Successful Conservative Management in Iatrogenic Tracheobronchial Injury

Transcription

1 Successful Conservative Management in Iatrogenic Tracheobronchial Injury Abel Gómez-Caro Andrés, MD, Francisco Javier Moradiellos Díez, MD, Pilar Ausín Herrero, MD, Vicente Díaz-Hellín Gude, MD, Emilio Larrú Cabrero, MD, PhD, Eduardo de Miguel Porch, MD, and José Luis Martín De Nicolás, MD Departments of Thoracic Surgery and Pneumology, Hospital Universitario 12 de Octubre, Madrid, Spain Background. The aim of this study was to describe and to assess the effectiveness of conservative treatment as the chosen treatment for managing iatrogenic tracheobronchial injuries (ITBI). Methods. Between January 1993 and December 2003, 33 tracheobronchial injuries were treated in our hospital. Eighteen (54.5%) were ITBI and 15 (45.5%) were traumatic noniatrogenic injuries. Of the ITBI patients, sex distribution was 15 (83%) females and 3 (17%) males with a mean age of years (range, 17 to 88 years). Fifteen (83.3%) of the injuries were caused by oro intubation and 3 (15.7%) by tracheotomy. The average diagnostic delay was hours. The mean injury size was cm (range, 1 to 4 cm). Nine (50%) injuries were located in the cervical trachea, 6 (33.3%) in the thoracic trachea, and 3 (16%) involved both trachea and main bronchi. Conservative treatment was chosen for 17 (94.4%) of the 18 cases. We performed surgical repair in only 1 case owing to progressive subcutaneous and increasing difficulty with mechanical ventilation. Results. No complications arose from the use of conservative treatment. Four patients (22%) died in our hospital, 3 of these of non ITBI-related causes. Mortality was not related to four variables: sex, diagnostic delay, location, or size of the ITBI. Fourteen of the 18 patients (77.7%) were discharged uneventfully, and the endoscopic and clinical follow-up examinations were satisfactory in all patients. Conclusions. Conservative treatment for ITBI is effective regardless of production, size, or site of the injuries. Surgical treatment is advisable in specific cases: rapid progression of subcutaneous and mediastinal, mediastinitis, and difficulty with mechanical ventilation. (Ann Thorac Surg 2005;79:1872 8) 2005 by The Society of Thoracic Surgeons Although iatrogenic tracheobronchial injuries (ITBI) are generally rare, such injuries are one of the most common causes of airway injury and are also a feared complication of general anesthesia. The incidence of ITBI is approximately 1 of every 20,000 oro intubations. Postmortem findings have indicated an incidence of 15% after emergency intubations [1]. The incidence of ITBI in double-lumen intubation is less than 1%, significantly higher than that of single-lumen intubation [2]. A surgical approach has traditionally been the most widely accepted treatment [3, 4]. Recently, a few conservative treatment series have been reported [5]. However, the percentage of conservatively treated patients has never exceeded 50% in the reported series [6]. The aim of this study is to assess the results of a conservative approach as the chosen treatment for selected ITBI cases independent of location or size of injury, or diagnostic delay. Accepted for publication Oct 4, Address reprint requests to Dr Gómez-Caro, Department of Thoracic Surgery, Hospital Universitario 12 de Octubre, Crta. Andalucía KM 5.400, Madrid, Spain; abelitov@yahoo.es. Material and Methods Between January 1993 and December 2003, 33 tracheobronchial injuries were diagnosed in our department. A retrospective review of the clinical records was carried out during this period. Eighteen (54.5%) were iatrogenic injuries and 15 (45.5%) were traumatic non-iatrogenic. Among the ITBI cases, 15 (83.3%) were women and 3 (17%) were men, with a mean age of years (range, 17 to 88 years). Diagnostic delay was defined as time between iatrogenic intubation and bronchoscopic confirmation of ITBI. The time of this delay was hours (range, 3 to 72 hours). Diagnosis was carried out by bronchoscopy in all cases. A computed tomographic scan of the chest was taken in all patients to detect mediastinal fluid collections and quick progression of mediastinal. Blood and serum analyses were performed daily to assess the patients clinical situation and progress. The reasons for carrying out intubation and intubation type are listed in Table 1. The mean size of injury measured by bronchoscopic vision was cm (range, 1 to 4 cm). The injury was located in the cervical trachea in 9 (50%) of the cases, the thoracic trachea in 5 (27.7%) of the cases, the trachea and 2005 by The Society of Thoracic Surgeons /05/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 Table 1. Patients Characteristics, Bronchoscopic Findings, and Outcome Patient No. Sex Age (y) Reason for Type/ Tracheostomy Presenting Symptoms 1 F 26 Appendicitis/NE Mediastinal 2 F 17 Trauma/NE Percutaneous 3 F 32 Nasal surgery/ne 4 F 66 Orthopedic surgery/ NE 5 M 88 Self-inflicted/E 6 F 69 Orthopedic surgery/ NE 7 F 71 Pneumonia/NE 8 M 49 Trauma/NE Percutaneous Dyspnea 9 F 86 Cholecystectomy/NE 10 F 38 Hysterectomy/NE Endoscopic Finding/Site CTM Tear, view of esophagus wall CTM Tear, view of esophagus wall Anfractuous borders CTM necrotic tissues Tear CTM CTM lesion, ulcerous borders TTM Lineal 11 F 74 Pneumonia/NE Tracheostomy Hemoptysis TTM lesion, ulcerous borders Length (cm) Delay of Diagnosis (hours) ICU/ Hospital Stay (days) Type of Bronchoscopic Revision After Critical Period (1 2 months) Outcome 4 8 3/6 Medical Scar Normal /72 Medical Scar Granulomous 4 7 1/6 Medical Normal 4 3 2/7 Medical Normal /3 Medical... Death /26 Surgical Plastic repair and /16 Medical Normal /31 Medical Scar Granulomous... Death /14 Medical... Death 4 3 5/8 Medical Normal /42 Medical Normal Death continued Ann Thorac Surg GÓMEZ-CARO ANDRÉS ET AL 2005;79: MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY 1873 GENERAL THORACIC

3 Table 1. Continued Patient No. Sex Age (y) Reason for Type/ Tracheostomy Presenting Symptoms 12 F 71 Trauma/NE Mediastinal 13 F 65 Peritonitis/NE 14 F 61 Peritonitis/NE 15 F 44 Maxillofacial surgery/ne 16 F 73 Orthopedic surgery/ NE 17 F 39 Internal saphenous stripping/ne Endoscopic Finding/Site TTM Lineal TTM and RMB Linear and lateral cartilaginousmembranous unction bronchial TTM and LMB Lateral cartilaginousmembranous junction bronchial TTM and LMB Lateral cartilaginousmembranous junction bronchial Hemoptysis TTM Posterior wall 18 M 7.1 Thoracic surgery/ne Doublelumen TTM Posterior wall Carina Posterior wall Length (cm) Delay of Diagnosis (hours) ICU/ Hospital Stay (days) Type of Bronchoscopic Revision After Critical Period (1 2 months) Outcome /23 Medical Normal /7 Medical Scar in RMB Stenosis noncritical RMB /7 Medical No scar Normal LMB /8 Medical Normal LMB /8 Medical Normal /7 Medical Normal /17 Medical No findings CTM cervical trachea membranous; E emergency intubation; LMB left main bronchus; NE nonemergency intubation; RMB right main bronchus; TTM thoracic trachea membranous GÓMEZ-CARO ANDRÉS ET AL Ann Thorac Surg MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY 2005;79:1872 8

4 Ann Thorac Surg GÓMEZ-CARO ANDRÉS ET AL 2005;79: MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY Table 2. Causes of Death After Iatrogenic Tracheobronchial Injury Patient No. Sex Age (y) Endoscopic Finding Localization 5 M 88 Cervical trachea membranous Tear, view of esophagus wall Anfractuous borders 6 F 69 Cervical trachea membranous necrotic tissues Tear 9 F 86 Cervical trachea membranous lesion ulcerous borders 11 F 74 Thoracic trachea membranous lesion ulcerous borders Delay (hours) and Finding in First Endoscopy 24 Medical improvement of ITBI in early endoscopy Cause of Death Peritonitis/MOF 24 Surgical dehiscence Mediastinitis/MOF 8 Medical improvement of ITBI in early endoscopy 74 Medical improvement of ITBI in early endoscopy 1875 Pneumonia/MOF Pneumonia/MOF GENERAL THORACIC TBI iatrogenic tracheobronchial injury; MOF multiorgan failure. main bronchi in 3 (16.6%) of the cases, and the carina in 1 (5.5%) of the cases. In all the patients there were one or more of the classic signs and symptoms such as subcutaneous, mediastinal, pneumothorax, hemoptysis, dyspnea, or difficulty with mechanical ventilation. Conservative management was the chosen treatment in all cases, independent of the injury size and location, diagnostic delay, or cause. Therefore, with early diagnosis, ITBI that were large in size or found in certain sites, or any other local circumstances, were not considered sufficient grounds for surgical treatment. Mechanical ventilation before diagnosis did not exclude patients from conservative treatment. Most of these injuries occurred during intubation for scheduled surgery, meaning that the extubation happened soon in 11 patients (61%). Only 4 patients were receiving mechanical ventilation when diagnosed with ITBI. Tracheoesophageal injuries were not treated conservatively as they were considered to be a completely different type of injury and were excluded from this study. All patients received broad-spectrum antibiotics and were carefully monitored for signs of air leaks in our unit or in the intensive care unit. Oro intubation was avoided when possible and was never used as a treatment for ITBI in our series. Surgery should be performed in cases with progressive subcutaneous or mediastinal, severe dyspnea requiring intubation, difficulty with mechanical ventilation, pneumothorax with an air leak through the chest drains, or mediastinitis. Endoscopic examinations were performed between the first and second weeks after diagnosis for visual evaluation of the ITBI healing process. Follow-up was performed in months 1 and 3, and then yearly for the following 5 years after hospital discharge if no other incidences were detected. Statistical analyses were carried out with the computer program package SPSS v.11 (SPSS, Inc, Chicago, IL). Qualitative variables were compared with the 2 statistical test or Fisher s exact test where appropriate, with a significance level of p less than Conservative management was effective in all 17 of the patients in which this was the chosen approach. Three patients died of non ITBI-related causes (Table 2). Mortality was not significantly associated with sex, diagnostic delay, or size or site of the injury (p 0.05). Endoscopic findings were described in Table 1. The ITBI in patient 18 occurred after double-lumen intubation and remained undetected throughout a right upper lobectomy. This injury was later detected during bronchoscopy performed for asymptomatic atelectasis after sputum retention. The patient responded well to conservative treatment. Only one case of ITBI occurred after emergency intubation (Figs 1, 2). Surgery was performed as the initial treatment for patient 6 because of the rapid progression of mediastinal and subcutaneous, in turn leading to dyspnea and intubation. A cervical approach with plastic suture and a were performed in a one-stage operation. This particular patient died because of suture dehiscence, mediastinitis, and multiorgan failure. Patient 5 was intubated after a suicide attempt that caused colon perforation and peritonitis followed by multiorgan failure after abdominal surgery. No ventilation problems were reported in this case, but this patient eventually died because of old age and injury-related causes. Patients 9 and 11 died of Results Fig 1. Chest roentgenogram of patient with iatrogenic tracheobronchial injury. We observed a large subcutaneous and mediastinal.

5 1876 GÓMEZ-CARO ANDRÉS ET AL Ann Thorac Surg MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY 2005;79: Fig 2. Thoracic computed tomographic scan from patient with iatrogenic tracheobronchial injury in a emergency intubation in a suicide attempt. A large mediastinal without intrathoracic collection was detected. The iatrogenic tracheobronchial injury was medically treated. pneumonia associated with mechanical ventilation and sepsis. These patients needed oro intubation or for sepsis (peritonitis and nosocomial pneumonia), and injury occurred during this procedure. No radiologic evidence of mediastinitis in thoracic computed tomographic scan or progressive ITBI signs and symptoms were noted in the following days. Bronchoscopic examinations of patients 9 and 11 showed marked improvement or even complete healing of the ITBI days after first exploration. However, these were older patients with many comorbidities and with lifethreatening disease. The mean follow-up time for discharged patients was months (range, 6 to 120 months) with a scheduled follow-up period of 60 months if no pathologic endoscopic findings were observed. Evidence of -related granulomas was found in 2 patients, who were subsequently treated with a laser. Asymptomatic stenosis of the right main bronchus was noted in 1 patient. Follow-up lasted for a longer period, and at 120 months this patient remained asymptomatic. Comment Iatrogenic tracheobronchial injuries are life-threatening complications, and the incidence of such injuries is increasing in hospital and extrahospital environments [7]. Known risk factors for ITBI include female sex and the increasing age of the patients undergoing surgery [8]. The high number of ITBI cases reported in our hospital during such a short time may be related to the fact that ours is a reference center for general thoracic surgery and interventional bronchoscopy where cases from all over the country are treated. Several other risk factors for iatrogenic airway injuries have been reported in the literature, such as emergency intubation, inadequate oro tube, inappropriate use of stylets, balloon overinflation damaging the mucosa, and others [3, 8]. We suggest that an injury directly caused by the oro tube usually occurs in the membranous cervicothoracic wall, whereas injuries caused by high pressure or volume ventilation, or sudden movements of the oro tube because of coughing or neck movements, occur in the membranous and cartilaginous portions of bifurcation and tracheobronchial region [6]. Direct injuries usually consist of small s and are more frequent after difficult intubation or in older patients (Fig 3) [8]. In our opinion, the appearance of distal injuries in the main bronchi caused by a single-lumen tube could be related to an overinflation of the airway during intubation and is not related to direct injuries as in the case of cervical s. In our series the minimum delay between diagnosis and procedure means that it is improbable for there to be another mechanism that is not a direct injury in patients with. Iatrogenic tracheobronchial injuries as a result of double-lumen intubation could be more related to excessive tube size [8, 9]. In our series, intubation or airway procedures were performed in all but one case and were not rated as difficult by the anesthesiologist in any case. Iatrogenic airway injuries are more common in women [2, 3, 6], probably because of oversizing of the tube and of the shorter average length in women. The most common sign observed in our series was subcutaneous. However, this symptom was only detected intraoperatively in 2 (11.1%) patients. Initial signs that were less frequently noted were radiologic mediastinal and hemoptysis. The diagnostic delay in our series is related to the fact that many patients (n 6) were transferred to our center from other hospitals with diagnoses yet unconfirmed. Nevertheless, this Fig 3. Thoracic computed tomographic scan of a 19-year-old woman with postintubation injury that was medically treated. Deep cervical can be observed.

6 Ann Thorac Surg GÓMEZ-CARO ANDRÉS ET AL 2005;79: MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY Table 3. Review of Recent Published Series on of Iatrogenic Tracheobronchial Injuries Series n Sex (M/F) Procedure (SL/DL/TC/IB) Localization (CT/TT/MB) Delayed Diagnosis Length (cm) Surgical Conservative Marty-Ane et al [13] 6 0/6 6/0/0/0 1/5/ hours (16.6%) 33 Kaloud et al [8] 12 2/10 7/4/1/0 7/3/2 0 5 days (0%) 25 Borasio et al [2] 10 1/9 8/2/0/0 6/4/0 0 5 days (30%) 0 Mussi et al [10] 11 1/10 7/4/0/0 3/4/ (18.8%) 0 Jougon et al [6] 14 2/12 12/2/0/0 3/9/2 0 6 days (50%) 7 Hoffman et al [3] 19 4/9 11/4/2/2 2/15/ hours (5.3%) 42 Current series 18 3/15 14/1/3/0 9/6/ hours (94.4%) Death (%) GENERAL THORACIC M/F Male/Female; SL/DL/TC/IB single-lumen/double-lumen//interventional bronchoscopy; CT/TT/MB cervical trachea/ thoracic trachea/main bronchus. delay period was similar to those of other published series [6, 8, 9]. Surgical management has been accepted as the classic therapeutic approach for ITBI [3, 4, 9, 10]. However, published cases [11, 12] and series [5, 6, 13] have also described the conservative management of such injuries (Table 3). Nevertheless, conservative management has usually been limited to small s, less than 2 cm in length, in the membranous trachea [5] or in patients with a delayed diagnosis or unstable clinical situation. Jougon and colleagues [6] reported a series in which conservative treatment was the chosen treatment in 50% of the cases. The decision to use conservative management has been based on the following criteria: injuries with a length equal to or less than 4 cm, significant diagnostic delay, and integrity of the tracheobronchial junction [14]. In our series we propose the following criteria to be used as guidelines for nonoperative management of intubation-related injuries: vital sign stability, no evidence of esophageal injury, no evidence of difficulty with mechanical ventilation if intubation is needed, no development of subcutaneous or mediastinal, and no signs of sepsis related to ITBI [11]. The diagnosis should be made as early as possible [15], preferably in the first 12 hours. In our series, only 36% of patients were diagnosed within this time. Even in these cases initial surgical treatment was not advised, and no increase in mortality or morbidity was observed. Injury size was not a determining factor when choosing the optimal treatment for our patients, although no injury longer than 4 cm was observed in our series. Contrary to other authors, we do not believe that injury size alone provides sufficient grounds for surgery as the immediate treatment. However, in injuries larger than 4 cm, we think that conservative treatment is risky for the patient [3, 6, 8, 10, 14, 16]. The decision to operate was made if massive subcutaneous, progressive mediastinal, tension pneumothorax, an air leak through the chest drain, or mediastinitis developed in the patient. Direct visualization of the intact esophageal wall through the injury does not warrant surgical repair in itself., as well as, should be avoided if possible for patients recommended for conservative management [6]. On the other hand, we know of other opinions that propose distal intubation beneath the injury as a conservative treatment in seriously compromised patients [17]; however, we prefer early extubation if this is possible. Post iatrogenic injuries are detected later, usually after encountering problems with mechanical ventilation, with aspiration of secretions, and with frequently associated tracheoesophageal fistulas. Posterior wall injuries are much more likely to occur among those injuries caused by percutaneous procedures [3, 18]. Broad-spectrum antibiotics covering the mixed bacterial populations of the tracheobronchial tree should be applied for at least 1 week [6, 11]. Long-term complications are rarely detected in endoscopic follow-up of medically [2, 6, 13] and surgically [19, 20] treated patients. In conclusion, treatment of ITBI should be carefully assessed, but we believe conservative management is recommended, particularly for ITBI patients with no associated esophageal injuries, no rapidly progressive subcutaneous or mediastinal, and no mediastinitis. If these guidelines are carefully followed, the conservative approach will have a good outcome, regardless of size, location, or diagnostic delay of the injury. We are indebted to Pedro A. Anton, MD, of Sant Creu and Sant Pau Pneumology Department for his help in the preparation and correction of this original article. References 1. Maxeiner H. Weichteilverletzungen an kehlkopf bei notfallmaessiger intubation. Anaesth Intensivmed 1988;29: Borasio P, Ardissone F, Chiampo G. Post-intubation rupture. A report on ten cases. Eur J Cardiothorac Surg 1997;12: Hofmann HS, Rettig G, Radke J, Neef H, Silber RE. Iatrogenic ruptures of the tracheobronchial tree. Eur J Cardiothorac Surg 2002;21: Meyer M. Iatrogenic 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 rupture. Ann Thorac Surg 2004;77: Jougon J, Ballester M, Choukroun E, Dubrez J, Reboul G, Velly JF. Conservative treatment for postintubation tracheobronchial rupture. Ann Thorac Surg 2000;69:

7 1878 GÓMEZ-CARO ANDRÉS ET AL Ann Thorac Surg MANAGEMENT OF IATROGENIC TRACHEOBRONCHIAL INJURY 2005;79: Rossbach MM, Johnson SB, Gomez MA, Sako EY, LaWayne Miller O, Calhoon JH. Management of major tracheobronchial injuries: a 28-years experience. Ann Thorac Surg 1998; 65: Kaloud H, Smolle-Juettner FM, Prause G, List WF. Iatrogenic ruptures of the tracheobronchial tree. Chest 1997;112: Massard G, Rouge C, Dabbagh A, Kessler R, Hentz JG, Roeslin N. Tracheobronchial lacerations after intubation and. Ann Thorac Surg 1996;61: Mussi A, Ambrogi MC, Ribechini A, Lucchi M, Menoni F, Angeletti CA. Acute major airway injuries: clinical features, management. Eur J Cardiothorac Surg 2001;20: Ross HM, Grant JG, Wilson RS, Burt ME. Nonoperative management of laceration during endo intubation. Ann Thorac Surg 1997;63: Molins L, Buitrago LJ, Vidal G. Conservative treatment of laceration secondary to endo intubation. Ann Thorac Surg 1997;64: Marty Ané CH, Picard E, Jonquet O, Mary H. Membranous rupture after endo intubation. Ann Thorac Surg 1995;60: Gabor S, Renner H, Pinter H, Sankin O, Maier A, Tomaselli F. Indications for surgery in tracheobronchial ruptures. Eur J Cardiothorac Surg 2001;20: Cassada DC, Munyikwa MP, Moniz MP, Dieter RA, Schuchmann GF, Enderson BL. Acute injuries of the trachea and major bronchi: importance of early diagnosis. Ann Thorac Surg 2000;69: Janni A, Menconi G, Mussi A, Ambrogi MC, Angeletti CA. Postintubation repair by cervicotomy and longitudinal tracheotomy. Ann Thorac Surg 2000;69: Marquette CH, Bocquillon N, Roumilhac D, Neviere R, Mathieu D, Ramon P. Conservative treatment of rupture. J Thorac Cardiovasc Surg 1999;117: Trottier SJ, Hazard PB, Sakabu SA, Levine JH, Troop BR, Thompson JA. Posterior wall perforation during percutaneous dilational : an investigation into its mechanism and prevention. Chest 1999;115: Balci AE, Eren N, Eren S, Ulku R. Surgical treatment of post-traumatic tracheobronchial injuries: 14-year experience. Eur J Cardiothorac Surg 2002;22: Richardson JD. Outcome of tracheobronchial injuries: a long-term perspective. J Trauma 2004;56:30 6. INVITED COMMENTARY The authors have described a compelling management approach to 18 patients treated over a 10-year period in their institution with iatrogenic tracheobronchial injury. It is important to stress that these are not trauma patients. This approach has been mentioned in the past mainly for those patients with compelling morbities such as age, cardiac disease, and other organ system dysfunction. This is one of, if not the largest, series of this particular injury and its management. The authors make a solid case for this type of management. They are to be congratulated for this series and for describing only one case of bronchial stenosis in good follow-up of this cohort. In our experience, a rate of stenosis after open repair this low is very satisfactory, especially when it was without symptoms. The deaths they saw seem to not be related to the treatment algorithm, but rather the disease leading to their need for intubation. We believe that they correctly outlined the reasons for a more traditional surgical approach. These remain massive air leak, associated esophageal injury, inability to ventilate adequately, and increasing subcutaneous. Likely size of the lesion over 4 cm in length will also prove to be an indication, although patients with this size lesion may well have one or more of the previously described signs or symptoms compelling urgent surgical attention. The main purpose of our commentary is to remind us all that this is a highly select group of patients who were not diagnosed with this problem on average more than 1 day after the likely injury occurred. We would be very reluctant to apply this management method to non-iatrogenic trauma patients. Having stressed this point, we congratulate the authors on this series and intend to consider this therapy from time to time in highly selected patients as they have. In the many other patients who present with large air leaks, ventilatory issues, esophageal injuries, and any sign of mediastinal sepsis, an immediate surgical approach will remain our standard. John Calhoon, MD Luis F. Angel, MD Daniel Martinez, MD Scott B. Johnson, MD Department of Surgery The University of Texas Health Science Center at San Antonio 7703 Floyd Curl Dr MSC 7841 San Antonio, TX calhoon@uthscsa.edu; angel@uthscsa.edu; martinezd4@uthscsa.edu; johnsons@uthscsa.edu 2005 by The Society of Thoracic Surgeons /05/$30.00 Published by Elsevier Inc doi: /j.athoracsur