CASE REPORTS Experience with an Ultrasound Stethoscope Eleni C. Vourvouri, MD, Don Poldermans, MD, Johan De Sutter, MD, Fabiola B. Sozzi, MD, Paolo Izzo, MD, and Jos R.T.C. Roelandt, MD, Rotterdam, The Netherlands Background: To test the diagnostic potential of the SonoHeart, a battery-powered hand-held ultrasound imaging device, in an outpatient clinic setting. Methods: A total of 114 patients with a variety of cardiac diseases were examined by 2 independent cardiologists with the hand-held device using the standard echocardiographic system (SE) as a reference. Global right ventricular (RV) and left ventricular (LV) function (scored as normal, mildly to moderately, or severely reduced) and internal cavity dimensions were assessed. Regional wall motion of 6 segments using a 2-point score (1 = normal wall motion, 2 = abnormal wall motion) was evaluated in 34 patients on-line. Results: There was a good agreement between the 2 imaging devices for evaluation of global LV (93%) and RV function (99%), regional wall motion (90%), dimensions of the LV (99%) and the RV (99%), and the left (96%) and right atria (99%). Furthermore, SonoHeart identified hypertrophic cardiomyopathy, pericardial effusion, and abnormalities of valves. Conclusion: The SonoHeart device allows rapid and accurate diagnosis, whenever needed in the outpatient clinic. (J Am Soc Echocardiogr 2002;15:80-5.) Cardiac ultrasound has become an important and STUDY PATIENTS AND METHODS effective noninvasive diagnostic imaging tool over the years, and better imaging performance has been realized with more expensive and complicated systems.To broaden the availability of ultrasound imaging and to increase its versatility in application, small, easy-to-use, and low-cost echocardiographic devices have recently been developed.These devices act like an ultrasound stethoscope (from the Greek words stethos = chest and skopein = see), being ultraportable and providing information beyond physical examination at the bedside. The aim of this study was to evaluate the potential and diagnostic accuracy of a prototype hand-held ultrasound system (SonoHeart System, SonoSite, Inc, Bothell, Wash) for the assessment of cardiac pathomorphology and global and regional function in an unselected outpatient population, using the results of high-end 2-dimensional (2D) echocardiographic equipment for performance comparison. From the Department of Cardiology, Thoraxcenter, Erasmus Medical Center Rotterdam, The Netherlands. Reprint requests: Prof Dr J.R.T.C. Roelandt, MD, Department of Cardiology, Thoraxcenter, Erasmus Medical Centre, Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands. (E-mail: roelandt@card.azr.nl). Copyright © 2002 by the American Society of Echocardiography. 0894-7317/2002/$35.00 + 0 27/4/115268 doi:10.1067/mje.2002.115268 80 The Ultrasound Stethoscope The SonoHeart (SonoSite Inc) hand-held ultrasound system (Figure 1) is a small hand-held ultrasound device measuring 33.8 × 19.3 × 6.1 cm and weighing 2.4 kg. It is equipped with a small (15 mm) 2 to 4 MHz phased-array broadband transducer and operates on a rechargeable lithium ion battery or alternating current.The 2D control settings comparable with a standard echocardiographic device and a color power Doppler flow mapping are integrated to the unit. Quantitative assessment of the heart is possible with inclusive callipers (Figure 2). SonoHeart has a storage memory of 50 images and can be connected to a video recorder, a printer, or an external monitor. Study Population We studied 114 unselected patients (72 men) with a mean age of 52 ± 17 years, referred to our outpatient clinic for left ventricular (LV) function assessment (n = 60), followup of known congenital heart disease (n = 13), follow-up of known valvular disease (n = 15), evaluation of LV hypertrophy (LVH) (n = 15), and evaluation of pericardial effusion (n = 11). Study Design All patients underwent 2 consecutive echocardiographic examinations by 2 different investigators: one examination by means of a standard echocardiographic system (SE), Hewlett-Packard (Sonos 5500, Andover, Mass) or Vingmed (System V, Horten, Norway), and the other by means of the Journal of the American Society of Echocardiography Volume 15 Number 1 Vourvouri et al 81 Table 1 Accuracy of left atrial size assessment (111 patients) Ultrasound stethoscope 2D standard echo <4 cm 4-6 cm >6 cm Normal Enlarged Severely enlarged 87 1 0 0 21 0 0 0 2 Kappa, 0.974 2D, Two-dimensional. Table 2 Accuracy of left ventricular size assessment (111 patients) Ultrasound stethoscope 2D standard echo <5.5 cm 5.5-7 cm >7 cm Normal Enlarged Severely enlarged 91 0 0 1 19 0 0 0 0 Kappa, 0.969 2D, Two-dimensional SonoHeart device. Each investigator was blinded to the results of the other investigator.The time spent for the bedside examination with SonoHeart was always less than 5 minutes. The patients were evaluated on-line for the following parameters: presence of pericardial effusion, left atrial (LA) enlargement (internal diameter >40 mm in the parasternal long axis view), LV enlargement (end-diastolic diameter >55 mm in the parasternal long axis view),LVH (defined as a septal thickness >12 mm), right atrial (RA) enlargement (internal transverse diameter >35 mm in the 4-chamber view), and right ventricular (RV) enlargement (end-diastolic transverse diameter >43 mm in the 4-chamber view).Also, gross morphologic assessment of the valves was performed. In all patients the global RV function and the LV ejection fraction (EF), and in 34 patients, who where referred for stress echocardiography, the regional wall motion were estimated. With both imaging techniques, the scoring was performed on-line during the examination procedure. Global LV function was defined as normal when the estimated EF was greater than 55%, mildly to moderately reduced when the estimated EF was 35% to 55%, and severely reduced when the estimated EF was less than 35%.The scoring of the regional wall motion for testing the efficacy and feasibility of the SonoHeart was simplified, by dividing the LV into 6 segments (anterior, inferior, septal posterior, septal anterior, lateral, and posterior wall) and by using a 2-point score: 1 for normal wall motion, including mild hypokinesia, and 2 for abnormal wall motion, including severe hypokinesia, akinesia, and dyskinesia. Figure 1 Photograph of SonoHeart device, hand-held ultrasound imager, used in this study. Statistical Analysis The agreement for segmental and global wall motion was assessed from 2 × 2 and 3 × 3 tables with weighted kappa statistics. Kappa values 0.4, between 0.4 and 0.75, and 0.75 or greater were considered to represent poor, fair-to-good, and excellent agreement respectively, based on Fleiss’s classification.1 RESULTS Visualization Three patients had poor echo windows with both imaging techniques and were excluded from the study, leaving 111 patients for analysis. Agreement The results for detection of LA and LV enlargement with both examination techniques are summarized in Tables 1 and 2. The agreement for RV and RA size was 99%. A thrombus in the RA was identified with both imaging techniques in 1 patient. Pericardial effusion was correctly assessed with SonoHeart in 11 patients and LVH (Figure 3) in 7 of Journal of the American Society of Echocardiography January 2002 82 Vourvouri et al A B Figure 2 Diastolic (A) and systolic (B) left ventricular long-axis views of normal subject obtained with SonoHeart device. Measurements of diameter of left ventricle at end-diastole (A) and end-systole (B) are shown. LA = Left atrium; LV = left ventricle; AO = aorta; RV = right ventricle; IVS = interventricular septum. transposition of the great arteries, and subvalvular aortic membrane. Evaluation of Global and Regional Left and Right Ventricular Systolic Function The agreement between both imaging techniques for all 111 patients was 93% for global and 89% for regional LV systolic function (Tables 3 and 4). The agreement for global RV systolic function was 99%. Interobserver Variability Figure 3 Long-axis view of patient with hypertrophic cardiomyopathy obtained with SonoHeart device (abbreviations as in Figure 2). The arrow indicates the increased septal thickness. 8, resulting in an agreement of 100% and 87.5%, respectively. In 1 patient pericardiocentesis was performed under SonoHeart guidance (Figure 4). In 20 patients, the morphologic abnormalities of the aortic (calcification, Figure 5) and the mitral valve (calcification, prolapse) were identified with the SonoHeart device and confirmed by the SE. Furthermore, the diagnosis of a chordal rupture of the mitral valve was made by SonoHeart and verified by the SE in a patient presenting with acute dyspnea in the outpatient clinic. Finally, the SonoHeart identified the morphologic characteristics of surgically corrected congenital cardiac abnormalities in 13 follow-up patients, including atrial septum defect (ASD) (Figure 6), ventricular septum defect, tetralogy of Fallot, Ebstein anomaly, There was a good correlation of the 2 independent observers’ evaluations of regional wall motion between the SE (81%) and the SonoHeart device (76%). DISCUSSION In the late 1970s, Roelandt et al2-4 introduced the first portable battery-powered 2D echocardiographic instrument to be used as part of the clinical examination. Today, this vision of the past has become a reality as a result of miniaturization and digital technology. The SonoHeart device is easy to use, is ultraportable with excellent imaging quality, and gives in addition quantitative information about dimensions within the heart. Our study demonstrates the utility of such a device for assessing pathomorphology and function of the heart, and the results are comparable with those obtained by SE. We included in our study a group of patients with corrected congenital heart disease to test the clinical potential of the SonoHeart device in identifying the complex disease seen in these patients. It has proven to be as reliable as the SE for the rapid Journal of the American Society of Echocardiography Volume 15 Number 1 A Vourvouri et al 83 B Figure 4 Parasternal long-axis view of patient with pericardial effusion before (A) and after pericardiocentesis obtained with SonoHeart device (B) (abbreviations as in Figure 2; PE = pericardial effusion). A B Figure 5 Parasternal (A) and apical (B) long-axis views of patient with calcified aortic valve stenosis with SonoHeart device (abbreviations as in Figure 2). detection or exclusion of a morphologic abnormality. This does not indicate though that it can be substituted for the SE because most cardiac congenital diseases require extensive 2D and Doppler analyses. We proved that SonoHeart can become an extension of physical examination enhancing our clinical senses and increasing the accuracy and sensitivity of diagnosis. The ultraportability and ease of use suggest that similar data and results can be obtained in other clinical scenarios, which are indicated in Table 5. It can be used as a screening tool for cardiac abnormalities such as hypertrophic cardiomyopathy, LVH, ASD, abdominal aorta aneurysm,5 mitral valve prolapse (MVP), or enlargement of the heart. Prior studies that used limited imaging protocols for most of these diseases (hypertrophic cardiomyopathy,6 LVH,7-9 MVP10) have provided evidence that a limited imaging study is feasible. The SonoHeart device can be used as a rapid diagnostic tool in emergency scenarios where rapid decisions are essential. Thus, giving immediate answers about EF and regional wall motion abnormalities, presence of pericardial fluid or tamponade, or the cause of a new murmur in unstable patients, SonoHeart can play a decision-making role. Taking into account that most of the patients are referred for echocardiography to answer a single clinical question (like follow-up of EF or pericardial effusion after pericardiocentesis or search for the source of embolism), a new strategy of limiting echo services starts to emerge in daily clinical practice.11 Thus, a focused imaging protocol could be used to screen for the referral disorder and to proceed to a full study only in the discovery of any abnormality. Such a limited echo strategy could be most effectively applied with the use of a hand-held ultrasound device such as SonoHeart. Recently, Bruce et al5 demonstrated the efficacy and high accuracy of that device in screening patients at risk for abdominal aortic aneurysms. The implementation of a limited echo or screening policy will lead to cost savings in terms of bene- Journal of the American Society of Echocardiography January 2002 84 Vourvouri et al A B Figure 6 A, Foreshortened apical 4-chamber view of patient with atrial septum defect of secundum type (arrow) obtained with SonoHeart device (abbreviations as in Figure 2; RA = right atrium). B, Same view obtained with SE system (Hewlett-Packard). Table 3 Agreement between left ventricular ejection fraction estimated by the SonoHeart device and a standard echocardiographic system SonoHeart estimation of EF Standard echo estimation of EF >55% 35%-55% <35% >55% 35%-55% <35% 61 5 0 1 31 2 0 0 11 Agreement, 93%. Kappa, 0.871. EF, Ejection fraction. Number of patients: 111. The numbers inside the table express the number of patients. Table 4 Agreement of regional wall motion analysis between the SonoHeart and a standard echocardiographic system SonoHeart Standard echo Normal Abnormal Normal Abnormal 92 15 5 92 Agreement, 90%. Kappa, 0.80. Number of patients: 34. Number of segments: 204. The numbers inside the table express the number of segments. fit to all subjects, reducing the time spent imaging healthy patients. Limitations For the evaluation of regional wall motion, the LV was divided into 6 segments and not into 16 seg- Table 5 Clinical utility of the hand-held ultrasound device Emergency department, CCU, and ICU Hospital ward rounds Outpatient clinic Surgical theater Cardiac catheter laboratory Private office practice ments, which is proposed by the American Society of Echocardiography. Because this was an on-line assessment and we had no recording facilities, we limited the analysis in 6 myocardial segments. The same segments were independently scored later from the SE and compared with the data of the handheld device.The hand-held device cannot substitute for the SE, but our purpose was to demonstrate the feasibility and efficacy of the SonoHeart device as a screening device in recognizing and distinguishing normal from abnormal wall motion.Within this concept, the study proved that such small devices potentially could be used in acute coronary syndromes to exclude/identify regional wall motion abnormalities. Furthermore, because of the small number of patients who were included, this study should be considered as a pilot study for later wall motion studies. Another limitation is related to the fact that we focused on 2D evaluation of anatomy and function of the heart, without including the color flow modality. Refinements in the technology of color power Doppler flow mapping in the device since beginning this study have made this now reliable, however. In Europe, most of the studies are being performed by cardiologists. The SonoHeart instrument should not be used by sonographers but by a trained cardiologist. This study was performed by a junior Journal of the American Society of Echocardiography Volume 15 Number 1 Vourvouri et al 85 staff member who had experience in echocardiography. Training and licensing for noncardiologists to use these devices will become an important issue in the future. 4. Conclusion 5. The hand-held ultrasound device, performing as a “real” stethoscope, makes ultrasound imaging an excellent tool immediately available for the diagnosis and assessment of cardiac patients whenever cardiac physical examination is indicated, improving the health care service. 6. 7. 8. We are grateful to Wim B. Vletter and Jackie McGhie for their expert technical assistance and to Eric Boersma, PhD, for statistical advice. 9. REFERENCES 10. 1. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: John Wiley; 1981. 2. Roelandt JRTC, Wladimiroff JW, Baars AM. Ultrasonic real time imaging with a hand-held scanner, II: initial clinical experience. Ultrasound Med Biol 1978;4:93-7. 3. Ligtvoet CM, Rijsterborgh H, Kappen L, Bom N. Real time 11. ultrasonic imaging with a hand-held scanner, I: technical description. Ultrasound Med Biol 1978;4:91-2. Roelandt JRTC, Bom N, Hugenholtz P. The ultrasound cardioscope: a hand-held scanner for real-time cardiac imaging. J Clin Ultrasound 1980;8:221-5. Bruce CJ, Spittell PC, Montgomery SC, Bailey KR, Tajik AJ, Seward JB. Personal ultrasound imager: abdominal aortic aneurysm screening. J Am Soc Echoc 2000;13:674-9. Weidenbener EJ, Krauss MD, Waller BF, Taliercio CP. Incorporation of screening echocardiography in the preparticipation exam. Clin J Sport Med 1995;5:86-9. Shep SG, Frohlich ED. Limited echocardiography for hypertensive left ventricular hypertrophy. Hypertension 1997;29: 560-3. Black HR, Weltin G, Jaffe CC. Limited echocardiogram: a modification of standard echocardiography for use in the routine evaluation of patients with systemic hypertension. Am J Cardiol 1991;67:1027-30. Shub C, Tajik AJ, Sheps SG. Value of two-dimensional echocardiography and Doppler examination in the assessment of hypertensive patients: a pilot study. J Am Soc Echocardiogr 1995;8:280-4. Kimura BJ, Scott R, Willis CL, DeMaria AN. Accuracy and cost-effectiveness of single-view echocardiographic screening for suspected mitral valve prolapse. Am J Medicine 2000; 108:331-3 Kimura BJ, DeMaria AN. Indications for limited echocardiographic imaging: a mathematical model. J Am Soc Echocardiogr 2000;13:855-61.