Despite the great number of intubations and diagnostic

Transcription

1 Management of Iatrogenic Tracheobronchial Injuries: A Retrospective Analysis of 29 Cases Thomas Schneider, MD, Konstantina Storz, MD, Hendrik Dienemann, MD, PhD, and Hans Hoffmann, MD, PhD Department of Thoracic Surgery, Thoraxklinik, University of Heidelberg, Heidelberg, Germany Background. Tracheobronchial injuries are rare but potentially life-threatening complications of endotracheal intubations or endobronchial interventions. This retrospective analysis discusses the criteria for the operative and nonoperative management of tracheal lacerations. Methods. From July 1996 to June 2006, 29 patients with iatrogenic tracheobronchial injuries were diagnosed at our institution. The injury occurred during single-lumen tube intubation in 14 patients and during double-lumen tube intubation in 2 patients. Two ruptures were diagnosed after surgical tracheostomy, eight after dilational percutaneous tracheostomy, and three after interventional bronchoscopy. Results. The lacerations in 11 patients were superficial or were sufficiently covered by the esophagus, and they underwent conservative management. Bronchoscopy revealed healing per primam in every case. Surgical repair was done in 18 patients (62%). The transtracheal approach was used for repair in 7 patients; a right-sided posterolateral thoracotomy was performed in 11 patients with lacerations affecting the lower third of the trachea. Three surgical patients died from causes unrelated to the tracheal injury. No clinically evident mediastinitis or postoperative tracheobronchial stenosis was observed. Conclusions. The decision for operative or nonoperative treatment of iatrogenic tracheobronchial lacerations is determined by the ventilating situation and the local extent of the injury. Nonoperative management of iatrogenic tracheobronchial injuries may be a save option in patients with uncomplicated ventilation, superficial or sufficiently covered tears, and moderate and nonprogressive emphysema. Immediate surgical repair remains warranted in those patients who require mechanical ventilation that cannot be delivered past the laceration. (Ann Thorac Surg 2007;83:1960 4) 2007 by The Society of Thoracic Surgeons Despite the great number of intubations and diagnostic or therapeutic endobronchial interventions, iatrogenic tracheal lacerations are rare complications. The reported incidence is approximately 0.005% for orotracheal intubations [1], with double-lumen intubations at 0.05% to 0.19% [2]. The true incidence may, however, be largely underestimated because of the unknown number of worldwide intubations or endobronchial procedures that are performed daily. In recent years, the increase in percutaneous dilational tracheostomies has led to a higher incidence of tracheal lacerations [3]. In addition to injuries that may happen during tube placement with either technique, cuff-related ruptures in the intubated and mechanically ventilated patient may also occur, although known to be very seldom. These injuries are often related to an over-inflation or excentric inflation of the cuff [2, 4]. Symptoms of severe tracheobronchial injuries are dyspnea, hemoptysis, (massive) soft-tissue or mediastinal emphysema, and pneumothoraces. The diagnosis often is delayed because tracheobronchial injuries are not always readily apparent. Tracheobronchoscopy will reveal the Accepted for publication Jan 22, Address correspondence to Dr Hoffmann, Department of Thoracic Surgery, Thoraxklinik am Universitätsklinikum Heidelberg, Amalienstrasse 5, Heidelberg, D Germany; hans.hoffmann@urz.uniheidelberg.de. localization and the extent of the laceration. Very superficial tears are treated conservatively; however, most cases of transmural lacerations require surgical repair [5]. The classic surgical approach is right-sided posterolateral thoracotomy for lacerations of the thoracic trachea. For repair of lacerations of the cervical trachea, the cervical transtracheal approach can be performed [6, 7]. Surgical repair is the classic treatment of tracheal lacerations, but conservative management in selected cases has recently been reported [8 14]. In this retrospective analysis of 29 cases, the criteria for operative and nonoperative management of tracheal lacerations are discussed. Patients and Methods From July 1996 to June 2006, 29 patients with iatrogenic tracheobronchial injuries were diagnosed at our institution, and their records were analyzed for patient profile, cause, and extent of tracheobronchial injury, clinical symptoms, chosen treatment option, treatment success or failure, and patient outcome. There were 20 women (69%) and 9 men (31%), with a mean age of 65 years (range, 17 to 89 years). The ethics committee of the University of Heidelberg approved this retrospective study (Approval ID: 080/2006). The need of a subsequent individual consent of patients whose documents were 2007 by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg SCHNEIDER ET AL 2007;83: IATROGENIC TRACHEOBRONCHIAL INJURIES Table 1. Cause of Laceration and Body Mass Index Patients (n) BMI (mean) Emergency intubation Female Male Elective intubation Double lumen Single lumen Tracheostoma Percutaneous dilational Surgical Endobronchial endoscopy/ intervention All patients BMI body mass index (kg/m 2 ). evaluated was waived because individuals were not identified within the study. Table 2. Localization of the Laceration (Endoscopy) Surgical (n 18) Conservative (n 11) Upper middle third Covered 1 1 Not covered 0 0 Middle distal third Covered 6 9 Not covered 6 0 Distal third main carina Covered 1 0 Not covered 0 0 Distal third, main carina main bronchus Covered 0 1 Not covered GENERAL THORACIC Etiology The etiology of all injuries was iatrogenic. The injury occurred in 10 patients during single-lumen tube intubation in an emergency situation, and in an elective situation in 4 patients. In 2 patients, the distal trachea was perforated by double-lumen tube intubation under elective conditions. Two cases were diagnosed after surgical tracheostomy (one in an in emergency situation), and eight after dilational percutaneous tracheostomy. All lacerations that were associated with dilational percutaneous tracheostomies occurred under elective conditions, and one case after surgical tracheostomy was associated with an emergency condition. In 3 patients, a tracheal rupture happened during interventional bronchoscopy procedures. Female patients after emergency intubation presented with mild-to-severe obesity, defined as a mean body mass index (BMI) of 28.7; all others were in the normal range, defined as a mean BMI of 22.9 to 24.9 (Table 1). In none of our intubated and mechanically ventilated patients was the cause for the tracheal injury cuff-related. All lacerations showed either the typical longitudinal linear tear in the paries membranaceous of the trachea or a perforation of the main carina leading into a longitudinal laceration of the main bronchus. Five iatrogenic injuries occurred at our institution, and 24 patients were referred from other hospitals. Endoscopy The median length of the injury was 4 cm (range, 1.5 cm to 8 cm). In 2 patients, the lacerations were located in the upper two thirds of the trachea and were covered by the esophagus. The lower two thirds of the trachea was the location for 21 ruptures, 15 of which were covered by esophagus or mediastinal tissue. Six were affecting the tracheal bifurcation, and in five, the split began in the distal third of the trachea and ended in the left main bronchus. Two cases affecting the main carina were covered, and there was an open perforation into the pleural cavity in four cases (Table 2). The median diagnostic delay between the intervention that caused the laceration and the bronchoscopic confirmation of the tracheal laceration was 14 hours (range, 0 to 120 hours). Most injuries were diagnosed within the first 24 hours after the laceration or patients were referred within that time lag. Clinical and Radiographic Findings Most patients (n 22) showed at least one of the classic symptoms, such as mediastinal emphysema or subcutaneous emphysema, but 6 with very superficial tears showed no symptoms or radiologic findings at all. Unilateral pneumothoraces were seen in 7 patients with lacerations of the distal trachea or the main carina, and a bilateral pneumothorax was present in 1 patient with destruction of the main carina (Table 3). Treatment Decision Conservative treatment was chosen in those patients who did not require mechanical ventilation or in whom mechanical ventilation was possible without any loss of tidal Table 3. Clinical and Radiographic Findings Total Surgical Conservative Patients Emphysema Mediastinal a 21 (72) 15 (83) 6 (55) Cutaneus a 11 (38) 9 (50) 2 (18) Pneumothorax a Unilateral 7 (24) 6 (33) 1 (9) Bilateral 1 (3) 1 (6) 0 None 6 (21) 2 (11) 4 (36.3) a Patients with multiple findings included.

3 1962 SCHNEIDER ET AL Ann Thorac Surg IATROGENIC TRACHEOBRONCHIAL INJURIES 2007;83: Table 4. Ventilatory Management Surgical, Conservative, Patients Mechanical ventilation Not required 0 (0) 3 (27.3) Required 18 (100) 8 (72.7) Tube placement Distal single-lumen 7 (38.9) 6 (54.5) tube, feasible Contralateral 6 (33.3) 2 (18.1) double-lumen tube, feasible Not feasible 5 (27.8) 0 (0) volume, and the emphysema was only mild and did not progress during ventilation. We decided for surgical treatment when mechanical ventilation was not possible, the subcutaneous or mediastinal emphysema was progressive,or there was an open perforation into the pleural cavity (Table 4). Mechanical ventilation was impractical in 5 patients because tube placement distal of the laceration was not possible; the laceration was located in the distal trachea in 1 patient, and the lacerations affected the main carina and the main bronchus in 4. The local extent of the injury and an extensive subcutaneous emphysema indicated surgery in 8 patients, of which 1 presented with a laceration longitudinally affecting the paries membranaceous in the upper and middle third over a distance of 7 cm. Five patients had uncovered perforations in the middle and distal third of the trachea and developed rapidly progressive mediastinal and subcutaneous emphysema. Surgery was indicated in all cases promptly after endoscopic confirmation of the laceration. In no case did we decide for surgery secondary to an attempt at conservative therapy. All patients received broad-spectrum antibiotics. Endoscopic examinations were performed at least twice after surgical treatment, and at least on day 1, 3, and 7 in the conservative group, depending on the healing process. Results Treatment Success and Failure Eleven patients underwent conservative management. These patients presented with the following symptoms: cutaneous emphysema in 2, mediastinal emphysema in 6, pneumothorax in 1, and no symptoms in 4 (Table 3). In all cases, the lacerations were either superficial or sufficiently covered by the esophagus. Two of these lacerations occurred at our institution: one happened when a tracheal canula was changed through a tracheostoma and the other was associated with the bronchoscopic dilation of a tracheal stenosis. Bronchoscopy was performed to control the healing processes and revealed healing per primam in every case. No tracheal stenosis occurred in the further follow-up. Surgical repairs were done on 18 patients (62%) presenting with a tracheobronchial laceration. Symptoms in this group were cutaneous emphysema in 9 patients, mediastinal emphysema in 15, unilateral pneumothorax in 6, bilateral pneumothorax in 1, and no symptoms in 2 (Table 3). Mechanical ventilation was nearly impossible in 1 patient, and the surgical repair was performed immediately after the bronchoscopic confirmation of the laceration. In 7 patients with lacerations of the upper and middle third of the trachea, a transtracheal approach as described by Jacobs and colleagues [6] and Angelillo- Mackinlay [7] was used for repair. After a cutaneous collar incision, a longitudinal midline incision was made in the anterior wall of the trachea, and a running absorbable suture was used to repair the laceration of the posterior wall. A right-sided posterolateral thoracotomy in the fourth intercostal space was performed in 11 patients with a laceration of the lower thirds of the trachea or a laceration affecting the tracheal bifurcation, including the right or left main bronchus. After incision of the mediastinal pleura, the destroyed tracheobronchial walls were reconstructed by interrupted (n 7) or running (n 4) absorbable sutures. The sutures were covered with vital material (esophagus or azygos vein). In 1 patient, perforation of the carina was discovered intraoperatively after elective double lumen intubation for planned thoracotomy, and was sutured immediately after diagnosis. Endoscopic controls of the healing process of the tracheal sutures were performed at least twice in every patient between days 5 and 12 after surgery. Management of Ventilation In patients in the group with conservative treatment who needed mechanical ventilation, the cuff of the singlelumen tube was placed distal of the laceration with sufficient reserve to the carina. The ventilation regimen was directed towards early extubation, but in all cases was dependent on associated diseases rather than the tracheal injury itself. Low tidal volumes and low positive end-expiratory pressure (PEEP) were used to avoid exacerbating the local situation under mechanical ventilation. There was also careful monitoring in the intensive care unit for signs of air leaks (loss of tidal volume). In the surgical group, if the laceration affected the tracheal bifurcation or the right or left main stem bronchus, the opposite bronchus was intubated with a corresponding double-lumen tube. When the transcervical approach was chosen for repair, the tracheal tube was removed after the transection of the anterior tracheal wall, and a tube for jet ventilation was placed transorally into the distal trachea to maintain gas exchange by means of jet-ventilation while the posterior tracheal wall lacerations were sutured. After the suturing was completed, the jet tube was replaced with a single-lumen tube positioned distal of the laceration. Finally, the transected anterior tracheal wall was sutured closed. When a right-sided posterolateral thoracotomy was required for repair, the single-lumen tube already in place was forwarded into the left main bronchus or a jet catheter was inserted during exploration and reconstruction of the tracheobronchial wall. Afterwards, the tube

4 Ann Thorac Surg SCHNEIDER ET AL 2007;83: IATROGENIC TRACHEOBRONCHIAL INJURIES was placed proximal of the laceration under endoscopic control. The patients were extubated as soon as possible, but in many cases, the underlying disease necessitated mechanical ventilation for a prolonged period of time. When the injury occurred during construction of a tracheostoma (surgical or dilational), a tracheostomy tube was passed through the stoma into the trachea and used for subsequent ventilation. Care was taken to maintain a low cuff pressure, however. Outcome All patients who were treated conservatively had local healing per primam and survived. No mediastinitis was observed, and no secondary surgical repair was necessary. The laceration of the trachea and the treatment modalities had only a little impact on the clinical course of the underlying diseases. Endoscopy confirmed good local healing, and there was no evidence of tracheobronchial stenosis in the further follow-up. After surgical repair, the tracheobronchial sutures healed per primam in all cases. No clinically evident mediastinitis or postoperative tracheobronchial stenosis was observed in any patients. Three patients in the surgical group died: 2 from multiorgan failure owing to sepsis unrelated to the tracheal injury and 1 from progressive ischemic cerebral insult. Comment Tracheobronchial injuries are rare but potentially lifethreatening complications of endotracheal intubations or endobronchial interventions. Our data in essence reconfirm the reported experiences of other authors, that the cause of the iatrogenic injury in most cases was single lumen intubation under emergency conditions or dilational tracheostomies [2, 13, 15]. The risk for tracheal lacerations seems to increase with difficult or emergency intubations, multiple vigorous attempts of an inexperienced anesthesiologist, or the inappropriate use of a stylet. Overweight patients are susceptible to an overestimation of the size of their endobronchial anatomy and the selection of oversized endotracheal tubes. Our criteria for choosing nonoperative management were uncomplicated mechanical ventilation without any loss of tidal volume and a laceration sufficiently covered by the esophagus and only mild emphysema with no progress during ventilation. Accordingly, our criteria for surgical treatment were insufficient mechanical ventilation or an open perforation into the pleural cavity or progressive subcutaneous or mediastinal emphysema. This algorithm corresponds with the recommendations of several other authors: 1963 Carbognani and colleagues [12] chose nonsurgical therapy in small, uncomplicated tears ( 2 cm) in stable patients. Gomez-Caro and colleagues [13] recommended conservative management in 17 cases of iatrogenic injury for patients without associated esophageal injuries, no rapidly progressive subcutaneous or mediastinal emphysema, and no mediastinitis [13]. Sippel and colleagues [15] reported 13 cases of tracheal rupture after endotracheal intubation; nonsurgical treatment was chosen in 2 of these patients with a minor laceration of the cervical trachea (1.5 and 3.0 cm). Although most publications of conservative management of tracheal injuries concern iatrogenic injuries, the criteria for conservative management seem to be useful also for noniatrogenic lacerations. Self and colleagues [16] reported nonoperative management in 2 patients with severe blunt chest wall and lung injuries and concomitant tracheobronchial laceration. After placement of the endotracheal tube distal of the laceration and ventilation with PEEP and low tidal volume, the tracheobronchial laceration healed in these 2 patients who were considered too critically ill to undergo thoracotomy. From our experience, conservative treatment is particularly appropriate in stable patients with a delayed ( 24 hours) diagnosis of the tracheobronchial injury, and this assessment was affirmed by the uncomplicated healing in all of our conservatively managed cases. On the other hand, the delay of the diagnosis itself may qualify patients for conservative management because of a confirmed stable clinical course; however, as stressed by others [16], low tidal volumes and PEEPs are mandatory with conservative management because positive airway pressures may exacerbate the condition. When the surgical approach to tracheobronchial injuries is indicated, the repair is traditionally performed through a right lateral thoracotomy [15]. However, for surgical repair in patients with a laceration of the upper or middle thirds of the trachea we choose the transcervical approach as described by Jacobs and colleagues [6] and Angelillo-Mackinlay [7]. We, and others, see the major advantage of the cervical approach in the smaller operative trauma compared with the alternate thoracotomy approach [12, 17]. The transcervical repair was also successful in a patient where the laceration was affecting the lower third of the trachea; constructing the suture was feasible by using video-assisted thoracic surgery (VATS), with instruments and visual control by the VATS optics. We performed a right-sided posterolateral thoracotomy in all other patients with the laceration localized in the lower third of the trachea or affecting the tracheal bifurcation, including the right or left main bronchus. Patients who required thoracotomy were characterized by an extended length of the laceration ( 7 cm) in the distal part of the trachea, destroyed tracheal bifurcation and main bronchus, or a high loss of tidal volume during mechanical ventilation when the laceration was smaller (4 to 7 cm). We did not consider VATS as an alternative to thoracotomy in these patients because even open surgery for reconstruction of anatomy can be challenging, and in our opinion, a suitable covering of the suture is essential. We used either esophagus or intercostal-muscle flap to cover the reconstructed trachea or bronchus. GENERAL THORACIC

5 1964 SCHNEIDER ET AL Ann Thorac Surg IATROGENIC TRACHEOBRONCHIAL INJURIES 2007;83: Conversely, patients with a distal laceration of a smaller extent that would be suitable for VATS may be manageable conservatively if the ventilatory situation is stable. For patients with lacerations affecting the upper two thirds of the trachea, and in selected cases even closer towards the tracheal bifurcation, the transcervical approach offers the advantage of minimal surgical trauma. Endoscopic stenting was not done in our patients. For the extended distal ruptures involving the tracheal bifurcation, a stent may not be able to cover the laceration completely; conversely, the expansion of a stent could result in a dilation of the rupture, impairing the local situation. The healing of the tracheal rupture could also induce adhesions with the stent that would make a later removal difficult. Few reports exist about acute stenting of tracheal lacerations. Madden and colleagues [18] reported two cases of a longitudinal posterior wall perforation after percutaneous tracheostomy treated by tracheal stenting as alternative to surgical therapy. Shimizu and colleagues [19] inserted a T-silicon stent through a tracheostomy over a laceration as alternative to a tracheostomy tube. With growing experience and favorable results after conservative therapy of tracheal injuries, our rate of surgical therapy declined. This retrospective review thus also represents our learning curve in the management of tracheal injuries. The extent of the mediastinal emphysema was an important criterion for indication toward surgery in the earlier patients, whereas the instability of ventilation management was the main criterion in the later cases. Pneumothorax per se may not be an indication for surgery if the patient can be sufficiently ventilated. Whenever possible, and the criteria for conservative treatment are redeemed, we decide for nonoperative therapy. In conclusion, nonoperative conservative management of iatrogenic tracheobronchial injuries is a safe option in patients with uncomplicated ventilation, superficial or sufficiently covered tears, and moderate and nonprogressive emphysema. Immediate surgical repair remains warranted in those patients who require mechanical ventilation that cannot be delivered past the laceration. References 1. Borasio P, Ardissone F, Chiampo G. Post-intubation tracheal rupture. A report on ten cases. Eur J Cardiothorac Surg 1997;12: Massard G, Rouge C, Dabbagh A, et al. Tracheobronchial lacerations after intubation and tracheostomy. Ann Thorac Surg 1996;61: Trottier SJ, Hazard PB, Sakabu SA, et al. Posterior tracheal wall perforation during percutaneous dilational tracheostomy: an investigation into its mechanism and prevention. Chest 1999;115: Fitzmaurice BG, Brodsky JB. Airway rupture from doublelumen tubes. J Cardiothorac Vasc Anesth 1999;13: Marty-Ane CH, Picard E, Jonquet O, Mary H. Membranous tracheal rupture after endotracheal intubation. Ann Thorac Surg 1995;60: Jacobs JR, Thawley SE, Abata R, Sessions DG, Ogura JH. Posterior tracheal laceration: A rare complication of tracheostomy. Laryngoscope 1978;88: Angelillo-Mackinlay T. Transcervical repair of distal membranous tracheal laceration. Ann Thorac Surg 1995;59: Jougon J, Ballester M, Choukroun E, Dubrez J, Reboul G, Velly JF. Conservative treatment for postintubation tracheobronchial rupture. Ann Thorac Surg 2000;69: Ross HM, Grant FJ, Wilson RS, Burt ME. Nonoperative management of tracheal laceration during endotracheal intubation. Ann Thorac Surg 1997;63: Hofmann HS, Rettig G, Radke J, Neef H, Silber RE. Iatrogenic ruptures of the tracheobronchial tree. Eur J Cardiothorac Surg 2002;21: Meyer M. Latrogenic tracheobronchial lesions--a report on 13 cases. Thorac Cardiovasc Surg 2001;49: Carbognani P, Bobbio A, Cattelani L, Internullo E, Caporale D, Rusca M. Management of postintubation membranous tracheal rupture. Ann Thorac Surg 2004;77: Gomez-Caro Andres A, Moradiellos Diez FJ, Ausin Herrero P, et al. Successful conservative management in iatrogenic tracheobronchial injury. Ann Thorac Surg 2005;79: Conti M, Pougeoise M, Wurtz A, et al. Management of postintubation tracheobronchial ruptures. Chest 2006;130: Sippel M, Putensen C, Hirner A, Wolff M. Tracheal rupture after endotracheal intubation: experience with management in 13 cases. Thorac Cardiovasc Surg 2006;54: Self ML, Mangram A, Berne JD, Villarreal D, Norwood S. Nonoperative management of severe tracheobronchial injuries with positive end-expiratory pressure and low tidal volume ventilation. J Trauma 2005;59: Lancelin C, Chapelier AR, Fadel E, Macchiarini P, Dartevelle PG. Transcervical-transtracheal endoluminal repair of membranous tracheal disruptions. Ann Thorac Surg 2000;70: Madden B, Datta S, Hussain I, McAnulty G. Tracheal stenting for rupture of the posterior wall of the trachea following percutaneous tracheostomy. Monaldi Arch Chest Dis 2001; 56: Shimizu J, Hirano Y, Ishida Y, Kinoshita T, Tatsuzawa Y, Kawaura Y. Use of a silicone T-tube for management of a tracheal injury in a patient with cervical blunt trauma. Jpn J Thorac Cardiovasc Surg 2003;51:541 4.