmyelinated branches of the Aaxons terminate at a postcapillary venule. The terminal axons are swollen and filled with mitochondria and vesicles. The axonal membrane is in direct contact to the basement lamella of the endothelial cell. A basement lamella may be completely absent around the axonal terminals. This arrangement leads us to suspect that the various afferent units fulfill different receptor functions to monitor various parameters of the chemical composition and/or mechanical condition of the innervated area. From this point of view, it has to be assumed that the different terminals have different transducer properties despite their rather uniform ultrastructural appearance. For instance, these terminals are the best candidates to sense the composition of the blood and the exchange of fluid at the post-capillary venule 4. Central projections of sensory nerves innervating the MMA and SSS Depending on the application site of Cholera toxin subunit b (CTb) or wheat germ agglutininhorseradish peroxidase conjugate (WGA-HRP) 295 Innervation of the Cerebral Dura Mater Xianli Lv to the adventitia of the MMA, labeled neurons were predominantly found ipsilaterally in the lateral part of the spinal dorsal horn of segments C1-C3 and in the caudal and interpolar parts of the spinal trigeminal nucleus 23. WGAHRP-labeled terminations were mainly located in laminae I and II, whereas CTb-labeled terminations were located in laminae III-V. These results indicate that sensory information from the MMA is transmitted through both trigeminal and cervical spinal nerve branches to a region in the central nervous system extending rostrally from the C3 dorsal horn to the interpolar part of the spinal trigeminal nucleus. Liu et al. 24 examined the central projections of the SSS sensory innervation in rat using transganglionic tract tracing techniques. CTb or choleragenoid-conjugated horseradish peroxidase (B-HRP) label both large and small diameter myelinated A-Àbers, which mainly originate from low threshold mechanoreceptors and terminate in the deep part (laminae III–V) of the spinal dorsal horn. WGA-HRP preferentially labels unmyelinated C-Àbers which terminate in the superÀcial layers (laminae I and II) and transmit nociceptive stimuli. more, unlike the C-Àbers, the fast A neurons that were mechanically insensitive could not be sensitized or activated by chemical stimuli, including 100 mM KCl. The Ànding that the fastest conducting neurons have the lowest mechanosensitivity is the reverse of what is found in the cutaneous innervation where the most sensitive mechanoreceptors are Aǃ Àbers. One characteristic that many of the fast conducting dural afferents share with cutaneous Aǃ Àbers is rapid adaptation. However, unlike in cutaneous afferents, rapid adaptation in dural afferents is associated with very high response thresholds and with rapidly fatiguing responses. Such responses might be suited for signaling transient mechanical stimuli such as would result from sudden head movements. The Aǃ fibers may play an important role in TCR and this should be further investigated. The Ànding of a substantial number of fastconducting dural afferents was surprising as it has generally been thought that the dura received a predominantly small-Àber sensory innervation 27, similar to other visceral tissues. Previously ADž was documented to be the afferent pathway for TCR. Subpopulations Sensory responses of dural primary afferent neurons Although the dural afferent population does not appear to mediate distinct sensory modalities, it does show a pattern of variation in mechanosensitivity as a function of conduction velocity that suggests the presence of subpopulations 25. The most marked difference was between neurons with conduction velocities more or less than 5 m/s, and consequently this conduction velocity was used to subdivide neurons in the A Àber range into slow A (1.5-5 m/s, corresponding to the low end of the ADž range) and fast A (5-25 m/s, which includes neurons in the Aǃ as well as the upper end of the ADž range), as discussed 25. Slow A neurons had the highest mechanosensitivity, in that they had the highest incidence of mechanosensitivity (97%) as well as the highest stimulus-response slopes and the lowest thresholds. Furthermore, a substantial number of C neurons (18%) had no baseline mechanosensitivity, although in some cases these mechanically insensitive neurons could be sensitized by inÁammatory mediators 26. In contrast, most of the Aǃ neurons had much lower mechanosensitivity than the ADž or C Àbers, and a relatively high percentage (33%) had no mechanosensitivity. Further296 Stimulation of dural afferent C-fibers and greater occipital nerve (GON) increased the background activity, extended the size of cutaneous trigeminal and cervical receptive fields, and decreased the thresholds to mechanical dural stimulation 24,28. These findings suggest that dural stimulation may lead to a central sensitization of nociceptive convergent second-order neurons with an increased responsiveness to stimulation of cervical afferents. This mechanism may be important in pain referral from cervical structures to the head and therefore have implications for most forms of primary headache 24,28. The locations of the recording sites of neurons responding to convergent input from the dura mater and the GON were confined to laminae V/VI of the C2 dorsal horn, which is consistent with other studies analyzing responses of convergent neurons to stimulation of dura mater and dural vessels 22,29-31. The location of the recorded neurons corresponds to the dorsal horn area that receives projections from the ophthalmic division of the trigeminal nerve 6, which constitutes the primary source www.centauro.it The Neuroradiology Journal 27: 293-298, 2014 - doi: 10.15274/NRJ-2014-10052 of afferents from the supratentorial dura mater . It has been shown that the trigeminal innervation of the cerebral circulation preferentially employs calcitonin gene-related peptide CGRP in favor of SP 31,34. A functional reflection of this can be seen in studies of neuropeptide release. Trigeminal ganglion stimulation in cat and humans increases cranial levels of both CGRP and SP while during migraine and cluster headache substance P release is not seen. In the same way, CGRP is preferentially released in subarachnoid hemorrhage. Similarly, SSS stimulation preferentially releases CGRP in the cat while it has also been shown that increases in facial blood flow following trigeminal ganglion stimulation in guinea pig are primarily due to release of CGRP. Direct stimulation of the nasociliary nerve, the cerebrovascular branch of the ophthalmic division of the trigeminal nerve, leads to a cerebral vasodilator effect that is inhibited by the CGRP antagonist CGRP. It is likely that the neuropeptide measurements represent local overflow from trigeminal nerves innervating the cerebral and extracerebral circulation. Certainly the release is likely to be local since exogenous CGRP only markedly alters cerebral blood flow after bloodbrain barrier disruption. Dural vessels are innervated by an adventitial plexus of sensory nerve fibers which contain vasodilator neuropeptides. These include SP, neurokinin A and CGRP, which are stored within vesicles of the naked nerve endings. Vasoactive neuropeptides can be released from perivascular sensory nerves by axon reflex mechanisms in response to a variety of stimuli (an axon reflex involves antidromic depolarisation within the collateral branches of a single sensory neuron in response to peripheral stimulation of one branch, e.g. the triple response). 22,30-33 Central mechanisms Expression of c-fos immunoreactivity within the brain or spinal cord can be used to map the distribution of neurons activated by noxious or non-noxious stimuli, and is a useful tool to study the brain’s response to painful events 31. The A11 nucleus, located in the posterior hy- pothalamus, provides the only known source of descending dopaminergic innervation for the spinal gray matter. Extracellular recordings were made in the trigeminocervical complex (TCC) in response to electrical stimulation of the dura mater. Receptive fields were characterized by mechanical noxious and innocuous stimulation of the ipsilateral ophthalmic dermatome. Stimulation of the A11 significantly inhibited peri-middle meningeal artery dural and a noxious pinch evoked firing of neurons in the TCC. This inhibition was reversed by the D(2) receptor antagonist eticlopride. Lesioning of the A11 significantly facilitated dural and noxious pinch and innocuous brush-evoked firing from the TCC. No dopamine receptors were present in the A11 nucleus itself. However, the A11 does contain dopamine and calcitonin gene-related peptide (CGRP) 35. Conclusions The trigemino-cardiac reflex during Onyx embolization for dural arteriovenous fistula may be caused by mechanical or chemical stimulus to the terminals of the unencapsulated Ruffini-like receptors stemming from A-axons in the dural connective tissue at sites of dural arteries and sinuses. Slow A (ADž) and fast A (Aǃ) neurons may play a role in the stimulus afferent pathway due to their higher mechanosensitivity and chemosensitivity. These afferent pathway nerves are the cholinergic innervation of the dura mater, which also contains vasoactive neuropeptides such as calcitonin gene-related peptide, substance P, and neurokinin A. Stimulation of meningeal sensory Àbers can evoke cerebral vasodilation through the peripheral release of neuropeptides, which play a role in headache pathogenesis. These myelinated A-Àbers terminate in the deep part (laminae III-V) of the spinal dorsal horn whose efferent pathway has been defined as the acetylcholinergic vagus nerve. The A11 nucleus, located in the posterior hypothalamus, providing the only known source of descending dopaminergic innervation for the spinal gray matter can inhibit the neurons in the spinal dorsal horn. 297 Innervation of the Cerebral Dura Mater Xianli Lv References 1 Lv X, Li Y, Lv M, et al. Trigemino-cardiac reflex in embolization of intracranial dural arteriovenous fistula. Am J Neuroradiol. 2007; 28 (9): 1769-1770. doi: 10.3174/ ajnr.A0675. 2 Lv X, Li Y, Jiang C, et al. The incidence of trigeminocardiac reflex in endovascular treatment of dural arteriovenous fistula with Onyx. Interv Neuroradiol. 2010; 16 (1): 59-63. 3 Arnold F. Dissert, departe cephalica nervi sympathici in homine. Heidelberg; 1826. 4 Andres KH, von Düring M. Interferenzphänomene am osmierten Präparat für die systematische elektronenmikroskopische Untersuchung. Mikroskopie. 1974; 3: 139-149. 5 Kemp WJ 3rd, Tubbs RS, Cohen-Gadol AA. The innervation of the cranial dura mater: neurosurgical case correlates and a review of the literature. World Neurosurg. 2012; 78 (5): 505-510. doi: 10.1016/j.wneu.2011.10.045. 6 Strassman AM, Weissner W, Williams M, et al. Axon diameters and intradural trajectories of the dural innervation in the rat. J Comp Neurol. 2004; 473 (3): 364376. doi: 10.1002/cne.20106. 7 Messlinger K, Dostrovsky JO, Strassman AM. Anatomy and physiology of head pain. In: Olesen J, Goadsby PJ, Ramadan NM, et al, editors. The headaches. Philadelphia, USA: Lippincott Williams & Wilkins; 2006. p. 95-96. 8 Adeeb N, Mortazavi MM, Tubbs RS, et al. The cranial dura mater: a review of its history, embryology, and anatomy. Childs Nerv Syst. 2012; 28 (6): 827-837. doi: 10.1007/s00381-012-1744-6. 9 Amenta F, Sancesario G, Ferrante F, et al. Acetylcholinesterase-containing nerve fibers in the dura mater of guinea pig, mouse, and rat. J Neural Transm. 1980; 47 (3): 237-242. doi: 10.1007/BF01250604. 10 Kimmel DL. Innervation of spinal dura mater and dura mater of the posterior cranial fossa. Neurology. 1961; 11: 800-809. doi: 10.1212/WNL.11.9.800. 11 Cavallotti D, Artico M, De Santis S, et al. Catecholaminergic innervation of the human dura mater involved in headache. Headache. 1998; 38 (5): 352-355. doi: 10.1046/j.1526-4610.1998.3805352.x. 12 Keller JT, Saunders MC, Beduk A, et al. Innervation of the posterior fossa dura of the cat. Brain Res Bull. 1985; 14 (1): 97-102. doi: 10.1016/0361-9230(85)90181-9. 13 Furness JB, Papka RE, Della NG, et al. Substance Plike immunoreactivity in nerves associated with the vascular system of guinea-pigs. Neuroscience. 1982; 7: 447-459. doi: 10.1016/0306-4522(82)90279-2. 14 Matsuyama T, Shiosaka S, Wanada A, et al. Fine structure of peptidergic and catecholaminergic nerve fibers in the anterior cerebral artery and their interrelationship: an immunoelectron microscopic study. J Comp Neurol. 1985; 235 (2): 268-276. doi: 10.1002/cne.902350209. 15 Dux M, Sántha P, Jancsó G. Capsaicin-sensitive neurogenic sensory vasodilatation in the dura mater of the rat. J Physiol. 2003; 552 (Pt 3): 859-867. doi: 10.1113/ jphysiol.2003.050633. 16 Strassman AM, Levy D. Response properties of dural nociceptors in relation to headache. J Neurophysiol. 2006; 95 (3): 1298-1306. doi: 10.1152/jn.01293.2005. 17 Keller JT, Marfurt CF. Peptidergic and serotoninergic innervation of the rat dura mater. J Comp Neurol. 1991; 309 (4): 515-534. doi: 10.1002/cne.903090408. 18 Keller JT, Marfurt CF, Dimlich RVW, et al. Sympathetic innervation of the supratentorial dura mater of the rat. J Comp Neurol. 1989; 290 (2): 310-321. doi: 10.1002/cne.902900210. 19 Motavkin PA, Chertok VM, Shul’ga SD. Effect of acetylcholine on mast cells of the dura mater. Bull Exp Biol Med. 1979; 87: 518-521. doi: 10.1007/BF00806702. 20 Rozniecki JJ, Dimitriadou V, Lambracht-Hall M, et al. Morphological and functional demonstration of rat dura mater mast cell-neuron interactions in vitro and 298 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 in vivo. Brain Res. 1999; 849 (1-2): 1-15. doi: 10.1016/ S0006-8993(99)01855-7. Ebersberger A, Takac H, Richter F, et al. Effect of sympathetic and parasympathetic mediators on the release of calcitonin gene-related peptide and prostaglandin E from rat dura mater, in vitro. Cephalalgia. 2006; 26 (3): 282-289. doi: 10.1111/j.1468-2982.2005.01035.x. Goadsby PJ, Knight YE, Hoskin KL, et al. Stimulation of an intracranial trigeminally-innervated structure selectively increases cerebral blood Áow. Brain Res. 1997; 751 (2): 247-252. doi: 10.1016/S0006-8993(96)01344-3. Bari AA, Pouratian N. Brain imaging correlates of peripheral nerve stimulation. Surg Neurol Int. 2012; 3 (Suppl 4): S260-S268. doi: 10.4103/2152-7806.103016. Liu Y, Broman J, Edvinsson L. Central projections of sensory innervation of the rat superior sagittal sinus. Neuroscience. 2004;129 (2): 431-437. doi: 10.1016/j.neuroscience.2004.07.045. Levy D, Strassman AM. Mechanical response properties of A and C primary afferent neurons innervating the rat intracranial dura. J Neurophysiol. 2002; 88 (6): 30213031. doi: 10.1152/jn.00029.2002. Levy D, Strassman AM. Distinct sensitizing effects of the cAMP/PKA second messenger cascade on rat dural mechanonociceptors. J Physiol. 2002; 538 (Pt 2). 2: 483-493. PenÀeld W, McNaughton F. Arch Neur Psych. 1940; 44 (1): 43-75. doi: 10.1001/archneurpsyc.1940. 022800700 51003. Bartsch T, Goadsby PJ. Increased responses in trigeminocervical nociceptive neurons to cervical input after stimulation of the dura mater. Brain. 2003; 126 (Pt 8): 1801-1813. doi: 10.1093/brain/awg190. Davis KD, Dostrovsky JO. Responses of feline trigeminal spinal tract nucleus neurons to stimulation of the middle meningeal artery and sagittal sinus. J Neurophysiol. 1988; 59 (2): 648-666. Burstein R, Yamamura H, Malick A, et al. Chemical stimulation of the intracranial dura induces enhanced responses to facial stimulation in brain stem trigeminal neurons. J Neurophysiol. 1998; 79 (2): 964-982. Hoskin KL, Zagami AS, Goadsby PJ. Stimulation of the middle meningeal artery leads to Fos expression in the trigeminocervical nucleus: a comparative study of monkey and cat. J Anat. 1999;194 (Pt 4): 579-588. doi: 10.1046/j.1469-7580.1999.19440579.x. Andres KH, von Düring M, Muszynski K, et al. Nerve fibres and their terminals of the dura mater encephali of the rat. Anat Embryol (Berl). 1987; 175 (3): 289-301. doi: 10.1007/BF00309843. Ebersberger A, Schaible H-G, Averbeck B, et al. Is there a correlation between spreading depression, neurogenic inflammation, and nociception that might cause migraine headache? Ann Neurol. 2001; 49 (1): 7-13. Macfarlane R. New concepts of vascular headache. Ann R Coll Surg Engl. 1993; 75 (4): 225-228. Charbit AR, Akerman S, Holland PR, et al. Neurons of the dopaminergic/calcitonin gene-related peptide A11 cell group modulate neuronal firing in the trigeminocervical complex: an electrophysiological and immunohistochemical study. J Neurosci. 2009; 29 (40): 12532-12541. doi: 10.1523/JNEUROSCI.2887-09.2009. Youxiang Li, MD Beijing Neurosurgical Institute and Beijing Tiantan Hospital Capital Medical University Tiantan, Xili, 6 100050, Beijing, China. E-mail: lvxianli000@163. com The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 www.centauro.it Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT WIESLAW L. NOWINSKI1, RYSZARD S. GOMOLKA1, GUOYU QIAN1, VARSHA GUPTA1, NATALIE L. ULLMAN2, DANIEL F. HANLEY2 1 2 Biomedical Imaging Lab, Agency for Science Technology and Research; Singapore Division of Brain Injury Outcomes, Johns Hopkins Medical Institutions; Baltimore, MD, USA Key words: stroke, hematoma, IVH, ICH, CLEAR, MISTIE, NCCT SUMMARY – Characterization of hematomas is essential in scan reading, manual delineation, and designing automatic segmentation algorithms. Our purpose is to characterize the distribution of intraventricular (IVH) and intracerebral hematomas (ICH) in NCCT scans, study their relationship to gray matter (GM), and to introduce a new tool for quantitative hematoma delineation. We used 289 serial retrospective scans of 51 patients. Hematomas were manually delineated in a two-stage process. Hematoma contours generated in the first stage were quantified and enhanced in the second stage. Delineation was based on new quantitative rules and hematoma profiling, and assisted by a dedicated tool superimposing quantitative information on scans with 3D hematoma display. The tool provides: density maps (40-85HU), contrast maps (8/15HU), mean horizontal/ vertical contrasts for hematoma contours, and hematoma contours below a specified mean contrast (8HU). White matter (WM) and GM were segmented automatically. IVH/ICH on serial NCCT is characterized by 59.0HU mean, 60.0HU median, 11.6HU standard deviation, 23.9HU mean contrast, –0.99HU/day slope, and –0.24 skewness (changing over time from negative to positive). Its 0.1st-99.9th percentile range corresponds to 25-88HU range. WM and GM are highly correlated (R2=0.88; p<10–10) whereas the GM-GS correlation is weak (R2=0.14; p<10–10). The intersection point of mean GM-hematoma density distributions is at 55.6±5.8HU with the corresponding GM/hematoma percentiles of 88th/40th. Objective characterization of IVH/ICH and stating the rules quantitatively will aid raters to delineate hematomas more robustly and facilitate designing algorithms for automatic hematoma segmentation. Our two-stage process is general and potentially applicable to delineate other pathologies on various modalities more robustly and quantitatively. Introduction Characterization of the density distribution of hematomas is essential in scan reading, manual delineation, and designing automatic segmentation algorithms. Intraventricular hemorrhage (IVH) or bleeding into the ventricular system often results from severe intracerebral hemorrhage (ICH); this complication has a mortality rate of 60-80% and only about 10% of patients recover with a good outcome 1. The Phase II trials promise better outcomes associated with catheter-based drug delivery and pharmacologic clot reduction. Objective, quan- titative volume measurement has been essential to characterize drug action and treatment goals 2. CLEAR III (Clot Lysis Evaluating Accelerated Resolution of Intraventricular Hemorrhage) and MISTIE III (Minimally Invasive Surgery plus rt-PA for Intracerebral Hemorrhage Evacuation) are Phase-III randomized, multi-center clinical trials testing the use of a recombinant tissue plasminogen activator (rt-PA) delivered by a small catheter for hemorrhagic stroke treatment 1,3. This treatment requires multiple non-contrast computed tomography (NCCT) scans to be acquired to monitor clot lysis. For each scan, a hematoma (IVH 299 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Wieslaw L. Nowinski A B Figure 1 Distributions of WM, GM and hematoma. A) Means as a function of examination number. B) Histograms (WM and GM mean distributions are approximated by Gaussians). 300 www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 jointly with ICH) is delineated, its volume measured, and treatment progression assessed. Standardizing these objective measurements for clinical practice is critical for clinically useful treatment assessment. The CLEAR and MISTIE trials use NCCT for hemorrhage evaluation, as this modality is fast, provides generally high contrast between tissue and blood, and is available in most hospitals and emergency departments. On NCCT brain scans, blood appears hyperdense, soft tissues isodense, and cerebrospinal fluid (CSF) hypodense. Some sources provide absolute ranges in Hounsfield units (HU) of blood, clotted blood and its fractions 4-10. However, definition and delineation of hematomas on NCCT is not straightforward because of their variations in shape, size, location, density (intensity), contrast and texture. Moreover, imaging factors, such as partial volume effect (volume averaging), fuzzy and low contrast borders, noise, beam hardening, motion artifacts, and head tilt can further complicate the delineation of hematomas. The purpose of this work is to characterize the density distribution of intraventricular and intracerebral hematomas in hemorrhagic stroke NCCT scans in terms of mean, median, standard deviation, maximum, full width at half maximum, skewness and kurtosis, both overall and per day. We also study the relationship between the density distributions of gray matter (GM) and hematoma. In addition, we formulate rules defining IVH and ICH on NCCT, propose a two-stage process for quantitative hematoma delineation, and introduce a new tool aiding quantitative hematoma delineation. Materials and Methods Materials Scan selection. This study was IRB-approved and HIPAA compliant. A cohort of 317 NCCT serial anonymized scans (acquired predominantly on a daily basis) of 53 IVH/ICH patients was selected from CLEAR IVH, CLEAR III and MISTIE II clinical trials. The patients were chosen to be representative of hemorrhages within different study cohorts, including IVH only, ICH only, both IVH and ICH, and ICH at originating in different locations. From this cohort, we excluded scans with 1) severe artifacts, such as beam hardening, to avoid distortion of white matter (WM), GM and hematoma density char- acteristics; 2) air within the hematoma to avoid distortion of hematoma characteristics (unless the air-occupying regions could be eliminated from hematoma); 3) cleared clots as their characteristics could not be determined; and 4) hematomas in subarachnoid spaces; 289 scans of 51 patients were included in this study. Patient/scan description. Patients’ mean age and standard deviation (SD) were 56.8±11.5 yrs; 55% of them were male and 45% female. Their scans were acquired between 2004.01.112012.04.14 and divided into eight groups for day 1, day 2, …, and day 8 and later. The scans were performed on GE, Philips, Siemens and Toshiba CT devices, and the X-Ray tube potential ranged between 120 and 140kV. The scan number per day was the following: one, 74(25.6%); two, 39(13.5%); three, 32(11.1%); four, 32(11.1%); five, 27(9.3%); six, 21(7.3%); seven, 14(4.8%); and eight+, 50(17.3%). The slice thickness varied between 2.10 and 7.48 mm with mean±SD 4.55±0.79 mm and the matrix was 512×512. Methods Hematoma delineation. IVH/ICH hematomas for all scans were delineated retrospectively (by WLN) in two stages, qualitative and quantitative (see also Appendix B for illustration). In the first qualitative stage, the hematoma region was considered a hyperdense area with HU approximately between 40 and 85. Isodense regions following the borders of the ventricular system (particularly in the posterior horns of the lateral ventricles with clear gravitational fluid leveling) were included in the delineated hematoma region. The exclusion areas corresponded to the catheter, bone, calcifications, and dura matter. To delineate hematomas on acquisition (axial) images interactively, we used the contour editor employed earlier to delineate ischemic infarcts 11. For the hematomas delineated in the first stage, regions outside the 0-100HU range were excluded as 1st/99th percentiles of the aggregated hematomas were 30/82HU (also confirmed by the literature, Section 4). We considered contrast between the hematoma and the surrounding structures by providing the mean contrast for each hematoma contour, and contrast maps for all scans. Contrast was calculated along horizontal and vertical lines of each slice. For a given pixel, its horizontal (vertical) contrast was computed as an absolute difference of the mean of HU of the consecutive n pixels to the left (top) and the mean of HU of 301 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Wieslaw L. Nowinski A 302 www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 B Figure 2 Hematoma distributions. A) Overall (top) and per day (average per case). B) Means with error bars as a function of examination day along with linear fit (day 2 and 3 increases are insignificant). the consecutive n pixels to the right (bottom), see an illustration in Appendix B, Figure 10. The mean contrast of the hematoma contour was calculated as the average of all pixels belonging to this contour. The overall (across all scans) hematoma mean contrast was calculated in two ways, as the average of hematoma contour mean contrasts or average of boundary pixel contrasts. The averages of hematoma contour mean contrasts were 15.4/12.7HU for the horizontal/vertical contrasts. The averages of hematoma boundary pixel contrasts were 16.1HU/12.5HU/16.6HU for the horizontal/vertical/higher of either contrast. Moreover, the contrast map of 15HU corresponded well visually to the hematoma boundaries. Therefore for further analysis, we took 15HU or higher as the hematoma-surrounding structures contrast. In the second quantitative stage, the he303 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Wieslaw L. Nowinski Figure 3 Distribution of intersection points of GM and hematoma histograms. matoma regions were enhanced quantitatively by considering their HU density, hematoma contrast and three-dimensional (3D) relationships. The contour editor was extended to calculate: 1) density maps for a specified range (taken here as 40-85HU) indicated by differently colored pixels within and outside this range (see Appendix B, Figure 8); 2) contrast maps for a given contrast (taken here as 15HU or higher), 3) mean horizontal and vertical contrasts for all hematoma contours; and 4) to determine all hematoma contours below a specified mean contrast (taken here as 8HU). Moreover, the contour editor reconstructs the hematoma in 3D and provides 3D-2D spatial synchronization for visual comparison (see Appendix B, Figure 9). Using this editor and extending the set of rules defining hematoma (see Appendix A), the hematoma regions delineated in the first stage were re-edited as follows. The areas of ”0HU and •100HU were eliminated from analyzed 304 regions. The density map for each scan was superimposed on it, and under-segmented and/or over-segmented hematoma contour boundaries were fit accordingly to this map (see Appendix B, Figure 8F). As the ranges of WM and GM were quite variable across all cases, there were situations (particularly for low means of WM and GM) in which clearly discernible hematomas were below 40HU. Then, the contrast map was applied to fit the re-edited hematoma boundary to the contrast of 15HU or higher (see Appendix B, Figures 7 and 8E). In three situations (regions), this contrast was still too high. First, isodense areas, that are within the ventricles (particularly when they form clear gravitational fluid leveling) or are located in the aqueduct, indicate hematoma. These areas usually have low signal and contrast, and fuzzy borders, particularly in the posterior horns of the lateral ventricles. For such regions, the density map was not applied and the contrast www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 Figure 4 WM to GM scatter plot. was reduced to 8HU. A processed hematoma contour was re-edited to fit its boundary to this contrast map and, if its mean contrast was 8HU or higher, this contour was retained; otherwise it was deleted (considered a partial volume effect region). Second, some isodense or slightly hyperdense areas within the parenchyma are questionable, as they could result from: 1) dorsal and/or ventral extension of hematoma, 2) partial volume effect, or 3) a mixture of both. Similar processing was applied in this case. Third, some hyperdense ICH close to the skull have unclear internal boundaries. In these cases, the density map typically indicates a false connection to the skull, the contrast map of 15HU does not show a boundary, whereas the contrast of 8HU does. The contrast width n was generally taken as 5. When the mean hematoma contrast of 8HU could not be obtained for n=5, this value was varied. For fuzzy hematomas, their contrast width was extended to 7. For very small (e.g., aqueductal) clots and narrow elongated clots, the contrast width was reduced to 3 to avoid inclusion of non-hematoma pixels in the calculation of hematoma-dependent contrast. Proximity of a catheter, calcification and/or air may distort the actual contrast value. Therefore, any hematoma boundary pixel, having in its contrast width pixel(s) in the range of ”0HU or •100HU was excluded from contrast calculation. A 3D surface modeled hematoma was displayed to evaluate its connectedness, shape inconsistency and/or artifacts. Every 3D component of the hematoma was mapped to the corresponding 2D contour(s). Neighboring 3D components, if required, were connected by extending their corresponding contours (particularly in the aqueduct, see Figure 9 in Appendix B). In shape inconsistency, the contours in neighboring slices were re-evaluated. In case of gaps between the neighboring and overlapping 305 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Wieslaw L. Nowinski Figure 5 GM to hematoma scatter plot. contours in the dorso-ventral direction (easily identifiable in 3D), the missing contours were created such as to meet the hematoma criteria. Characterization of hematoma, WM and GM. Overall and per day hematoma mean, median, SD, maximum, Full Width at Half Maximum (FWHM), skewness, kurtosis, and percentiles were calculated. In addition, overall and per day WM and GM mean, median, and SD were computed automatically for each scan applying the algorithm 13. We also studied: 1) correlation between WM and GM distributions; 2) correlation between GM and hematoma distributions; and 3) hematoma contrast distribution. Statistical analysis The MATLAB®/7.8.0.347 software tool 13 was used for statistical analysis. Means, medians, SDs, skewness, kurtosis and percentiles were calculated to characterize hematoma distribution. WM, GM and hematoma means were 306 tested by one-way ANOVA test to detect significant differences. Least Significance Difference test was used to confirm the significance between means of WM-GM, WM-hematoma and GM-hematoma, and differences between hematoma means at respective days. The differences were considered significant if p<0.05 after Bonferroni’s correction. Pearson’s linear and Spearman’s range correlation coefficients were calculated between means of WM-GM and GM-hematoma. Results The IVH/ICH definition criteria were formulated (summarized in Appendix A) and 8,205 hematoma contours set on all 289 scans. The contour editor has been extended to provide the density and contrast maps, see Appendix B (Figures 7-10). Tables 1-3 present the characteristics of WM, GM and hematoma (over- www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 A B C Figure 6 Horizontal, vertical and higher of both hematoma contrast distribution plots with the contrast width of: A) 3 pixels; B) 5 pixels (default); and C) 7 pixels. 307 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Wieslaw L. Nowinski Table 1 Means (A), medians (B) and standard deviations (C) of WM, GM and hematoma (overall and per day) in HU. Parameter Overall Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8+ A Hematoma 59.0 59.9 60.4 60.7 58.5 57.1 56.7 55.7 53.3 Gray matter 37.3 37.8 38.0 37.5 36.9 36.8 37.0 36.4 36.5 White matter 28.3 28.7 28.7 28.7 28.2 28.0 27.9 27.5 27.6 60 61.0 62.0 62.0 59.0 57.0 57.0 56.0 53.0 Gray matter 36.8 36.7 37.7 36.8 36.2 36.9 36.8 36.8 36.3 White matter 27.9 28.4 28.3 28.3 28.1 28.1 27.3 27.9 27.3 11.6 11.6 11.3 11.5 11.6 11.4 10.9 11.2 11.2 Gray matter 6.1 6.1 6.4 6.1 5.9 6.1 6.3 6.2 6.1 White matter 5.2 5.2 5.3 5.2 5.1 5.2 5.1 5.0 5.1 B Hematoma C Hematoma Table 2 Hematoma 0th (minimal), 0.01st, 0.1st, 0.5th, 1st, 5th, 25th, 50th, 75th, 95th, 99th, 99.5th, 99.9th, 99.99th and 100th (maximal) percentiles (overall and per day) in HU. Percentile Overall Day 1 Day 2 Day 3 Day 5 Day 6 Day 7 Day 8+ 0 5 5 6 9 8 8 8 12 11 0.01 18 18 19 21 17 19 18 20 18 0.1 25 25 26 27 24 25 24 25 24 0.5 30 31 31 32 30 30 29 30 28 1 33 33 34 35 32 32 32 32 30 5 39 40 40 41 39 39 38 38 36 25 50 51 52 52 50 49 49 47 45 50 60 61 62 62 59 57 57 56 53 75 68 69 69 69 67 66 65 65 62 95 76 77 77 78 76 75 74 73 72 99 82 82 82 83 82 80 79 78 77 99.5 84 84 83 85 84 82 81 79 79 99.9 88 88 87 89 89 87 84 82 83 99.99 93 93 92 95 94 92 90 86 88 100 100 100 100 100 100 100 99 90 100 all and per day). Means, medians and SDs of WM, GM and hematoma are in Table 1. Table 2 provides hematoma percentiles and Table 3 hematoma maximum, FWHM, skewness and kurtosis. Figure 1 shows HU distributions of WM, GM and hematoma. Figure 2 presents overall and per day hematoma distributions. Figure 3 illustrates the relationships between GM and hematoma distributions, while Table 4 lists HU values with GM and hematoma percentiles in the neighborhood of the GM-hematoma distribution intersection. Figures 4 and 5 present 308 Day 4 scatter plots of WM-GM and GM-hematoma. Figure 6 shows hematoma contrast distribution plots. ANOVA test revealed significant differences between the WM, GM and hematoma means [F=3574.48, p<0.0001]. Least significant difference test showed the means of WM-GM, WM-hematoma and GM-hematoma are significantly different (p<10–10). There were no significant differences between hematoma means at days 1-4 (p>0.05). The means±SDs of horizontal/vertical/higher of both contrasts were: 17.0 ± 10.7/13.3 ± 9.7/20.8 ± 9.9, 19.7 ± 11.6/15.9 ± www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 Table 3 Hematoma characterization: maximum in HU (max), full width at half maximum in HU (FWHM), skewness and kurtosis. Overall Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8+ Max 67.2 67.2 68.2 66.9 65.1 64.5 59.1 66.1 46.9 FWHM 31.3 30.3 24.2 29.0 30.9 32.9 30.2 33.4 32.0 Skewness –0.24 –0.28 –0.41 –0.29 –0.20 –0.09 –0.18 –0.11 0.08 Kurtosis 2.44 2.50 2.58 2.46 2.52 2.40 2.56 2.25 2.36 Table 4 HU values with corresponding GM and hematoma percentiles in the neighborhood of intersection of the overall GM and hematoma distributions. Crossing point Density [HU] Percentiles 46 48 50 52 54 55.55 58 60 61 Hematoma 16 20 25 30 35 40 45 50 54 Gray matter 68 72 76 80 84 88 93 97 99 10.9/23.9 ± 10.6, and 20.9 ± 12.1/17.4 ± 11.5/25.5 ± 10.8 for the contrast width 3, 5 and 7. Discussion Hematoma, WM and GM characterization. This study advances our knowledge in quantitative characterization of hematoma on NCCT, and its relationship to WM and GM. IVH/ICH hematoma on serial NCCT, whose delineation was assisted by applying the density and contrast maps and 3D-2D spatial correlation, is characterized by 59.0HU mean, 60.0HU median, 11.6HU standard deviation, 23.9HU mean contrast and –0.99HU/day slope (i.e., the density decrement per day). The overall hematoma distribution gains its maximum at 67.2HU with 31.3HU FWHM, 2.44 kurtosis and –0.24 skewness (which changes in time from negative to positive, Figure 2a). The hematoma 0.1st-99.9th percentile range corresponds to the 25-88HU range. Low density hematoma areas resulted mostly from CSF embedded into hematoma and the hematoma-CSF partial volume effect, whereas areas with ”0HU from air penetrating the brain due to an inserted catheter. Areas with •85HU resulted mainly from the partial volume effect in the regions when hematoma neighbored a catheter, bone, or calcifications. Other studies report flowing blood of 51HU 8; blood 52HU 14, 56HU 5, 50-60HU 4; venous whole blood 55±5HU 10; acute blood 56-76HU 9, 50100HU 15; fresh blood clot 70-74HU 5; clotted blood 76HU 14, 60-80HU (70HU mean) 5, 70- 90HU 4, 80±10HU 10, 81HU 8; white clot 40HU 5; retracted clot 84-86HU 5; and hemorrhage 5070HU 6,7, 40-60HU immediate and 60-80HU over one hour (or described as an acute) 6,16, 5080HU 9, 60-100HU 17, 40-80HU 18, 80-100HU after the core of the hemorrhage becomes denser 6. Note that 50/90HU correspond to hematoma 25th/99.95th percentiles. The effect of decreased hematoma density with hematoma aging is known 6,19. Here it was measured and quantified. The overall means±SDs of WM and GM are 28.3±5.2HU and 37.3±6.1HU, and their ratio is 0.76±0.02. These results (i.e., calculated automatically by applying the algorithm 12 in the presence of hematoma) are consistent with the results reported in the literature. Other studies report normal WM/GM means±SDs as 29/35HU 20, 31.8 ± 2.3/38.7 ± 2.2HU 21, and 29.8±3.3/33.2±2.6HU for males and 30.1±3.5/33.0±3.3HU for females 22; and WM/GM ratio for normal brain of 0.76±0.14 23. WM and GM are highly correlated (R2=0.88) whereas the GM-hematoma correlation is weak (R2=0.14) (both are significant, p<10–10). The hematoma distribution substantially overlaps with that of GM (Figure 3, Table 4). The mean GM-hematoma intersection point is at 55.6±5.8HU with the corresponding GM/ hematoma percentiles of 88th/40th. This clearly indicates that manual hematoma segmentation is difficult and automatic density-based segmentation impossible. Implications for patient care. The hematoma boundary is set here based on new rules, hematoma profiling, and employing a dedicated 309 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT tool developed in-house. Stating these (inclusion, exclusion and enhancement) rules explicitly and quantitatively will aid different raters to delineate hematomas more robustly and facilitate designing algorithms for automatic hematoma segmentation. This is particularly important taking into account a substantial overlap in density between hematoma and GM. The hematoma delineation process is made quantitative by providing the human observer with the density and contrast maps between the hematoma and surrounding structures superposed on a scan along with synchronization of a 3D hematoma model with a scan. We have proposed a two-stage process for hematoma delineation. Having the first stage hematoma contours allows them in the second stage to be quantified, displayed in 3D, and superimposed on the density and contrast maps for further enhancement. The more accurate segmentation of hematoma results in its both more accurate volume and shape. It is well known that hematoma volume is essential for prognosis 24. To estimate hematoma volume, the so-called ABC/2 method is often employed (also in the CLEAR and MISTIE clinical trials) based on the measurements of the three maximal diameters of the hematoma 25 (i.e., without marking its borders). This method, though rapid, leads to substantial error 26,27. The accurately delineated series of hematomas result in more accurate volumes, facilitate making quantitative decisions (e.g., stopping the treatment when the hematoma volume is reduced by 80%), and enable checking for potential rebleeding by comparing the shapes and sizes of consecutive hematomas in 3D. Though traditionally CT was considered the modality of choice for hemorrhagic stroke diagnosis, MRI has been demonstrated to be as accurate as CT 28 and this work can be extended to support delineation of hematomas on MRI. Moreover, although this work focuses on IVH/ ICH on NCCT, the computer-assisted approach proposed here to delineate hematomas is general and potentially applicable to delineate other pathologies on various modalities more robustly and quantitatively. Parameter setting. The initial 40-85HU range overlaps with the ranges reported in the literature and corresponds approximately to the 5th99.7th percentile range, so it covers the majority of hematoma. The densities outside this range are handled by the enhancement criteria (see Appendix A). 310 Wieslaw L. Nowinski The 15HU contrast resulted from the quantitative contrast analysis between the hematoma and surrounding structures, and the contrast map of 15HU corresponded well to the hematoma boundaries. As the ventricular system has a larger antero-posterior than lateral extent, the mean horizontal contrast is higher, whereas there are more contour points with low vertical contrast (Figure 6). Therefore, the higher of these two contrasts better characterizes the hematoma distribution. Note that all three contrast curves intersect at 15HU (Figure 6). The 8HU contrast was observed in true hematomas in posterior horns with clear leveling and a clotted aqueduct. We applied the same contrast threshold to separate a hematoma from partial volume effect regions. Note that the 8HU difference is also recommended in European Guidelines on Quality Criteria for Computed Tomography to test image uniformity and measure CT numbers 29. Limitations. The scans were acquired using multiple sites’ CT scanners without an effort to standardize the HU across sites. Only routine clinical calibration procedures were employed by each site. Gender-, age-, and ethnicity-specific analyses were not performed in this study. The thresholds taken here (40/15/8HU) are absolute, although the hematoma characteristics vary across acquisitions and time. Potentially more accurate hematoma delineation could be achieved by expressing these thresholds as functions of WM or GM (Figures 4 and 5). It is questionable how to include in or exclude from the delineated hematoma the regions having a mixture of pixels within and outside the specified density range. To handle this situation quantitatively requires the tool to be extended, e.g., to calculate the inside/outside ratio and include a considered region if this ratio is •0.5. Future work. The results of this work are being used to enhance automatic segmentation of hematomas in NCCT in our previous algorithms. We are also in the process of analyzing the contrast of ischemic infarcts on NCCT from the study 11. Further studies are planned to correlate the current results with hematoma volume (measured by the clinicians and automatically), and to analyze IVH and ICH separately. Acknowledgement This research was funded by Biomedical Research Council, ASTAR, Singapore. www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 References 1 Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage Phase III (CLEAR III). 2008. Available at: www.cleariii.com. Accessed March 17, 2014. 2 Naff N, Williams MA, Keyl PM, et al. Low-dose recombinant tissue-type plasminogen activator enhances clot resolution in brain hemorrhage: the intraventricular hemorrhage thrombolysis trial. Stroke. 2011; 42 (11): 3009-3016. doi: 10.1161/STROKEAHA.110.610949. 3 Ziai WC, Tuhrim S, Lane K, et al. A multicenter, randomized, double-blinded, placebo-controlled phase III study of Clot Lysis Evaluation of Accelerated Resolution of Intraventricular Hemorrhage (CLEAR III). Int J Stroke. 2013. doi: 10.1111/ijs.12097. 4 Hofer M. CT teaching manual: a systematic approach to CT reading. Stuttgart: Thieme; 2007. 5 New PFJ, Aronow S. Attenuation measurements of whole blood and blood fractions in computed tomography. Radiology. 1976; 121 (3 Pt 1): 635-640. 6 Osborn AG. Diagnostic imaging. Brain. 2nd ed. Salt Lake City, UT: Amirsys; 2010. 7 Grumme TH. Cerebral and spinal computed tomography. 3rd ed. Berlin: Blackwell Wissenschafts-Verlag; 1998. 8 Gibby WA. X-Ray computed tomography. In: Zimmerman RA, Gibby WA, Carmody RF, eds. Neuroimaging: clinical and physical principles. New York, NY: Springer; 2000. doi: 10.1007/978-1-4612-1152-5_1. 9 Grossman RI, Yousem DM. Neuroradiology: the requisites. 2nd ed. St. Louis, Mo: Mosby; 2003. 10 Wegener OH. Whole body computerized tomography. 2nd ed. Boston, MA: Blackwell Scientific Publications; 1993. 11 Nowinski WL, Gupta V, Qian GY, et al. Automatic detection, localization and volume estimation of ischemic infarcts in non-contrast CT scans: method and preliminary results. Invest Radiol. 2013; 48 (9): 661-670. doi: 10.1097/RLI.0b013e31828d8403. 12 Gupta V, Ambrosius W, Qian G, et al. Automatic segmentation of cerebrospinal fluid, white and gray matter in unenhanced computed tomography images. Acad Radiol. 2010; 17 (11): 1350-1358. doi: 10.1016/j.acra.2010. 06.005. 13 MATLAB, The MathWorks Inc. Version 7.8.0.347 (R2009a). Natick, MA, USA; 2009. Available at: www. mathworks.com. Accessed March 17, 2014. 14 Norman D, Price D, Boyd D, et al. Quantitative aspects of computed tomography of the blood and cerebrospinal fluid. Radiology. 1977; 123 (2): 335-338. 15 Adams JG. Emergency medicine. Philadelphia, PA: Elsevier; 2008. 16 Osborn AG. Osborn’s brain: imaging, pathology, and anatomy. 1st ed. Salt Lake City, UT, USA: Amirsys Pub; 2013. 17 Lev MH, Gonzalez RG. Angiography and CT perfusion imaging. In: Toga AW, Mazziotta CJ, eds. Brain mapping: the methods. 2nd ed. Amsterdam: Academic Press; 2002. doi: 10.1016/B978-012693019-1/50019-8. 18 Creasy JL. Dating Neurological Injury: A forensic guide for radiologists, other expert medical witnesses, and attorneys. New York, NY, USA: Springer; 2011. doi: 10.1007/978-1-60761-250-6. 19 Messina AV, Chernik NL. Computed tomography: the “resolving” intracerebral hemorrhage. Radiology. 1976; 118 (3): 609-613. 20 Weinstein MA, Duchesneau PM, MacIntyre WJ. White and gray matter of the brain differentiated by computed tomography. Radiology. 1977; 122 (3): 699-702. 21 Fullerton GD, Zagzebski JA, eds. Medical physics of CT and ultrasound: tissue imaging and characterization. Medical Physics Monograph No. 6. New York, NY: American Association of Physicists in Medicine, American Institute of Physics; 1980. 22 Cala LA, Thickbroom GW, Black JL, et al. Brain density and cerebrospinal fluid space size: CT of normal volunteers. Am J Neuroradiol. 1981; 2 (1): 41-47. 23 Choi SP, Park HK, Park KN, et al. The density ratio 24 25 26 27 28 29 of grey to white matter on computed tomography as an early predictor of vegetative state or death after cardiac arrest. Emerg Med J. 2008; 25 (10): 666-669. doi: 10.1136/emj.2007.053306. Hemphill JC 3rd, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke. 2001; 32 (4): 891-897. doi: 10.1161/01.STR.32.4.891. Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996; 27 (8): 1304-1305. doi: 10.1161/01.STR.27.8.1304. Sheth KN, Cushing TA, Wendell L, et al. Comparison of hematoma shape and volume estimates in warfarin versus non-warfarin-related intracerebral hemorrhage. Neurocrit Care. 2010; 12 (1): 30-34. doi: 10.1007/s12028009-9296-7. Kosior JC, Idris S, Dowlatshahi D, et al. Quantomo: validation of a computer-assisted methodology for the volumetric analysis of intracerebral haemorrhage. Int J Stroke. 2011; 6 (4): 302-305. doi: 10.1111/j.17474949.2010.00579.x. Kidwell CS, Chalela JA, Saver JL, et al. Comparison of MRI and CT for detection of acute intracerebral hemorrhage. JAMA. 2004; 292 (15): 1823-1830. doi: 10.1001/ jama.292.15.1823. Menzel HG, Schibilla H, Teunen D. European guidelines on quality criteria for computed tomography. Publication no EUR 16262 EN. Luxembourg: European Commission; 1999. Wieslaw L. Nowinski, DSc, PhD Director and Principal Scientist Biomedical Imaging Lab Agency for Science Technology and Research Singapore Tel.: (65) 6478-8404 Fax: (65) 6478-9049 E-mail: wieslaw@sbic.a-star.edu.sg 311 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT Appendix A: Definition of IVH and ICH on NCCT The IVH and ICH definition is based on three sets of criteria: inclusion, exclusion, and enhancement. Wieslaw L. Nowinski 4. Hyperdense areas corresponding to calcifications. 5. Hyperdense areas corresponding to the dura matter. 6. Hyperdense areas in the subarachnoid space. 7. Isodense or slightly hyperdense areas with the contrast lower than 8HU. Inclusion Criteria 1. Hyperdense areas with HU value ranging between 40-85HU. Generally, this range varies and is time- and acquisition parameter-dependent. 2. Isodense areas within the ventricles (irrespective of whether they form clear gravitational fluid leveling or not) provided that the mean contrast between the hematoma and its surrounding structures is not lower than 8HU. 3. Isodense or slightly hyperdense areas within the parenchyma dorsal and/or ventral to the hematoma resulting from the partial volume effect with the mean contrast between the hematoma and its surrounding structures not lower than 8 HU. Exclusion Criteria 1. Hyperdense areas corresponding to the catheter. 2. Slightly hyperdense areas above or below the catheter resulting from the partial volume effect. 3. Hyperdense areas corresponding to bone. 312 Enhancement Criteria 1. The hematoma determined by applying the inclusion and exclusion criteria can be extended by including areas lower than 40HU, or reduced by excluding areas equal and higher than 40HU depending on the contrast between it and the surrounding tissues taken here as 15HU or higher. 2. Isodense regions surrounded by CSF, the choroid plexus excluded, should be included as hematoma. 3. Isodense regions resulting from the partial volume effect may be included as hematoma depending on knowledge of anatomy (e.g., that of the aqueduct) and adjacent slices (by studying their 3D relationships). 4. Isodense areas lower than 40HU surrounded by hematoma are to be included as hematoma provided that they are not CSF regions. 5. Catheter track hemorrhages, presented as hyperdensities surrounding the catheter and not resulting from the partial volume effect from the below and above catheter areas, are to be included as hematoma. www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 Figure 7 shows an NCCT scan of a hematoma with fuzzy borders, the density map (between 40-85HU) superimposed onto the scan, and the contrast map (15HU or higher) superimposed onto the scan. Figure 8 illustrates the use of the density and contrast maps in the process of hematoma delineation. Figure 9 illustrates the 3D-2D relationships used to identify undersegmented regions of the hematoma. Figure 10 illustrates the definition of the vertical and horizontal contrasts. A B Figure 7 Illustration of the density and contrast maps. Each map is represented by circles (of the constant size), each circle corresponding to a pixel. A): an NCCT image of a hematoma with fuzzy borders. B): the density map (with each pixel represented by a green circle for a given density range (set here between 40-85HU)) superimposed onto the scan. C): the contrast map (with each pixel represented by a colored circle (blue – horizontal contrast, yellow – vertical contrast, and red – either horizontal or vertical contrast) for a given contrast range (set here as 15HU or higher)) superimposed onto the scan. C Appendix B: Illustration of the Contour Editor with the Density and Contrast Maps 313 Characterization of Intraventricular and Intracerebral Hematomas in Non-Contrast CT A B C Wieslaw L. Nowinski D E F Figure 8 Illustration of the use of the density and contrast maps in the process of hematoma delineation. A) Original scan with the hematoma. B) Hematoma delineated in the first qualitative stage (the contour is set by manipulating the control points (marked in yellow on the contour marked in magenta). C) Density map (in green and red) superimposed on the contoured hematoma; the red pixels neighboring the hematoma border and lying inside the hematoma indicate the over-segmented regions, whereas the green pixels lying outside the hematoma indicate the under-segmented regions. D) Hematoma delineation enhanced by the density map (in the ideal case, the over- and under-segmented regions should be eliminated, which is indicated in the density map by the absence of the red and green pixels (circles) in the hematoma border); note: to increase the accuracy of editing, the image was highly magnified (in comparison to image C) resulting in smaller green circles outside the hematoma. E) Contrast map along with the delineated hematoma. F) Both contours (images C,D) magnified to highlight the differences between them (the red arrows point to the under-segmented regions and the green arrows to the over-segmented regions). 314 www.centauro.it The Neuroradiology Journal 27: 299-315, 2014 - doi: 10.15274/NRJ-2014-10042 Figure 9 Illustration of the 3D-2D relationships used here to identify under-segmented regions of the hematoma. The 3D hematoma (right) shows a missing connection (the aqueduct) between the third and the fourth ventricles, and the cross on the scan (left), whose location correspond to that in 3D, indicates its position (i.e., these two crosses are used for spatial 3D-2D synchronization). Figure 10 Illustration of the definition of the vertical (left) and horizontal (right) contrasts. 315 The Neuroradiology Journal 27: 316-321, 2014 - doi: 10.15274/NRJ-2014-10047 www.centauro.it Giant Arachnoid Granulations Mimicking Pathology A Report of Three Cases BART DE KEYZER, SVEN BAMPS, FRANK VAN CALENBERGH, PHILIPPE DEMAEREL, GUIDO WILMS Department of Radiology, UZ Leuven, Campus Gasthuisberg; Leuven, Belgium Key words: giant arachnoid granulations, CT, MR SUMMARY – We describe three cases of incidentally found lesions in the dural venous sinuses on magnetic resonance imaging, that other had erroneously considered pathological entities. We made the diagnosis of giant arachnoid granulations. The differential diagnosis with thrombosis or intrasinusal tumoral lesions was easily made on the basis of three typical radiological features of the granulations: the hyperintensity of the lesions on FLAIR, a blood vessel within the lesion and bone erosion. Introduction Arachnoid granulations (AG) are extensions of the arachnoid membrane into the dural venous sinuses, serving to drain the cerebrospinal fluid (CSF) from the subarachnoid space into the venous system. In some people, these extensions grow to become “giant” AG 1-4. We describe three new cases of giant AG, two in the transverse sinus and one in the superior sagittal sinus. Case 1 A 50-year old woman underwent MRI of the brain at another institution because of pain and numbness in the left V2 division of the trigeminal nerve. On this MR scan, no lesion was found to explain the trigeminal neuralgia. In the lateral aspect of the left lateral sinus an ovalar lesion was found partially obstructing the lumen of the sinus. The lesion was hyperintense on the axial and coronal T2-weighted sequences (Figure 1A,B) with evidence of a central strongly hypointense tubular structure, probably representing a flow void in a vascular structure. The lesion was also hyperintense on the axial CISS and gradient-echo sequence. On the FLAIR se316 quence (Figure 1C), there was no attenuation and the lesion remained hyperintense. There was no diffusion restriction in the lesion. Intravenous contrast was not administered. Since this lesion was read as a tumour, the patient was referred to the neurosurgery department at our hospital for a second opinion. When studying the MR together with the neurosurgical staff, we suggested the diagnosis of a giant AG and proposed a contrast-enhanced CT for confirmation. CT showed a hypodense (density of +/- 20 Hounsfield units) non-enhancing nodular lesion in the left transverse sinus, in the lateral third, at the level of the transition to the sigmoid sinus (Figure 1D,E). There was scalloping of the adjacent temporal bone (Figure 1E). A small enhancing vein was visible in the superior part of the lesion (Figure 1D,E). These findings confirmed the diagnosis of a giant AG. The patient was reassured she had no tumour and did not require further treatment. Case 2 A 36-year-old man had a known history of right parietal haemorrhages since his youth with several repetitive episodes. During a new Bart De Keyzer recent haemorrhage, on imaging at another institution a lesion was found in the right transverse sinus, interpreted as a thrombosis and patient was put on anticoagulation. The patient was referred to our institution for a second opinion and the images were shown to us. On MRI an ovalar structure was found in the right transverse sinus. This structure was hyperintense on T2-weighted images and the FLAIR sequence and was hypointense on T1. On FLAIR a rounded intralesional area of flow void was seen (Figure 2A). There was no enhancement after contrast (Figure 2B). There was erosion of the inner table of the skull, as confirmed by subsequent CT. Diagnosis of a giant AG was made. The real cause of the haemorrhages is still under investigation. Case 3 A 12-year-old boy complained of three episodes of sudden headache accompanied by blurred vision, diplopia and neck pain. There was no history of previous headaches. Clinical neurological examination was normal. MRI of the brain was performed to rule out a Chiari malformation. The cerebellar tonsils had a normal location. In the superior sagittal sinus a large lesion was present with T2 hyperintensity and with a central linear area of flow void (Figure 3A,B). This structure enhanced strongly whereas the rest of the lesion did not enhance at all (Figure 3C). Again a tumoral lesion was suspected but we made the diagnosis of a giant AG. Since the headache was not continuous and given the absence of papillary oedema or other signs of intracranial hypertension, it was considered that there was no relation between the attacks of headache and the AG. Discussion Arachnoid granulations (AG), first described by Antonio Pacchioni in 1705, are cerebrospinal fluid-filled protrusions extending from the subarachnoid space into the venous sinuses through apertures in the dura 2,4,5. They represent distended arachnoid villi and are macroscopically visible 1. The prevalence of AG increases with age 6. AG normally measure a few millimetres. In Giant Arachnoid Granulations Mimicking Pathology some cases they grow to fill and dilate the dural sinuses even with expansion of the inner table of the skull and are then called “giant AG” 1 . There is no consensus on the definition of “giant” in the literature. Sometimes a large AG is called “giant” when larger than 1 cm 4, but Kan et al. referred to AG as “giant” when they fill the lumen of a dural sinus, causing local dilatation or filling defects 1. Most commonly AG are found in the transverse sinuses, particularly within the middle and lateral portions of the sinus, with a slight left predominance 6. The second most common location is the superior sagittal sinus, but they can be found everywhere in the dural venous sinuses 2,4. There is no difference in sex distribution 6. Mostly, giant AG are discovered as incidental findings without any relation to the patient’s symptoms as in our three cases. In some cases, however, they fill and dilate the dural sinuses, causing symptoms of increased intracranial pressure based on venous hypertension secondary to partial sinus occlusion. They usually present with symptoms of headache 1,7,8. Endovascular treatment with stents might be necessary in case of invalidating symptoms 9. The imaging characteristics of normal and giant AG are well-known and can be considered typical 4,5. On skull radiography, they can be visible as radiolucent zones or they can cause impressions on the inner table of the calvaria 4. The CT density of granulations in the literature varies from hypodense to isodense with the brain parenchyma 4,6. On MR imaging, the arachnoid granulations are iso- to hypointense relative to brain parenchyma on T1weighted images. They all show a hyperintense signal on T2-weighted images 4,6. The fluid in the granulations is mostly not attenuated on a FLAIR sequence, but remains hyperintense, most likely due to pulsation artifacts from the adjacent sinus and differing CSF flow characteristics within the AG 2. This was the case in all three of our patients. In the series reported by Trimble et al. 4, approximately 80% of giant arachnoid granulations contain CSF-incongruent fluid on at least one MR image and nearly half contain fluid that does not parallel CSF on at least two sequences. FLAIR is the most reliable technique, differing in signal intensity with CSF in 100% of cases 4. On angiographic studies such as CT angiography, MR angiography or catheter angiography, AG appear as ovoid filling defects in the dural venous sinuses in the venous phase 2,5. Vascular structures presumed to be veins can 317 Giant Arachnoid Granulations Mimicking Pathology Bart De Keyzer A B C D 318 www.centauro.it E The Neuroradiology Journal 27: 316-321, 2014 - doi: 10.15274/NRJ-2014-10047 F Figure 1 Patient 1. Giant arachnoid granulation in the left transverse sinus. Transverse (A) and coronal (B) T2-weighted images. An ovalar hyperintense structure (arrow) can be seen in the left transverse sinus at the transition to the sigmoid sinus. Intralesional flow void (arrowhead). C) Transverse FLAIR. There is attenuation of the content of this “lesion” (arrow) that remains hyperintense. Transverse (D) and coronal (E) enhanced CT. The “lesion” (arrow) is hypodense with a central enhancing vascular structure (arrowhead). E) Coronal CT with bone windows. Note the scalloping of the overlying bone (arrow). A B Figure 2 Patient 2. Giant arachnoid granulation in the right transverse sinus. A) Transverse FLAIR. Note the hyperintense “lesion” (arrow) in the left transverse sinus at the transition to the sigmoid sinus. Central flow void pointing to a vascular structure (arrowhead). B) Transverse gadolinium-enhanced image. The hypointense ovalar lesion (arrow) in the transverse sinus is confirmed. The vascular structure less well visible due to volume-averaging. 319 Giant Arachnoid Granulations Mimicking Pathology A Bart De Keyzer B C Figure 3 Patient 3. Giant arachnoid granulation in the superior sagittal sinus. Transverse (A) and sagittal (B) T2-weighted image. Huge ovalar hyperintense structure (arrow) within the superior sagittal sinus in the under third. Intralesional vascular flow-void (arrowhead). C) Sagittal gadolinium-enhanced image. Confirmation of the granulation (arrow) and the vascular structure (arrowhead). be seen as focal linear contrast enhancement on enhanced CT or MR or as linear flow voids on non-enhanced MR as seen in two of our patients. It seems that larger granulations are more likely to contain these internal veins 2,4,6. Non-vascular soft tissue has also been reported in giant AG, and was variously interpreted 320 as stromal collagenous tissue, hypertrophic arachnoid mesangial cell proliferation, or invaginating brain tissue 4. AG can be mistaken for pathologic processes in the dural venous sinuses. It is important to distinguish these benign giant AG from more serious dural venous sinus pathology such as thrombosis and www.centauro.it The Neuroradiology Journal 27: 316-321, 2014 - doi: 10.15274/NRJ-2014-10047 neoplasia to avoid unnecessary invasive procedures, because of their infrequency and large size. The differential diagnosis with thrombosis can be made because thrombosis usually involves an entire segment of sinus or multiple sinuses, and can extend into cortical veins, while AG produce focal, well-defined defects . It has to be mentioned that fresh thrombus is hyperdense on CT and hyperintense on T1weighted MR, whereas AG have other imaging characteristics 8. The differential diagnosis with tumour can be made because of the shape, the lack of contrast enhancement and the lack of diffusion restriction 1,3,4. 7,8 References 1 Kan P, Stevens EA, Couldwell WT. Incidental giant arachnoid granulation. Am J Neuroradiol. 2006; 27: 1491-1492. 2 Leach JL, Meyer K, Jones BV, et al. A large arachnoid granulations involving the dorsal superior sagittal sinus: findings on MR imaging and MR venography. Am J Neuroradiol. 2008; 29: 1335-1339. doi: 10.3174/ajnr.A1093. 3 Mamourian AC, Towfighi J. MR of giant arachnoid granulation, a normal variant presenting as a mass within the dural venous sinus. Am J Neuroradiol. 1995; 16 (4 Suppl): 901-904. 4 Trimble CR, Harnsberger HR, Castillo M, et al. “Giant” arachnoid granulations just like CSF? Not. Am J Neuroradiol. 2010; 31: 1724-1728. doi: 10.3174/ajnr.A2157. 5 Roche J, Warner D. Arachnoid granulations in the transverse and sigmoid sinuses: CT, MR, and MR angiographic appearance of a normal anatomic variation. Am J Neuroradiol. 1996; 17 (4): 677-683. 6 Leach JL, Jones BV, Tomsick TA, et al. Normal appearance of arachnoid granulations on contrast-enhanced CT and MR of the brain: differentiation from dural sinus disease. Am J Neuroradiol. 1996; 17: 1523-1532. 7 Chin SC, Chen CY, Lee CC, et al. Giant arachnoid granulation mimicking dural sinus thrombosis in a boy with headache: MRI. Neuroradiology. 1998; 40 (3): 181183. doi: 10.1007/s002340050564. 8 Choi HJ, Cho CW, Kim YS, et al. Giant arachnoid granulation misdiagnosed as transverse sinus thrombosis. J Korean Neurosurg Soc. 2008; 43: 48-50. doi: 10.3340/jkns.2008.43.1.48. 9 Zheng H, Zhou M, Zhao B, et al. Pseudotumor cerebri syndrome and giant arachnoid granulation: treatment with venous sinus stenting. J Vasc Interv Radiol. 2010; 21: 927-929. doi: 10.1016/j.jvir.2010.02.018. Guido Wilms, MD, PhD Department of Radiology UZ Leuven, Campus Gasthuisberg Herestraat 49, 3000 Leuven, Belgium E-mail: guido.wilms@uz.kuleuven.ac.be 321 The Neuroradiology Journal 27: 322-326, 2014 - doi: 10.15274/NRJ-2014-10040 www.centauro.it Progressive Multifocal Leukoencephalopathy: a Rare Cause of Cerebellar Edema and Atypical Mass Effect A Case Report CHRIS OJEDA1, RACHID ASSINA2, MAUREEN BARRY3, ADA BAISRE4, CHIRAG GANDHI2 1 Biomedical Engineering, 2 Neurosurgery Department, 3 Radiology Department, 4 Pathology Department, Rutgers New Jersey Medical School; Newark, NJ, USA Key words: leukoencephalopathy, progressive multifocal, MRI, mass effect SUMMARY – Progressive multifocal leukoencephalopathy (PML) is an opportunistic demyelinating disease of the CNS caused by the JC papovavirus (JCV). Demyelination due to oligodendrocyte death leads to multifocal, asymmetric lesions. MRI is a valuable tool for detecting and differentiating PML from other neuropathies. Radiographically, PML classically presents as bilateral, subcortical white matter lesions with a lack of brain atrophy. As the disease progresses, lesions become larger and coalesce to become confluent. Minor edema and mass effect are infrequently described and the presence of significant mass effect suggests an alternative diagnosis. In our case, a patient demonstrated atypical marked infratentorial mass effect. Bilaterally, cerebellar lesions with associated mass effect were observed, as well as effacement of cerebellar folia and partial effacement of the fourth ventricle. The diagnosis of PML was confirmed with a biopsy of the right cerebellar lesion showing classic PML histology, with JCV DNA detection by polymerase chain reaction in the biopsy material. Introduction Progressive multifocal leukoencephalopathy (PML) is a demyelinating disorder caused by infection by the JC papovavirus (JCV). Although more than 70% of the adult population in the United States carries the virus, JCV is an opportunistic infection that primarily affects those with acquired immunodeficiency syndrome (AIDS) undergoing immunosuppressive therapy. PML is characterized by the destruction of infected oligodendrocytes resulting in myelin breakdown and the destruction of white matter 1. PML has a variable clinical presentation depending on the location of the lesion, but most commonly presents with neurological symptoms including cognitive deterioration, apraxia, visual deficits and motor problems that develop over weeks. Pathological findings include focal demyelination with macrophage infiltrates and viral particles in the nuclei of oligodendrocytes 322 with a ground-glass appearance. MRI reveals widespread asymmetric, multifocal areas of hypointensity on T1 and hyperintensity on T2. Mass effect and enhancement are typically absent or mild, whereas marked mass effect suggests an alternative diagnosis 2. In this case, the radiological presentation of PML with atypical infratentorial marked mass effect is described. Case Report A 45-year-old woman with a medical history significant for poorly managed HIV, anemia, and malnutrition presented to the emergency department with a seven-day history of weakness, fatigue and decreased appetite. Physical examination revealed altered mental status, visual deficits, limited range of motion and difficulty with sitting and standing balance. She was alert, oriented only to self, her motor strength was 5/5 throughout her right upper Chris Ojeda Progressive Multifocal Leukoencephalopathy: a Rare Cause of Cerebellar Edema and Atypical Mass Effect and lower extremities, 3/5 and 2/5 on her left upper and lower extremities respectively. Sensation was decreased on her left side. Reflexes were +2 throughout with negative Hoffman and clonus. Hematology workup showed a CD4 count of 14 cells/—l (normal = 300-1400 cells/—l), and lymphocyte count of 95 cells/—l (normal = 1000-3500 cells/—l). A contrast-enhanced MRI revealed an ill-defined T2 hyperintense confluent lesion in the right cerebellar hemisphere with extension to the brainstem (Figure 1). The T2 hyperintense signal extended superiorly to the midbrain and internal capsule and inferiorly throughout the pons (Figures 2). There was associated mass effect with effacement of cerebellar folia and partial effacement of the fourth ventricle. There was also an ill-defined T2 signal in the left cerebellar hemisphere. There was no abnormal enhancement of the lesion. A few millimeters, non-enhancing T2 hyperintense lesions were also noted in the cerebral hemispheres, subcortical and periventricular white matter. Stereotactic needle biopsy of the right cerebellar lesion revealed a well circumscribed area of demyelination containing several foamy macrophages, perivascular lymphocytes, oligodendrocytes with ground-glass enlarged nuclei, large reactive astrocytes some with bizarre appearance in a background with relative preservation of axons (Figure 3). JCV DNA was detected by PCR in formalin-fixed paraffin-embedded tissue from the brain biopsy. Once the diagnosis of PML was established the patient began highly active antiretroviral therapy (HAART) and was given prophylactic antibiotics. Figure 1 Axial T2 MRI showing large confluent regions in the right cerebellum causing 4th ventricle effacement. Smaller focal lesions are seen in the left cerebellum and pons. Mass effect is seen bilaterally. Discussion PML is a progressive, opportunistic demyelinating disease of the CNS caused by the JC virus whose prevalence has increased in tandem with the rise in AIDS. Demyelination due to oligodendrocyte death leads to multifocal, asymmetric lesions. PML classically presents as bilateral, subcortical white matter lesions with a lack of brain atrophy. As the disease progresses, lesions become larger, coalesce and become confluent along with the appearance of some brain atrophy 3. Lesions are most commonly observed supratentorially in the parietooccipital and frontal lobes, and involve the periventricular region and centrum semiovale. Most cases demonstrate no or mild ventricular Figure 2 Axial T2 MRI showing the upper boundary of lesions at the level of the thalamus and posterior internal capsule. 323 Progressive Multifocal Leukoencephalopathy: a Rare Cause of Cerebellar Edema and Atypical Mass Effect A B C D E F Chris Ojeda Figure 3 A) Abundant foamy macrophages mixed with enlarged oligodendrocyte nuclei showing a ground-glass appearance. A small vessel is also shown in the center with perivascular mononuclear cell infiltrate. H&E 400×. B) Foamy macrophages, enlarged oligodendrocytes and reactive, bizarre astrocytes. H&E 400×. C) Normal myelinated white matter to the left, contrasts with the pale area showing loss of myelin to the right. LFB-PAS, 100×. D) Several macrophages are highlighted with a CD68 immunohistochemical stain, 200×. E) Relative preservation of axons is noted with a neurofilament immunohistochemical stain, 200×. F) SV40 immunostain showing nuclear reactivity in the infected cells. H&E 400×. dilation. Infratentorial lesions of the pons, midbrain and middle cerebellar peduncle are seen in addition to supratentorial lesions in many cases 4. On MRI, lesions have a scalloped shape and appear hypointense on T1 and hyperintense on T2 relative to normal white matter. 324 Mass effect and enhancement are uncommon in untreated PML, but temporary enhancement of PML lesions has been reported as a response to HAART. Diffusion-weighted MRI demonstrates a high signal in newer lesions and at the advancing edge of large lesions 5. www.centauro.it The Neuroradiology Journal 27: 322-326, 2014 - doi: 10.15274/NRJ-2014-10040 Restricted diffusion in older lesions and at the core of large lesions has been observed 3. Atypical presentations of PML have been reported in the literature focusing on lesion distribution, imaging manifestation and as the result of therapy. In a study of 48 HIVpositive patients pretreatment by Post et al., supratentorial and infratentorial lesions were reported in 58.3% of patients, while Berger et al. reported that solely infratentorial lesions are only observed in 20% of cases. Post et al. also reported unilateral white matter lesions in 8.3% of pretreatment patients. Gray matter involvement can be seen in up to 50% of patients but rarely occurs in regions other than the thalamus, basal ganglia, or cortical gray matter 3. Typically, PML lesions do not enhance, but faint peripheral enhancement has been reported 6. While minor edema and mild mass effect are infrequently observed, a marked mass effect has been noted as a differentiating feature suggesting an alternative diagnosis to PML 4,7-9. Focal hemorrhage in PML lesions is rare 3,10. Certain monoclonal antibodies, such as Natalizumab, Efalizumab, and Rituximab, used to treat autoimmune diseases such as multiple sclerosis, have been associated with cases of PML. MRI findings in monoclonal antibody-associated PML are similar to those of classic PML. The two major differences include the presence of destructive cavitary lesions and a significantly higher rate of gadolinium enhancement at diagnosis 11. In our patient, the largest T2 hyperintense lesion burden was atypically located primarily infratentorially with a large associated mass effect. The largest confluent lesion extended from the right cerebellar hemisphere involving the right aspect of the pons extending to the midbrain and internal capsule. A smaller left cerebellar lesion with mass effect was also noted. Both white and gray matter were af- fected and no abnormal enhancement was observed with gadolinium contrast. Nonspecific areas of T2 hyperintensity in supratentorial white matter were also seen. This abnormal signal, along with mild cerebral volume loss was assumed to represent changes due to HIV encephalopathy. Based on the initial MRI appearance, notably the mass effect and posterior fossa infiltrating lesion, a neoplasm such as primary lymphoma or an infectious process such as toxoplasmosis were suggested. Slight edema and contrast enhancement have occasionally been reported in the initial period after starting HAART as a result of improved immune response. Our patient was not on antiretroviral therapy at or prior to presentation making this post-treatment scenario an unlikely cause of edema and mass effect. Additionally, our patient presented in an immunocompromised state with a CD4 count of 14 cells/—l (normal = 300-1400 cells/—l), and lymphocyte count of 95 cells/—l (normal = 10003500 cells/—l). Diagnosis was confirmed by the morphological findings, including the characteristic triad of: well demarcated foci of demyelination, oligodendrocytes with large, bizarre, hyperchromatic nuclei with a ground-glass appearance. Additionally there was positive SV40 immunostain and JCV DNA was detected on the brain biopsy specimen. Conclusion We present a unique case of an atypical radiographic presentation of PML in an immunocompromised patient showing a cerebellar lesion with a widespread mass effect. The PML diagnosis should be included in the differential workup of posterior fossa infiltrating lesions with mass effect. 325 Progressive Multifocal Leukoencephalopathy: a Rare Cause of Cerebellar Edema and Atypical Mass Effect Chris Ojeda References 1 Malhotra A, Sidhu R. Progressive multifocal leukoencephalopathy. Eur J Radiol Extra. 2006; 57 (2): 35-39. doi: 10.1016/j.ejrex.2005.11.005. 2 Weissert R. Progressive multifocal leukoencephalopathy. J Neuroimmunol. 2011; 231 (1-2): 73-77. doi: 10.1016/j.jneuroim.2010.09.021. 3 Shah R, Bag A, Chapman RP, et al. Imaging manifestations of progressive multifocal leukoencephalopathy. Clin Radiol. 2010; 65 (6): 431-439. doi: 10.1016/j. crad.2010.03.001. 4 Post M, Constantin Y, Simpson D, et al. Progressive multifocal leukoencephalopathy in AIDS: are there any MR findings useful to patient management and predictive of patient survival? AIDS Clinical Trials Group, 243 Team. Am J Neuroradiol. 1999; 20 (10): 1896-1906. 5 Küker W, Mader I, Nägele T, et al. Progressive multifocal leukoencephalopathy: value of diffusion-weighted and contrast-enhanced magnetic resonance imaging for diagnosis and treatment control. Eur J Neurol. 2006; 13 (8): 819-826. doi: 10.1111/j.1468-1331.2006.01362.x. 6 Thurnher M, Post J, Rieger A, et al. Initial and followup MR imaging findings in AIDS-related progressive multifocal leukoencephalopathy treated with highly active antiretroviral therapy. Am J Neuroradiol. 2001; 22 (5): 977-984. 7 Berger J, Kaszovitz B, Post M, et al. Progressive multifocal leukoencephalopathy associated with human immunodeficiency virus infection: A review of the literature with a report of sixteen cases. Ann Intern Med. 1987; 107 (1): 78-87. doi: 10.7326/0003-4819-107-1-78. 8 Karahalios D, Breit R, Dal Canto M, et al. Progressive multifocal leukoencephalopathy in patients with HIV infection: lack of impact of early diagnosis by stereotactic brain biopsy. J Acquir Immune Defic Syndr. 1992; 5 (10): 1030-1038. 326 9 Berger J, Mucke L. Prolonged survival and partial recovery in AIDS-associated progressive multifocal leukoencephalopathy. Neurology. 1988; 38 (7): 10.1212/ WNL.38.7.1060. 10 Berger J, Pall L, Lansha D, et al. Progressive multifocal leukoencephalopathy in patients with HIV Infection. J Neurovirol. 1998; 4 (1): 59-68. doi: 10.3109/13550 289809113482. 11 Tan C, Koralnik I. Progressive multifocal leukoencephalopathy and other disorders caused by JC virus: clinical features and pathogenesis. Lancet Neurol. 2010; 9 (4): 425-437. doi: 10.1016/S1474-4422(10)70040-5. Chirag D. Gandhi MD, FACS FAANS Neurological Surgery 90 Bergen Street, Suite 8100 Newark, NJ 07103 USA Tel.: (973) 972-9626 Fax: (973) 972-0222 E-mail: gandhich@njms.rutgers.edu The Neuroradiology Journal 27: 327-333, 2014 - doi: 10.15274/NRJ-2014-10037 www.centauro.it Varicella Zoster CNS Vascular Complications A Report of Four Cases and Literature Review FRANCISCO CHIANG1, THEERAPHOL PANYAPING1, GUSTAVO TEDESQUI1, DANIEL SOSSA1, CLAUDIA COSTA LEITE1,2, MAURICIO CASTILLO1 1 2 Department of Radiology, Neuroradiology Division, University of North Carolina at Chapel Hill; Chapel Hill, NC, USA Department of Radiology, University of São Paulo, School of Medicine; São Paulo, Brazil Key words: varicella zoster, complication, vasculitis, MRI, imaging SUMMARY – This study explored the neurologic vascular complications of varicella zoster virus (VZV). We describe four patients presenting at our institution with neurologic involvement by VZV. MR and MRA studies of the intracranial arterial circulation in the head were read by board-certified radiologists using standard clinical procedures. On MRI, three patients had acute infarcts and in two instances irregularities and narrowings of vessels were visible. Many of these complications are recognized to be due to a vasculopathy affecting small or large vessels and resulting in cerebral infarctions and rarely hemorrhages. The pattern of cerebral infarction and vascular abnormalities is not specific and resembles those of vasculitis/vasculopathy from other causes. The central nervous system (CNS) vascular complications of VZV should be considered in the patients with simultaneous primary or prior VZV infection whose imaging studies show cerebral infarction and/ or vasculitic appearing intracranial arteries. Introduction Systemic varicella zoster virus (VZV) infection is very common. Although rare, neurological complications of primary infection are well-described in the literature and with the advent of the polymerase chain reaction (PCR) the recognized spectrum of neurological disorders associated with VZV has widened. The VZV virus can infect a broad variety of cells in the CNS, including neurons, oligodendrocytes, meningeal cells, ependymal cells and the blood vessel wall cells. The wide range of susceptible cells explains the diversity of clinical and pathological nervous system manifestations of VZV. Lesions caused by small-vessel vasculopathy show central cavitation and macrophage influx secondary to the initial ischemic event or to the additional damage caused by VZV infection of astrocytes and neurons. Disruption of the internal elastic lamina in cerebral arteries infected with VZV can result in a weakened vessel wall that leads to dolichoectasia or aneurysm with subarachnoid or intracerebral hemorrhage, or to arterial dissection 1. Accurate recognition of the vascular complications of CNS Varicella Zoster is important as early treatment with antiviral agents may be beneficial. Here we describe four patients with vascular complications due to VZV infection with emphasis in their imaging characteristics and a pertinent literature review. Case 1 A 26-year-old African-American man was diagnosed with systemic lupus erythematosus and three years ago developed end stage kidney disease requiring dialysis. One year ago he was diagnosed with zoster ophthalmicus and VZV meningitis successfully treated with retroviral medication. Four months later he reported right eye swelling, pain, and rash consistent with VZV recurrence with right trigeminal distribution confirmed by CSF PCR. MRI and MRA of the brain revealed severe stenosis (•70% stenosis) of the right MCA near its origin with generalized asymmetric narrowing of multiple arteries consistent with vasculi327 Varicella Zoster CNS Vascular Complications A Francisco Chiang B Figure 1 A) TOF 3D reconstruction shows severe stenosis (•70% stenosis) of the right MCA at its origin, moderate irregular stenosis of the left MCA and bilateral stenosis of the A1 segments, worse on the right. B) FLAIR axial image shows abnormal hyperintensity in the right centrum semiovale. tis. A subacute right occipital lobe infarct was demonstrated (Figure 1). The patient is stable without any additional brain symptoms. regularities in the right MCA (Figure 3). VZV infection was diagnosed by CSF PCR and the patient was started on antiviral antibiotics and steroids. Clinically she improved and was discharged home. Case 2 A three-year-old boy with a history of medulloblastoma resection was admitted with seizures and fever. One-year before admission, he had a herpetic skin rash on the left trunk and the diagnosis of VZV infection was made. Contrastenhanced MRI performed at that time demonstrated a posterior pontine subacute infarct (Figure 2). The patient was treated with acyclovir therapy for 20 days. No follow-up is available. Case 3 A 28-year-old woman presented with a rapid onset encephalopathy and altered mental status. Brain MRI showed bilateral basal ganglia T2-weighted bright lesions with high DWI signal and restricted diffusion on ADC maps. MRA of the circle of Willis showed mild ir328 Case 4 An 11-month-old boy with previously normal neurodevelopment and motor skills presented with right hemiparesis 15 days after VZV infection. MRI showed left basal ganglia acute infarcts with restricted diffusion on ADC maps. No abnormalities were identified on MRA of the circle of Willis (Figure 4). At 24 hours follow-up, the hemiparesis had improved greatly and after viral antibiotics he was discharged. Materials and Methods Multiplanar, multisequence MR was done including DWI in all cases. Intracranial arterial circulation MRA was performed using www.centauro.it The Neuroradiology Journal 27: 327-333, 2014 - doi: 10.15274/NRJ-2014-10037 A B C Figure 2 A) Axial T2-weighted image shows increased signal intensity in the posterior pons anteriorly to the 4th ventricle. B) Pre (left) and post contrast-enhanced (right) sagittal images show heterogeneous enhancement of the lesion in the posterior pons. Note post-operative cerebellar changes from previous medulloblastoma resection. C) DWI image shows the high signal focus in the posterior pons corresponding to the area of infarction. The ADC maps did not show restricted diffusion suggesting an subacute infarct. 329 Varicella Zoster CNS Vascular Complications A Francisco Chiang B C Figure 3 A) Axial FLAIR image shows abnormal hyperintensities in the caudate head bilaterally and in the left lentiform nucleus. B) DWI image shows a hyperintense signal in corresponding regions. The ADC maps showed restricted diffusion suggesting an acute infarct. C) TOF 3D reconstruction of the brain shows severe stenosis (•70% stenosis) at the distal M1 segment of the right MCA and A1 segment of the left ACA. the 3D TOF (time of flight) sequence. In the 11-month-old patient the field-of-view was changed from 28 to 26 cm, and the slice thickness was decrease from 7 to 5 mm compared to the MRA parameters of the adult patients. The studies were read by board-certified radiologists using standard clinical procedures. Conventional angiography was not performed in these patients. 330 Discussion Varicella zoster (VZV) is a human αherpesvirus. The primary infection causes chickenpox or varicella after which it becomes latent in the brain, dorsal nerve roots, and autonomic ganglia and may reactivate to cause herpes zoster. Nervous system complications can follow either primary VZV infection or its www.centauro.it The Neuroradiology Journal 27: 327-333, 2014 - doi: 10.15274/NRJ-2014-10037 A B Figure 4 A) Sequential DWI images show a high signal in the left caudate and putamen and restricted diffusion in the ADC maps consistent with acute infarcts. B) TOF 3D reconstruction shows no abnormality of the intracranial vessels. reactivation 2. After reactivation, VZV occasionally invades the spinal cord or cerebral arteries producing severe neurological manifestations such as myelitis, encephalitis, aseptic meningitis, acute cerebellar ataxia, Reye syndrome, Ramsay Hunt syndrome, and rarely stroke 2,3. All of these complications are recognized to be due to vasculitis affecting small and/or large vessels. Large and small vessel vasculopathies are due to direct infection of the vessel wall and vasculitis causing thrombosis resulting in cerebral infarctions and rarely hemorrhages 3. VZV has been detected in large and small CNS blood vessels using PCR. VZV DNA can be detected in arteries of both the anterior and posterior circulations but not in the brain substance suggesting that VZV infection is prima331 Varicella Zoster CNS Vascular Complications rily a vasculopathy that affects blood vessels. CNS vasculopathy due to VZV may involve small and large vessels depending on the immune status of the patients. Large vessel-associated encephalitis occurs mostly in immunocompetent patients while small vessel-mediated encephalitis is found in immunodeficient patients 4. Large vessel vasculitis is also called “unifocal vasculopathy” or “granulomatous arteritis” and has usually an acute onset. Small vessel vasculitis, also called “multifocal vasculopathy”, commonly has a subacute presentation 2-4. Both small and large vessel disease can occur together in HIV-positive patients 3. Serious manifestations arise when VZV invades the spinal cord or cerebral arteries after reactivation causing severe disease especially myelitis and focal vasculopathies. Neurologic involvement may also develop in the absence of rash. Neurological complications of primary VZV infection are rare and those associated with reactivation are even less common. With the introduction of PCR, the recognized clinical spectrum of acute and chronic neurological disorders associated with VZV reactivation has widened 2,3. Special conditions like advanced age or immunosuppression that produce some degree of immune compromise may lead to virus reactivation. The possible manifestations of this reactivation are diverse and include multifocal VZV vasculopathy. VZV vasculopathy is probably not unusual given that VZV affects >50% of individuals by 80 years of age, increasing the risk of stroke by 30% within the following year. Nearly 40% of such patients have no history of zoster or varicella rash. Thus, VZV must be considered a possible cause of TIA or stroke either in adults or children even without a history of rash. When a rash has occurred, there is often a long delay from its onset to the occurrence of neurologic symptoms 4,5. Pathological and virological analysis of affected arteries from cases of clinically unifocal vasculopathy after zoster or varicella reveal multinucleated giant cells, Cowdry type A inclusion bodies, and herpesvirus particles as well as VZV DNA antigen in affected vessels 3,6. Most patients with VZV vasculopathy have findings on brain imaging. MRI reveals multiple cortical infarcts, infarcts at gray/white matter junctions, and deep infarcts in central gray and white matter as well as in the brainstem. Most lesions are ischemic and in the subacute phase enhance partially or completely with 332 Francisco Chiang gadolinium 3,6. Catheter angiography or MRA may reveal narrowing in the middle and anterior cerebral arteries in patients with infarcts. A large vessel vasculitis of the circle of Willis can lead, in some instances, to hemorrhagic infarcts 5. Occlusion of the retinal vessels or posterior circulation arteries can also occur 7. Rarely, subarachnoid hemorrhage can occur due to VZV vasculopathy 8. In our cases, the patients were not immunocompromised and one of them had varicella recurrence along right trigeminal nerve distribution, which was confirmed by CSF PCR detection. In accordance with the previous discussion, all have large vessel-related disease. Brain MRI in the first, third and fourth patients showed acute infarcts in the basal ganglia. In our first patient there was also a small subacute infarction at the right occipital lobe and in our second patient a posterior pontine infarct. MRA of the brain in three patients showed a high-grade stenosis of the right MCA near its origin with generalized asymmetric small calibers of multiple vessels and a small right occipital lobe infarction in case 1 and high-grade stenosis of the distal M1 segment of the right MCA in case 3 with acute bilateral basal ganglia infarctions. MRA may not have the spatial resolution needed to detect abnormalities in all affected arteries, as in our fourth case where the MRA was normal. The presence of brain infarctions in both deep gray and cortical-subcortical areas, accompanying the vascular abnormality on MRA, was consistent with large vessel vasculopathy with focal vascular involvement. In these patients the evaluation of CSF may reveal moderate lymphocytic predominant pleocytosis, red blood cells, and mildly elevated protein with IgG oligoclonal bands against VZV. PCR analysis of CSF is a sensitive and specific test for VZV DNA. Detection of VZV antibody in CSF, even in the absence of PCR-amplifiable VZV DNA, supports the diagnosis in cases of suspected VZV infection of the nervous system 9,10. IgG antibody against VZV is the diagnostic method of choice as it is better than VZV DNA PCR. This is due to the time lag between the varicella infection and the onset of vascular symptoms making VZV DNA undetectable in CSF after 14-50 days of infection 10,11. The few protean complications associated with VZV include aneurysms, vascular ectasia formation, and carotid dissection due to infec- www.centauro.it The Neuroradiology Journal 27: 327-333, 2014 - doi: 10.15274/NRJ-2014-10037 tion of the tunica media of the arteries with disruption of their internal elastic lamina. VZV vasculopathy may be associated with other complications such as subarachnoid hemorrhage, spinal cord infarction, peripheral artery disease, and polyneuritis cranialis 8,12. The vasculitic changes can be partly reversed with the use of acyclovir and methylprednisolone 8. Early diagnosis of CNS and peripheral nervous system complications of VZV reactivation is important as aggressive treatment with intravenous acyclovir can be beneficial. The use of intravenous acyclovir for 14 days accompanied by a short course of steroid for five days reduces inflammation in cerebral vessels 3,8,12. Conclusion Reactivation of VZV occasionally invades the cerebral arteries producing severe neurological manifestations such as VZV-related focal vasculopathies. MRI and MRA are essential to diagnose VZV vasculopathy disclosing brain infarcts sometimes with accompanying stenosis or occlusion of major cerebral arteries especially those of the circle of Willis. The presence of anti-VZV antibody in CSF is strongly suggestive of VZV even in the absence of PCR-amplifiable VZV DNA. Early diagnosis of more serious CNS VZV reactivation is important as treatment with intravenous acyclovir is beneficial. References 1 Kleinschmidt-DeMasters BK, Gilden DH. VaricellaZoster virus infections of the nervous system: clinical and pathologic correlates. Arch Pathol Lab Med. 2001; 125 (6): 770-780. 2 Gilden D. Varicella Zoster Virus and CNS Syndromes. Herpes. 2004; 11 (Suppl 2): 89A-94A. 3 Nagel MA, Cohrs RJ, Mahalingam R, et al. The varicella zoster virus vasculopathies. Clinical, CSF, imaging, and virologic features. Neurology. 2008; 70 (11): 853-60. doi: 10.1212/01.wnl.0000304747.38502.e8. 4 Nagel MA, Gilden D. The challenging patient with varicella-zoster virus disease. Neurol Clin Pract. 2013; 3 (2): 109-117. doi: 10.1212/CPJ.0b013e31828d9f92. 5 McKelvie P, Collins S, Thyagarajan D. Meningoencephalomyelitis with vasculitis due to varicella zoster virus: a case report and review of the literature. Pathology. 2002; 34 (1): 88-93. doi: 10.1080/00313020120105705. 6 Gilden DH, Mahalingam R, Cohrs RJ, et al. The protean manifestations of varicella-zoster virus vasculopathy. J Neurovirol. 2002; 8 (Suppl. 2): 75-79. doi: 10.1080/13550280290167902. 7 Nagel MA, Russman AN, Feit H. VZV ischemic optic neuropathy and subclinical temporal arterial infection without rash. Neurology. 2013; 80 (2): 220-222. doi: 10.1212/WNL.0b013e31827b92d1. 8 Jain R, Deveikis J, Hickenbottom S, et al. Varicellazoster vasculitis presenting with intracranial hemorrhage. Am J Neuroradiol. 2003; 24 (5): 971-974. 9 Verma R, Lalla R, Patil TB. Extensive extracranial and intracranial varicella zoster vasculopathy. BMJ Case Rep. 2012; 2012. doi: 10.1136/bcr-2012-006845. 10 Russman AN, Lederman RJ, Calabrese LH, et al. Multifocal varicella zoster virus vasculopathy without rash. Arch Neurol. 2003; 60 (11): 1607-1609. doi: 10.1001/archneur.60.11.1607. 11 Nagel MA, Forghani B, Mahalingam R, et al. The value of detecting anti-VZV antibody in CSF to diagnose VZV vasculopathy. Neurology. 2007; 68 (13): 1069-1073. doi: 10.1212/01.wnl.0000258549.13334.16. 12 Gilden D, Cohrs RJ, Mahalingam R, et al. Varicella zoster virus vasculopathies: diverse clinical manifestations, laboratory features, pathogenesis and treatment. Lancet Neurol. 2009; 8 (8): 731-740. doi: 10.1016/ S1474-4422(09)70134-6. Dr Francisco Chiang Neuroradiology Division Department of Radiology University of North Carolina at Chapel Hill 101 Manning Drive Chapel Hill, NC 27154, USA E-mail: franciscochiang@gmail.com 333 The Neuroradiology Journal 27: 334-338, 2014 - doi: 10.15274/NRJ-2014-10038 www.centauro.it Histoplasmosis Brain Abscesses in an Immunocompetent Adult A Case Report and Literature Review ANA INES ANDRADE1, MAREN DONATO1, CARLOS PREVIGLIANO2, MARDJOHAN HARDJASUDARMA2 1 2 Department of Radiology, CIMED; La Plata, Buenos Aires, Argentina Department of Radiology, Louisiana State University Health Sciences Center; Shreveport, LA, USA Key words: Histoplasma capsulatum, ring enhancing, fungal abscess, immunocompetent individual SUMMARY – We describe the case of a 62-year-old man, who presented with a new onset of focal seizures of his right leg. There were no other clinical symptoms, and laboratory results were normal. Brain magnetic resonance imaging revealed multiple lesions, two supratentorial lesions were ring-enhancing. The brain biopsy tissue showed Histoplasma capsulatum abscesses. He improved on treatment with Amphotericin B. This case is reported since cerebral ring-enhancing lesions are rarely associated with histoplasmosis, which is also rare in an immunocompetent individual. We review the literature and discuss the radiologic and pathologic findings of this case compared with previous reports. Introduction Histoplasmosis is an infectious disease caused by the dimorphic fungus Histoplasma capsulatum 1. Humans are infected via inhalation of airborne microconidia, which the wind can carry for miles 2. Once in the lung, the microconidia are converted into pathogenic yeast forms that disseminate hematogenously to multiple organs, including the brain, spinal cord, and meninges 3. Histoplasmosis brain abscess is commonly seen in immunocompromised individuals 2. In addition, Histoplasma capsulatum is a neurotropic dimorphic fungus that causes self-limiting systemic mycosis in endemic regions while extra pulmonary manifestations are uncommon and typically asymptomatic 4. Central nervous system involvement is clinically recognized in 10% to 20% of cases of progressive histoplasmosis. However, in rare cases it may be an isolated finding especially in immunocompetent individuals 5. Meningitis and histoplasmoma formation are the common clinical manifestations of central nervous system histoplasmosis, whereas cerebritis and abscess are rare. To the best of our 334 knowledge, no cases of histoplasmosis brain abscesses in immunocompetent individuals have been reported. Case Report A 62-year-old man presented with a new onset of focal seizures of his right leg. This patient was not immunocompromised and serology for human immunodeficiency virus was negative. He did not have any history suggestive of deficient cellular immunity. He denied involvement with recreational activities that could have predisposed him for Histoplasma infection, or exposure to bat or bird droppings. His absolute lymphocyte count was normal. The cerebral spinal fluid culture was negative for an active infection or malignant cells. Computed tomography of the chest, abdomen and pelvis were negative for any malignancy. Magnetic resonance (MR) scan at the time showed multiple infra and supratentorial compartmental lesions, two supratentorial lesions were ring-enhancing, surrounding extensive vasogenic edema and restricted diffusion, in- Ana Ines Andrade A Histoplasmosis Brain Abscesses in an Immunocompetent Adult B Figure 1 A,B) T1-weighted axial and coronal images after contrast administration showing the cerebellar lesions and pachymeningeal thickening. dicating purulent content. There was also meningeal enhancement after contrast administration. Differential diagnosis like left frontal abscess coexisting with a metastatic deposit or a metastatic deposit with abscesses was included. Pathology evaluation of the brain biopsy was consistent with histoplasma capsulatum abscesses. The patient was started on oral amphotericin B. He was also started on dexamethasone as an anticerebral edema measure. The patient’s condition improved on treatment. rounded by extensive vasogenic edema. A ringenhancing lesion in the vertex on the left side with extensive vasogenic edema was seen and also a lesion attached to the dura in the dorsal frontal pole on the left. The abscess walls of the left ring-enhancing lesions presented a hypointense signal in susceptibility-weighted imaging indicating probable hemosiderin deposits in the capsule (Figure 3). The ring-enhancing lesions had restricted diffusion and MR spectroscopy showed prominent lipid and lactate peaks with increased choline (Cho) indicating purulent content (Figure 4). Imaging Findings MR characteristics of the case included multiple infra and supratentorial compartmental lesions. The cerebellar lesions appeared solid in the right vermis, a second lesion was adjacent to the lateral recess of the fourth ventricle and the third in the left cerebellar hemisphere in the paravermian area (Figure 1). There were also lesions in the right hippocampus, left posterior frontal orbital gyrus, left insula, left occipital lobe, right caudate and left caudate nucleus (Figure 2). A prominent ring-enhancing lesion was seen in the left prefrontal area sur- Pathologic Findings Brain biopsy tissue showed Histoplasma capsulatum abscesses (Figure 5). Discussion and Literature Review Histoplasma capsulatum is an endemic fungus in certain regions of North America and Latin America, including the Ohio and Mississippi river valleys of the United States. Central nervous system involvement occurs in 5 to 20% 335 Histoplasmosis Brain Abscesses in an Immunocompetent Adult A B Ana Ines Andrade C Figure 2 A-C) T1-weighted axial images after contrast administration showing the multiple supratentorial solid lesions. A B C Figure 3 A) Susceptibility-weighted axial image showing a round lesion surrounded by extensive vasogenic edema and hypointense signal in the wall on the left prefrontal area, indicating probable hemosiderin deposits in its capsule. B,C) T1-weighted images before and after contrast administration with prominent ring-enhancing lesions. A B C Figure 4 A-C) Diffusion sequences showing restricted diffusion, indicating purulent content. 336 www.centauro.it The Neuroradiology Journal 27: 334-338, 2014 - doi: 10.15274/NRJ-2014-10038 A B Figure 5 A) Photomicrograph shows multiple organisms in histiocytes with a pale zone suggesting a capsule (hematoxylin-eosin stain). B) Photomicrograph shows silver-stained intracytoplasmic round bodies of Histoplasma capsulatum in foamy histiocytes (Gomori methenamine silver stain). of cases of disseminated histoplasmosis and is more common in those with underlying immunosuppressive disorders 6,7. Disseminated histoplasmosis, especially with a ring-enhancing brain lesion, is uncommon in an immunocompetent individual 8. Few patients (5%-10%) with disseminated histoplasmosis will develop central nervous system (CNS) infection, and only 25% of these will develop neurological symptoms 7. Imaging findings include hydrocephalus, histoplasmoma, vasculitis with infarctions and leptomeningeal enhancement 9. In one case series of isolated CNS histoplasmosis, 90% of patients had hydrocephalus and two patients had focal enhancement on T1 imaging 10. This case series of 11 patients included two children age three and 15 years. Recently, another child in a Histoplasma endemic region was reported to have isolated histoplasma meningoencephalitis. T1 MRI demonstrated meningeal enhancement with multiple non-enhancing lesions consistent with infarction 11. Projections into the cavity from the wall of an abscess with low apparent diffusion coefficient and no enhancement have been described as a characteristic of fungal etiology 12. Our case had multiple lesions, two of them with ring-enhancement, restricted diffusion and thickened wall with probable hemosiderin deposits. These finding have not been reported histoplasmosis abscesses to date. Conclusion While pulmonary involvement of histoplasmosis in immunosuppressed patients is common, systemic presentation of this fungal infection in immunocompetent patients is rare and self-limiting. Isolated CNS histoplasmosis is exceedingly rare 13. To our knowledge, this is the first case in the literature of multiple histoplasmosis brain abscesses presenting with ring-enhancing lesions, irregular wall thickening and hemosiderin deposits inside. CNS fungal disease in immunocompetent hosts is unusual and requires a high index of suspicion for diagnosis. Therefore, in Histoplasma endemic regions, physicians should include CNS histoplasmosis for an early diagnosis and adequate treatment. 337 Histoplasmosis Brain Abscesses in an Immunocompetent Adult Ana Ines Andrade References 1 Adderson E. Histoplasmosis. Pediatr Infect Dis J. 2006; 25 (1): 73-74. doi: 10.1097/01.inf.0000196922.46347.66. 2 Saccente M. Central nervous system histoplasmosis. Curr Treat Options Neurol. 2008; 10 (3): 161-167. doi: 10.1007/s11940-008-0017-x. 3 Medoff G, Kobayashi GS, Painter A, et al: Morphogenesis and pathogenicity of Histoplasma capsulatum. Infect Immun. 1987; 55 (6): 1355-1358. 4 Parihar A, Tomar V, Ojha BK, et al. Magnetic resonance imaging findings in a patient with isolated histoplasma brain abscess. Arch Neurol. 2011; 68 (4): 534-535. doi: 10.1001/archneurol.2011.59. 5 Zalduondo FM, Provenzale JM, Hulette C, et al. Meningitis, vasculitis, and cerebritis caused by CNS histoplasmosis: radiologic-pathologic correlation. Am J Roentgenol. 1996; 166 (1): 194-196. doi: 10.2214/ ajr.166.1.8571874. 6 Assi MA, Sandid MS, Baddour LM, et al. Systemic histoplasmosis: a 15-year retrospective institutional review of 111 patients. Medicine (Baltimore). 2007; 86 (3): 162-169. doi: 10.1097/md.0b013e3180679130. 7 Wheat LJ, Batteiger BE, Sathapatayavongs B. Histoplasma capsulatum infections of the central nervous system. A clinical review. Medicine (Baltimore). 1990; 69 (4): 244-260. doi: 10.1097/00005792-199007000-00006. 8 Subramanian S, Abraham OC, Rupali P, et al. Disseminated histoplasmosis. J Assoc Physicians India. 2005; 53: 185-189. 9 Schuster JE1, Wushensky CA, Di Pentima MC. Chronic primary central nervous system histoplasmosis in a healthy child with intermittent neurological manifestations. Pediatr Infect Dis J. 2013; 32 (7): 794-796. doi: 10.1097/INF.0b013e31828d293e. 338 10 Schestatsky P, Chedid MF, Amaral OB, et al. Isolated central nervous system histoplasmosis in immunocompetent hosts: a series of 11 cases. Scand J Infect Dis. 2006; 38 (1): 43-48. doi: 10.1080/00365540500372895. 11 Threlkeld ZD, Broughton R, Khan GQ, et al. Isolated Histoplasma capsulatum meningoencephalitis in an immunocompetent child. J Child Neurol. 2012; 27 (4): 532535. doi: 10.1177/0883073811428780. 12 Luthra G, Parihar A, Nath K, et al. Comparative evaluation of fungal, tubercular, and pyogenic brain abscesses with conventional and diffusion MR imaging and proton MR spectroscopy. Am J Neuroradiol. 2007; 28 (7): 1332-1338. doi: 10.3174/ajnr.A0548. 13 Nguyen FN, Kar JK, Zakaria A, et al. Isolated central nervous system histoplasmosis presenting with ischemic pontine stroke and meningitis in an immunocompetent patient. JAMA Neurol. 2013; 70 (5): 638-641. doi: 10.1001/jamaneurol.2013.1043. Ana Ines Andrade, MD Resident, Department of Radiology CIMED-La Plata 416, 5th street, La Plata Buenos Aires. Argentina Tel.: (+54) 9-221-571-7037 E-mail: anainesandrade_21@hotmail.com The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 www.centauro.it Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis ATCHAYARAM NALINI1, AARON DE SOUZA1, JITENDER SAINI2, KANDAVEL THENNARASU3 1 Department of Neurology, 2 Department of Neuroimaging and Interventional Radiology, 3 Department of Biostatistics, National Institute of Mental Health and Neurosciences; Bangalore, India Key words: cysticercosis, magnetic resonance imaging, T2 relaxometry, albendazole, perilesional gliosis SUMMARY – We describe the evolution of quantitative T2 relaxometry values on serial MRI in patients with a solitary cerebral cysticercal lesion (SCCL), and determine whether albendazole therapy affects T2 relaxation (T2R) values. Patients with new-onset seizures and MRI-confirmed SCCL were randomized to treatment with albendazole and antiepileptics (“treatment group”) or antiepileptics only (“controls”). Serial MRI including T2 relaxometry was performed at baseline, three, six, 12, and 24 months. Of 123 patients recruited, 81 had more than three MRI scans (treatment group: 37; controls: 44; 58 patients had five scans). The lesion wall at baseline showed a mean T2R value of 152.3 ms, centre 474.9 and perilesional parenchyma 338.5 ms. These were significantly higher than those from normal parenchyma (114 ms). Over time, most sharply in the initial three months, T2R values fell but even at 24 months, they remained above those from normal parenchyma. A slight increase in T2R values from the lesion centre at six months was thought to represent the initiation of gliosis. In the treatment group, T2R values approached normal at 24 months, while controls had persistently higher T2R values. The decline in T2R values at six months was more prominent in the treatment group. T2R values at baseline and at three months differed significantly depending on the stage of the lesion, being higher in stage 2 SCCL. T2R values from SCCL declined over 24 months, being significantly higher in earlier stages of degeneration. A mild increase after six months may be due to the initiation of gliosis. T2R values appear to decline faster in patients who receive albendazole. Introduction Cysticercosis is the most common parasitic disease of the central nervous system, affecting millions of individuals in both the developing countries of Asia, Africa and Latin America and in developed countries. It is caused by the encysted larval stage of the pork tapeworm Taenia solium 1. Imaging in neurocysticercosis (NC) often shows single small ring-enhancing lesions, the diagnosis of which has always been a problem as they have to be differentiated from other ring lesions 2. Computed tomography and magnetic resonance imaging (MRI) are useful investigations for diagnosing NC 3, but MRI is considered the investigation of choice for studying the evolution of cyst as well as the effect of medication 4,5. Although the natural history of NC has been studied using serial CT and MRI 6, newer quantitative techniques have been infrequently used to prospectively study the evolution of these lesions. Magnetization transfer imaging (MTI) has proven useful in delineating gliosis around the SCCL which is seen as a hyperintense signal barely visible on conventional T2-weighted imaging 7-10. T2 relaxometry is a magnetic resonance imaging (MRI) morphometric technique, which provides a quantitative and objective means of comparing normal and pathological relaxation characteristics 11. Although the evolution of the cysticercal lesion on MTI has been described 10, the corresponding changes in T2 relaxometry over time remain unknown. In addition, even though NC 339 Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis is generally treated with cysticidal drugs such as albendazole and praziquantel, the effect of albendazole on the evolution of the SCCL and the appearance of adjacent perilesional gliosis is unclear. The present study is part of a prospective randomized controlled trial of albendazole in a cohort of patients with SCCL and new-onset seizures. We previously reported the effect of cysticidal therapy with albendazole on the evolution of the cysticercal lesion on serial conventional MRI 6 and on MTI 10. This paper presents the evolution of quantitative T2 relaxation (T2R) values over a period of 24 months in patients with SCCL. We also describe the effect of cysticidal therapy with albendazole on the T2R properties of these lesions. Materials and Methods Patients were prospectively enrolled at the National Institute of Mental Health and Neurosciences, a tertiary national referral centre for neurological diseases in India. After obtaining Atchayaram Nalini written informed consent, all patients presenting with new-onset focal or generalized seizures were subjected to plain and contrast CT scan of the brain followed by contrast MRI in every case. Patients were excluded from the study if they had evidence of other lesions on CT /MRI, a history of epilepsy, received albendazole or praziquantel, significant neurologic deficits, raised intracranial pressure or seizures refractory to acute treatment. Those with an MRIconfirmed solitary cysticercal lesion in the cerebral parenchyma were included in the study. All patients were imaged using a circular polarised head coil on a 1.5T system (Magnetom Vision, Siemens, Erlangen, Germany) according to a pre-determined protocol, which included pre- and post-contrast fast spin echo (SE) T1weighted sequences (time to pulse repetition [TR] 650, time to echo [TE] 12, number of excitations [n] 1), proton-density imaging (TR 4800, TE 22, n 1), T2-weighted imaging (TR 4800, TE 90, n 1), FLAIR (TR 9000, TE 119, inversion time [TI] 1200 ,n 1), CISS-3D sequences (TR 12.25, TE 5.9, n 1), and proton (H1) MR spec- Table 1 Comparison of baseline demographic and clinical characteristics between treatment and control groups. Variable Treatment group (n=37) p value Age in years (mean ± SD) 19.5 ± 9.8 17.8 ± 11.5 0.428 Sex (M:F) 2.4:1 0.8:1 0.091 73.1 64.7 0.239 97.5 88.4 0.227 2.5 11.6 Mean area of oedema in PLP (mm ) 1125.7 1141.2 0.764 Cyst location (%) 57.8 33.3 8.9 0 61.7 31.9 4.3 2.1 0.594 1.46 0.83 0.227 Mean cyst size (mm2) Cyst stage (%) Stage 2 † Stage 3 2 Frontal Parietal Temporal Occipital Duration of illness before enrolment (months) † Controls (n= 44) Classification proposed by us previously 6. Table 2 Evolution of T2R values over time in the lesion wall, centre and perilesional area. Time of MRI a b N Mean T2R values from the lesion Mean T2R values from normal-appearing tissue Wall Centre Baseline 76 152.3a,b 474.9a,b 338.5a,b 123.2 128.8 3 mo 68 132.3 174.6a,b 183.0a,b 124.9 131.4 6 mo 72 127.2 178.7 a 149.9 a,b 122.1 128.6 12 mo 74 136.3 155.3 a,b 172.1 a 126.1 130.6 24 mo 54 124.3 137.3 a 169.7 a,b 109.2 124.1 a Perilesional area White matter Grey matter : Significant difference between this value and the normal-appearing white matter by paired samples t test. : Significant difference between this value and the normal-appearing grey matter by paired samples t test. 340 www.centauro.it The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 troscopy. Imaging in the axial plane was performed using 5-mm slice thickness with an interslice gap of 0.5 mm, 24 × 24 cm field of view, and a matrix size of 256 × 256. All patients had post-gadolinium (0.1 mmol/kg) studies. T2 relaxometry were performed in all patients using a dual echo sequence (TR=5500 ms; TE=90 ms and 22 ms). The images were transferred to an installed software programme. T2 relaxometry maps were generated manually from the measured image intensities at the level of the lesion. T2 values were calculated in the cyst, perilesional parenchyma (PLP) and normal white matter (WM). Patients were imaged at least 24 hours after the last seizure. The stage of the lesion was determined based on the criteria proposed by us previously 6. The patients were randomized to two groups by means of a random-number table consisting of 200 random numbers: one group received antiepileptic drugs only (usually phenytoin or carbamazepine/oxcarbazepine) and the second group was given albendazole 15 mg/kg/day in two divided doses for 28 days in addition to antiepileptic medication. All patients were compliant with the medication. Repeat imaging studies following the same MRI protocol were carried out at three, six and 12 months after the first evaluation. The imaging procedures were funded by the Indian Council for Medical Research. At one month after enrolment in the study and at every followup thereafter, at the time of obtaining repeat imaging the patients were evaluated regarding recurrence of seizures, any side-effects of treatment, compliance with therapy and the presence of fresh symptoms. The antiepileptic drug was tapered and stopped at the end of a twoyear seizure-free period. Patients with more than 12 months’ follow-up were included in the final analysis. Two independent trained observers and a neuroradiologist who were blinded to treatment allocation analysed the MRI for cyst location, size, presence of scolex, extent and severity of perilesional oedema, contrast enhancement and cyst persistence, resolution or calcification and T2 relaxometry parameters on follow-up scans. The lesion was considered to have resolved when it disappeared without any evidence on pre- and post-contrast MRI. Descriptive statistics included percentages, mean and standard deviation. The T2 relaxation parameters obtained on serial MRI over 24 months were tabulated, and the distribution of variables between the treatment and control groups was subjected to multivariate analysis to detect differences in evolution of the lesion and PLP changes attributable to albendazole. Chi-square test was used to study categorical variables, and one-way analysis of variance (ANOVA) was used for continuous variables, including the effect of the stage of the lesion on T2 relaxation values. Results A total of 123 patients with new-onset seizures due to SCCL were recruited, of whom 81 underwent at least four serial MRI scans with T2 relaxometry at baseline, three, six and 12 months after enrolment. Fifty-eight patients had an additional fifth MRI at 24 months after enrolment. Thirty-seven patients had received albendazole (treatment group) and 44 had not (control group). All lesions were cortical or at the cortico-subcortical junction. Demographic data are presented in Table 1, and there was no statistically significant difference between the treatment and control groups. All cysts at enrolment were active, in stages 2 or 3 as per the classification proposed by us previously 6. Serial quantitative T2 relaxometry The normal white matter had a T2 value of 114 ms in our study. The average T2 value of stage 2 lesions (corresponding to Escobar’s colloid-vesicular stage) 6,12 was 321 ms, that of stage 3 (granular nodular) lesions was 389 ms and that of stage 4 (nodular calcified) lesions was 84 ms. Normal grey matter had a slightly higher mean T2 value than white matter. The wall of the lesion on the initial MRI showed a mean T2 value of 152.3 ms, while the centre of the lesion and the perilesional parenchyma had values of 474.9 and 338.5 ms respectively. These differed significantly from normal white and grey matter. Over time, the T2 values of these regions declined significantly. On serial MRI, T2 relaxometry values in the wall declined over the first three months and then remained fairly constant, while those from the centre of the lesion and in the perilesional parenchyma showed a progressive decrease with a sharp decline in the first three months (p <0.01 for the mean T2 values on serial MRI compared to the baseline values for the wall and centre of the lesion, and p <0.001 for the mean T2 values on serial MRI compared to the baseline values for the perilesional parenchyma). The T2 values 341 Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis Atchayaram Nalini Figure 1 Cysticercal lesion on conventional MRI and a T2-relaxometry map. of the lesion centre and the perilesional parenchyma remained significantly higher than normal white and grey matter even 24 months after enrolment, while those of the lesion wall were similar to those of normal brain parenchyma from the second MRI onwards (Table 2). Figure 1 shows a cysticercal lesion on conventional MRI and a T2-relaxometry map, while Figures 2 and 3 show serial T2R images and T2R maps in patients who received and did not receive albendazole. Effect of albendazole therapy on T2 relaxation values There was no significant difference between the treatment and control group in the overall T2 values from the centre of the lesion. However there was a significant difference in the decline of the T2 value between the treatment group and the control group between the third and sixth months. The T2 value of the treatment group decreased more rapidly compared to that of controls during this period, and approached the normal range, whereas that of the control group remained high. Similarly, at the 24-month scan the difference between the T2 values at the centre of the lesion showed a trend towards significance (p=0.054) with T2 values of the treatment group approaching normal whereas those of the control group continued to be high. The T2 values in the wall of the lesion showed a similar trend, but the differences between the treatment and control groups were less significant. Similarly, the T2 value of the perilesional area showed a sus342 tained decrease for the first six months with a sharp fall over the first three months, but without any difference between the two groups (Table 3). Effect of lesion stage on T2 values T2R values of the lesion wall at baseline differed significantly depending on the stage of the lesion (for lesions in stage 2, mean T2R value was 161.8, while for those in stage 3 it was 134.5, p=0.029 by one-way ANOVA). Similarly, the T2R values for cyst contents and the perilesional area, and the average T2R values for the entire cyst and the area adjacent to the cyst were higher in stage 2 lesions than in stage 3 at baseline and at three months (Table 4). Lesions at stage 2 at baseline also had higher mean T2R values in the perilesional area and up to 12 months after enrolment. At six months, there was a significant difference in T2R values only in the perilesional area, but in later MRIs no difference was seen in T2R values in lesions at different stages of degeneration. Discussion This cohort of patients with new-onset seizures due to SCCL was subjected to serial MRI including T2 relaxometry to study the evolution of T2 relaxation values over two years. Although the diagnosis of neurocysticercosis was not histologically confirmed — patients being diagnosed based on accepted radiological crite- www.centauro.it The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 0 mo 3 mo A 24 mo 12 mo 0 mo B 6 mo 3 mo 12 mo 6 mo 24 mo Figure 2 Serial T2R images in patients who received (A) and did not receive albendazole (B). 343 Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis 0 mo 6 mo 0 mo 12 mo Atchayaram Nalini 3 mo 24 mo A 0 mo 6 mo 0 mo 12 mo 3 mo 24 mo B Figure 3 Serial T2R maps in patients who received (A) and did not receive albendazole (B). 344 www.centauro.it The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 ria — this is within the norm for such studies as routine biopsy for a single ring-enhancing lesion is no longer considered justifiable. Patients were randomly allocated to albendazole treatment or a control group in order to evaluate the effect of anthelmintic therapy on this evolution. We demonstrated that the lesion wall, centre and PLP showed significantly higher T2R values compared to normal brain parenchyma; that these values declined over time, particularly in the initial three months; and that the T2R values differed significantly according to the stage of degeneration of the cysticercal lesion. Albendazole therapy did not significantly affect the evolution of T2R values in our cohort, although T2R values remained higher in patients who did not receive albendazole, even 24 months after enrolment. Quantitation of T2 relaxation time has been used to assess focal damage, which may not be readily apparent on conventional MRI. T2 relaxation of the central nervous system is shown in animal models to consist of distinct components corresponding to extra-axonal water protons, axonal protons, and protons possibly associated with mobile lipids in myelin sheaths 13 , T2 signal intensity is increased (reflecting a focal increase in T2 relaxation) in areas of oedema and tissue damage, where more free water is associated with relative expansion of the extracellular space 14. On the other hand, T2 relaxation times decrease with increasing protein concentration 15. T2 relaxometry represents an objective and sensitive way of detecting this change in T2 relaxation. In T2 relaxometry, T2 relaxation is computed from a series of MR images obtained at the same TR (time of repetition) and different echo time 13. This technique is simple and rapid, utilizes data acquired for routine proton-density and T2W imaging, and can be implemented on standard clinical MR systems. The measurements can be easily ac- quired and are reproducible 16. It is useful for the systematic quantitative study of patients, particularly when routine T2-weighted MRI is unremarkable 17,18. The method has been successfully used to evaluate patients with mesial temporal sclerosis and demyelinating disorders such as multiple sclerosis 19. Quantitation of T2 relaxation time in tissues that appear normal to gross image inspection in patients with temporal lobe epilepsy has shown significantly higher T2 values compared to the normal tissue (around 19% increased over normal side) 20-22 . Structural changes in normal-appearing white matter (NAWM) have been demonstrated in patients with multiple sclerosis using T2 relaxometry 23. Beta-amyloid plaque deposits in patients with Alzheimer’s disease may be visualized by T2 relaxometry 24. This technique has also been used to study patients with NCC and has shown promise in differentiating NCC from tuberculomata 18, as well as quantifying gliosis around healing and healed cysticercal granulomata 25. To the best of our knowledge, there are no previous reports of either serial imaging of NCC lesions with T2 relaxometry or the T2 relaxation values of different stages of NCC. In our study, the majority of lesions were in stage 2 at the time of initial evaluation, at three and six months most were in stage 3 while at later evaluations most lesions were in stage 4 or had resolved. These radiologic stages roughly correspond to the four pathologic stages described by Escobar 6,12. In the vesicular stage, the cyst contains clear fluid with little or no inflammation. At this stage, the lesion is expected to show high T2 relaxation values. As there is no inflammation and gliosis at this stage, the perilesional parenchyma is expected to show normal T2 values. In our study, there were no lesions in the vesicular stage as we recruited only patients with symptomatic SCCL. Due to the Table 3 Effect of albendazole on T2 relaxation values from the cysticercal lesion and the perilesional parenchyma. Time of MRI N Mean T2R values from the lesion Wall Treatment Control Treatment Control p Centre Perilesional area Treatment Control p Treatment Control p Baseline 35 41 147.5 156.3 0.365 379.4 558.8 0.155 326.7 348.6 0.563 3 mo 29 39 130.3 133.8 0.657 160.5 185.1 0.230 165.2 196.2 0.220 6 mo 32 40 114.8 137.1 0.016 125.8 221.0 0.016 141.9 156.3 0.054 12 mo 32 42 135.8 136.7 0.95 143.6 164.2 0.336 180.2 165.9 0.953 24 mo 25 29 115.7 131.8 0.118 115.8 156.7 0.054 157.5 180.6 0.146 345 MRI at Baseline MRI at 3 months Mean T2R p 7 108.1 0.133 4 22 117.2 189.6 Res 23 161.0 66 157.0 Total 52 135.4 0.029 3 18 181.9 0.583 3 7 155.6 p 0.632 3 18 138.0 0.162 3 138.7 4 27 144.3 14 186.5 Res 21 67 182.4 Total 0.001 3 35 156.3 0.018 3 35 203.2 163.6 4 18 10 202.6 Res Res 5 133.4 Total Total 62 179.1 4 367.0 4 344.5 3 43 4 Mean T2R p Lesion centre 2 42 572.4 0.413 2 3 19 377.5 Total 61 511.7 Stage Stage Stage Stage 2 43 385.6 0.002 2 3 19 241.4 3 43 169.7 4 18 156.4 4 27 197.3 4 22 152.1 Total 62 341.4 4 10 221.1 Res 14 124.9 Res 22 134.3 Res 23 189.9 Res 5 122.0 Total 67 149.8 Total 67 172.5 Total 52 169.3 Total 62 186.9 4 249.8 0.011 3 35 160.9 0.723 3 18 144.6 0.779 3 7 105.8 2 43 272.8 0.041 2 3 19 187.8 3 44 150.4 4 18 140.2 4 27 156.7 4 22 115.5 Total 62 246.8 4 10 163.6 Res 14 158.7 Res 22 167.7 Res 23 160.3 Res 5 135.7 Total 67 154.9 Total 67 157.8 Total 52 134.0 Total 63 157.7 4 293.4 0.021 3 35 148.8 0.173 3 18 178.6 3 6 152.7 2 43 424.9 0.001 2 3 19 239.3 3 44 164.7 4 18 147.3 4 27 196.4 4 21 143.7 Total 62 368.0 4 10 233.3 Res 14 128.2 Res 22 136.5 Res 23 174.7 Res 5 130.2 Total 67 144.1 Total 67 171.9 Total 50 159.0 Total 63 181.0 Res: resolved. p value calculated for effect of lesion stage on the mean T2R value by one-way ANOVA. 0.60 0.081 0.057 0.194 Atchayaram Nalini Average of perilesional area N Mean T2R p p N MRI at 24 months N Mean T2R Mean T2R Stage Average of lesion MRI at 12 months N N Parameter Perilesional area MRI at 6 months Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis 346 Table 4 Effect of lesion stage on T2R values. www.centauro.it The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 lack of inflammation around a live cyst, these are generally asymptomatic 26. In the vesicular colloid stage, the wall of the lesion thickens and the protein content of cyst fluid rises, containing abundant inflammatory cells. In the granular nodular stage, the lesion shows a further increase in the protein content with gelatinous granulomatous semisolid material in the centre of lesion. The wall of the lesion thickens further at this stage and perilesional gliosis sets in. In the nodular-calcific stage, the lesion is calcified with abundant perilesional gliosis. As would be expected from the evolution of pathological changes in a degenerating cysticercus, lesions at the initial MRI showed high T2 relaxation values in the centre of the lesion, due to the fluid content. Jayakumar et al. found that the mean T2R value of cysticercal cysts was 617 ms (range, 305-1365) which is very similar to the value from the centre of the lesion in our cohort. They also demonstrated that T2 relaxometry reliably differentiated NCC from tuberculomata, in which the T2R values range from 83 to 290 ms. However, a wide range of T2R values was seen in their study, which was explained as resulting from the metachronic stages of cyst evolution through varied substrates of oedema and/or gliosis with resultant variations in hydration states 16. Our cohort showed a progressive, statistically significant decline in the T2R value over time and as the lesions progressed through the various stages of degeneration. This is due to the increasing albumin and glycoprotein content of the cyst fluid and mineralization of the cyst 12,15,27-29 . We demonstrated a significant relation between the stage of the lesion and the T2R value, particularly when differentiating stage 2 from later stages. The T2R value was highest in stage 2 and declined as the lesion progressed to later stages. Lesions with greater values of T2 relaxation were more hyperintense on T2W images, corresponding to stage 2; the less hyperintense lesions showed lower T2 relaxation values and corresponded to the later stages of cysticercal degeneration 16. However, even at 24 months T2R values from the lesion centre remained higher than normal brain parenchyma. Similarly, the perilesional parenchyma showed high T2R values in the initial MRI due to perilesional oedema resulting from inflammation around the degenerating cysticercus, although vasculitis, infiltration and gliosis may also be responsible 30. As inflammation resolved, T2R values fell over time and with progression from stage 2 to later stages. However, T2R val- ues from the PLP remained significantly higher than normal parenchyma even after oedema resolved, due to gliosis around the lesion. Increased T2 relaxation values reflect the severity of reactive gliosis due to neuronal loss 12. In an earlier study on patients with NCC, T2R values in PLP were higher in patients with visible gliosis on MTI than those without. However, even though the T2R value increased by as much as 25%, the gliosis could not be seen on conventional T2-weighted images 25. The lesion wall appeared relatively hypointense, and had a lower T2R value than the lesion centre or the PLP. The reason for this relative hypointensity is not known, but is thought to be due to paramagnetic substances in the cyst wall 29. In our patients, T2R values in the centre and wall of the lesion declined in the first three months in both the control and treatment groups. After six months, patients receiving albendazole therapy showed a T2R value approaching normal while values in the control group continued to be high, albeit reduced from the baseline. Subsequently — one year from enrolment — the T2R values increased in both groups, possibly indicating the initiation of gliosis. This increase in T2R value was more prominent in the lesion centre, suggesting that gliosis may commence at the centre of the lesion. Two years after enrolment, T2R values from the lesion centre and PLP were lower in the treatment group than in the controls, but this difference did not reach statistical significance. This may indicate an effect of albendazole therapy on the development of gliosis. In a previous paper studying the same cohort of patients, we showed that gliosis in the PLP as visualized on MTI may be influenced by treatment with albendazole, but again did not demonstrate statistical significance 10. However, gliosis is an ongoing process and continues to form even two years after the onset of degeneration. Thus, it is foreseeable that a similar study with imaging after a longer period of follow-up will reveal more gliosis, and better conclusions may be possible. A previous study using MTI and T2 relaxometry to evaluate the PLP in patients with NCC reported an inverse relationship between the T2R value and the magnetization transfer ratio (MTR). In patients with gliosis on MTI, highly significant differences were found between mean T2R values determined from the perilesional and normal contralateral regions. A linear regression fit analysis showed a highly significant inverse correlation between perile347 Quantitative Serial T2 Relaxometry: A Prospective Evaluation in Solitary Cerebral Cysticercosis sional T2 values and MT ratios 25. In our previous paper on MTI in SCCL, the MTR from the lesion increased progressively from the baseline to the 24-month MRI 10, and the present study confirms an inverse relation with T2R time. The highest MTR values have been reported in healing T2 hypointense lesions while vesicular lesions showed the lowest values. Intermediate values were reported in healing hyperintense lesions 31. In view of these findings, the T2R value is expected to decrease with advancing degeneration, which was confirmed in this study. Strong correlations between T2R time and MTR have been reported in the abnormal white matter but not NAWM in the brains of patients with multiple sclerosis 32. Significant inverse correlations were noted between T2R values and MTR (P<0.001) in tuberculomata 33,34. The role of treatment with albendazole in cases of single ring-enhancing lesions continues to be controversial. Our results suggest a possible effect of such treatment on the evolution Atchayaram Nalini of cysticercal single ring-enhancing lesions, as we observed higher T2R values in patients who did not receive albendazole compared to those who did. However, the clinical importance of this radiological finding remains unclear as previous papers studying the same cohort demonstrated that there was no significant difference between the two groups in terms of overall seizure control 35,36. Further studies with longer follow-up may resolve this question. In conclusion, T2R values from SCCL declined on serial MRI over 24 months, and were significantly higher in earlier stages of degeneration. A mild increase after six months may be due to the initiation of gliosis within the SCCL as well as in the PLP. T2R values appear to decline faster in patients who received anthelmintic therapy, but more studies involving a larger number of patients with longer follow-up will probably clearly demonstrate an effect of albendazole treatment on the T2R values from the cysticercal lesion. References 1 Del Brutto OH, Rajshekhar V, White AC Jr., et al. Proposed diagnostic criteria for neurocysticercosis. Neurology. 2001; 57 (2): 177-183. doi: 10.1212/WNL.57.2.177. 2 Rajshekhar V. Etiology and management of single small CT lesions in patients with seizures: understanding a controversy. Acta Neurol Scand. 1991; 84 (6): 465-470. doi: 10.1111/j.1600-0404.1991.tb04996.x. 3 Noujaim SE, Rossi MD, Rao SK, et al. CT and MR imaging of neurocysticercosis. Am J Roentgenol. 1999; 173 (6): 1485-1490. doi: 10.2214/ajr.173.6.10584787. 4 Garcia HH, Del Brutto OH. Imaging findings in neurocysticercosis. Acta Trop. 2003; 87 (1): 71-78. doi: 10.1016/S0001-706X(03)00057-3. 5 de Souza A, Nalini A, Srikanth SG. Solitary cerebral parenchymal cysticercosis: a prospective comparative study with computed tomography and magnetic resonance imaging. Neurol India. 2013; 61 (6): 639-643. doi: 10.4103/0028-3886.125272. 6 de Souza A, Nalini A, Kovoor JM, et al. Natural history of solitary cerebral cysticercosis on serial magnetic resonance imaging and the effect of albendazole therapy on its evolution. J Neurol Sci. 2010; 288 (1-2): 135-141. doi: 10.1016/j.jns.2009.09.018. 7 Pradhan S, Kathuria MK, Gupta RK. Perilesional gliosis and seizure outcome: a study based on magnetization transfer magnetic resonance imaging in patients with neurocysticercosis. Ann Neurol. 2000; 48 (2): 181-187. 8 Gupta RK, Kathuria MK, Pradhan S. Magnetisation transfer magnetic resonance imaging demonstration of perilesional gliosis--relation with epilepsy in treated or healed neurocysticercosis. Lancet. 1999; 354 (9172): 4445. doi: 10.1016/S0140-6736(99)00881-8. 9 Gupta RK, Kathuria MK, Pradhan S. Magnetization transfer MR imaging demonstration of presumed perilesional gliosis around a healed / healing neurocysticercosis and its relationship with epilepsy. ISMRM. 2000. Available at: http://cds.ismrm.org/ismrm-2000/ PDF1/0016.pdf. 348 10 de Souza A, Nalini A, Kovoor JM, et al. Prospective quantitative imaging study by magnetisation transfer for appearance of perilesional gliosis in solitary cerebral cysticercal lesion. Neuroradiol J. 2010; 23 (5): 574-589. 11 Hoehn-Berlage M, Bockhorst K. Quantitative magnetic resonance imaging of rat brain tumors: in vivo NMR relaxometry for the discrimination of normal and pathological tissues. Technol Health Care. 1994; 2 (4): 247-254. 12 Escobar A. The pathology of neurocysticercosis. In: Palacios E, Rodriguez-Carbajal J, Taveras J (eds). Cysticercosis of the central nervous system. Springfield, IL,USA: Charles C Thomas; 1983. p. 27-54. 13 Gadian GD. Magnetic resonance approaches to the identification of focal pathophysiology in children with brain disease. Dev Sci. 2002; 5 (3): 3279-3285. doi: 10.1111/1467-7687.00367. 14 Petropoulos H, Sibbitt WL Jr, Brooks WM. Automated T2 quantitation in neuropsychiatric lupus erythematosus: a marker of active disease. J Magn Reson Imaging. 1999; 9 (1): 39-43. 15 Blackmore CC, Francisca CW, Bryant RG. Magnetic resonance imaging of blood and clots in vitro. Invest Radiol. 1990; 25 (12): 1316-1324. doi: 10.1097/00004424199012000-00009. 16 Jayakumar PN, Srikanth SG, Chandrashekar HS, et al. T2 relaxometry of ring lesions of the brain. Clin Radiol. 2007; 62 (4): 370-375. doi: 10.1016/j.crad.2006.09.017. 17 Bottomley PA, Hardy CJ, Argersinger RE, et al. A review of 1H nuclear magnetic resonance relaxation in pathology: are T1 and T2 diagnostic? Med Phys. 1987; 14 (1): 1-37. doi: 10.1118/1.596111. 18 Poon CS, Henkelman RM. Practical T2 quantification for clinical application. J Magn Reson Imaging. 1992; 2 (5): 541-553. doi: 10.1002/jmri.1880020512. 19 Grünewald RA, Jackson GD, Connelly A, et al. MR detection of hippocampal disease in epilepsy: factors influencing T2 relaxation time. Am J Neuroradiol. 1994; 15 (6): 1149-1156. www.centauro.it The Neuroradiology Journal 27: 339-349, 2014 - doi: 10.15274/NRJ-2014-10054 20 Jackson GD, Connelly A, Duncan JS, et al. Detection of hippocampal pathology in intractable partial epilepsy: increased sensitivity with quantitative magnetic resonance T2 relaxometry. Neurology. 1993; 43 (9): 17931799. doi: 10.1212/WNL.43.9.1793. 21 Bernasconi A, Bernasconi N, Caramanos Z, et al. T2 relaxometry can lateralize mesial temporal lobe epilepsy in patients with normal MRI. Neuroimage. 2000; 12 (6): 739-746. doi: 10.1006/nimg.2000.0724. 22 Namer IJ, Bolo NR, Sellal F, et al. Combined measurements of hippocampal N-acetyl-aspartate and T2 relaxation time in the evaluation of mesial temporal lobe epilepsy: correlation with clinical severity and memory performances. Epilepsia. 1999; 40 (10): 1424-1432. doi: 10.1111/j.1528-1157.1999.tb02015.x. 23 Neema M, Goldberg-Zimring D, et al. 3T MRI relaxometry detects T2 prolongation in the cerebral normalappearing white matter in multiple sclerosis. Neuroimage. 2009; 46 (3): 633-641. doi: 10.1016/j.neuroimage.2009.03.001. 24 Schenck JF, Zimmerman EA. High-field magnetic resonance imaging of brain iron: birth of a biomarker? NMR Biomed. 2004; 17 (7): 433-445. doi: 10.1002/nbm.922. 25 Kumar R, Gupta RK, Rathore RK, et al. Multiparametric quantitation of the perilesional region in patients with healed or healing solitary cysticercus granuloma. Neuroimage. 2002; 15 (4) :1015-1020. doi: 10.1006/ nimg.2001.1036. 26 Pal DK, Carpio A, Sander JW. Neurocysticercosis and epilepsy in developing countries. J Neurol Neurosurg Psychiatry. 2000; 68 (2): 137-143. doi: 10.1136/jnnp.68.2.137. 27 Del Brutto OH, Wadia NH, Dumas M, et al. Proposal of diagnostic criteria for human cysticercosis and neurocysticercosis. J Neurol Sci. 1996; 142 (1-2) :1-6. doi: 10.1016/0022-510X(96)00130-X. 28 Chatterjee KD. Helminthology. In: Parasitology, protozoology and helminthology. 12th ed. Calcutta: Chatterjee Medical Publishers; 1981. p 119. 29 Hussein FMY, Alhajri FA, Buriki KB, et al. Neurocysticercosis in Kuwait: Computerized tomography and magnetic resonance imaging findings. Kuwait Med J. 2003; 35: 187-191. 30 Martinez HR, Rangel-Guerra R, Elizondo G, et al. MR imaging in neurocysticercosis: a study of 56 cases. Am J Neuroradiol. 1989; 10 (5): 1011-1019. 31 Kathuria MK, Gupta RK, Roy R, et al. Measurement of magnetization transfer in different stages of neurocysticercosis. J Magn Reson Imaging. 1998; 8 (2): 473-479. doi: 10.1002/jmri.1880080231. 32 Papanikolaou N, Papadaki E, Karampekios S, et al. T2 relaxation time analysis in patients with multiple sclerosis: correlation with magnetization transfer ratio. Eur Radiol 2004; 14 (1): 115-122. doi: 10.1007/s00330-0031946-0. 33 Vasudev MK, Jayakumar PN, Srikanth SG, et al. Quantitative magnetic resonance techniques in the evaluation of intracranial tuberculomas. Acta Radiol. 2007; 48 (2): 200-206. doi: 10.1080/02841850601067678. 34 Neema M, Stankiewicz J, Arora A, et al. T1- and T2based MRI measures of diffuse gray matter and white matter damage in patients with multiple sclerosis. J Neuroimaging. 2007; 17 (Suppl 1): 16S-21S. doi: 10.1111/j.1552-6569.2007.00131.x. 35 de Souza A, Thennarasu K, Yeshraj G, et al. Randomized controlled trial of albendazole in new onset epilepsy and MRI confirmed solitary cerebral cysticercal lesion: effect on long-term seizure outcome. J Neurol Sci. 2009; 276 (1-2):108-114. doi: 10.1016/j.jns.2008.09.010. 36 de Souza A, Nalini A, Kovoor JM, et al. Perilesional gliosis around solitary cerebral parenchymal cysticerci and long-term seizure outcome: a prospective study using serial magnetization transfer imaging. Epilepsia. 2011; 52 (10): 1918-1927. doi: 10.1111/j.1528-1167.2011.03189.x. Aaron de Souza, MD Department of Neurology Manipal Hospital Goa Dr E Borges Road Dona Paula, Panjim Goa 403004 India Tel.: +91-997-5542655 Fax: +91-832-3002555 E-mail: adesouza1@gmail.com 349 The Neuroradiology Journal 27: 350-355, 2014 - doi: 10.15274/NRJ-2014-10050 www.centauro.it Effect of Electromagnetic Radiation on the Coils Used in Aneurysm Embolization XIANLI LV, ZHONGXUE WU, YOUXIANG LI Department of Interventional Neuroradiology, Beijing Tiantan Hospital and Beijing Neurosurgical Institute, Capital Medical University; Beijing, China Key words: electromagnetic radiation, coils, aneurysm SUMMARY – This study evaluated the effects of electromagnetic radiation in our daily lives on the coils used in aneurysm embolization. Faraday’s electromagnetic induction principle was applied to analyze the effects of electromagnetic radiation on the coils used in aneurysm embolization. To induce a current of 0.5mA in less than 5 mm platinum coils required to stimulate peripheral nerves, the minimum magnetic field will be 0.86 —T. To induce a current of 0.5 mA in platinum coils by a hair dryer, the minimum aneurysm radius is 2.5 mm (5 mm aneurysm). To induce a current of 0.5 mA in platinum coils by a computer or TV, the minimum aneurysm radius is 8.6 mm (approximate 17 mm aneurysm). The minimum magnetic field is much larger than the flux densities produced by computer and TV, while the minimum aneurysm radius is much larger than most aneurysm sizes to levels produced by computer and TV. At present, the effects of electromagnetic radiation in our daily lives on intracranial coils do not produce a harmful reaction. Patients with coiled aneurysm are advised to avoid using hair dryers. This theory needs to be proved by further detailed complex investigations. Doctors should give patients additional instructions before the procedure, depending on this study. Introduction Alternating magnetic fields induce electric fields inside living organisms. Significant progress has been made in describing biological effects resulting from these interactions throughout the world1. Much of this effort has been directed towards electric fields of power frequencies (i.e. 50 Hz and 60 Hz) because the power frequency field in China is 50Hz and in the USA it is 60Hz. Other low frequencies have also been examined and research has been expanded to include magnetic fields, both alone and in conjunction with electric fields2. Although it is now clear that extreme-lowfrequency (ELF) fields (1-300 Hz) can cause biological effects3, the mechanisms of such interactions on the coils used in aneurysm embolization are largely unknown and the clinical implications have yet to be determined. This study evaluates the effects of electromagnetic radiation in our daily lives, such as mobile 350 phones, computers and TV, on the patients treated with aneurysm coiling. Methods Faraday’s law of induction states that timevarying magnetic fields generate electric fields through induction. Faraday’s law: (1) where: - = the induced electromotive force in the opposite direction to the changing rate of the magnetic flux N = the number of turns of the coil ¨ϕ = the magnetic flux change [T⋅S2] ¨t = time[s] Therefore, a conductor exposed to ELF magnetic fields will also be exposed to an induced electric field from this course. For harmonic Xianli Lv, Effect of Electromagnetic Radiation on the Coils Used in Aneurysm Embolization (sinusoidal) Àelds and a circular contour in a homogeneous conductor, the magnitude of the electric Àeld (E) depends on the magnetic flux density (B), field frequency ( f ) derived from Faraday’s law of induction and radius of the induction loop (R) and can be expressed mathematically as 1,4: E = πRfB (2) The induced electric field causes current to flow in any conductive body. These currents, called eddy currents, circulate in closed loops that tend to lie in planes perpendicular to the direction of the magnetic field. The distribution of current in a three-dimensional volume is frequently specified using the current density vector J (A/m2). Ohm’s law relates the induced current density (J ) in an electric field (E) as follows 5: J = σE (3) where the constant proportionality σ is called the electrical conductivity of the medium. The σ of various living tissues lies in the range of 0.01–1.5 S/m6. The σ of the platinum (Pt) coils used for aneurysm coiling is 9.4×106 S/m7. Since σ Pt >>σtissue, it has been found valid to use Eqn. (2) and (3) in calculating the current produced within the intra-aneurysmal platinum coil mass exposed to the external field. The current density is related to other factors: J = σ πRfB (4) The induced current flow in the platinum coil mass circulates in closed loops in planes perpendicular to the direction of the magnetic field. I = JS (5) where I is induced current flow and S is the area of planes perpendicular to the direction of the magnetic field. The lowest thresholds for perception According to the mechanism of indirect interaction of electromagnetic fields 8, the electric charge accumulated on the platinum coils will flow to the tissue of the person. In the frequency less than 100 kHz, the current flow from the platinum coils to the living tissue may result in the stimulation of muscles and/or peripheral nerves. Threshold values for these effects are frequency-dependent, with the lowest threshold for peripheral nerve responses occurring at frequencies between 10 and 100 Hz 8. The 50 percentile threshold currents for their occurrence are given in Table 1. Table 1 Ranges of threshold currents for indirect effects, including children, women and men 8 Indirect effect Threshold current (mA) at frequency 50/60 Hz 1 kHz 100 kHz 1 MHz Touch perception 0.2-0.4 0.4-0.8 25-40 25–40 Pain on finger contact 0.9-1.8 1.6-3.3 33-55 28–50 Painful shock/let-go threshold 8-16 12-24 112-224 Not determined Severe shock/breathing difficulty 12-23 21-41 160-320 Not determined Results The frequencies and magnetic flux density of electromagnetic radiation from hair dryers, computers and TV were reported by a Chinese newsman (Table 2) 9. Magnetic induction of currents can be modeled using a simple global approximation of a coiled aneurysm. A globe with the semi major axes has the same R. The radius of aneurysms usually ranges from 1 mm to 15 mm. For instance, the radiuses of 1 mm, 5 mm, 10 mm, and 15 mm were used for calcu- lation (Table 3). Using equation (4) the maximum electric field Em induced in this model can be shown to occur when the magnetic flux density vector B is horizontal and perpendicular to the front of the coil mass. The actual value of Em is 1.5 × 109 RB [equation (2)], where f is the frequency of the field in Hz. Equation (4) can be used to estimate the induced current density from Em. Using a platinum conductivity of 9.4 × 106 S/m7, the maximum induced current flows are presented in Table 3. 351 Effect of Electromagnetic Radiation on the Coils Used in Aneurysm Embolization A B C D Xianli Lv, Figure 1 A) 3D reconstruction angiogram of the left carotid artery injection. B) Lateral view of the fluoroscopic image after coil embolization. C) Lateral view of the left internal carotid artery angiogram at 54-month follow-up. D) Lateral view of the fluoroscopic image after left internal carotid artery occlusion using coils. E) Lateral view of the left carotid artery angiogram after occlusion of the left internal carotid artery. F) Frontal view of the right internal carotid artery angiogram after occlusion of the left internal carotid artery. 352 www.centauro.it The Neuroradiology Journal 27: 350-355, 2014 - doi: 10.15274/NRJ-2014-10050 E F Table 2 The frequencies and magnetic flux density of electromagnetic radiation from hair dryers, computers and TVs reported by a Chinese newsman 9 Household appliances Frequency (Hz) Magnetic flux density (—T) Hair dryer 50 7.16 Computer (the front 0.5 meters, boot instant) 50 0.17 TV (the front 0.5 meters, boot instant) 0.12 50 Table 3 Maximum induced current densities in coil masses with different radiuses Household appliances Induced electric current in platinum coil masses (mA) R = 1 mm R = 5 mm R = 10 mm R = 15 mm Hair dryer 0.03 4.1 33.2 111.98 Computer (the front 0.5 meters, boot instant) 0.0008 0.10 0.79 2.66 TV (the front 0.5 meters, boot instant) 0.0006 0.07 0.56 1.88 The minimum values It is now possible to estimate the aneurysm radius, R, that will produce internal current flow exceeding the 0.5 mA thresholds discussed above. Aneurysms less than 10 mm in diameter represent more than 80% of all intracranial aneurysms 10. To induce a current of 0.5 mA in platinum coils in R ” 5 mm aneurysm required to stimulate peripheral nerves, the minimum magnetic field will be 0.86 —T. To induce a current of 0.5 mA in platinum coils by a hair dryer, the minimum radius of aneurysm is 2.5 mm (5 mm aneurysm). To induce a current of 0.5 mA in platinum coils by a computer or TV, the minimum radius of aneurysm is 8.6 mm (approximate 17 mm aneurysm). The minimum magnetic field is much larger than the flux densities produced by computer and TV, while the minimum aneurysm radius is much larger than most aneurysm sizes to levels produced by computer and TV. 353 Effect of Electromagnetic Radiation on the Coils Used in Aneurysm Embolization Case illustration A 48-year-old woman presented with ophthalmoplegia on the left side. Cerebral angiograms showed a large intracavernous internal carotid artery aneurysm. The aneurysm size was 21 mm × 21 mm × 20 mm (Figure1A). The aneurysm was treated with coil embolization (Figure 1B). This patient complained of left facial numbness, headache and extraocular muscle spasm while booting a computer and TV. A control angiogram was obtained at 54-month follow-up and showed a neck remnant of the aneurysm (Figure 1C). The aneurysm was treated with parent internal carotid artery occlusion using coils (Figure 1D-F). Discussion Intracranial aneurysms are common, with a prevalence of 0.5% to 6% in adults, according to angiography and autopsy studies 11. Endovascular coil embolization is an option for treatment of ruptured and unruptured intracranial aneurysms 12. In daily life, we are surrounded by electromagnetic radiation, mainly from household appliances such as hair dryers, computers, TVs and mobile phones. According to the principle of electromagnetic induction, platinum coils exposure to ELF electric and magnetic fields may involve effects within the coils caused by the induction of internal current flow 3. This problem has not been noted and investigated previously. In this study, we found that induced current flow in most platinum coils produced by exposure to daily ELF is much lower than the reported threshold level. Most coiled aneurysms (more than 80%) cannot produce induced current flow to stimulate the nervous tissue using computers and TVs. Coiled aneurysms greater than 17 mm may produce current using computers and TVs sufficient to excite the nervous tissue. We think that nerves must not be irritated under the above conditions because such a case is very rare in clinical practice. We only encountered one patient complaining of neurological symptoms coinciding with computer and TV booting in five years (< 1/1000). We thought this might be due to the aneurysm wall, pial membrane, cerebrospinal fluid, and nerve sheath composing a barrier between the platinum coils and the nerve tissue. But a large aneurysm close to the nerve tissue, as in our case of a large cavernous aneurysm compressing the cerebral nerve, would 354 Xianli Lv, produce induced current to stimulate the cerebral nerve. The threshold recommended is 50% percentile values, which indicates that nerve stimulating symptoms must not appear even when the threshold is reached. The thresholds for effects (perception, shock, etc.) are generally higher for men than for women and children, though there are also individual differences 13. Ours was a female patient. Although there is no nerve tissue stimulation symptoms in patients exposed to fields above the threshold, this may explain headaches when the patients use a computer or watch TV. Mobile phone For frequencies below about 100 kHz, an established interaction mechanism is the stimulation of muscles and/or peripheral nerves by induced currents 8. For higher frequencies such as mobile phones with a 1800-2000MHz frequency and 55.29 —W/cm2 power density 14 , thermal interactions predominate 15. At the lower frequencies, much less of the electromagnetic field is absorbed by biological systems. Thermal interactions occur at energy levels much higher than interactions due to induced currents. Therefore, thermal interactions are usually of little interest for fields at levels to which people are exposed 8. Limitations An example of a valid use of Eq. (4) is a conductive solution in a cylindrical dish in a uniform magnetic Àeld parallel to the cylinder axis 2 . The model is too simple and reflection and refraction of electromagnetic radiation caused by human tissue is not considered. The thermal effect caused by interaction between electromagnetic radiation and coils is also not considered. But in scientific research, simplifying and idealizing will make the problem easier to explain and understand. This result needs further detailed calculation and simulation analysis, complex investigations to prove it. Despite this limitation, it appears reasonable to quantify the exposure conditions in terms of dosimetric measures that are well grounded in physics, namely the induced electric Àeld and current. Evaluations of induced Àelds and currents are useful in comparing previous experimental results. Their parameters can also be used in scaling exposure levels of human beings. Fur- www.centauro.it The Neuroradiology Journal 27: 350-355, 2014 - doi: 10.15274/NRJ-2014-10050 thermore, there are well-known physiologic effects that are related to the induced electric Àelds and currents in conductors, which need to be considered for occupational exposure to strong Àelds. Protective measures against electromagnetic field coupling In virtually all cases there is no practical way of shielding against exposure to ELF magnetic fields. Thus the only practical protective method is to limit exposure, either by limiting access of personnel to areas where magnetic fields are strong or by limiting to safe levels all magnetic fields to which people could be exposed. Conclusions At present, the effects of electromagnetic radiation in our daily lives on intracranial coils do not produce a harmful reaction of the human body. Patients with coiled aneurysm have been advised to avoid using hair dryer. This theory needs to be proved by further detailed complex investigations. Doctors may give patients additional instructions after the procedure, depending on this study. Acknowledgements We would like to thank the two anonymous reviewers and the editor for their comments. References 1 Stuchly MA, Xi W. Modelling induced currents in biological cells exposed to low-frequency magnetic fields. Phys Med Biol. 1994; 39 (9): 1319-1330. doi: 10.1088/00319155/39/9/001. 2 Polk C. Physical mechanisms for biological effects of low field intensity ELF magnetic fields biological effects of magnetic and electromagnetic fields. In: Ueno S, ed. Biological effects of magnetic and electromagnetic fields. New York: Plenum Press; 1996. p 63-83. doi: 10.1007/978-0-585-31661-1_5. 3 Tenforde TS, Kaune WT. Interaction of extremely low frequency electric and magnetic fields with humans. Health Phys. 1987; 53 (6): 585-606. doi: 10.1097/ 00004032-198712000-00002. 4 Stuchly MA. Low-frequency magnetic fields: dosimetry, cellular, and animal effects. In: Bronzino JD. ed. The biomedical engineering handbook, second edition. Boca Raton: CRC Press LLC; 1999. 5 International Commission on Non-Ionizing Radiation Protection. Guidelines for limiting exposure to timevarying electric and magnetic fields (1 HZ to 100 kHZ). Health Phys. 2010; 99 (6): 818-836. 6 Anderson LE, Kaune WT. Nonionizing radiation protection. Electric and magnetic fields at extremely low frequencies. WHO Reg Publ Eur Ser. 1988; 25: 175-243. 7 Technical data for Platinum. http://periodictable.com/ Elements/078/data.html. 8 UNEP/WHO/IRPA. Electromagnetic fields (300Hz300GHz). Environmental health criteria 137. Geneva, World Health Organization: United nations environmental programme/World Health Organization/International Radiation Protection Association; 1993. 9 Ma Jing. Household appliances radiation. Beijing TV station. 10 Beck J, Rohde S, Berkefeld J, et al. Size and location of ruptured and unruptured intracranial aneurysms measured by 3-dimensional rotational angiography. Surg Neurol. 2006; 65 (1): 18-25; discussion 25-7. doi: 10.1016/j.surneu.2005.05.019. 11 Schievink WI. Intracranial aneurysms. N Engl J Med. 1997; 336 (1): 28-40. doi: 10.1056/NEJM199701023 360106. 12 Johnston SC, Higashida RT, Barrow DL, et al. Recommendations for the endovascular treatment of intracranial aneurysms: a statement for healthcare profession- als from the Committee on Cerebrovascular Imaging of the American Heart Association Council on Cardiovascular Radiology. Stroke. 2002; 33 (10): 2536-2544. doi: 10.1161/01.STR.0000034708.66191.7D. 13 Bernhart JH. On the rating of human exposition to electric and magnetic fields with frequencies below 100 kHz. Ispra, Italy: Commission of European Communities, Joint Research Center; 1983. 14 The electromagnetic wave detection. http://www.library.com.tw/emf/check.htm. 15 International Commission on Non-Ionizing Radiation Protection. Guidelines for limiting exposure to timevarying electric, magnetic, and electromagnetic fields (up to 300 GHz). Health Physics. 1998; 74 (4): 494-522. Youxiang Li, MD Beijing Neurosurgical Institute Beijing Tiantan Hospital Capital Medical University Beijing Neurosurgical Institute No.6, Tiantan Xili, Chongwen Beijing, 100050 P.R.China Tel.: 86-10-67098850 Fax: 86-10-67018349 E-mail: lvxianli000@163.com 355 The Neuroradiology Journal 27: 356-360, 2014 - doi: 10.15274/NRJ-2014-10035 www.centauro.it Early Endovascular Treatment of Aneurysmal Subarachnoid Hemorrhage Complicated by Neurogenic Pulmonary Edema and Takotsubo-Like Cardiomyopathy ANDREA MANTO1, ANGELA DE GENNARO2, GAETANA MANZO2, ANTONIETTA SERINO1, GAETANO QUARANTA3, CLAUDIA CANCELLA4 1 Neuroradiology Unit, 3 Cardiology Unit, 4 Anesthesia and Reanimation Unit, Umberto I Hospital; Nocera Inferiore, Salerno, Italy 2 Department of Biomorphological and Functional Sciences, Federico II University; Naples, Italy Key words: endovascular treatment, aneurysmal subarachnoid hemorrhage, Takotsubo cardiomyopathy, neurogenic pulmonary edema SUMMARY – Aneurysmal subarachnoid hemorrhage (SAH) may be associated with acute cardiopulmonary complications, like neurogenic pulmonary edema (NPE) and Takotsubo-like cardiomyopathy (TCM). These dysfunctions seem to result from a neurogenically induced overstimulation of the sympathetic nervous system through the brain-heart connection and often complicate poor grade aneurysmal SAH. The optimal treatment modality and timing of intervention in this clinical setting have not been established yet. Early endovascular therapy seems to be the fitting treatment in this particular group of patients, in which surgical clipping is often contraindicated due to the added risk of craniotomy. Herein we describe the case of a woman admitted to the emergency department with aneurysmal SAH complicated by NPE-TCM, in which early endovascular coiling was successfully performed. Our case, characterized by a favorable outcome, further supports the evidence that early endovascular treatment should be preferred in this peculiar clinical scenario. Introduction Cardiopulmonary dysfunctions, in particular neurogenic pulmonary edema (NPE) and Takotsubo-like cardiomyopathy (TCM), may complicate aneurysmal subarachnoid hemorrhage (SAH) 1,2. NPE is appreciated in 2-29% of SAH patients and is more frequently associated with poor grade SAH 2. TCM is reported to complicate 4-15% of SAH 3. It is still debated in literature whether TCM should be considered a distinct clinical entity or a manifestation of neurogenic stunned myocardium 4, so we use the term “Takotsubo-like cardiomyopathy” to differentiate it from the idiopathic form, that by definition should not have an extracardiac cause. Although NPE and TCM have been considered separate entities in the past, a shared common pathophysiologic mechanism was 356 recently demonstrated 5. A neurogenically induced overstimulation of the sympathetic nervous system through the brain-heart connection may cause endothelial damage, increased pulmonary vascular permeability and a characteristic myocardial injury through an excitotoxic mechanism 5. No evidence-based guidelines indicating the optimal modality and timing of treatment for SAH patients complicated with NPE-TCM have been developed yet. However, endovascular therapy was recently recognized as the preferable treatment in these patients 6. Moreover, medical treatment, performed to prevent and attenuate cerebral vasospasm, is crucial in these patients. Currently, triple-H therapy represents the most widely adopted strategy. However, implementation of this therapy in patients with neurogenic pulmonary edema and Takotsubo-like cardiomyopathy is contraindicated. Andrea Manto Early Endovascular Treatment of Aneurysmal Subarachnoid Hemorrhage Complicated by Neurogenic Pulmonary Edema... Figure 1 Unenhanced brain CT shows SAH involving the interhemispheric fissure and right sylvian fissure; hemorrhage into the left lateral ventricle is associated. Figure 2 CT angiogram demonstrates an aneurysm (maximum diameter: 6 mm, neck: 3 mm) of the M1 segment of the right middle cerebral artery. Figure 3 Left ventriculography reveals the ampulla-shaped morphology of the left ventricle, a characteristic sign of Takotsubo-like cardiomyopathy. Hence, prompt endovascular treatment to minimize the risk of secondary SAH and treat active vasospasm is beneficial in these patients. Herein we describe a case of aneurysmal SAH complicated by NPE-TCM, in which early endovascular treatment was performed with a favorable outcome. Case Report A 42-year-old woman was admitted to the emergency department with loss of consciousness (grade 6 on the Glasgow Coma Scale). Her relatives reported she had been suffering from headache for two days. 357 Early Endovascular Treatment of Aneurysmal Subarachnoid Hemorrhage Complicated by Neurogenic Pulmonary Edema... Unenhanced brain CT, performed on a 64 slice scanner (Aquilion, Toshiba), showed SAH involving the right frontal and parietal sulci, the interhemispheric fissure, the right sylvian fissure, pre-pontine and peri-mesencephalic cisterns, along with left lateral intraventricular hemorrhage (Fisher scale grade 4) (Figure 1). CT angiogram revealed a ruptured aneurysm of the right M1 segment with a maximum diameter of 6 mm and a neck of 3 mm (Figure 2). Two hours later the patient developed dyspnea and was found to be hypoxemic; a chest radiograph revealed pulmonary edema and tracheal intubation was performed. On ECG, atrial fibrillation was present and severe systolic dysfunction (EF 25-30%) along with akinesis of the apex and of the distal left ventricular walls were demonstrated on echocardiography. Troponin I and BNP were found to be 1.25 ng/ ml (nv < 0.06 ng/ml) and 777.6 pg/ml (nv < 100 pg/ml), respectively. Consequently, acute heart failure was diagnosed. Medical management was started (ACEi, β blockers, diuretics and vasopressin) and the patient’s cardiac output progressively improved. On the second day of admission, her vital parameters were made stable, so coronarography could be performed to exclude an acute myocardial infarction. Coronary arteries were found to be unharmed, while left ventriculography demonstrated the typical ampulla-shaped morphology of the left ventricle (Figure 3), leading to a diagnosis of Takotsubo-like cardiomyopathy. Immediately afterwards, endovascular embolization of the ruptured aneurysm with detachable coils could be performed (Figures 4 and 5). An unenhanced head CT, performed on the first postoperative day, revealed the success of the procedure without any complications. An echocardiographic control, performed four days after the intervention, showed a complete recovery of her cardiac function (EF 65%), and a chest radiograph, obtained on the same day, revealed an evident reduction of the signs of pulmonary edema. She was finally discharged on the 14th postoperative day with a favorable outcome. Discussion SAH-induced cardiopulmonary complications are thought to result from an increased central sympathetic activity. In particular, neurogenic pulmonary edema (NPE) seems to originate from a lesion around 358 Andrea Manto the medulla oblongata: compression of the dorsal and solitary tract nuclei suppressing sympathetic activity may cause acute changes in pulmonary vascular permeability 2. Indeed, NPE is more frequently associated with poor grade SAH patients with ruptured posterior circulation aneurysms 7. Takotsubo-like cardiomyopathy (TCM), characterized by reversible left ventricular dysfunction producing a typical ampulla-shaped morphology on the ventriculogram, ECG abnormalities, mildly increased cardiac markers and an absence of coronary artery disease 4, may result from a neurogenically induced overstimulation of the sympathetic nervous system through the brain-heart connection 5. The severe stress induced by SAH may promote the release of corticotropin releasing factor from the hypothalamus, causing an increase in plasma catecholamines and a cardiac hypersensitivity to sympathetic stimulation 8. Myocardial contraction band necrosis is the characteristic anatomopathological result of this overstimulation, been the proof of a sympathetic discharge. As a consequence of the common pathogenesis, i.e. neurogenic overstimulation of the sympathetic nervous system, SAH patients complicated with NPE tend to develop concomitant TCM more frequently 2. Whether the treatment for SAH patients complicated with NPE-TCM should be surgical or endovascular and if the intervention should be performed in the acute stage or should be delayed, is not still clear. Aneurysm clipping versus coiling does not have a differential effect on the risk of troponin release and left ventricular dysfunction after SAH, so the choice of the modality of aneurysm treatment does not affect the risk of developing cardiac injury 9. Yabumoto et al. have insisted on the early surgical management of the ruptured aneurysm and symptomatic treatment of NPE, claiming that NPE should not be an obstacle to radical intervention when cardiorespiratory control can maintain the minimal anesthetic limit 10. However, several recent reports indicated an early endovascular procedure as the treatment of choice in these patients, who are often considered poor surgical candidates, reporting a good prognosis 2,6,11. The advantages of endovascular treatment are the minor invasivity, the immediate possibility of performance after the diagnostic angiogram and the shorter duration of the procedure. The timing of the intervention is related to the patient’s cardiopulmonary sta- www.centauro.it The Neuroradiology Journal 27: 356-360, 2014 - doi: 10.15274/NRJ-2014-10035 Figure 4 Digital subtraction angiography (DSA) (oblique projection) confirms the aneurysm of the right M1, well demonstrating its size, orientation and neck. Figure 5 DSA (same projection) testifies the success of the embolizing procedure, showing complete obliteration of the lumen of the aneurysm by detachable coils. tus; the procedure should be delayed until hemodynamic stability is achieved. If patients considered unfit for endovascular therapy may be treated with surgery immediately or if the intervention should be deferred for at least two weeks remains unclear 6. This clinical scenario is complicated further by the fact that in approximately half of the patients with NPE-TCM neither surgical nor endovascular intervention in the acute phase may be feasible 6. Medical treatment is based on the prevention and control of cerebral vasospasm, which is estimated to complicate up to 70% of all SAH and remains a major cause of morbidity and mortality. However, much controversy exists generally over the prevention and pharmacological treatment of this fearful complication. A general consensus exists only for the oral administration of nimodipine (a calcium channel blocker) to all patients suffering from SAH to improve the neurological outcome, without an effect on cerebral vasospasm 12. The use of triple-H therapy (hypertension, hypervolemia and hemodilution) in the prevention and treatment of cerebral vasospasm is still controversial 12. This strategy is intended to increase cerebral blood flow through the expansion of intravascular volume and the reduction of blood viscosity and is performed after the aneurysm treatment. Hypertension may be achieved by volume expansion alone or with the addition of vasopressor medications, such as phenylephrine or dopamine. Hemodilution is based on the achievement of a hematocrit goal of 30-35%, which is considered an optimal balance between oxygen-carrying capacity and blood viscosity 13. Recent American Heart/ Stroke Association guidelines suggest maintenance of euvolemia for vasospasm prevention and recommend induced hypertension for patients with active cerebral vasospasm with a normal cardiopulmonary status 13. Indeed, this therapeutic strategy is contraindicated in patients with SAH and cardiopulmonary dysfunction, as the cardiocirculatory status may be worsened by induced hypertension. Indeed, cardiogenic shock with stunned myocardium may be induced by triple H therapy even in patients with normal cardiopulmonary function 14. Consequently, balloon angioplasty is considered the fitting treatment of cerebral vasospasm in this particular clinical scenario 13. In our case, the early recognition and treatment of NPE-TCM was crucial for correct planning of the endovascular procedure, successfully performed after the improvement in cardiac function. We suggest that if SAH-related, reversible cardiopulmonary abnormalities are promptly detected, an early endovascular procedure should be preferred once a stable hemodynamic function or initial signs of recovery are demonstrated, leading to the best prognosis. 359 Early Endovascular Treatment of Aneurysmal Subarachnoid Hemorrhage Complicated by Neurogenic Pulmonary Edema... Conclusions NPE and TCM represent possible complications of aneurysmal SAH. An early diagnosis and prompt treatment of these reversible disorders are essential for Andrea Manto a good prognosis and for the correct planning of the aneurysmal treatment. We suggest early endovascular embolization as the most appropriate procedure in SAH patients complicated with NPE-TCM, reporting an excellent prognosis in our case. References 1 Mayer SA, Lin J, Homma S, et al. Myocardial injury and left ventricular performance after subarachnoid hemorrhage. Stroke. 1999; 30 (4): 780-786. doi: 10.1161/01. STR.30.4.780. 2 Inamasu J, Nakatsukasa M, Mayanagi K, et al. Subarachnoid hemorrhage complicated with neurogenic pulmonary edema and Takotsubo-like cardiomyopathy. Neurol Med Chir (Tokyo). 2012; 52 (2): 49-55. doi: 10.2176/nmc.52.49. 3 Ako J, Honda Y, Fitzgerald PJ. Takotsubo-like left ventricular dysfunction. Circulation. 2003; 108 (23): e158. doi: 10.1161/01.CIR.0000102942.24174.AE. 4 Abe Y, Kondo M. Apical ballooning of the left ventricle: a distinct entity? Heart. 2003; 89 (9): 974-976. doi: 10.1136/heart.89.9.974. 5 Lee VH, Oh JK, Mulvagh SL, et al. Mechanism in neurogenic stress cardiomyopathy after aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2006; 5 (3): 243249. doi: 10.1385/NCC:5:3:243. 6 Jain R, Deveikis J, Thompson BG. Management of patients with stunned myocardium associated with subarachnoid hemorrhage. Am J Neuroradiol. 2004; 25 (1): 126-129. 7 Ochiai H, Yamakawa Y, Kubota E. Deformation of the ventrolateral medulla oblongata by subarachnoid hemorrhage from ruptured vertebral artery aneurysms causes neurogenic pulmonary edema. Neurol Med Chir (Tokyo). 2001; 41 (11): 529-534; discussion 534-535. doi: 10.2176/nmc.41.529. 8 Masuda T, Sato K, Yamamoto S, et al. Sympathetic nervous activation and myocardial damage immediately after subarachnoid hemorrhage in a unique animal model. Stroke. 2002; 33: 1671-1676. doi: 10.1161/01. STR.0000016327.74392.02. 9 Miss JC, Kopelnik A, Fisher LA, et al. Cardiac injury after subarachnoid hemorrhage is independent of the type of aneurysm therapy. Neurosurgery. 2004; 55 (6): 1244-1251; discussion 1250-1251. doi: 10.1227/01. NEU.0000143165.50444.7F. 10 Yabumoto M, Kuriyama T, Iwamoto M, et al. Neurogenic pulmonary edema associated with ruptured intracranial aneurysm: case report. Neurosurgery. 1986; 19 (2): 300-304. doi: 10.1227/00006123-198608000-00025. 360 11 Meguro T, Terada K, Hirotsune N, et al. Early embolization for ruptured aneurysm in acute stage of subarachnoid hemorrhage with neurogenic pulmonary edema. Interv Neuroradiol. 2007; 13 (Suppl. 1): 170-173. 12 Connolly ES, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2012; 43 (6): 1711-1737. doi: 10.1161/STR.0b013e3182587839. 13 Adamczyk P, He S, Amar AP, et al. Medical management of cerebral vasospasm following aneurysmal subarachnoid hemorrhage: a review of current and emerging therapeutic interventions. Neurol Res Int. 2013; 2013: 462491. doi: 10.1155/2013/462491. 14 Taccone FS, Lubicz B, Piagnerelli M, et al. Cardiogenic shock with stunned myocardium during triple-H therapy treated with intra-aortic balloon pump counterpulsation. Neurocrit Care. 2009; 10 (1): 76-82. doi: 10.1007/ s12028-008-9135-2. Andrea Manto, MD Department of Neuroradiology Umberto I Hospital Viale S. Francesco 2 Nocera Inferiore 84014 Salerno, Italy Tel.: +390819213875 Fax: +390819213874 E-mail: and.manto@libero.it The Neuroradiology Journal 27: 361-364, 2014 - doi: 10.15274/NRJ-2014-10039 www.centauro.it An Unusual Case of Isolated Hypoglossal Nerve Palsy Secondary to Osteophytic Projection from the Atlanto-Occipital Joint SATYA NARAYANA PATRO1, CARLOS TORRES1, ROY RIASCOS2 1 2 Department of Radiology, University of Ottawa, The Ottawa Hospital; Ottawa, ON, Canada Department of Radiology, The University of Texas Medical Branch; Galveston, TX, USA Key words: hypoglossal nerve palsy, CT, MRI, osteophyte, atlanto-occipital joint SUMMARY – We describe an unusual and rare case of isolated left hypoglossal nerve palsy secondary to compression from a prominent degenerative osteophyte from the left atlanto-occipital joint. The hypoglossal nerve is a purely motor cranial nerve innervating the tongue musculature. Palsy of the hypoglossal nerve is frequently associated with other cranial nerve palsies and can be related to vascular, neoplastic, infectious or traumatic conditions. Isolated hypoglossal nerve palsy is quite rare and very few cases have been reported in the literature to date. Introduction Hypoglossal nerve palsy is not uncommon and is frequently associated with other cranial nerve palsies. However, isolated hypoglossal nerve palsy is rare and very few cases have been reported in the literature to date. Herein, we describe a rare case of isolated left hypoglossal nerve palsy secondary to compression from a prominent osteophyte arising from the left atlanto-occipital joint. Case Report A 92-year-old right-handed woman presented with tongue weakness and slurred speech. She also complained of heaviness of the tongue with some difficulty in swallowing solid food with occasional choking. She had no history of headache, neck pain, cervical radiculopathy or visual disturbance. She could walk unaided and did not have balance problems. The patient’s medical history revealed diabetes mellitus and hypertension. On examination, the patient had atrophy of the left side of the tongue and deviation of the tongue to the right side. The other cranial nerves were intact. Based on the patient’s clinical presentation and neurological examination, the diagnosis of isolated left 12th cranial nerve palsy was established. The case was discussed with a neuroradiologist who recommended a gadoliniumenhanced MRI of the brain with emphasis on the skull base to rule out the most common causes of isolated hypoglossal nerve palsy such as: medullary infarct involving the hypoglossal nucleus, and primary or secondary tumors potentially compressing one of the segments of the 12th cranial nerve. MRI of the skull base showed asymmetric tongue proper, slightly smaller on the left side with associated subtle fatty infiltration and T2W hyperintensity suggestive of denervation atrophy (Figure 1A). In particular, no skull base mass lesion was identified and there was no evidence of a medullary lesion or abnormality. Close observation of the skull base revealed a bony osteophyte projecting ventral to the hypoglossal canal, in a position to compress the nerve (Figure 1B,C). A high resolution CT scan of the skull base was performed to confirm the MRI finding. It showed an osteophyte arising from the left lateral mass of the C1 vertebral body, projecting ventral to the left hypoglossal canal (Figure 2). Of note, there was no relevant 361 An Unusual Case of Isolated Hypoglossal Nerve Palsy Secondary to Osteophytic Projection from the Atlanto-Occipital... A B Satya Narayana Patro C Figure 1 A) Axial T2W fat sat image of skull base and oral cavity reveals atrophy and T2W hyperintensity in the left tongue proper (white arrows). B,C) Axial and coronal T1W post-contrast images demonstrate hypointense osteophyte (dotted black arrow) projecting to the ventral aspect of the left hypoglossal canal (dotted white arrow). A B C Figure 2 A,B,C) High resolution CT of the skull base in the axial, coronal and sagittal planes shows osteophyte (black arrow) from the left lateral mass of C1 projecting to the ventral aspect of the left hypoglossal canal (white arrow). O: Occipital condyle, C1: lateral mass of C1. occupational hazard that could explain this finding. Considering the patient’s age and the iatrogenic risk, the neurosurgery service recommended not to operate on this patient. Discussion The hypoglossal nerve is a purely motor cranial nerve innervating the intrinsic and extrinsic muscles of the tongue. It can be divided into five anatomical segments. The hypoglossal nucleus (medullary/nuclear segment) is situated in the hypoglossal trigone in the floor of the fourth ventricle and has a spinal extension. The hypoglossal nerve fibers come out of the 362 medulla as three to 15 roots between the inferior olivary nucleus and the pyramid. These roots fuse together in the medullary cistern to form the hypoglossal nerve, which passes anterolaterally and inferiorly to enter the hypoglossal canal (extramedullary/intracranial segment). The nerve then travels through the bony hypoglossal canal (basicranial segment) to exit out of the skull base. It runs anteriorly and inferiorly between the internal carotid artery and the internal jugular vein along with the glossopharyngeal and vagus nerves (nasopharyngeal/oropharyngeal carotid space segment), and finally courses anteriorly and superiorly into the tongue (sublingual segment). The hypoglossal nerve receives fibers from the www.centauro.it The Neuroradiology Journal 27: 361-364, 2014 - doi: 10.15274/NRJ-2014-10039 C-1 nerve which then exit to form the superior root of the ansa cervicalis 1. Palsy of the hypoglossal nerve is uncommon and is frequently associated with other cranial nerve palsies (IX, X and XI) 2. There are multiple causes of hypoglossal nerve palsy, which can be divided into various subgroups such as (i) Vascular: consisting of medullary infarct with involvement of the hypoglossal nucleus, AV fistula, vertebral and carotid artery dissection/aneurysm and vasculitis, (ii) Neoplasm: consisting of metastatic lesions of the skull base, glioma, meningioma and glomus tumor, (iii) Infectious/inflammatory: consisting of acute poliomyelitis, retropharyngeal infection, sarcoidosis, (iv) Trauma: consisting of fracture of the occipital condyle and odontoid process subluxation, (v) Autoimmune: consisting of diabetes and multiple sclerosis. Out of all these, malignant tumors are the most common cause of hypoglossal nerve palsy 3,4. The etiologies of hypoglossal nerve palsy can also be divided on the basis of the anatomical segments involved, as described by Thompson and Smoker 1. Isolated hypoglossal nerve palsy is quite rare and represents a diagnostic challenge in everyday clinical practice. It has been associated with various pathologies such as hypoglossal nerve schwannomas, dural arteriovenous fistulas, enlarged emissary veins of the hypoglossal canal, aneurysms of the stump of a persistent hypoglossal artery, occipital condyle fractures, arachnoid cysts, juxta facet synovial cyst, metastatic lesions to the skull base, internal carotid and vertebral artery dissections, cervical rheumatoid arthritis and Arnold-Chiari malformation 5-7. In certain cases, no apparent cause was found. Cervical osteophytes are commonly noted in diseases such as diffuse idiopathic skeletal hyperostosis (Forestier’s disease), ankylosing spondylitis, and cervical spondylosis. Degenerative changes in the cervical spine are very common in the elderly population and usually present with neck and radicular pain due to compression of the radicular nerves. Sometimes the cervical osteophytes can lead to unusual presentations, like vertebrobasilar insufficiency, dysphagia, cough and aspiration 8,9. To the best of our knowledge, only two cases of isolated hypoglossal nerve palsy secondary to an atlanto-occipital joint osteophyte have been reported in the literature 10. Our case demonstrates an osteophyte from the atlanto-occipital joint projecting into the external orifice of the hypoglossal canal, resulting in mechanical Figure 3 Cartoon illustrating the different segments of the hypoglossal nerve from the medulla to the tongue. Focal magnified view shows a large osteophyte compressing the basicranial segment of the left hypoglossal nerve. compression of the basicranial segment of the hypoglossal nerve (Figure 3). Systematic imaging should be performed while evaluating the cause of hypoglossal nerve pathology. MRI with contrast of the skull base and brain focusing on the hypoglossal nerve and its course is the most commonly performed examination. In the case of a suspected vascular cause, an MR angiogram is usually helpful to delineate various entities such as flameshaped arterial occlusion, double arterial lumen, or aneurysmal dilation of the carotid of vertebral artery. High resolution basicranial CT is the preferred imaging modality in evaluating osseous pathologies such as degenerative osteoarticular disease or cervico-occipital hinge trauma etiology 1. Conclusion Cervical degenerative osteophytes are very common. The clinical presentation of patients with cervical osteophytosis is quite broad and ranges from common symptoms like radicular and neck pain to unusual presentations like vertebrobasilar insufficiency. An isolated hy363 An Unusual Case of Isolated Hypoglossal Nerve Palsy Secondary to Osteophytic Projection from the Atlanto-Occipital... poglossal nerve palsy caused by an osteophytic projection is extremely rare. One such case is presented here. The combination of MRI and Satya Narayana Patro high resolution CT scan of the skull base is key to diagnose this rare cause of 12th cranial nerve palsy. References 1 Thompson EO, Smoker WR. Hypoglossal nerve palsy: a segmental approach. Radiographics. 1994; 14 (5): 939958. doi: 10.1148/radiographics.14.5.7991825. 2 Combarros O, Alvarez de Arcaya A, Berciano J, et al. Isolated unilateral hypoglossal nerve palsy: nine cases. J Neurol. 1998; 245 (2): 98-100. doi: 10.1007/ s004150050185. 3 Pedro Alves. Imaging the hypoglossal nerve. Eur J Radiol. 2010; 74 (2): 368-377. doi: 10.1016/j.ejrad.2009. 08.028. 4 Keane J. Twelfth-nerve palsy. Analysis of 100 cases. Arch Neurol. 1996; 53 (6): 561-566. doi: 10.1001/archneur.1996.00550060105023. 5 Mujic A, Hunn A, Liddell J, et al. Isolated unilateral hypoglossal nerve paralysis caused by an atlanto-occipital joint synovial cyst. J Clin Neurosci. 2003; 10 (4): 492-495. doi: 10.1016/S0967-5868(03)00083-3. 6 Omura S, Nakajima Y, Kobayashi S, et al. Oral manifestations and differential diagnosis of isolated hypoglossal nerve palsy: report of two cases. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1997; 84 (6): 635-640. doi: 10.1016/S1079-2104(97)90365-6. 7 Demisch S, Lindner A, Beck R, et al. The forgotten condyle: delayed hypoglossal nerve palsy caused by fracture of the occipital condyle. Clin Neurol Neurosurg. 1998; 100 (1): 44-45. doi: 10.1016/S0303-8467(97)00111-X. 8 Mourand I, Azakri S, Boniface G, et al. Teaching NeuroImages: intermittent symptomatic occlusion of the vertebral artery caused by a cervical osteophyte. Neurology. 2013; 80 (5): e54. doi: 10.1212/WNL.0b013e31827 f0eeb. 364 9 Lee TH, Lee JS. High-resolution manometry for oropharyngeal dysphagia in a patient with large cervical osteophytes. J Neurogastroenterol Motil. 2012; 18 (3): 338339. doi: 10.5056/jnm.2012.18.3.338. 10 Patron V, Roudaut PY, Lerat J, et al. Isolated hypoglossal palsy due to cervical osteophyte. Eur Ann Otorhinolaryngol Head Neck Dis. 2012; 129 (1): 44-46. doi: 10. 1016/j.anorl.2011.01.006. Carlos Torres, MD Assistant Professor of Radiology Program Director Neuroradiology Department of Radiology University of Ottawa The Ottawa Hospital 1053 Carling Avenue, K1Y 4E9 Ottawa, ON, Canada Tel.: +1(613) 737-8899 Ext:12036 Fax: +1 (613) 737-8207 E-mail: catorres@toh.on.ca The Neuroradiology Journal 27: 365-367, 2014 - doi: 10.15274/NRJ-2014-10043 www.centauro.it Letter to the Editor CT Angiography Source-Images and CT Perfusion: Are They Complementary Tools for Ischemic Stroke Evaluation? Response to: Reliability of CT Perfusion in the Evaluation of Ischaemic Penumbra. NICOLA MORELLI1,2, EUGENIA ROTA1, PAOLO IMMOVILLI1, ILARIA IAFELICE1, EMANUELE MICHIELETTI2, DONATA GUIDETTI1, JOHN MORELLI3 Neurology Unit, Guglielmo da Saliceto Hospital; Piacenza, Italy Radiology Unit, Guglielmo da Saliceto Hospital; Piacenza, Italy 3 Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine; Baltimore, MD, USA 1 2 Key words: CT perfusion, CT angiography, source-images ischemic stroke We read with great interest the study of Alves et al. 1 demonstrating that the qualitative evaluation of CBV and MTT maps obtained with computed tomography perfusion (CTP) imaging may overestimate the true ischemic penumbra in ischemic stroke. The authors conclude that CBV abnormalities are reliable markers for irreversible ischemia. They demonstrate that CBV may, rarely, overestimate the ischemic “core” and that MTT tends to overestimate the extent of final infarct areas, as these values do not differentiate the true “at risk” penumbra from benign oligemia. The authors also highlight some important methodological considerations related to the use of perfusion analysis software. Software applications from different vendors do not generate equivalent quantitative perfusion results. Caution should thus be exercised when interpreting quantitative/qualitative CTP measures, because these values may vary considerably, depending on the post-processing software. The major mathematical technique utilized for calculation of perfusion parameters is the deconvolution approach. Deconvolution can be performed using several methods. The classic deconvolution method is termed “standard singular value decomposition” (sSVD). This technique is robust, and its results are independent of the underlying vascular anatomy (i.e. intra/extracranial stenoses and/or leptomeningel collaterals). However, sSVD is sensitive to delays in the arrival of the contrast agent. When such delays occur, MTT and CBV are overestimated and CBF is underestimated. The use of abnormal map values may lead to overestimation of the ischemic penumbra by including non-ischemic brain regions into delayed contrast agent arrival calculations. To overcome these difficulties, new deconvolution methods have been developed to minimize the effects of bolus delay and dispersion, including delay-corrected deconvolution (dSVD) and block-circulant deconvolution (bSVD) algorithms. As a general rule, the use of a delayinsensitive deconvolution method is recommended when imaging stroke patients, a population in whom arterial stenoses are common. The role of computed tomography angiography (CTA) in the interpretation of the core and the ischemic penumbra was not discussed in the Alves et al. article. Several groups have suggested that CTA source images (CTA-SI), like DWI MRI, may sensitively delineate tissue destined to infarct in spite of successful recanalization. The superior accuracy of CTA-SI in 365 Letter to the Editor Nicola Morelli A B C D Figure 1 A 78-year-old man with ischemic stroke of unknown onset presented to the emergency department with left hemiplegia. A) Volume rendering CTA (cranio-caudal view) shows occlusion of the right cerebral artery in the M1-M2 segment. B) CTA-SI at level of the basal nuclei with standard window width and center level (W/L) (286/66) is also illustrated. The CTA protocol acquisition occurs at an approximate steady state of contrast in the brain arteries and parenchyma so that the image is predominantly blood volume-weighted. Image review should be performed with narrow window width and center level display settings to maximize gray-white matter differentiation, thereby improving the detection of subtle, hypodense ischemic regions. C) With the same CTASI viewed with narrow W/L (38/48), it is possible to better appreciate the hypodense area of ischemic origin in the right cerebral hemisphere attributable to infarct “core” (irreversibly damaged tissue). D) A follow-up unenhanced CT scan clearly delineates the extent of ischemia, corresponding to the region identified with the CTA-SI in Figure 1C. Accurate inspection of CTA-SI with narrow windows enables definition of the ischemic core with the additional advantage of whole brain coverage. 366 www.centauro.it The Neuroradiology Journal 27: 365-367, 2014 - doi: 10.15274/NRJ-2014-10043 identifying infarct “core” compared with unenhanced CT has been unequivocally established in multiple studies 2,3. Theoretical modeling indicates that CTA-SI obtained using early generation protocols generates predominantly blood volume-weighted rather than blood flowweighted images. Due to the relatively slow acquisition, images acquired with these protocols reflect an approximate steady state of contrast in the brain arteries and parenchyma. However, newer, faster MDCT CTA-SI protocols, such as those used with 64-slice scanners, with injection rates up to 7 mL/s and short preparation delay times, change the temporal shape of the “time-density” infusion curve, eliminating the near-steady state during the timing of the CTA-SI acquisition. Hence, using current generation CTA protocols on faster state-of-the art MDCT scanners, the CTA-SI maps typically achieve more flow- rather than volume-weighted images. With early-generation CTA protocols, in the absence of early recanalization, CTA-SI typically defines minimal final infarct size and, hence, like DWI MRI, can be used to identify an “infarct core” in the acute setting (Figure 1 A-D). CTA-SI subtraction maps, obtained by coregistration and subtraction of the unenhanced head CT from the CTA-SI, result in quantitative blood volume maps of the en- tire brain and are particularly appealing for clinical use because, unlike quantitative firstpass CT perfusion maps, they provide whole brain coverage. Thus, radiologists should interpret CTP maps in conjunction with CTA-SI data. Finally, it is again worth underscoring that with the newer, more flow-weighted CTA protocols, not every acute CTA-SI hypodense ischemic lesion is destined to infarct; a substantial portion of a CTA-SI hypodense lesion may reflect “at-risk” ischemic penumbra, or even benign oligemia 3. Moreover, Kamalian et al. 4 reported that CBF maps optimally correlate with admission diffusion-weighted imaging. The accuracy of CBF versus CBV in determining the infarct core is supported by the fact that calculation of CBV is typically more sensitive to time-density curve truncation than is CBF. It is critical for radiologists who utilize PCT for stroke imaging to be familiar with the postprocessing software implemented at their institution and to analyze comprehensive datasets from advanced stroke imaging (CT, CTA, CTP and relative source image). The optimal method to delineate the core from the ischemic penumbra using CTP maps has not yet been established, and further studies should be performed to better understand the hemodynamic phenomena that occur in the acute phases of ischemic stroke. References 1 Alves JE, Carneiro A, Xavier J. Reliability of CT perfusion in the evaluation of the ischaemic penumbra. Neuroradiol J. 2014; 27 (1): 91-95. doi: 10.15274/NRJ2014-10010. 2 Schramm P, Schellinger PD, Fiebach JB, et al. Comparison of CT and CT angiography source images with diffusion-weighted imaging in patients with acute stroke within 6 hours after onset. Stroke. 2002; 33 (10): 2426-2432. doi: 10.1161/01.STR.0000032244.03134.37. 3 Lin K, Rapalino O, Law M, et al. Accuracy of the Alberta Stroke Program Early CT Score during the first 3 hours of middle cerebral artery stroke: comparison of noncontrast CT, CT angiography source images, and CT perfusion. Am J Neuroradiol. 2008;29 (5): 931-936. doi: 10.3174/ajnr.A0975. 4 Kamalian S, Kamalian S, Maas MB, et al. CT cerebral blood flow maps optimally correlate with admission diffusion-weighted imaging in acute stroke but thresholds vary by postprocessing platform. Stroke. 2011; 42 (7): 1923-1928. doi: 10.1161/STROKEAHA.110.610618. Nicola Morelli, MD Neurology Unit and Radiology Unit Guglielmo da Saliceto Hospital Via Taverna 49 29121 Piacenza, Italy Tel.: 00390523303310 Fax: 00390523303322 E-mail: n.morelli@inwind.it 367 The Neuroradiology Journal 27: 368, 2014 - doi: 10.15274/NRJ-2014-10044 www.centauro.it Response to Letter to the Editor "CT Angiography Source-Images and CT Perfusion: Are They Complementary Tools for Ischemic Stroke Evaluation?" JOSÉ EDUARDO ALVES, ÂNGELO CARNEIRO, JOÃO XAVIER Neuroradiology Department, Centro Hospitalar do Porto; Porto, Portugal Key words: CT perfusion, CT angiography source-images, cerebral blood volume, endovascular treatment We appreciate Morelli et al.’s comment on our recent article 1 and we do agree that the interpretation of computed tomography angiography source images (CTA-SI) may be a valuable tool in the assessment of acute ischaemic stroke. It has been shown that CTA-SI, compared with non-enhanced CT scans, improve both the detection of early ischaemic changes and the prediction of final infarct extent 2,3. Furthermore CTA-SI offer substantial more brain coverage than most common computed tomography perfusion (CTP) protocols. However, as the authors pointed out, recent papers revealed that infarct core estimation on CTA-SI is highly dependent on the CTA acquisition protocol. Pulli et al. have shown that current CTA protocols, designed to speed imaging acquisition (with a shorter time interval between contrast material injection and brain scanning), are associated with significant overestimation of infarct size on CTA-SI 4. Moreover, when comparing CTP-CBV maps with CTA-SI, CBV appears to be a significantly more sensitive marker for early irreversible ischaemic damage and a more accurate predictor of final infarct volume 5,6. Finally, we feel that, nowadays, the main challenge in the initial evaluation of acute ischaemic stroke patients is not the assessment of the ischaemic core (which can reliably be done by DWI, CTP and CTA-SI) but the accurate delineation of the ischaemic penumbra. In fact, there is currently no fast, easily accessible and reproducible way of differentiating tissue at-risk of infarction from areas of benign oligaemia. Only a correct evaluation of the “real” ischaemic penumbra will enable a proper selection of patients for endovascular treatment. References 1 Alves JE, Carneiro A, Xavier J. Reliability of CT perfusion in the evaluation of the ischaemic penumbra. Neuroradiol J. 2014; 27 (1): 91-95. doi: 10.15274/NRJ2014-10010. 2 Camargo EC, Furie KL, Singhal AB, et al. Acute brain infarct: detection and delineation with CT angiographic source images versus nonenhanced CT scans. Radiology. 2007; 244 (2): 541-548. doi: 10.1148/radiol.2442061028. 3 Aviv RI, Shelef I, Malam S, et al. Early stroke detection and extent: impact of experience and the role of computed tomography angiography source images. Clin Radiol. 2007; 62 (5): 447-452. doi: 10.1016/j. crad.2006.11.019. 4 Pulli B, Schaefer PW, Hakimelahi R, et al. Acute ischemic stroke: infarct core estimation on CT angiography source images depends on CT angiography protocol. Radiology. 2012; 262 (2): 593-604. doi: 10.1148/ radiol.11110896. 368 5 Lin K, Rapalino O, Law M, et al. Accuracy of the Alberta Stroke Program Early CT Score during the first 3 hours of middle cerebral artery stroke: comparison of noncontrast CT, CT angiography source images, and CT perfusion. Am J Neuroradiol. 2008; 29 (5): 931-936. doi: 10.3174/ajnr.A0975. 6 Parsons MW, Pepper EM, Chan V, et al. Perfusion computed tomography: prediction of final infarct extent and stroke outcome. Ann Neurol. 2005; 58 (5): 672-679. doi: http://dx.doi.org/10.1002/ana.20638. José Eduardo Alves Neuroradiology Department Hospital de Santo António - Centro Hospitalar do Porto Largo Prof. Abel Salazar 4099-001 Porto, Portugal Tel.: 00351 222 077 500 E-mail: zeedualves@gmail.com The Neuroradiology Journal 27: 369, 2014 - doi: 10.15274/NRJ-2014-10048 www.centauro.it Letter to the Editor 28 March 2014 Dear Professors Krejza and Leonardi, The paper by Colla, et al. 1 regarding use of the WEB II device in wide neck basilar tip aneurysms was most timely. The paper presented 4 cases each with excellent outcomes for these difficult complex aneurysm cases. Interestingly, in all 4 cases, a different antiplatelet regimen was utilized both pre- and postprocedurely. In our combined experience we have performed 49 cases (26 in Beaujon, 23 in Cologne) with the new WEB devices in similar complex aneurysms. Recently, we instituted a standard practice of pre- and postprocedure antiplatelet therapy for WEB cases that we feel balances the need to minimize the risk of procedural and postprocedural hemorrhagic complications with the potential for thromboembolic complications or possible use of adjunctive intraluminal devices. For less complex unruptured cases to be treated with the WEB, we pretreat with ASA 100 mg/ day for 3-5 days as we do for coil cases. No postprocedure antiplatelet therapy is prescribed unless otherwise warranted. For unruptured complex aneurysms, especially those which are large with wide necks we pretreat the patient with double antiplatelet therapy (ASA, 100 mg/day; clopidogrel, 75 mg/day) for a minimum of 4 days. This regimen minimizes the risks of thromboembolism in these difficult complex cases. Most important, it allows for the potential use of adjunctive intraluminal devices should that option be deemed necessary. Hwang 2 found that a double regimen significantly reduced thromboembolic events when coiling similar complex aneurysms and did not increase the risk of postprocedural hemorrhage. This practice was also supported by Rahme, et al. 3 in unruptured cases in which coils and stents may be used. For ruptured cases we follow standard coiling practice and recommend no preprocedure antiplatelet treatment. Should adjunctive use of a stent be necessary in complex ruptured aneurysms, GP IIb/IIIa inhibitors are administered that are replaced by ASA and Plavix in the subacute phase. The decision to put the patient under double antiplatelet therapy postprocedurally will depend on whether there is WEB protrusion into the parent vessel or a suspicion for thrombus formation outside the occlusion plane of the aneurysm. This is similar to coils, where if there are no coils protruding into the parent vessel, we do not use antiplatelet therapy; if there is significant coil protrusion, we prescribe double antiplatelet therapy for a limited period, usually up to 90 days. If clot formation occurs during a WEB procedure but resolves after WEB detachment, we utilize no GP IIb/IIIa inhibitors and no antiplatelet therapy. If a clot remains visible following device detachment, we administer IV abciximab (Beaujon) or Tirofiban (Cologne) followed by double antiplatelet therapy for 30 days. Testing for response to ASA and Plavix is done by means of a point-of-care system (VerifyNow). Sincerely, Professor Laurent Spelle Professor Thomas Liebig References 1 Colla R, Cirillo L, Princiotta C, et al. Treatment of wide-neck basilar tip aneurysms using the Web II device. Neuroradiol J. 2013; 26 (6): 669-677. Epub 2013 Dec 18. 2 Hwang G, Jung C, Park SQ, et al. Thromboembolic complications of elective coil embolization of unruptured aneurysms: the effect of oral antiplatelet preparation on periprocedural thromboembolic complication. Neurosurgery. 2010; 67 (3): 743-748; discussion 748. 3 Rahme RJ, Zammar SG, El Ahmadieh TY, et al. The role of antiplatelet therapy in aneurysm coiling. Neurol Res. 2014; 36 (4): 383-388. Epub 2014 Feb 20. 369 The Neuroradiology Journal 27: 370, 2014 - doi: 10.15274/NRJ-2014-10049 Response to the Editor Response to Letter to the Editor Antiplatelet Therapy and the WEB II Device Professors Spelle and Liebig are correct to highlight the differences in antiplatelet therapy in the paper by Colla, et al 1. They recommend a regimen of ASA 100 mg/day for 3-5 days without post-procedure antiplatelet therapy unless otherwise warranted for less complex unruptured WEB cases. This is a less intense regimen than the dual antiplatelet therapy recommended for other devices such as the Enterprise, Neuroform, or Pipeline. The recommendation is based on the expectation that the surface interface of the WEB II is no more thrombogenic than standard mass of randomly distributed embolization coils. Their recommendation follows that of Klisch et al. in the initial clinical experience with the WEB II in two patients 2. It is based on Professors Spelle’s and Liebig’s unpublished experience with 49 cases using WEB devices in similar less complex aneurysms. For unruptured complex aneurysms, they recommend pretreatment with double antiplatelet therapy (ASA, 100 mg/day; clopidogrel, 75 mg/day). As noted, this allows for the potential use of adjunctive intraluminal devices if necessary. Since the advent of stent treatment for intracranial aneurysms, attention to preventive therapy for procedural and post-treatment thrombosis has been increasingly necessary. This is particularly important in cases of unruptured aneurysms since preventive treatments need to be safer than the natural history of disease. Nevertheless, little consensus for the ideal antiplatelet regimen currently exists for most endovascular devices. Faught et al. recently surveyed neurointerventional surgeons on antiplatelet practices, finding considerable heterogeneity of practice 3. In addition, as highlighted by Professors Spelle and Liebig, different devices likely have different propensities for formation of thrombus and emboli. Consequently, optimal regimens may well differ for differing devices. Certainly the ideal regimen would be the one with the maximal chance of preventing thrombus formation while minimizing the chance of hemorrhagic complications. If no post procedure thrombotic events occur with WEB II devices in the absence of antiplatelet therapy, this would certainly be ideal. While more peer-reviewed experience is needed with the WEB II device to confirm that post-procedure antiplatelet treatment is unnecessary, the points made by Professors Spelle and Liebig are valuable in providing initial guidance toward minimizing risks associated with the use of this new device. Robert Hurst University of Pennsylvania, Philadelphia References 1 Colla R, Cirillo L, Princiotta C, et al. Treatment of wide-neck basilar tip aneurysms using the Web II device. Neuroradiol J. 2013; 26 (6): 669-677. 2 Klisch J, Sychra V, Strasilla C, et al. The Woven EndoBridge cerebral aneurysm embolization device (WEB II): initial clinical experience. Neuroradiology. 2011; 53 (8): 599-607. doi: 10.1007/s00234-011-0891-x. 3 Faught RW, Satti SR, Hurst RW, et al. Heterogeneous practice patterns regarding antiplatelet medications for neuroendovascular stenting in the USA: a multicenter survey. J Neurointerv Surg. 2014. doi: 10.1136/neurintsurg-2013-010954. 370 The Neuroradiology Journal 27: 371, 2014 - doi: 10.15274/NRJ-2014-10058 www.centauro.it Erratum The Neuroradiology Journal 27: 133-137, 2014 - doi: 10.15274/NRJ-2014-10025 www.centauro.it Brain targets: can you believe your own eyes? STEFANONI G1, TIRONI M2, TREMOLIZZO L2, FUSCO ML2, DIFRANCESCO JC2, PATASSINI M3, FERRARESE C2, APPOLLONIO I2 Unfortunately, one of the authors names, DiFrancesco JC, was incorrectly listed as Di Francesco J instead of DiFrancesco JC, in the original publication of this paper. Published on line: 18 April 2014 © Centauro s.r.l. 2014 on site: www.theneuroradioloyjournal.it Printed on: 30 April 2014 © Centauro s.r.l. 2014 371 The Neuroradiology Journal 27: 372-373, 2014 - doi: 10.15274/NRJ-2014-10057 AINR - Newsletter In memoriam Dr Mario Savoiardo MARIA CONSUELO VALENTINI, LUDOVICO D’INCERTI Mario Savoiardo, a leading figure in world neuroradiology, died on 30th January 2014. He had suffered a long illness with dignity and in silence, never losing his enthusiasm for and dedication to a discipline he continued until the last. Mario died with characteristic discretion and the peace of mind he deserved. Italian neuroradiology would like to remember Mario to those who knew him. Mario was born on 5th July 1939 in Corsico, near Milano. He graduated in medicine in 1965 and then continued his studies, specialising in neurology. After marrying Maria Riccarda, to whom he would remain devoted for the rest of his life, the young couple moved to the USA, where Mario spent three years as a resident at Boston University (1967-1969). These years were to prove crucial for his clinical approach and scientific training. It was in Boston that Mario first became interested in neuroradiology, thanks to Marjorie LeMay, one of the pioneers of neuroradiology who remained a lifelong friend. Mario’s beloved twin daughters Cristina and Silvia were born in Boston. In late 1971, Mario returned to Milan to devote himself to neuroradiology and began his career at the “C. Besta” National Neurology Institute run by Prof. Guido Lombardi whom he had met during a conference held in New York. Mario was awarded his postgraduate specialisation in diagnostic radiology at Padua University in 1973. Five years later he went to Canada, where he worked in the paediatric neuroradiol372 Mario Savoiardo è scomparso. Una delle figure più autorevoli e insigni della neuroradiologia mondiale ci ha lasciato il 30 Gennaio 2014, dopo una malattia lunga, vissuta con dignità e silenzio e con l’entusiasmo e la dedizione alla sua disciplina espressi sino ai suoi ultimi giorni. Mario se ne è andato con discrezione, come sapeva fare, e serenamente, come meritava. La Neuroradiologia Italiana vuole ricordarlo a quelli che lo hanno conosciuto. Mario nasce il il 5 luglio 1939 a Corsico, vicino a Milano. Laureato in Medicina nel 1965 ha proseguito i suoi studi con la specializzazione in Neurologia. Dopo aver sposato Maria Riccarda, a cui resterà legato per il resto della vita, si trasferisce con lei negli USA, dove trascorrerà 3 anni come resident alla Boston University (1967-1969), fondamentali per la sua impostazione clinica e la formazione scientifica. Lì nasce l’interesse per la Neuroradiologia, grazie all’incontro con Marjorie LeMay, neuroradiologa, cui rimarrà legato da profonda stima e affetto. A Boston nascono Cristina e Silvia, le sue amatissime figlie gemelle. Alla fine del ’71 torna a Milano per dedicarsi definitivamente alla Neuroradiologia. Inizia così la sua carriera all’Istituto Nazionale Neurologico “C. Besta”, sotto la direzione del Prof. Guido Lombardi, conosciuto tempo prima in occasione di una conferenza a New York. Si specializza in Radiologia Diagnostica presso l’Università degli Studi di Padova nel 1973. Nel 1979 si reca in Canada, a Toronto, dove frequenta la Neuroradiologia pediatrica del Sick Children Hospital assieme a Pepe Scotti e a Rina Tadmor, con i quali rimane legato da forte ami- AINR News ogy department at the Hospital for Sick Children in Toronto with Pepe Scotti and Rina Tadmor, who became close friends. He returned to the USA in 1987 to work in Philadelphia with Robert Zimmerman. Mario was director of the Neuroradiology Unit at the Besta Institute in Milan until March 2005 and after that served in an advisory capacity until shortly before his death. Over the years he published hundreds of scientific papers, many of which were milestones in neuroradiology research on topics such as the angiography of posterior fossa lesions, vascular territories of the brain stem, the diagnosis of optic pathway gliomas, the diagnostic work-up of intracranial hypotension and many studies on neurodegenerative and metabolic diseases. Mario Savoiardo was a member of the executive council of the Italian Association of Neuroradiology for many years and was the guiding force for more than one generation of neuroradiologists. Despite being an internationally renowned neuroradiologist, Mario was always generous with his time, ready and willing to apply his expertise and intuitive skill to help solve complex cases. Enthusiasm was one of Mario’s salient features. He tackled every aspect of his daily work with enthusiasm. Enthusiasm fired the subtlety of his quick mind, backed by a formidable memory that made him one of the most wellinformed diagnostic neuroradiologists of his generation. These gifts were flanked by discretion and openness: never patronizing, he was and always willing to lend a hand, a trait that made him both well-known and well-loved by his colleagues the world over. From Mario’s enthusiasm led him to be intolerant of approximation and compromise. He lived by the principle of giving his all and demanding the same of others, no half measures accepted. Mario was known to most people as a gifted doctor, clinician and researcher, less for his artistic streak expressed through drawing, photography and painting. During congresses, he would often be seen to put pencil to paper and quickly sketch the face or profile of a speaker or fellow participant. He was a lifelong painter of some renown and lately had come to appreciate the sensitivity and effects of print engraving. cizia, e poi ancora nel 1987 lavora a Philadelphia, da Robert Zimmerman. All’Istituto Besta rimane come Direttore dell’U.O. di Neuroradiologia fino al marzo 2005 e poi ancora come consulente fino a pochi giorni prima di andarsene per sempre. Negli anni pubblica centinaia di lavori scientifici, molti dei quali pietre miliari della Neuroradiologia come l’angiografia delle lesioni della fossa posteriore, i territori vascolari del cervelletto e del tronco encefalico, la diagnostica dei gliomi delle vie ottiche, la diagnostica dell’ipotensione liquorale e i numerosi studi sulle malattie neurodegenerative e metaboliche. Mario Savoiardo è stato membro del Consiglio Direttivo dell’Associazione Italiana di Neuroradiologia per molti anni, ma soprattutto ha illuminato più di una generazione. Pur essendo neuroradiologo di fama, conosciuto a livello internazionale, era sempre pronto e disponibile a risolvere casi complessi con la sua competenza, il suo intuito e la sua costante disponibilità e generosità. Una dote, meglio di altre, esprime gli aspetti salienti della sua personalità: la passione. Con passione, infatti, affrontava quotidianamente ogni aspetto del suo lavoro. Con passione alimentava la finezza e la rapidità nel ragionamento, supportato dalla capacità memoria che faceva di lui uno dei più colti e fini diagnosti del mondo neuroradiologico. Queste doti erano abilmente contenute in un atteggiamento discreto, mai supponente, sempre disponibile. Era, per questo, noto e amato da molti colleghi in tutto il mondo. Sempre dalla passione scaturiva la sua intolleranza verso l’approssimazione e il compromesso. Il suo “modus vivendi’ era dare ed esigere il massimo da se stessi e dagli altri, con molto rigore e senza sconti. Mario era noto ai più per le sue grandi qualità di medico, clinico e ricercatore, lo era forse meno per la sua grande vena artistica che esprimeva con il disegno, la fotografia e la pittura. Durante i congressi non era raro vederlo prendere la matita e schizzare, con mano veloce, il volto o il profilo del relatore o del vicino di banco. La pittura poi l’ha accompagnato sempre, con esiti tutt’altro che banali; ultimamente aveva anche assaporato la sensibilità e la vibrazione della stampa incisoria. In ultimo ha saputo anche utilizzare la penna e raccontare nel suo libro “Piccoli Ricordi” episodi della sua vita da medico, della 373 AINR News Mario was also a keen writer and his book “Piccoli Ricordi” (Excerpts from Memory) recounts experiences of his life working with patients and episodes with friends with characteristic humour and delicacy. Despite his many professional commitments, Mario Savoiardo had a full family life as husband, father and grand-father. He taught us all to defend our work firmly and with authority. but also elegance, aware of the importance of passing on to the next generation the habit of analysis and comparison, generously dispensing words of wisdom and encouragement. Mario leaves neuroradiology his profound cultural and moral legacy. The condolences and warmth of all those who knew, appreciated and loved him go to his widow Maria Riccarda, his daughters and grandchildren. 374 sua esperienza nel rapporto con i pazienti e di vicende con gli amici, con l’umorismo e la delicatezza che lo contraddistinguevano. Nonostante gli innumerevoli impegni professionali imprescindibili, Mario Savoiardo ha costantemente saputo mantenere viva in famiglia la sua presenza di compagno, padre e nonno. Ha insegnato a tutti come sia importante difendere il proprio operato con fermezza, autorevolezza ed anche eleganza, nella consapevolezza di quanto sia decisivo sapere trasmettere ai giovani l’abitudine all’analisi e al confronto, dispensando loro incoraggiamenti in modo parsimonioso ma senza avarizia. Mario lascia alla Neuroradiologia la sua profonda eredità culturale e morale. Alla moglie Maria Riccarda, alle figlie e ai nipoti va il saluto e il calore di tutti coloro che lo hanno conosciuto, apprezzato e amato. The Neuroradiology Journal 27: 375-378, 2014 www.centauro.it Books The Human Brain in 1969 Pieces 2.0 Structure, Vasculature, Tracts, Cranial Nerves, Systems, Head Muscles, and Tracts Wieslaw Nowinski, Beng Choon Chua $ 349.99 - € 299.99 - Publication Date: December 2013 - 0 pp - CD-ROM ISBN (Americas): 9781626230194 The Human Brain in 1969 Pieces, version 2.0 is a highly sophisticated, visually stunning 3D neuroanatomy atlas. Innovative and incredibly detailed, yet easy to navigate, this product allows every clinician and educator in neuroradiology, neurosurgery, neurology, and neuroscience to explore and better understand the intricacies of the human brain. About 2,000 detailed components identify every area of the brain from the spinal cord to tiny vessels. The modular dashboard allows the user to see one structure at a time or in any combination, turn off structures, rotate the brain, pan across the brain, see structures as labeled or unlabeled, and much more. Features of the new edition: Head muscles and glands Cerebral vertebrae A new, resizable interface that conforms to your screen size Additional cranial nerve and vessels content Labeling of 3D cuts and triplanar images Enhanced functionality and visual refinements 375 Books www.centauro.it Books Pocket Atlas of Sectional Anatomy Volume I: Head and Neck Computed Tomography and Magnetic Resonance Imaging Torsten B. Moeller - Emil Reif $ 47.99 - € 34,99 - Publication Date: November 2013 - 4th Edition - 340 pp, 792 illustrations Paperback / softback - ISBN (Americas): 9783131255044 This comprehensive, easy-to-consult pocket atlas is renowned for its superb illustrations and ability to depict sectional anatomy in every plane. Together with Volumes II and III, it provides a highly specialized navigational tool for all clinicians who need to master radiologic anatomy and accurately interpret CT and MR images. Special features of Pocket Atlas of Sectional Anatomy: Didactic organization in two-page units, with high-quality radiographs on one side and brilliant, full-color diagrams on the other Hundreds of high-resolution CT and MR images made with the latest generation of scanners (e.g., 3T MRI, 64-slice CT) Color-coded schematic drawings that indicate the level of each section Concise, easy-to-read labeling of all figures Sectional enlargements and magnified views for easy classification of anatomic structures Updates for the 4th edition of Volume I: New cranial CT imaging sequences of the axial and coronal temporal bone Expanded MR section, with all new 3T MR images of the temporal lobe and hippocampus, basilar artery, cranial nerves, cavernous sinus, and more New arterial MR angiography sequences of the neck and additional larynx images Compact, easy-to-use, highly visual, and designed for quick recall, this book is ideal for use in both the clinical and classroom settings. 376 www. centauro. it The Neuroradiology Journal 27: 375-378, 2014 Books Pocket Atlas of Sectional Anatomy Volume II: Thorax, Heart, Abdomen and Pelvis Computed Tomography and Magnetic Resonance Imaging Torsten B. Moeller - Emil Reif $ 47.99 - € 34,99 - Publication Date: September 2013 - 4th Edition - 344 pp, 611 illustrations Paperback / softback - ISBN (Americas): 9783131256041 This comprehensive, easy-to-consult pocket atlas is renowned for its superb illustrations and ability to depict sectional anatomy in every plane. Together with Volumes I and III, it provides a highly specialized navigational tool for all clinicians who need to master radiologic anatomy and accurately interpret CT and MR images. Special features of Pocket Atlas of Sectional Anatomy: Didactic organization in two-page units, with high-quality radiographs on one side and brilliant, full-color diagrams on the other Hundreds of high-resolution CT and MR images made with the latest generation of scanners (e.g., 3T MRI, 64-slice CT) Color-coded schematic drawings that indicate the level of each section Concise, easy-to-read labeling of all figures Sectional enlargements and magnified views for easy classification of anatomic structures Updates for the 4th edition of Volume II: CT imaging of the chest and abdomen in all 3 planes: axial, sagittal, and coronal New sections on MRI and CT of the heart and MR angiography New back-cover foldout featuring pulmonary and hepatic segments and lymph node stations Follows standard international classifications of the American Heart Association for cardiac vessels and the AJCC/UICC for mediastinal lymph nodes Compact, easy-to-use, highly visual, and designed for quick recall, this book is ideal for use in both the clinical and classroom settings. 377 Books www.centauro.it Books TIA as Acute Cerebrovascular Syndrome Editor(s): Uchiyama S. (Tokyo), Amarenco P. (Paris), Minematsu K. (Osaka), Wong K.S.L. (Hong Kong) USD 209.00 - CHF 267,00 - Bibliographic Details - Frontiers of Neurology and Neuroscience, Vol. 33 - VIII + 166 p., 25 fig., 7 in color, 14 tab., hard cover/online, 2014 - Status: available ISBN: 978-3-318-02458-6 - e-ISBN: 978-3-318-02459-3 The latest knowledge of TIA as a medical emergency Transient ischemic attack (TIA) is well known to be a prodromal syndrome of ischemic stroke. However, TIA is easily neglected or underestimated by patients or even general physicians because the symptoms naturally disappear without treatment. Despite this, early after the onset of TIA the patients are at very high risk of stroke. As it is not possible to differentiate TIA from acute ischemic stroke (AIS) only by the duration of symptoms, both TIA and AIS should be recognized on the same spectrum of acute ischemic syndrome in the central nervous system. This book presents the new concept ‘acute cerebrovascular syndrome’ (ACVS), which includes both TIA in acute settings and AIS. The publication covers all topics of TIA in ACVS, which includes the definition, concept, etiology, epidemiology, symptomatology, risk scores, neuroimaging, neurosonology, acute management, primary and secondary prevention, and guidelines. Written by leading international experts in the field, the publication presents valuable and essential information for neurologists, general practitioners, neurosurgeons, radiologists, students, and nurses, in both clinical practice and research. 378 Subscription order form FThe Neuroradiology Journal for 1 year (6 issues + suppl.) =63<4,5V-,)9<(9@  PAPER VERSION ITALY FREGULAR: EURO 190.- NEURORADIOLOGY JOURNAL + INTERVENTIONAL NEURORADIOLOGY EURO 275.- FSPECIAL PRICE THE *,5;(<96:YS)636.5( )PTLZ[YHSL7VZ[L0[HSPHULZWH:WLKPUHW+3JVU]PU3U‡HY[JVTTH+*))6 ,\YV PAPER VERSION OTHER COUNTRIES FREGULAR: $,15$VVRFLD]LRQH,WDOLDQDGL1HXURUDGLRORJLD DQG 7KH1HXURUDGLRORJLVWVRI$OSH$GULD $156$OEDQLDQ1HXURUDGLRORJLFDO6RFLHW\ 3$1563DQ$UDE1HXUR5DGLRORJ\6RFLHW\ 5DGLRORJLFDO6RFLHW\RI6DXGL$UDELD'LYLVLRQRI1HXURUDGLRORJ\ ,615,QGLDQ6RFLHW\RI1HXURUDGLRORJ\ ,QGRQHVLDQ6RFLHW\RI1HXURUDGLRORJ\ 1HXURUDGLRORJ\6HFWLRQRIWKH5DGLRORJ\6RFLHW\RI,UDQ ,VUDHOL6RFLHW\RI1HXURUDGLRORJ\ &ROOHJHRI5DGLRORJ\0DOD\VLD EURO 220.- NEURORADIOLOGY JOURNAL + INTERVENTIONAL NEURORADIOLOGY EURO 330.- FSPECIAL PRICE THE 2I¿FLDO-RXUQDORI 1HXURUDGLRORJ\6HFWLRQ3DNLVWDQ3V\FKLDWU\5HVHDUFK&HQWHU 6HFWLRQRI1HXURUDGLRORJ\3ROLVK5DGLRORJLFDO6RFLHW\ 7KH1HXURUDGLRORJLVWVRI5RPDQLD 6HFWLRQRI1HXURUDGLRORJ\RI6HUELDDQG0RQWHQHJUR 6,/$16RFLHGDG,EHUR/DWLQR$PHULFDQDGH1HXURUUDGLRORJLD 1HXURUDGLRORJ\6HFWLRQRI6LQJDSRUH5DGLRORJLFDO6RFLHW\ 6ORYHQLDQ6RFLHW\RI1HXURUDGLRORJ\ 1HXURUDGLRORJLFDO6RFLHW\RI52&7DLZDQ 76157XUNLVK6RFLHW\RI1HXURUDGLRORJ\ FInterventional DIGITAL VERSION Neuroradiology for 1 year (4 issues + suppl.) INTERVENTIONAL NEURORADIOLOGY VOLUME 17 - No. 2 - JUNE 2011 Trimestrale - Poste Italiane s.p.a. - Sped. in abbonamento postale 70%, DCB - Bologna FREGULAR: EURO 125.- NEURORADIOLOGY JOURNAL + INTERVENTIONAL NEURORADIOLOGY EURO 187.- FSPECIAL PRICE THE CENTAURO S.r.l., BOLOGNA Euro 46,00 ISSN 1591-0199 DIGITAL VERSION FSTUDENT: 17 N 129 272 2011 95.- NEURORADIOLOGY JOURNAL + INTERVENTIONAL NEURORADIOLOGY EURO 142.- FSPECIAL PRICE THE Vl 2 P EURO Journal of Peritherapeutic Neuroradiology, Surgical Procedures and Related Neurosciences Official Journal of: WFITN - World Federation of Interventional and Therapeutic Neuroradiology AAFITN - Asian & Australasian Federation of Interventional & Therapeutic Neuroradiology SAWITN - South American Working Group in Interventional and Therapeutic Neuroradiology Journal sponsored by JSNET - Japanese Society of Neuro Endovascular Therapy Interventional Neuroradiology is published in cooperation with the American Journal of Neuroradiology  I enclose a cheque / money order payable to Centauro srl  Swift transfer to Centauro srl - Unicredit Banca - Swift code: UNCRITM1BA2 IBAN code: IT 74 D 02008 02435 000002933096  Please charge my  Visa  Diners  Master Card  Carta Si  Amex Expiry date Credit Card number Name (please print) Address City / State / Post code Phone Fax E-mail Please compile for your digital subscription activation User Password Date Signature  Please send invoice to: Please return this form to: CENTAURO S.R.L. - Via del Pratello, 8 - I-40122 Bologna - Italy - Fax +39.051.220099 Students: Please enclose a copy of your student registration form Dear Subscriber – In relation to the provisions of DLgs 196/03, we assure you that your personal information (name, surname, qualification, profession and address) currently stored in our database shall be used solely for the purposes of sending business letters and notices from publisher to subscriber. In accordance with clause 7 of DLgs 196/03, you are entitled to refuse to authorise any use of the information in our possession for purposes other than those prescribed by law. THE NEURORADIOLOGY JOURNAL Founded by: Marco Leonardi, with the support of Aldo Benati, Luigi Bozzao, Gianni Boris Bradač, Alberto Calabrò, Aristide Carella, Gian Carlo Dal Pozzo, Giovanni Di Chiro, Raffaele Elefante, Ugo Pasquini, Kurt Pardatscher, Angelo Passerini, Marco L. Rosa, Ugo Salvolini, Giuseppe Scialfa, Giuseppe Scotti, Auguste Wackenheim Editorial Committee Jaroslaw Krejza - Editor-in-Chief E-mail: jkrejza@me.com - krejza@rad.upenn.edu Marco Leonardi, European Editor E-mail: marco.leonardi@centauro.it Kay Yamada, Asia - Pacific Editor E-mail: kyamada@koto.kpu-m.ac.jp Consulting Editors - JK Advisor Board Ferenc A. Jolesz, Elias R. Melhem, James Provenzale International Associate Editors Umair Rashid Chaudry, Sossio Cirillo, Kavous Firouznia, Dorith Goldsher, Wan-Yuo Guo, Rashad Hamdi, Francis Hui, Benny Huwae, Eric Günter Klein, Sattam S. Lingawi, Ljubomir Markovic, Zoran Milosevic, Yuko Ono, Julian Opincariu, R.V. Phadke, Norlisah Mohd Ramli, Marek Sasiadek, E. Turgut Tali, Jean Tamraz, Eduardo Gonzalez Toledo, Gjergji Vreto Section Editors Interventional Neuroradiology Alexander Norbash, Robert Hurst, IizukaYuo Iizuka Clinical Editors Viken Babikian Editorial Boards Wieslaw Nowinski, David Liebeskind, Donald O-Rurke, Peter LeRoux, Riyadh Al-Okaili, Jerry Glikson, Edward Herskovits, Haris Sair, Felix Wehrli Consulting Biostatisticians Patrizia Agati Scientific Committee, International Reviewers AINR - Associazione Italiana di Neuroradiologia Sossio Cirillo, Chairman Cosma F. Andreula, Luigi M. Fantozzi, Mario Savoiardo, Giuseppe Scotti The Neuroradiologists of Alpe-Adria Alberto Beltramello, Chairman Francesco Causin, Zoran Milosevic, Alain Tournade ANRS - Albanian Neuroradiological Society Gjergji Vreto, Chairman Arben Rroji, Eugen Enesi, Fatmir Bilaj, Denis Qirinxhi PANRS – Pan Arab NeuroRadiology Society Jean Tamraz, Chairman Mohamed Hiari, Najat Boukhrissi, Sami Slaba, Eman Bakhsh Radiological Society of Saudi Arabia, Division of Neuroradiology Sattam S. Lingawi, Chairman Ibrahim A. Al-Oraini, Riyad Al-Okaily Egyptian Society of Neuroradiology Rashad Hamdi, Chairman Yasser Abdel Azeem, Hany Lotfy, Sahar Saleem ISNR - Indian Society of Neuroradiology RajendraV. Phadke, Chairman A.K. Gupta, N. Khandelwal, S.B. Gaikwad, S. Joseph H. Mahajan, N. Chidambaranathan PUBLISHING COMMITEE PUBLISHER Nicola Leonardi Indonesian Society of Neuroradiology Sri Andreani Utomo, Chairman Rustiadji, Benny Huwae, Anggraini Dwi Sensuati Neuroradiology Section of the Radiology Society of Iran Kavous Firouznia, Chairman Kh Bakhtavar, M. Saburi Deilami, H. Ghanaati, H. Hadizadeh, H. Hashemi, A. Radmehr, A. Sedaghat, J. Jalal Shokouhi, B. Taheri, K. Vessal, Z. Miabi Israeli Society of Neuroradiology Dorith Goldsher, Chairman Moshe J Gomori Japanese Scientific Committee Yuko Ono, Chairman Osamu Abe, Masahiro Ida, Yoshihiro Toyama, Key Yamada College of Radiology Malaysia Norlisah Mohd Ramli, Chairman Jeyaledchumy Mahadevan, Sobri Muda, Adam Pany, Kartini Rahmat, Khairul Azmi, Shahizon Mukari, Sharifah Aishah Neuroradiology Section Pakistan Psychiatry Research Center Umair Rashid Chaudry, Chairman Khuram Shafiq Khan, Najum ud Din, M.A. Raheem Section of Neuroradiology Polish Radiological Society Marek Sasiadek, Chairman Monika Figatowska-Bekiesi̾ska, Marek Stajgis, Andrzej Urbanik Romanian Scientific Commitee Iulian Opincariu, Chairman Dafin F. Muresanu, Khaled Abu Arif, Daria Abu Arif Section of Neuroradiology of Serbia and Montenegro Ljubomir Markovic, Chairman Svetlana Milosevic Medenica, Slobodan Cirkovic, Tatjana Stosic-Opincal, Milanka Raicevic, Dragan Stojanov SILAN - Sociedad Ibero Latino Americana de Neurorradiologia Eduardo Gonzalez Toledo, Chairman Alejandro Berenstein, Ramon Figueroa, Rafael Rojas, Teresa Sola Martinez Neuroradiology Section of Singapore Radiological Society Francis Hui, Chairman Tchoyoson Lim, Yih-Yian Sitoh, Winston Lim Slovenian Society of Neuroradiology Zoran Milosevic, Chairman Miha Skrbec, Jernej Knific, Igor Kocijancic, Nuska Pecaric Meglic, Ales Koren, Katarina Surlan, Tomaz Seruga The Neuroradiological Society of Taiwan Ho Fai Wong, Chairman WanYuo Guo, Shu-Hang Ng, Clayton Chi-Chang Chen, Chi-Jen Chen Ping-Hong Lai, Chung-Jung Lin, Sandy Cheng-Yu Chen TSNR – Turkish Society of Neuroradiology E. Turgut Tali, Chairman Saruhan Cekirge, Canan Erzen, Civan Islak, Naci Kocer, Ozenc Minarecy, Isil Saatci Interventional Neuroradiology Committee Yuo Iizuka, Chairman Augusto Goulão, Alfredo Casasco, Hiro Kiyosue, Michael Söderman LANGUAGE EDITOR Anne P. Collins, B.A. - E-mail: collins@iol.it MANAGING EDITOR Elisabetta Madrigali - E-mail: elisabetta@centauro.it ADVERTISING Serena Preti - E-mail: serena.preti@centauro.it Centauro s.r.l. - Via del Pratello, 8 - I-40122 Bologna Tel: ..39/051/227634 - Fax: ..39/051/220099 Subscription Form and Information: www.centauro.it Online Submissions: http://nrj.edmgr.com Single issue: € 30,00 - Back issue: ONLY DIGITAL VERSION http://www.centauro.it/store.htm Printed by: Abc Tipografia Srl - Sesto fiorentino Firenze, Italy Reg. Trib. di Bologna n. 7413 del 20-02-2004 Indexed in: EMBASE, Scopus, Google Scholar - ISSN 1971-4009 THE NEURORADIOLOGY JOURNAL INSTRUCTIONS FOR AUTHORS First formulated in 1988 - Revised in April, 2014 The Neuroradiology Journal (NRJ) is a clinical practice journal documenting the cur rent state of diagnostic and interventional neuroradiology worldwide. NRJ publishes original clinical observations, descriptions of new techniques or procedures, case reports, and articles on the ethical and social aspects of related health care. Original research published in NRJ is related to the practice of neuroradiology. Submissions suitable for the Journal include observational studies, clinical trials, epidemiological work, reports on health services and outcomes, and advances in applied (translational) and/or basic research. The instructions for submission of articles to NRJ follow the Uniform Requirements for Manuscripts Submitted to Biomedical Journals of the International Committee of Medical Journal editors (ICMJE, http://www.icmje.org), if not otherwise indicated below. DOI From 2014 all articles published in NRJ and INR will be assigned a DOI number. Authors are requested to specify the DOI number of publications when available in the References of papers submitted to NRJ and INR. Articles approved for publication in our journals are immediately published in digital version in our Publications Calendar without page numbers but the relative DOI ensures citation accuracy. The exact page numbers are added when the complete journal issues are published. For any further information please contact: marco.leonardi@centauro.it Research and Publication Ethics Conflict of interest policies: The Neuroradiology Journal (NRJ) upholds high standards of integrity and ethical conduct in research, and related communications. It is important that the editors, authors, and reviewers conduct themselves in accordance with rigorous standards and transparent policies for addressing potential conflicts of interest. Herein, we delineate what constitutes a potential conflict of interest for the NRJ as it relates to editors, authors, and reviewers. Those found in violation of these policies may be subject to sanctions as determined by the NRJ editors. Editorial conflicts of interest: The NRJ editors are responsible for maintaining high standards in evaluating contributions and maintaining the integrity of the Journal. In the interest of establishing full transparency, editors are obliged to disclose any and all potential conflicts of interest to the NRJ. We have determined two tiers of potential conflict and corresponding actions to be taken. The editors will report changes to their potential conflicts as they occur. An annual formal review of all disclosures will be performed in the evaluation of compliance. The first tier of potential conflicts for editors: (1) Ownership. If an editor currently has direct ownership of equity in a private or public company in the health care field of $10,000 or more (including restricted stock; the market price of all options, vested or unvested; and warrants), a first-tier potential conflict must be declared. Interests held by immediate family members (spouse or children) of the editor are included. This does not apply to ownership of mutual funds, where the editor does not directly control the purchase and sale of stocks. - (2) Income. If an editor has received $10,000 or more per annum of income from any single private or public company in the health care field in the preceding calendar year, a first-tier potential conflict must be declared. This includes any and all sources of financial benefit, including, but not limited to, consultancy, speaking fees, royalties, licensing fees, retainers, salary (including deferred compensation), honoraria, service on advisory boards, and providing testimony as an expert witness. Income generated by immediate family members (spouse or children) of the editor is included. - (3) Research support. If an editor’s research was funded by $50,000 or more per annum from a private or public company in the health care field in the preceding fiscal year, including funding for personnel working within the laboratory, a first-tier potential conflict must be declared. If an editor declares a first-tier potential conflict relating to (1), (2), or (3), this information will be published on the NRJ website. An editor will be considered to be in conflict if a manuscript is funded solely by an organization with which the editor has a potential conflict, regardless of whether a research institution employs the authors. The second tier of potential conflicts for editors: (4) Relationship with a company. If an editor had a relationship with a private or public company in the health care field wherein the editor received some compensation for services, but the total amount of income was between $1,000 and $9,999 for the preceding calendar year, a second-tier potential conflict must be disclosed. This includes, but is not limited to, any compensation as detailed above in (2). - (5) Relatives. If an editor has a close relative other than a spouse or child (sibling or parent) employed by or with a significant financial interest in a private or public company in the health care field, a second-tier potential conflict must be declared. - (6) Prospective employment. If an Editor is negotiating with, has arranged prospective employment with, or is expected to initiate a significant financial relationship as defined in (1), (2), or (3) with a private or public company in the health care field, a second-tier potential conflict must be declared. - (7) Personal. Editors will be required to declare a second-tier potential conflict if a manuscript is submitted by a close personal contact (former student, fellow or mentor, for example) or a recent collaborator (over the last 3 years). Relevant collaborations may include co-authoring a research article or serving as co-investigators on a grant. - (8) Competition. Editors will be required to declare a second-tier potential conflict if a submitted manuscript presents data that are highly relevant to a manuscript the Editor has under review or in press elsewhere. Editors are prohibited from using unpublished information from the manuscripts under consideration by the NRJ to further their own research, nor can they use new information gained from unpublished manuscripts for financial gain. - (9) Personal benefit. The editor must avoid making a decision on a manuscript if he or she could benefit personally from its disposition. The second tier of potential conflicts will necessitate only internal disclosure to the editorial board. These potential conflicts will not be published, but they will be known to the NRJ staff and other NRJ editors. The editor in potential conflict will not make decisions related to the manuscript. All editors will have access to a list of the first- and second-tier potential conflicts. Editor in Chief is responsible for recording and updating all potential conflicts. The Editor in Chief reviews any NRJ editorial staff potential conflicts. We are aware that other potential issues may arise, and these will be evaluated by the Editor in Chief on a case-by-case basis. NRJ editors are discouraged from serving as editors for other neuroradiology journals for which they would make final decisions on manuscripts. All such editorial duties for other journals must be approved by the Editor in Chief. In order to avoid even the appearance of potential favoritism to institutional colleagues, manuscripts from Editors’ institutions will not be handled by the editorial board at large, but instead in a separate process. In these circumstances, a specific Editor will be the only editor privy to the manuscript and, if the manuscript is sent for review, will work with an outside consultant to formulate a decision. Author Conflicts of Interest All authors are expected to disclose all financial relationships that could undermine the objectivity, integrity, or perceived value of a publication. The editors will keep the potential conflicts in mind while evaluating the manuscripts. Authors must disclose all potential conflicts as described below even if they believe their conflict is not germane to the content of the submitted paper (these correspond to the first tier of potential conflicts defined for editors). Such potential conflicts will be published in a footnote if the manuscript is ultimately accepted. It is the responsibility of the corresponding author to gather the list of potential conflicts from each author and to communicate the list of all potential conflicts to the editors with the submission. Potential conflicts to be disclosed by authors: (1) Ownership. If an author currently has direct ownership of equity in a private or public company in the health care field of $10,000 or more (including restricted stock; the market price of all options, vested or unvested; and warrants), a first-tier potential conflict must be declared. Interests held by immediate family members (spouse or children) of the author are included. This does not apply to ownership of mutual funds, where the author does not directly control the purchase and sale of stocks. - (2) Income. If an author has received $10,000 or more of income per annum from any single private or public company in the health care field in the calendar year preceding the date of the original submission, a potential conflict must be declared. This includes any and all sources of financial benefit, including, but not limited to, consultancy, speaking fees, royalties, licensing fees, retainers, salary (including deferred compensation), honoraria, service on advisory boards, and providing testimony as an expert witness. Income generated by immediate family members (spouse or children) of the author are included. - (3) Research support. If an author’s research was funded by $50,000 or more per annum from a private or public company in the health care field in the fiscal year preceding the date of the original submission, including funding for personnel working within the laboratory, a potential conflict must be declared. Reviewers’’ Conflicts of Interest Reviewers should exclude themselves in cases where there is a material potential conflict of interest, financial or otherwise. We ask that reviewers inform the editors of any potential conflicts that might be perceived as relevant as early as possible following invitation to participate in the review, and we will determine how to proceed. Disclosing a potential conflict does not invalidate the comments of a reviewer, it simply provides the editors with additional information relevant to the review. We ask reviewer to use their judgment in responding to our request for full disclosure, basing their response to the editors on the same financial criteria applied to authors and editors, as described above. - 1. Author responsibility for originality: The corresponding author acknowledges responsibility for the integrity of the manuscript, assures the originality of the paper, and guarantees that submitted manuscripts do not contain previously published material and are not under consideration for publication elsewhere. If the submitted manuscript builds on or includes parts of previously published articles, authors are encouraged to enclose copies of the articles with the new submission. The Editors reserve the right to request the original data obtained in the investigation. - 2. Registration of clinical trial re- search: Any research that includes clinical trials should be registered with the primary national clinical trial registration authority accredited by WHO (http://www.who.int/ictrp/network/primary/en/index.html) or ICMJE. - 3. Disclosure statement: This is not intended to prevent authors with potential conflicts of interest from contributing to NRJ. Rather, the Journal will place on record any relationship that may exist with disclosed, or competing, products or firms. Disclosed information will be held in confidence during the review process and the Editors will examine any printed disclosure accompanying a published article. Authors are responsible for notifying the Journal of financial arrangements including, but not limited to, agreements for research support including provision of equipment or materials, membership of speaker bureaus, consulting fees, or ownership interests. It is important that disclosure statements be updated promptly to reflect any new relationships that arise after initial submission of the manuscript. If the study is supported by a commercial sponsor, the authors must document the input of the sponsoring agency in study design, data collection, analysis of results, interpretation of data, and report writing. It is important to specify whether the sponsors could have suppressed or influenced publication if the results were negative or detrimental to the product they produce. Authors should also state if the company was involved in the original study design, the collection and monitoring of data, analysis and/or interpretation, and/or the writing and approval of the report. - 4. Patient anonymity and informed consent: It is the author's responsibility to ensure that a patient's anonymity is carefully protected and to verify that any experimental investigation with human subjects reported in the manuscript was performed with informed consent and followed all the guidelines for experimental studies with human subjects required by the institution(s) with which all the authors are affiliated. Patients have a right to privacy that should not be infringed without informed consent. Identifying information, including patients' names, initials, or hospital numbers, should not be published in written descriptions, images, and pedigrees unless the information is essential for scientific purposes and the patient (or parent or guardian) gives written informed consent for publication. Informed consent for this purpose requires that a patient who is identifiable be shown the manuscript to be published. Authors should identify Individuals who provide writing assistance and disclose the funding source for this assistance. Identifying details should be omitted if they are not essential. Complete anonymity is difficult to achieve, however, and informed consent should be obtained if there is any doubt. If identifying characteristics are altered to protect anonymity, authors should provide assurance that alterations do not distort scientific meaning and editors should be noted. Permissions: Authors must submit written permission from the copyright owner (usually the publisher) to use direct quotations, tables, or illustrations that have appeared in copyrighted form elsewhere, along with complete details about the source. Any permissions fees that might be required by the copyright owner are the responsibility of the authors requesting use of the borrowed material and not the NRJ. - 5. Statement of human and animal rights: When reporting experiments on human subjects, authors should indicate whether the procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2005. If doubt exists whether the research was conducted in accordance with the Helsinki Declaration, the authors must explain the rationale for their approach, and demonstrate that the institutional review body explicitly approved the doubtful aspects of the study. When reporting experiments on animals, authors should be asked to indicate whether the institutional and national guide for the care and use of laboratory animals was followed. All biomedical research performed on subjects should be in accordance with international ethic rules and approved by local ethic committees. Randomized clinical trial reports must be written in accordance with the CONSORT reporting guidelines (The Consolidated Standards of reporting Trials). 6. Duplicate/ Redundant publication: NRJ only accepts manuscripts describing original research. The editorial office of the NRJ does not accept duplicate submission or redundant publication. Redundant (or duplicate) publication is publication of a paper that overlaps substantially with a paper already published in print or electronic media as defined by the updated ICMJE guidelines that cover allegations of scientific misconduct. When submitting a paper, authors should make a full statement to the Editor on all submissions and previous reports that might be regarded as redundant publication of the same or similar work. If authors believe that their manuscript may be considered redundant, they should address this issue in a letter to the Editor accompanying the submission. The authors should also explain in the letter how their report overlaps already published material, or how it differs. Copies of such published material should be included with the submitted paper to help the Editor examine the possibility of redundant publication. If redundant publication is attempted without such notification, authors should expect editorial action to be taken. At the very least, rejection of the manuscript may be expected. - 7. Authorship: The Editors consider authorship to belong to those persons who accept intellectual and public responsibility for the statements made and results reported. By submitting a manuscript for publication, each author acknowledges having made a substantial contribution to the concept and design of the study, the analysis and interpretation of the results, and the writing of the paper, in addition to having approved the final submitted version. Authorship should not be attributed to Departmental Chairs not directly involved in the study, to physicians or technicians who provided routine services, or to technical advisors. A group study should carry the group name and reference contributing authors in the Acknowledgments. Peer Review Process Papers are accepted on the understanding that they are subject to peer review, editorial revision, and, in some cases, comment by the Editors. Manuscripts are examined by independent peer reviewers. Articles and other material published in the Journal represent the opinions of the authors and should not be construed to reflect the opinions of the publisher. Language The official language of NRJ is English. Most papers in NRJ have been written by non-native English speakers, and they will also be read by many non-native speakers. For the purposes of clarity, it is strongly recommended that authors not fluent in English have their manuscript professionally edited for English usage prior to submission. A professional editor will improve the English to ensure that author meaning is clear and to identify problems that require author review. Authors are invited to contact the NRJ language editor Anne Collins (collins@iol.it) for referral to professional English language editing services. Authors should make contact with and arrange payment to their chosen editing service directly. Please note that the use of such a service is at the author's own expense and risk and does not guarantee that the article will be accepted for publication. Centauro Publishers does not receive any commission or other benefit from editing services. Accordingly, Centauro Publishers neither endorses nor accepts any responsibility or liability for such editing services. Copyright and Purchase of Offprints All articles published in NRJ are protected by copyright, which embraces the exclusive right to reproduce and distribute the article (e.g., as offprints), as well as all translation rights. No material published in the Journal may be reproduced without written permission from the publisher. NRJ offers authors a complimentary hard copy and .pdf file of the issue in which their article is published, for personal use. To purchase offprints of articles published by NRJ please contact Centauro Srl for a quotation at serena.preti@centauro.it. Publication Types Original Research Articles are full-length research papers, which are favored by NRJ . Articles cover topics relevant to clinical studies, and may include both basic and experimental work. Review Articles are comprehensive, state-of-the-art papers focusing on important clinical problems and should address a specific topic in a scholarly manner. Such articles may be invited by the Editor or may be unsolicited reviews. Case Reports/Technical Notes should be unusually educational and medically important. Although NRJ preferentially encourages submission of full-length Original Research Articles, the Journal will consider publication of a limited number of concise Case Reports and Technical Notes. Editorials are usually invited by the Editor. Editorial Comments are specific comments on the articles published in NRJ and are usually invited by the Editor. Letters to the Editor contain constructive comments or criticism of a specific paper published by NRJ . Letters dealing with subjects of general interest within the field of NRJ , or personal opinions on a specific subject within the ambit of NRJ , may also be accepted. Electronic Submission of Manuscripts NRJ provides an electronic submission system (Editorial Manager) and review process to promote expeditious peer review. Manuscripts should be submitted electronically to the following URL: http://www.editorialmanager.com/nrj/. Questions about manuscripts under consideration may be addressed to the editorial office. The corresponding author listed on the manuscript must complete the forms in Editorial Manager, entitled Conflicts of Interest/Disclosures, Copyright Transfer Agreement, Financial Support, Exclusive Publication Statement, and Author Contribution Form. Submissions not containing completed forms by the corresponding author will not progress to peer review. Prior to submitting any paper, please follow the instructions given below. Please note that an author must have an e-mail address to use the online submission system. Manuscript preparation Introductions that Satisfy Reader Expectations – Overview: Once you have a revised draft, you need to ensure that your Introduction frames it, so that your readers will understand where you are taking them. The Introduction should orient readers and motivate them to read the rest of the paper. The Introduction must also make a contract with the reader that a question will be answered. – Functions: – To awaken the reader’s interest. – To be informative enough to prepare readers to understand your paper. – Content and Organization: 1. Common ground « Context « Relevant background « Orients the reader. 2. Disruption « Problem « Gap in Knowledge, Question « Motivates the reader. 3. Resolution « Response « Promise of an answer « Makes a contract with the reader. 1. Common Ground: States the consensus, shared understanding, common ground in the field, what’s know and not known. Gives the reader context and provides relevant background information, not a literature review. 2. Disruption: States the problem/the question that the paper addresses. Conveys the significance, i.e. the cost of leaving problem unsolved, or the benefit of solving it. 3. Resolution: Implicitly promises that you will present your answer in the Results and Discussion. Try not to state the answer; that has the effect of closing off the paper rather than leading into it. When drafting: Although the Introduction unfolds in the above order, 1-2-3, when you’re drafting, write it in a 3-2-1 order. By writing the Introduction in this order, you are sure to work through and clearly set up your main point, the question that it answers, and to include only relevant background information. When revising: “The key is to think like a reader -- readers expect and need a sense of structure. Since readers read each sentence in light of how they see it contributing to the whole, when you revise it makes sense to diagnose first the largest elements of the paper, then focus on the coherence of your paragraphs, the clarity of your sentences, and only last on matters of spelling and punctuation. Of course, in reality, no one revises so neatly; all of us revise as we go. But, it is useful to keep in mind that when you revise from the top down, from global structure to sections to paragraphs to sentences to words, you are more likely to discover useful revisions than if you start at the bottom with words and sentences and work up”. Booth WC, Colomb GG, Williams JM, et al. The Craft of Research. Chicago: University of Chicago Press; 2008. Original research should be organized in the customary format, as described below. The text of the manuscript should be submitted as a single document with the following sections (in order): Author Information Page (First page) Full title of the article, authors’ names, highest academic degree earned by each author, authors’ affiliations, name and complete address for correspondence, address for reprints if different from address for correspondence, fax number, telephone number, and e-mail address. Acknowledgments and Funding Page (Second page) The Acknowledgments section lists all funding sources for the research of the study, and details substantive contributions of individuals. The authors must reveal all possible Conflicts of Interest/Disclosures. Title Page (Third page) The full title, itemized list of the number of tables, the number and types (color or black-and-white) of figures, and three-to-five key words for use as indexing terms should be included. Appropriate key words should be selected from the Medical Subject Heading. The word count of the text should be specified. Summary A summary of up to 250 words should summarize the problems presented and describe the studies undertaken, results and conclusions. Since the abstract must be explicative, the abbreviations must be reduced to a minimum and explained. References should not be cited in the abstract. Text Typical main headings include Introduction, Materials and Methods, Results, Discussion, and Conclusions. Abbreviations must be defined at first mention in the text, tables and figures. The complete names and short addresses of manufacturers of any equipment used in Materials and Methods must be supplied. If animals are used in experiments, state the species, the strain, the number of animals used, and any other pertinent descriptive characteristics. If human subjects or patients are employed, provide a table with relevant characteristics. When describing surgical or neurointerventional procedures on animals, identify the pre-anesthetic and anesthetic agents used, and state the amount or concentration and the route and frequency of administration of each agent. Generic names of drugs must be given. Manuscripts that describe studies on humans must indicate that the study was approved by an Institutional Review Committee and that all subjects gave informed consent. Reports of studies on both animals and humans must indicate that all procedures followed were in accordance with institutional guidelines. References Citations should be listed in order of appearance in the text, and between square brackets [ ]. References must be listed at the end of the text in the order of citation. Journal titles should be abbreviated according to Index Medicus. For citation rules not specified here, authors should refer to the NLM Style Guide for Authors, Editors, and Publishers (http://www.nlm. nih.gov/citingmedicine). All references must be checked by the author(s). Ê To the kind attention of the authors, you’re gently invited to pair your article’s references with the corresponding doi Journal article: Rossini R, Angiolillo DJ, Musumeci G, et al. Aspirin desensitization in patients undergoing percutaneous coronary interventions with stent implantation. Am J Cardiol. 2008; 101: 786-789. doi: 10.1016/j. amjcard.2007.10.045 Journal article if the number of authors is more than six: list the Àrst three authors followed by et al. De Luca G, Verdoia M, Binda G, et al. Aspirin desensitization in patients undergoing planned or urgent coronary stent implantation. A single-center experience. Int J Cardiol. 2013; 167 (2): 561-563. doi: 10.1016/j.ijcard.2012.01.063. Entire book: Valavanis A. Medical radiology: interventional neuroradiology. Heidelberg: Springer Verlag; 1993. Part of book if number of authors is more than two: Bonneville JF, Clarisse J, et al. Radiologie interventionnelle. In: Manelfe C ed. Imagerie du rachis et de la moelle. Paris: Vigot Editeur; 1989. p. 761-776. Tables Each table must be typed on a separate sheet and double-spaced, if possible. Tables should be numbered using the Arabic system, followed by a brief informative title. Use type of the same font and size as employed in the text. Include footnotes at the bottom of each table. Tables must be numbered in the order cited in the text. Tables should not duplicate data given in the text or figures. Figure Legends Provide figure legends on a separate sheet. Legends must be doublespaced, and figures must be numbered in the order cited in the text. Figures Figures should preferably be submitted online in .tiff format to: http:// www.editorialmanager.com/nrj/. The time required to send files will vary depending on the number of figures, but image resolution must not be reduced to decrease transmission time. When labeling the figures, please ensure that the label corresponds to the figure number. Digital images (originals or images acquired by a scanner) must meet the following criteria: Black-and-white figures: Images must be acquired using the grey scale with a minimum resolution of 300 pixels per inch or 150 pixels per cm. Images must have a base of at least 8.1 cm for one item or a minimum base of 16.9 cm for several items. Color figures: Images must be acquired using the full color CMYK method with a minimum resolution of 300 pixels per inch or 150 pixels per cm. Images must have a base of at least 8.1 cm for one item or a minimum base of 16.9 cm for several items. The RGB method is recommended for video reproductions only, as the quality of printed figures is poor. Images must be saved in .tiff format. Images in .jpg format are not acceptable as details tend to be lost upon scanning, even at high resolution. Image definition also depends on the enlargement factor. Thus, a large low-resolution image can be proportionally reduced (by 24%) for publication, thereby permitting optimal presentation in print. However, enlargement of a small high-resolution image will highlight all flaws, yielding a pixelated effect. Do not submit figures already paged in Word, PowerPoint, or other documents, or images inserted in web pages. Such images are of low resolution and are unsuitable for printing. Figures in .dcm (Dicom) format may be submitted as .dcm or .tiff files. The editorial office will process such images for printing. Illustrations may be compressed using the StuffIt, Aladdin, or Zip programs. Do not label an image with arrows, numbers, or letters, indicate on a duplicate copy or a sketch where such indications are desired. Authors are advised to refer to the Journal guidelines when formatting their work. Otherwise the editorial office may return a submission to the authors for improvement before any Editor is assigned. If it is not possible to send figures via the Internet, images may be sent by express courier in one of the following formats: Original x-ray films, slides, or glossy or opaque prints. Clearly indicate on the back of the image the first author’s name and the number corresponding to the figure caption and citation in the text. Figures may be submitted on a CD-ROM or DVD, copied in ISO9006 format legible on both PC and MAC. Floppy disks are not recommended. Clearly indicate on one side of the figure both the top of the figure and the figure number. Do not label the actual image with arrows, numbers, or letters, but rather indicate the top and figure number on a duplicate copy or on a sketch. Do not cut or attach figures with adhesive tape or use paper clips. If images are sent offline, thumbnails of the images should be uploaded as a submission item online, together with the manuscript, to allow the publishers to reserve space for the original highresolution images. Authors are reminded that manuscripts must be sent online via Editorial Manager even when authors choose off-line figure submission. All off-line material should be sent to the editorial office: In case of difficulties or problems please contact: Prof. Marco Leonardi at marco.leonardi@centauro.it The Neuroradiology Journal Centauro S.R.L. Via del Pratello, 8 - I-40122 Bologna – Italy E-mail: marco.leonardi@centauro.it Rivista di proprietà EDIZIONI DEL CENTAURO s.r.l. - Edita da CENTAURO s.r.l. - Sede legale: Via del Pratello, 8 – I-40122 Bologna C.F. e P. IVA IT 01896531207 – Cap. soc. Euro 100.000,00 i.v. – Reg. Impr. BO-R.E.A. BO n. 397358 www.centauro.it THE NEURORADIOLOGY JOURNAL Conflicts of Interest/Disclosures, Copyright Transfer Agreement, Financial Support, Exclusive Publication Statement, and Author Contribution Form Date: _______________ With respect to the article entitled: _________________________________________________________ The authors certify that: (Conflicts of Interest/Disclosures) The corresponding author must disclose all relevant financial, personal, and/or professional relationships with other people or organizations related to the subject of this article. Is there any actual or potential conflict of interest in the subject of this article? No (___). If yes, please give details in the space below. (Copyright Transfer Agreement) All rights, title, and interest in the manuscript, including copyright ownership, are transferred to Centauro Srl Publisher, Bologna. Authors reserve all proprietary rights (such as patent rights) other than copyright and the right to use parts of this article in their own subsequent work as soon as the article is accepted for publication. The article may not be published elsewhere without permission from the publisher. (Financial Support) If the authors or their institution(s) is/are affiliated with any organization having a financial interest (including the provision of financial support) in the subject of this article, the authors should indicate (a) the name of the industry/institution providing support for the study, (b) the type of affiliation of the organization (stockholder, consultant, donor of honorarium, granting body, source of other financial or material support, provider of equipment, or other support category), and (c) that the authors had complete control of the data and information submitted for publication. Any other non-financial conflict of interest should also be disclosed to the Editor, with the understanding that the information may be published if deemed appropriate by the Editor. No, I (We) do not have any conflict of interest related to this article (___). If yes, (a) (_________) (b) (_________) (c) (_________). (Exclusive Publication Statement) None of the material in this manuscript has been published previously or submitted for publication elsewhere prior to appearance in NRJ. (Author Contribution Form) The corresponding author must certify that all authors participated in and contributed sufficiently to the conception and design of the work, and analysis of the data, as well as the writing of the manuscript, and assume public responsibility for the content of the paper. The corresponding author also certifies that all authors have reviewed the final version of the manuscript and approved it for publication. Name (Print) Signature Date This signed statement must be received before the manuscript can be processed for publication. Please print, fill and send by e mail or fax to Marco Leonardi, +39 051 220099 – marco.leonardi@centauro.it The Neuroradiology Journal Centauro s.r.l. - Via del Pratello, 8 - I-40122 Bologna - Italy - Fax: +39.051.220099 - E-mail: marco.leonardi@centauro.it DELTAWIND ™ microcoil technology The revolutionary DELTA WIND™ technology features a unique rounded triangular shaped wind which provides hundreds of natural micro-deflection points, allowing the coil to change direction at will. The result is a coil with amazing softness, stability and dense packing ability. DELTAPLUSH ™ microcoil DELTAPLUSH™mircocoils are engineered with the softest.0013” wire ever used by Micrus, providing the flexibility to find empty spaces and the softness to thoroughly fill. DELTAPLUSH™- the ideal finishing coil Relative softness Comparison of Finishing Coils* Spring Constant Soft More Soft DELTAPLUSH™ Microcoil Competitive Coil A ULTIPAQ® Microcoil Competitive Coil A DELTAPLUSH measures 11% softer than Comparative Coil A and 43% softer than Comparative Coil B* * Data on File, Micrus Endovascular Corporation d istrib ute d b y Via Ne rvia no , 31 20020 La ina te (Mi) te l. +39 02 93305.1 fa x.+39 02 93305400 www.a b me d ic a .it More Innovation. More Validation. Better Outcomes. Solitaire 2 Revascularization Device ce ™ Delivering more to transform acute ischemic stroke treatment. MORE RIGOROUSLY TESTED With more than 1,000 patients enrolled across 6 diferent studies, the Solitaire ™ 2 device is the most extensively researched mechanical thrombectomy device available.1-6 BETTER In every study, patient outcomes improved across all Modiied Rankin Scale, with up to 62% achieving mRS 0-2 at 90 days.1-7 LOWEST he Solitaire ™ 2 device has been proven to offer the lowest observed mortality rate in any signiicant prospective study of mechanical thrombectomy devices.1-7 NEUROLOGICAL OUTCOMES MORTALITY REFERENCES 1. Pereira VM, Gralla J, Davalos A, et al. Prospective, multicenter, single-arm study of mechanical thrombectomy using Solitaire Flow Restoration in acute ischemic stroke. Stroke. 2013;44(10):2802-2807. 2. Saver JL, Jahan R, Levy EI, et al; SWIFT Trialists. Solitaire flow restoration device versus the Merci Retriever in patients with acute ischaemic stroke (SWIFT): a randomised, parallel-group, non-inferiority trial. Lancet. 2012;380(9849):1241-1249. 3. Jahan R, Liebeskind D, Nogueira R, et al; for SWIFT Investigators. Abstract 163: TICI success rates in SWIFT: comparison between randomized arms and correlation to 90 day neurologic outcome. Stroke. 2013;44:A163. 4. Dávalos A, Pereira VM, Chapot R, et al; Solitaire Group. Retrospective multicenter study of Solitaire FR for revascularization in the treatment of acute ischemic stroke. Stroke. 2012;43(10):26992705. 5. Schroth G. Endovascular stroke therapy in Bern: 20 years of experience. Presented at: 12th Congress of the World Federation of Interventional and Therapeutic Neuroradiology; November 9–13, 2013; Buenos Aires, Argentina. 6. Nguyen T, Malisch T, Castonguay A, et al. O-004 Balloon guide catheter improves recanalisation, procedure time, and clinical outcomes with Solitaire in acute stroke: analysis of the NASA Registry. J NeuroIntervent Surg. 2013;5:A2-A3. 7. Nogueira RG, et al. Trevo versus Merci retrievers for thrombectomy revascularisation of large vessel occlusions in acute ischaemic stroke (TREVO 2): A randomised trial. Lancet. 2012:DOI:10.1016/ S0140-6736(12)61299-9. The Solitaire™ 2 Revascularization Device is intended to restore blood flow by removing thrombus from a large intracranial vessel in patients experiencing ischemic stroke within 8 hours of symptom onset. Patients who are ineligible for intravenous tissue plasminogen activator (IV t-PA) or who fail IV t-PA therapy are candidates for treatment. Indications, contraindications, warnings and instructions for use can be found on the product labeling supplied with each device. CAUTION: Federal (USA) law restricts this device to sale by or on the order of a physician. COVIDIEN, COVIDIEN with logo, and Covidien logo are US and internationally registered trademarks of Covidien AG. Other brands are trademarks of a Covidien company. ©2014 Covidien. MK1027012014A Learn more at covidien.com/solitaire