Management of the Retained Knife Blade

World J Surg (2010) 34:1648–1652 DOI 10.1007/s00268-010-0514-4 Management of the Retained Knife Blade Sanju Sobnach • Andrew Nicol • Hassed Nathire Delawir Kahn • Pradeep Navsaria • Published online: 2 March 2010  Société Internationale de Chirurgie 2010 Abstract Background The retained knife blade is an unusual and spectacular injury. The aim of this study was to review our experience with the management of such injuries. Methods A retrospective chart review of patients with retained knife blades treated at Groote Schuur Hospital Trauma Centre from January 1996 to December 2007 was undertaken. Results Thirty-three patients with retained knife blades were identified. Site of wound entry was the thorax in 13 patients (40%), the neck and back in 7 patients (21%) each, upper and lower extremities in 4 (12%), and the face and abdomen in 1 patient (3%) each. Thirty patients (91%) were hemodynamically stable on admission; two (6%) presented with wound abscesses, and one patient (3%) with active bleeding required emergency surgery. All 33 blades were extracted after clinical and radiological assessment. Simple withdrawal of the blade was possible in 19 cases (58%) and the likelihood of post-extraction bleeding was only 5%. Thirteen patients (40%) required an open surgical approach through dissection of the entry wound, laparotomy, or thoracotomy. Video-assisted thoracoscopic removal was used in one case. Retained thoracic blades were significantly associated with postoperative sepsis (P = 0.0054). There were no deaths. Presented at the 36th Annual Meeting of the Surgical Research Society of Southern Africa, Cape Town, Republic of South Africa, 3-4 July 2008. Published in abstract form as ‘‘The management of the retained knife blade at a major trauma center’’ [S Afr J Surg 2009; 47:25]. S. Sobnach
A. Nicol
H. Nathire
D. Kahn
P. Navsaria (&) Trauma Center, Groote Schuur Hospital, University of Cape Town, Observatory, 7925 Cape Town, South Africa e-mail: pradeep.navsaria@uct.ac.za 123 Conclusions All impacted knife injuries require careful clinical and radiological assessment. Simple withdrawal can be performed safely in the emergency room provided potential life-threatening vascular and solid organ injuries have been excluded. There should be a low threshold for investigating and treating patients with retained intrathoracic blades for postoperative sepsis. Introduction The retained knife blade remains an uncommon injury. Diagnostic and management challenges are often encountered with the extraction of impacted knives. Unplanned extraction of the retained knife can result in massive hemorrhage, hemodynamic deterioration, and death [1]. Most centers have very limited experience with this uncommon injury and there are only isolated case reports and small series published in the literature on the subject [2–8]. We therefore reviewed our experience with the management of retained knife blades in a high-volume level-one urban trauma center. Patients and methods The records of all patients who underwent surgery for a retained knife blade in the Trauma Centre at Groote Schuur Hospital, Cape Town, from January 1996 to December 2007 were retrospectively reviewed. Demographic data, injury pattern and presentation, investigations, associated injuries, treatment, duration of hospital stay, complications, and outcome were sought. Injury severity was categorized using the Revised Trauma Score (RTS). All patients were managed along Advanced Trauma Life Support (ATLS) World J Surg (2010) 34:1648–1652 guidelines. Patients who were in shock, either from stab wounds or concomitant injuries, were taken for emergency surgery. Hemodynamically stable patients and those who stabilized after simple resuscitation underwent further evaluation. Patients with retained intracranial and vertebral column blades were managed by the neurosurgical service. The site of entry of the blade was assessed clinically and radiographically. Anterioposterior (AP) and lateral X-rays were performed in all patients to determine the position of the retained knife. When plain films were equivocal or the retained knife was close to vital structures, the location was determined more precisely using computed axial tomography (CAT) scanning. Contrast swallow was performed in patients with suspected esophageal injury. Angiography was used when a vascular injury could not be ruled out on CAT scanning. Our unit has limited experience with CT angiography (CT-A) since it was introduced in our trauma center only after the completion of this study. We are currently conducting a prospective study to investigate the use of CT-A in the management of penetrating neck injuries. All patients were managed in the operating room (OR) under general or local anesthesia. Surgical approach included simple extraction, wound exploration and extraction, or open operation and extraction defined as follows: • • • Simple extraction: retained knife was withdrawn along its line of entry with no further surgical intervention. This approach was reserved for patients whose clinical examination and minimal investigations excluded solid organ/hollow viscous and neurovascular injuries. Wound exploration and extraction: entry wound was surgically extended and the retained blade was extracted under direct vision. Open operation and extraction: retained blades involving deeper structures and those not externally visible required dissection of the wound entry, laparotomy, or thoracotomy followed by extraction under direct vision. Bleeding after extraction of the embedded knife and management thereof were documented. Statistical analysis was performed using SPSS v15.0. The v2 test and Fisher’s exact test were used to compare proportions; a two-sided P \ 0.05 was considered statistically significant. 1649 injury-to-admission interval of 4 (range = 1–89) days. The median total hospital stay was 3 (range = 2–65) days. The site of wound entry was the thorax in 13 patients (40%), the neck in 7 patients (21%), the back in 7 patients (21%), upper and lower extremities in 4 patients (12%), and the face and abdomen in 1 patient (3%) each. Clinical presentation and blade characteristics are summarized in Table 1. All but one patient presented hemodynamically stable. One patient with an impacted knife injury to the neck presented with active bleeding and required emergency surgery which showed a lacerated subclavian artery. The blade was carefully withdrawn along its line of entry and the artery primarily repaired. In 25 patients (76%), the entire knife (handle and blade) was present, and in 5 patients the blade was palpable subcutaneously. Two patients presented with abscess formation at the entry site, with the blade palpable subcutaneously in both cases. CT of the chest was performed in three patients (9%) and showed lung parenchyma injuries in two cases and no injury in one. One patient needed a contrast swallow which confirmed an esophageal injury. Ten patients (30%) required angiography; the only positive finding was an arteriovenous fistula (AVF) between the left internal iliac artery and vein in one patient. The AVF was radiologically embolized and the retained blade was removed under direct vision after wound exploration in the operating room. The surgical approach used in the 33 patients is documented in Table 2. Thirty-one patients (94%) required general anesthesia and local anesthesia was used in two patients (6%). During surgery, 18 patients were put into the supine position, 14 in the lateral position, and 1 was operated on in the prone position. Ventilation was achieved through a single-lumen endotracheal tube in 23 patients (70%), double-lumen endotracheal tube in 5 cases (15%), laryngeal mask airway in 3 patients (9%), and face-mask oxygen in 2 patients (6%). Associated injuries are listed in Table 3. Post extraction bleeding was seen in only one Table 1 Presentation and retained blade characteristics in 33 patients Clinical presentation and retained knife characteristics N (%) Hemodynamically stable 32 (97) Whole knife in situ (25) BPS, no abscess formation (5) BPS, abscess formation (2) Results During the study period, 33 patients with retained knives were identified. There were 32 males (97%) and 1 female with a median age of 29 (range = 15–50) years. The median Revised Trauma Score was 7.84 (range = 5.64– 7.84). Six patients (18%) presented late, with a median Active bleeding (whole knife in situ) Fractured/bent blade 1 (3) 15 (45) Impacted in Bone Solid organ Other (body cavity, muscle, blood vessel) 10 (30) 3 (9) 15 (45) BPS blade palpable subcutaneously 123 1650 World J Surg (2010) 34:1648–1652 Table 2 Surgical management of 33 retained knife blades Procedure N (%) Simple extraction 19 (58) Open operation 10 (30) Thoracotomy 5 Neck exploration 2 Supraclavicular subclavian vessel exposure 1 Infraclavicular subclavian/axillary vessel exposure 1 Laparotomy 1 Wound exploration and extraction 3 (9) Video-assisted thoracoscopy 1 (3) Table 3 Injuries management associated with retained blades and their Injury N (%) Management Hemopneumothorax 8 (24) Tube thoracostomy Vascular 4 (12) Arteriovenous fistula 1 Radiological embolization Subclavian artery 1 Primary repair Axillary vein 1 Ligation Internal jugular vein 1 Ligation Liver (grade 2 liver injury) Stomach (1-cm anterior wall gastric perforation) 2 (6) 1 (3) Laparotomy and drainage Laparotomy and primary repair Esophagus (2-cm perforation, \50% circumference) 1 (3) Primary repair and drainage Brachial plexus (transection) 1 (3) Lung parenchyma (grade 2 lung injury) 1 (3) Primary repair by hand surgeon Tube thoracostomy patient with a stab wound to the neck. This patient was hemodynamically stable on admission with a clinical examination unremarkable for any signs of vascular injury. At neck exploration a transected internal jugular vein was ligated. Complications were limited to five patients (15%) with thoracic injuries. Retained thoracic blades were significantly associated with postoperative sepsis when compared to all other impacted knives (38 vs. 0%, P = 0.0054). Three patients developed empyemas, two of whom required repeat tube thoracostomy and one formal pleural washout via thoracotomy. Wound sepsis and pneumonia were seen in the other two cases and were managed with local wound care and antibiotics, respectively. There were no deaths. Discussion Some of the largest published series dealing with retained blades have emanated from South Africa. Grobbelaar and 123 Knottenbelt [4] described a series of 11 patients with retained blades in the face at our trauma center, and Hudson [5] reported 4 similar cases managed by simple extraction. The spectacular appearance of this injury along with its rarity often leads to the management focusing on the retained implement rather than on initial resuscitation of the patient [9, 10]. Clinical examination should be performed systematically; assessment of the extent and depth of the entry wound along with the surface anatomy can point to specific organ and neurovascular injuries. A high index of suspicion for a retained blade should be entertained when the site of wound entry is more than superficial and the whole knife has not been recovered. Detailed history from paramedics or witnesses can be helpful in determining whether the knife was intact after stabbing [1]. Delayed presentation is not unusual and occurred in six patients (18%), of whom four presented with subcutaneous swellings and two with wound abscesses. Significant injury-to-admission intervals of 12 and 89 days were noted in the two patients with wound abscesses. In a recent South African study, Van Lierop et al. [7] made a similar observation, reporting two patients with retained blades where the diagnosis was delayed by more than 8 weeks. An abnormal delay in presentation is associated with adverse outcomes and can lead to mediastinal abscesses, empyema, aneurysm of major vessels, osteomyelitis, intra-abdominal sepsis, and septicemia [11, 12]. This highlights the importance of routine radiographic examination for all deep stab wounds so that early diagnosis of retained blades can be made. Spiral CT may help in defining the relation between the retained knife and major visceral structures, thus facilitating safer extraction. Angiography must be performed when there are inconclusive CT scans. It is also extremely valuable for planning the surgical approach and can be used for selective embolization of vascular injuries. Patients with retained blades and major vessel injuries can be stable hemodynamically on admission because the blade can act as a tamponade and prevent hemorrhage. All but one of our patients were hemodynamically stable on admission. In a series of 154 stab wounds, Hanoch et al. [1] reported hemodynamic collapse in four patients following unplanned extraction of retained knives by the victim or inexperienced medical personnel; intraoperative and postmortem findings confirmed the tamponading effect of the blade in all four cases. Retained implements have also been described to provide a tamponading effect in solid organs such as the liver, spleen, and kidney [11]. As in previously described studies [1], simple extraction was sufficient for the management of the majority (58%) of our patients; the rate of post-extraction bleeding was only 5% (1 in 19 patients). Impalement of knife blades into bony structures occurred in up to 30% of our patients; heavy instruments such as World J Surg (2010) 34:1648–1652 Kocher’s or Robert’s forceps were useful in extracting these impacted knives. Following any impacted knife injury, clothing of the victim can be carried within the body with extensive debris along the path of impalement. Tissue along the wound entry is dirty and contused and requires adequate debridement and irrigation to prevent infection. Tetanus prophylaxis and broad-spectrum antibiotics are therefore essential components of care [9]. We managed intraperitoneal injuries through laparotomy, while thoracic injuries mandated thoracotomy for exposure of the greater vessels. In all cases of suspected vascular injury, surgical approach allowed for good proximal and distal control of vascular structures prior to removal of the retained implement. Utmost care is necessary when extracting the knife to prevent further aggravation of organ or vascular injuries by rotation of the blade as well as injury to the surgeon by its sharp edges [11, 13]. Reparative rather than resectional surgery was sufficient for the management of visceral and solid organ injuries during our study. The advent of video-assisted thoracic surgery (VATS) as a minimally invasive technique has elevated thoracoscopy to a diagnostic and therapeutic tool for traumatic intrathoracic lesions in hemodynamically stable patients at our center [14, 15]. VATS has proven to be safe, less invasive than thoracotomy, and extremely effective in the extraction of retained knives in previous studies [16–18]. It can simultaneously be used to diagnose lung, pericardial, and diaphragmatic injuries caused by the injurious event. Retained knife blades have rarely been reported in the literature, apart from case reports [2–7]. There are currently no published guidelines describing whether retained blades should be withdrawn in the emergency room (ER) or the OR. Although our study is limited by its small number of patients, our experience suggests that all patients should be investigated prior to extraction of the retained blade. Minimal investigations should include AP and lateral X-rays where appropriate, and the injury complex should guide further invasive investigations. Simple extraction can be attempted in the ER since the likelihood of post-extraction bleeding is only 5% in this subgroup of patients. However, the procedure should be performed with an open surgical pack to manage a potential catastrophe. For example, the only patient who bled after simple withdrawal of the knife blade was found to have a transected internal jugular vein at formal neck exploration, which required ligation. This active bleeding initially could have been managed in the ER by placing the patient in the Trendelenburg position to minimize the risk of air embolism and applying manual pressure. Foley catheter balloon tamponade would have also been helpful in controlling initial hemorrhage prior to taking the patient to the OR [19]. Should a patient require any procedure more invasive than simple withdrawal, the 1651 extraction of the retained knife blade must be performed in the OR. Our study also shows that there is a significant association (P = 0.0054) between retained thoracic blades and postoperative sepsis. These patients therefore should be monitored closely postoperatively with a low threshold for investigating and treating sepsis. Anesthetic management and resuscitation of any patient with a knife in situ can be challenging [20]. The injury complex considerably determines each surgical procedure, and patient positioning may pose a challenge for the anesthetist and the surgeon. Surgical positioning for anesthesia should be guided by the most likely injuries but can often be limited by the presence of the retained implement. It has been suggested that patients can be induced during anesthesia in the lateral position, although this technique can take longer due to lack of familiarity with its use. Fiberoptic intubation has been used successfully in patients who need to be anesthetized in the prone position [21]. Adequate communication between the surgeon and the anesthetist prior to the surgery therefore is essential. Conclusions Initial management of all patients with retained knife blades should be conducted along standard ATLS guidelines, with careful clinical and radiological assessment being paramount prior to the extraction of impacted knives. Simple withdrawal of retained knife blades can be performed safely in the emergency room since the likelihood of post-extraction bleeding is only 5%; however, potential lifethreatening vascular and solid organ injuries must be excluded prior to knife withdrawal. Adequate facilities (such as an open surgical tray or an OR on standby) must also be readily available to manage any surgical catastrophe following simple withdrawal. Surgical exploration of any wound for the extraction of an impacted knife blade should be performed in a controlled environment of the OR. Retained thoracic knife blades are significantly associated with postoperative sepsis and these patients must be monitored very closely after surgery. There is a place for the use of minimally invasive techniques in carefully selected patients. References 1. Hanoch J, Feigin E, Pikarsky A et al (1996) Stab wound associated with terrorist attacks in Israel. JAMA 276:388–390 2. Wilson A, McNatt S (2006) A unique zone II neck injury. J Trauma 60:1378 3. Frangos SG, Ben-Arie E, Bernstein MP et al (2006) Thoracic stab wound with impaled knife. J Trauma 60:1379 123 1652 4. Grobbelaar A, Knottenbelt JD (1991) Retained knife blades in stab wounds of the face: is simple withdrawal safe? Injury 22: 29–31 5. Hudson DA (1992) Impacted knife injuries of the face. Br J Plast Surg 45:222–224 6. Daya NP, Liversage HL (2004) Penetrating stab wound injuries to the face. SADJ 59:55–59 7. Van Lierop AC, Raynham O, Basson O et al (2008) Retained knife blades in the ear, nose and throat: three cases. J Laryngol Otol 3:1–5 8. Taylor AG, Peter JC (1997) Patients with retained transcranial knife blades: a high-risk group. J Neurosurg 87:512–515 9. Thomson BN, Knight SR (2000) Bilateral thoracoabdominal impalement: avoiding pitfalls in the management of impalement injuries. J Trauma 49:1135–1137 10. Kelly IP, Attwood SE, Quilan W et al (1995) The management of impalement injury. Injury 26:191–193 11. Madhok BM, Roy DD, Yeluri S (2005) Penetrating arrow injuries in Western India. Injury 36:1045–1050 12. Fingleton LJ (1987) Arrow wounds to the heart and mediastinum. Br J Surg 74:126–128 13. Karger B, Sudhues H, Kneubuehl BP et al (1998) Experimental arrow wounds: ballistics and traumatology. J Trauma 45:495–501 123 View publication stats World J Surg (2010) 34:1648–1652 14. Navsaria PH, Vogel RJ, Nicol AJ (2004) Thoracoscopic evacuation of retained posttraumatic haemothorax. Ann Thorac Surg 78:282–286 15. Navsaria PH, Nicol AJ (2006) Video-assisted thoracoscopic pericardial window for penetrating cardiac trauma. S Afr J Surg 44:18–20 16. Bar I, Rivkind A, Deeb M et al (1998) Thoracoscopically guided extraction of an embedded knife from the chest. J Trauma 44:222–223 17. Burach JH, Amilraj EA (2005) Thoracoscopic removal of a knife impaled in the chest. J Thor Cardiovasc Surg 130:1213–1214 18. Williams CG, Haut ER, Ouyang H et al (2005) Video-assisted thoracic surgery removal of foreign bodies after penetrating chest trauma. J Am Coll Surg 202:848–852 19. Navsaria P, Maximilien T, Nicol A (2006) Foley catheter balloon tamponade for life-threatening haemorrhage in penetrating neck trauma. World J Surg 30:1265–1268 20. McCarthy T, Lassota-Korba B, O’Leary T et al (2007) Chest compressions for a patient in cardiac arrest after penetrating trauma with a knife still in situ. Emerg Med J 24:596–597 21. Lipp M, Mihaljevic V, Jakob H et al (1993) Fibreoptic intubation in the prone position. Anesthesia in a thoracoabdominal knife stab wound. Anaesthetist 42:305–308