Bronchial Stump Coverage With a Pedicled Pericardial Flap: An Effective Method for Prevention of Postpneumonectomy Bronchopleural Fistula

Transcription

1 Bronchial Stump Coverage With a Pedicled Pericardial Flap: An Effective Method for Prevention of Postpneumonectomy Bronchopleural Fistula Shahrokh Taghavi, MD, Gabriel M. Marta, MD, Georg Lang, MD, Gernot Seebacher, MD, Gunther Winkler, MD, Katharina Schmid, MD, and Walter Klepetko, MD Department of Cardiothoracic Surgery and Clinical Institute of Pathology, Medical University of Vienna, Vienna, Austria Background. Bronchopleural fistula is a serious complication after pneumonectomy. The aim of this retrospective study was to investigate the efficacy of bronchial stump reinforcement with a pedicled flap of whole pericardium. Methods. The bronchial stump of 93 consecutive patients who underwent pneumonectomy between July 1988 and March 2003 was covered with a pedicled pericardial flap. Pneumonectomy was performed for primary lung cancer in 89.2% of patients. The study patients received concomitant extensive mediastinal lymphadenectomy, resection of adjacent structures (aorta, vena cava, thoracic wall), and neoadjuvant or planned adjuvant chemotherapy or radiotherapy, or both. Operative and perioperative complications were recorded, and patients were followed up for a mean of months (range, 9 to 126). Accepted for publication June 11, Address reprint requests to Dr Klepetko, Department of Cardiothoracic Surgery, Medical University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria; walter.klepetko@medunivie.ac.at. Results. Perioperative mortality was 4.3% (n 4; pulmonary embolism, sepsis, cardiac arrest, and sudden death in 1 patient each). Perioperative complications occurred in 2 patients: renal failure and hemiplegia in 1 patient and cardiac tamponade in 1 patient. The latter complication, caused by tight reconstruction of the pericardium, was directly related to the applied method and required reoperation. No evidence of postpneumonectomy bronchopleural fistula was observed perioperatively and during the whole follow-up. One-year and 2-year survival was 65.7% and 44.8%, respectively. Conclusions. Bronchial stump reinforcement with a pericardial flap is a highly effective method for preventing postpneumonectomy bronchopleural fistula in selected patients. (Ann Thorac Surg 2005;79:284 8) 2005 by The Society of Thoracic Surgeons Postpneumonectomy bronchopleural fistula (PBPF) remains the most serious complication after pneumonectomy and occurs with an incidence of 0% to 12% (Table 1). It leads to a number of life-threatening situations, such as aspiration of infectious fluid from the pleural cavity, pneumonia of the remaining lung, and infection of the pleural cavity followed by empyema. Surgical technique is clearly related to its occurrence, and different approaches with regard to the optimal closure of the bronchial stump have been described [1 4]. Besides the different attempts to optimize the technique of bronchial stump closure, it has been emphasized that additional coverage with surrounding tissue might decrease the incidence of PBPF [3, 5]. However, it is still unclear whether reinforcement of the bronchial stump should be performed in every patient and, especially, what is the particular value of flaps for prevention of PBPF. Different biological materials such as pleura [6], intercostal muscle [7], pericardial fat pad [5, 8], diaphragm [9], vena azygos in case of a right-sided pneumonectomy [6], and pericardiophrenic pedicles [5] have been used for such a prophylactic coverage. Particularly, the use of the patient s own pericardium has been the preferred method for bronchial stump coverage in our department for many years. The aim of this retrospective study was therefore to analyze the resulting large series of patients and to describe the efficacy of coverage of the bronchial stump after pneumonectomy with a pericardial flap for prevention of PBPF. Patients and Methods A total of 697 patients underwent pneumonectomy between July 1988 and March 2003 at our department (381 left-side and 316 right-side pneumonectomies). Of these, patients were included in our study if they had a standard resection procedure of the main bronchus, did not undergo concomitant pleuropericardophrenectomy, and when the bronchial stump was covered with a pericardial flap as described below. Additional resection procedures besides the pneumonectomy such as resection of the thoracic wall or resection of the greater vessels did not exclude patients from the study. Ninety-three patients 2005 by The Society of Thoracic Surgeons /05/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg TAGHAVI ET AL 2005;79:284 8 BRONCHIAL STUMP COVERAGE WITH PERICARDIAL FLAP 285 Table 1. Incidence of Postpneumonectomy Bronchopleural Fistula (PBPF) According to Different Authors Author (Year) Reference Number of Patients Incidence of PBSF (%) Sarsam (1989) Asamura (1992) Weissberg (1992) Conlan (1995) Al-Kattan (1995) Wright (1996) Hollaus (1997) Klepetko (1999) Sirbu (2001) Javadpour (2003) (69 male, 24 female; mean age, 54.5 years) finally entered the analysis. Indications for pneumonectomy were primary lung cancer in 89.2% (n 83), other malignancies in 5.4% (n 5), and benign diseases in 5.4% (n 5). Fifty-seven patients (61.3%) underwent right side pneumonectomy and 36 patients (38.7%) underwent left side pneumonectomy. Operative reports and postoperative courses with regard to major events and complications were recorded. Underlying histology, TNM stage for primary lung cancers, and preoperative or postoperative chemotherapy or radiation therapy, or both, was documented (Tables 2 and 3). Of the total of 93 patients, 69 (74.2%) received some form of additional therapy (irradiation or chemotherapy), either alone or in various combinations. In 19 patients (20.4%), preoperative neoadjuvant therapy was given; the remaining 50 patients (53.8%) had postoperative treatment with or without induction therapy. All patients discharged from hospital were seen in our outpatient department 1 month after surgery and thereafter twice yearly by the referring pulmologist. For this analysis, office records, written questionnaires, or direct telephone Table 2. Indications for Pneumonectomy (n 93) and Underlying Histology Diagnosis Histology Patients n % Lung cancer Squamous cell carcinoma Adenocarcinoma Small cell carcinoma Large cell carcinoma Other malignancies Metastasis Lymphoma Malignant thymoma 1 1 Benign diseases Bronchiectasis Infectious diseases Table 3. Tumor Stage Distribution in Patients With Primary Lung Cancer (n 83) Tumor Stage Patients (n 83) n % I II IIIa IIIb contact were employed. Follow-up ranged from 9 to 126 months (mean, ) and was 97% complete (3 patients were lost to follow-up). Surgical Technique CLOSURE OF THE BRONCHUS. Bronchial stump closure was performed with commercial mechanical staplers (Ethicon and Auto Suture) in all patients. Stapling was performed by approximation of the membranous and the cartilaginous portion of the bronchus, as suggested before [10]. Tumor negativity of resection margins was ensured by histologic examination of frozen sections. The bronchial stump was then checked for air leakage with 30 cm H 2 O sustained airway pressure. LYMPH NODE DISSECTION. In all patients with primary lung cancer, a complete systematic mediastinal lymphadenectomy was routinely added to the resection procedure. Typically, the subcarinal and tracheobronchial lymph nodes were resected en bloc with the lung. This was followed by complete dissection of the other remaining mediastinal lymph node compartments. ADDITIONAL RESECTION PROCEDURES. In 14 patients (15%), additional resection procedures were performed (aorta, n 5; superior vena cava, n 3; thoracic wall, n 6). Resection of the aorta was performed with cardiopulmonary bypass, and details about these complex procedures have been published elsewhere [11]. Reconstruction of the thoracic wall was performed with polytetrafluoroethylene (Gore-Tex, W. L. Gore and Associates, Inc, Flagstaff, AZ). COVERAGE OF THE BRONCHUS. A generous flap of the anterolateral pericardium, pedicled at its cranial part with or without inclusion of the phrenic vessels and measuring approximately 4 12 cm, was prepared. This technique was applied regardless of whether the pericardium had been opened during the resection procedure. The flap was attached caplike over the bronchial stump with numerous single mattress stitches of 4-0 polydioxanon (PDS) (Johnson and Johnson Intl, Woluwe, Belgium) (Fig 1). In all patients, the resulting defect in the pericardium was reconstructed with Vicryl mesh (Johnson and Johnson Intl, Brussels, Belgium). Results Perioperative Period Median hospital stay was 6 days (range, 3 to 65). Perioperative mortality was 4.3%; 4 patients died within 30 days

3 286 TAGHAVI ET AL Ann Thorac Surg BRONCHIAL STUMP COVERAGE WITH PERICARDIAL FLAP 2005;79:284 8 Fig 2. Kaplan-Meier survival analysis of 93 patients who underwent pneumonectomy and had bronchial stump coverage with a pericardial flap, between July 1988 and March Numbers in parentheses represent patients at risk. of surgery. The causes of death were pulmonary embolism, multiorgan failure due to sepsis, cardiac arrest, and sudden death in 1 patient each. Significant perioperative complications occurred in 2 patients (2.2%). One patient had renal failure and hemiplegia after additional resection of the aorta; the other patient had cardiac tamponade due to tight reconstruction of the pericardium. That was the only patient with a complication directly related to the method who required reoperation, and the problem of cardiac compression was overcome by insertion of a larger Vicryl mesh. Postoperative recovery of this patient was uneventful. Supraventricular tachyarrhythmia occurred in 16 patients (17%), and was successfully managed pharmacologically in all of them. Fig 1. (A) A pedicled pericardial flap is attached, caplike, to a left bronchial stump with single mattress stitches of 4-0 polydioxanon. (PP pericardiophrenic patch; BS bronchial stump; Es esophagus.) (B) Schematic illustration of left mediastinum with pericardiophrenic patch to cover left bronchial stump. Long-Term Follow-Up Sixteen patients (17.2%) died within 6 months postoperatively. In these patients, causes of death were infection (43%), tumor progression (35%), and other causes (cardiovascular, renal, 22%). For the studied patients, survival was 65.7% at 1 year, 44.8% at 2 years, and 23% at 3 years (Fig 2). In the long-term follow-up, the overwhelming cause of death was tumor recurrence. No case of PBPF occurred during the entire outpatient follow-up period. Late empyema developed in 1 patient 2 months after operation. No evidence of bronchopleural fistula was detected at broncoscopy, and the patient was treated with open window thoracostomy. In 2 patients, tumor recurrence was detected at the bronchial stump, which, however, did not result in stump insufficiency. Neither of these 2 patients underwent reoperation, as other systemic metastases were present at the same time in both. Comment Postpneumonectomy bronchial fistula remains one of the most serious complications after pneumonectomy, and there is proven evidence that a number of patient-related factors as well as factors related to the operative tech-

4 Ann Thorac Surg TAGHAVI ET AL 2005;79:284 8 BRONCHIAL STUMP COVERAGE WITH PERICARDIAL FLAP 287 nique are important for its development. In particular, patients undergoing pneumonectomy, who in addition receive adjuvant or neoadjuvant therapies, do have a clearly elevated risk [12]. The extent of the surgical resection and the need for additional therapies are determined by oncologic principles, however, and direct influence on the prevention of PBPF can therefore be made only by the applied technique for closure and coverage of the bronchial stump. A large number of publications have dealt with this problem in the past, and especially the need for bronchial stump coverage has been emphasized repeatedly [13, 14], although no prospective randomized trial on this question has ever been published. Several years ago, we reviewed our personal experience with routine coverage of the postpneumonectomy stump with various tissues [15]. The flaps used included pleura, azygos vein, intercostal muscle, pericardial fat pad, and pericardial flaps; and the overall reported incidence of PBPF was as low as 0.8%. Since then, our preferred technique for bronchial stump coverage has been the use of a pedicled pericardial flap, and the purpose of this paper is now to review the results achieved with this particular technique in a much larger group of patients. The use of a flap of pericardium in thoracic surgery was first described as an alternative method to the pericardial fat graft by Brewer and associates [8] as early as Anderson and Miller [5, 6] later on have used this technique in different clinical situations, such as repair of tracheoesophageal fistulas, sleeve lobectomies, tracheal anastomosis, and extended pneumonectomies. The present paper represents the largest published series of patients in whom pericardial flaps were used for coverage of postpneumonectomy stumps. Two different techniques have been applied in these patients. In some of them, the bronchial stump was covered with pedicled pericardium; in the remaining patients, a pericardial flap including the pericardiophrenic vessels was used. At the beginning of our experience, we used pericardiophrenic flaps only in those patients whose phrenic nerve had to be sacrificed for oncologic reasons. Theoretically, phrenic nerve dysfunction after pneumonectomy might have an impact on the functional behavior of the contralateral diaphragm as well. However, no studies have investigated the functional difference of a pneumonectomy with or without phrenic nerve injury until now. If any difference could be expected at all, this would be of importance only in the early postoperative period, because later on the diaphragm becomes completely fixed on the pneumonectomy side. In our initial series, no particular functional disadvantage of the loss of the phrenic nerve after pneumonectomy was observed, and therefore the phrenic nerve was sacrificed on purpose later on in a number of patients to allow harvesting of a pedicled pericardiophrenic flap, which owns the potential advantage of a better blood supply. Attachment of the flaps to the bronchial stump was performed caplike with single stitches of 4-0 PDS, in a way that covered the stump completely, without necessarily decreasing its blood supply. Right-sided flaps usually were brought into the thoracic cavity behind the superior vena cava to avoid functional narrowing of the vessel. In all patients, the resulting defect in the pericardium was reconstructed with a Vicryl mesh [16], which was sewn in to prevent herniation of the heart through the resulting defect. The potential side effects that can be expected from such a procedure are arrhythmias in the postoperative period, infection of the foreign material, and cardiac tamponade in case of tight reconstruction. In this series, these specific complications occurred at a low rate. The incidence of postoperative supraventricular tachyarrhythmia was 17% (n 16), which was within the range described in literature [17]. Intrathoracic infection resulting in empyema was observed in 1 patient only, and it must remain speculative whether this was related to the use of foreign material. The patient was treated by thoracic wall fenestration and the bronchial stump remained closed during the whole treatment period. Even more, in 5 additional patients with concomitant aortic resection and prosthetic reconstruction, no infectious complication of the vascular graft occurred, possibly owing to the beneficial use of the pericardial flap [11]. The only serious method-related complication that was observed in 1 patient was cardiac tamponade early after the operation. Tamponade was caused by tight reconstruction of the pericardium, which most likely was performed during a temporary hypovolemic status. During postoperative normalization of the filling volume, symptoms of tamponade occurred. The patient was taken into the operating room, and the Vicryl mesh was exchanged for a larger one, thereby overcoming all symptoms. As mentioned before, controversy exists about the need for and the benefit from coverage of the bronchial stump. Asamura [3] concluded in his review of more than 2,300 patients after lung resection that further investigation should be performed to answer whether prevention of PBPF by tissue coverage is of benefit. Wright and colleagues [1] attributed the low incidence of 3.1% PBPF to their coverage technique in the discussion of their results with 256 patients after pneumonectomy, in whom the bronchial stump was routinely covered with autologous tissue. They used pleural flaps and pericardial fat pad flaps in the vast majority of their patients. However, of the 8 cases of PBPF described by them, 3 had been covered with pleura, 2 with omentum, 2 with pericardial fat pad, and 1 with intercostal muscle. That gives evidence that none of the methods applied, not even the technique of omentum pull-up, can offer complete protection against development of PBPF. Choice of the autologous tissue for coverage seems, therefore, to be of crucial importance for optimal results. Pleural flaps, although being the most frequently used structure [6], usually have the disadvantage that they are extremely thin and sometimes lack adequate blood supply. Intercostal muscle flaps have been used in some institutions [7]. It was not reported that harvesting of this type of flap would result in any disadvantage. However, vascularization at the end of operation sometimes can be poor,

5 288 TAGHAVI ET AL Ann Thorac Surg BRONCHIAL STUMP COVERAGE WITH PERICARDIAL FLAP 2005;79:284 8 Fig 3. Bronchoscopic view of a dehiscent bronchial stump, covered by pericardial patch. The patch remains intact and prevents communication of the bronchial system with the left thoracic cavity. despite careful dissection before introduction of the rib retractor. Mineo and coworkers [9] have reported excellent results with the use of a diaphragmatic flap to reinforce the bronchial stump after pneumonectomy. Pedicled omental flaps have widely been used for coverage of tracheobronchial defects and empyema [18]. Both techniques have the disadvantage of extending the thoracic operation into the abdomen. The favorable results of the use of a pedicled flap of pericardium in our study and the low incidence of specific complications observed suggest that bronchial stump reinforcement with this technique is a highly effective method for prevention of PBPF especially in patients at risk for bronchial healing problems. Addendum Since the submission of this manuscript, we have performed a left pneumonectomy in a 61-year-old man, and the bronchial stump was covered with a pedicled flap of pericardium. The postoperative course of the patient was complicated by pneumonia of the remaining lung, and he required mechanical ventilation for a total of 8 weeks. Five weeks after pneumonectomy, broncoscopy revealed an opening of the bronchial stump, which was effectively prevented by the pericardial patch from communicating with the thoracic cavity (Fig 3). Without further specific treatment, the patient continued to improve and was discharged from hospital. This experience underlines the efficacy of pericardial patch coverage to prevent a communication between the bronchial system and the thoracic cavity. References 1. Wright CD, Wain JC, Mathisen DJ. Postpneumonectomy bronchial fistula after sutured bronchial closure: incidence, risk factors and management. J Thorac Cardiocvasc Surg 1996;112: Sarsam MAI, Moussali H. Technique of bronchial closure after pneumonectomy. J Thorac Cardiovasc Surg 1989;98: Asamura H, Naruke T, Tsuchiya R. Bronchopleural fistulas associated with lung cancer operations. J Thorac Cardiovasc Surg 1992;104: Akoi T, Ozeki Y, Watanabe M. Cartilage folding for main bronchial stapling. Ann Thorac Surg 1998;65: Anderson TM, Miller JI. Surgical technique and application of pericardial fat pad and pericardiophrenic grafts. Ann Thorac Surg 1995;59: Anderson TM, Miller JI. Use of pleura, azygos vein, pericardium and muscle flaps in tracheobronchial surgery. Ann Thorac Surg 1995;60: Mineo TC, Ambrogi V, Pompeo E. Comparison between intercostal and diaphragmatic flap in the surgical treatment of early bronchopleural fistula. Eur J Cardiothorac Surg 1997;12: Brewer LA, King EL, Lilly LJ. Pericardial fat graft reinforcement. J Thorac Cardiovasc Surg 1953;26: Mineo TC, Ambrogi V. Early closure of the postpneumonectomy bronchopleural fistula by pedicled diaphragmatic flaps. Ann Thorac Surg 1995;60: Sweet RH. Closure of the bronchial stump following lobectomy or pneumonectomy. Surgery 1945;18: Klepetko W, Wisser W, Bîrsan T. T4 lung tumors with infiltration of the thoracic aorta: is surgery reasonable? Ann Thorac Surg 1999;67: Yamamoto R, Tada H, Kishi A. Effects of preoperative chemotherapy and radiation therapy on human bronchial blood flow. J Thorac Cardiovasc Surg 2000;119: Algar FJ, Alvarez A, Aranda JL, Salvatierra A, Baamonde C, Lopez-Pujol FJ. Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis. Ann Thorac Surg 2001;72: Deschamps C, Bernard A, Nichols FC, et al. Empyema and bronchopleural fistula after pneumonectomy: factors affecting incidence. Ann Thorac Surg 2001;72: Klepetko W, Taghavi S, Pereslenyi A, et al. Impect of different coverage techniques on incidence of postpneumonectomy stump fistula. Eur J Cardiothorac Surg 1999;15: Liermann A, Lachat M, von Segesser LK, Turina M. Resorbable pericardial replacement an experimental study. Helv Chir Acta 1992;58: Amar D, Roistacher N, Burt M, Reinsel RA, Ginsberg RJ, Wilson RS. Clinical and echocardiographic correlates of symptomatic tachydysrhythmias after noncardiac thoracic surgery. Chest 1995;108: Okada M, Tsubota N, Yoshimura M, Miyamoto Y, Yamagishi H, Satake S. Surgical treatment for chronic pleural empyema. Surg Today 2000;30: Weissberg D, Kaufman M. Suture closure versus stpaling of bronchial stump ín 304 lung cancer operations. Scand J Thorac Cardiovasc Surg 1992;26: Conlan AA, Lukanich JM, Schutz J. Elective pneumonectomie for benign lung disease: modern-day mortality and morbidity. J Thorac Cardiovasc Surg 1995;110: Al-Kattan K, Cattelani L, Goldstraw P. Bronchopleural fistula after pneumonectomy for lung cancer. Eur J Cardiothorac Surg 1995;9: Hollaus PH, Lax F, El-Nashef BB. Natural history of bronchpleural fistula after pneumonectomy: a review of 96 cases. Ann Thorac Surg 1997;63: Sirbu H, Busch T, Aleksic I, Schreiner W, Oster O, Dalichau H. Bronchopleural fistula in the surgery of non-small cell lung cancer: incidence, risk factors, and management. Ann Thorac Cardiovasc Surg 2001;7: Javadpour H, Sidhu P, Luke DA. Bronchopleural fistula after pneumonectomy. Ir J Med Sci 2003;172:13 5.