Early and Late Airway Complications After Lung Transplantation: Incidence and Management

Early and Late Airway Complications After Lung Transplantation: Incidence and Management Vibhu R. Kshettry, MD, Timothy J. Kroshus, MD, Marshall I. Hertz, MD, David W. Hunter, MD, Sara J. Shumway, MD, and R. Morton Bolman III, MD Division of Cardiovascular and Thoracic Surgery, Pulmonary and Critical Care Medicine, and Department of Radiology, University of Minnesota, Minneapolis, Minnesota

Background. Airway anastomosis complications continue to be a source of morbidity for lung transplant recipients. Methods. This study analyzes incidence, treatment, and follow-up of airway anastomotic complications occurring in 127 consecutive lung transplant airway anastomoses (77 single lung and 25 bilateral sequential lung). Complications were categorized as stenosis (11), granulation tissue (8), infection (7), bronchomalacia (5), or dehiscence (3). Follow-up after treatment ranged from 6 months to 4 years. Results. Nineteen airway anastomosis complications (15.0%) occurred in 18 patients. Telescoping the airway anastomosis reduced the complication rate to 12 of 97 (12.4%), compared with 7 of 30 (23.3%) for omental

wrapping, (p 5 0.15). Complications developed in 13 of 77 single-lung airway anastomoses (16.9%) versus 6 of 50 bilateral sequential lung recipients (12.0%). Treatment consisted of stenting (9 airway anastomoses), bronchodilation (8), laser debridement (4), rigid bronchoscopic debridement (2), operative revision (2), and growth factor application (2). There was no difference in actuarial survival between patients with or without airway anastomosis complications (p 5 1.0). Conclusions. Airway anastomosis complications can be successfully managed in the immediate or late postoperative period with good outcome up to 4 years after intervention. (Ann Thorac Surg 1997;63:1576 – 83) © 1997 by The Society of Thoracic Surgeons

I

ued occurrence of airway anastomosis complications, frequently becoming manifest weeks to months after transplantation, suggests that there are multiple events involved in the pathogenesis of airway anastomosis problems.

nitial attempts at human lung transplantation were hampered by a high frequency of impaired healing of the bronchial anastomosis. Disruption of the airway anastomosis was the major cause of death in the majority of patients who survived at least 2 weeks after transplantation [1]. Bronchial complications have been attributed to ischemia of the donor bronchus. The bronchial arterial circulation is not reestablished during transplantation, and rearterialization via recipient bronchial arteries requires 1 to 2 weeks after the operation [2]. Therefore, the viability of the donor bronchus is initially dependent upon retrograde collaterals from the pulmonary artery [3]. Early investigators have advocated revascularization of the transplant bronchial arteries [4, 5]. However, this proved to be technically demanding and complicated the procedure. Currently, many simpler operative techniques are used to minimize bronchial anastomotic complications. These include shortening the donor bronchial stump to two or less cartilaginous rings proximal to the upper lobe takeoff [6], reinforcing the anastomosis with a vascularized tissue pedicle such as the omentum [7] or intercostal muscle pedicle flap [8], and using the intussuscepting bronchial anastomosis technique [9, 10]. These maneuvers have helped in improving bronchial anastomosis healing. Despite these advances, the contin-

Accepted for publication Dec 17, 1996. Address reprint requests to Dr Kshettry, Minneapolis Heart Institute, 920 E 28th St, Suite 420, Minneapolis, MN 55407.

© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

Material and Methods Patients A retrospective review of 119 consecutive patients undergoing pulmonary transplantation from June 1989 through December 1994 was conducted to determine the incidence, treatment, and outcome of patients in whom airway anastomosis complications developed in the posttransplantation period. One hundred two patients met the study criteria, surviving at least 3 months after transplantation, including 77 single-lung (SL) and 25 bilateral sequential lung (BSL) transplants, for a total of 127 airway anastomoses at risk. Those patients who died within 3 months of causes related to airway anastomotic complications were included in this analysis. Seventeen patients who did not survive greater than 3 months died of causes unrelated to their airway anastomoses and thus were excluded from this analysis.

Operative Technique Operative techniques for each type of transplantation have previously been reported [11]. The first 26 patients (22 SLs and 4 BSLs for a total of 30 airway anastomoses) 0003-4975/97/$17.00 PII S0003-4975(97)00327-5

Ann Thorac Surg 1997;63:1576 – 83

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

Table 1. Patient Airway Complications by Anastomotic Technique Type of Tx

Telescoped (n 5 97)

Omental Wrap (n 5 30)

Right SL Left SL BSL

5 (5.2%) 3 (3.1%) 3a (4.1%)

3 (10%) 2 (6.7%) 2 (6.7%)

Total

12 (12.4%)

7 (23.3%)

a

One of these patients had bilateral airway complications. All other airway complications in BSL transplant recipients involved one of the two airways. BSL 5 bilateral single-lung transplant; Tx 5 transplant.

SL 5 single-lung transplant;

to receive transplants at our institution underwent a simple continuous suture technique with an omental wrap to reinforce the anastomosis. Subsequently, our airway anastomosis technique was modified to a telescoping technique, without the use of omental reinforcement. Since that time a total of 76 patients have undergone lung transplantation, including 55 SLs and 21 BSLs for a total of 97 airway anastomoses. All anastomoses were performed with nonabsorbable monofilament suture.

Immunosuppression Standard induction triple immunosuppression therapy was instituted in the immediate preoperative period consisting of cyclosporine (3 to 6 mg/kg), azathioprine (2.5 mg/kg), and perioperative methylprednisolone (500 mg) immediately before graft perfusion followed by 125 mg every 8 hours for three doses thereafter. Before 1990, steroids were withheld for 14 days after transplantation following perioperative administration. Patients were then given an oral steroid taper. From 1990 to the present time this taper was started immediately after perioperative coverage consisting of prednisone at 0.5 mg z kg21 z day21 and tapered to 0.1 mg z kg21 z day21. Cyclosporine was administered on a daily basis to maintain serum trough levels of 200 to 300 ng/mL as measured by high-pressure liquid chromatography (whole blood method). Azathioprine was administered daily at 2.5 mg z kg21 z day21 and adjusted to maintain a white blood cell count greater than 4,000/mL.

Infection Prophylaxis Perioperative antibiotics were routinely administered for the first 7 days after transplantation and adjusted based on intraoperative culture data obtained from the donor allograft bronchus at time of implantation. Fungal and bacterial prophylaxis were instituted indefinitely in the posttransplantation period consisting of daily trimethoprim-sulfamethoxazole and nystatin. Management of bacterial infections was directed by culture results. Anastomotic fungal infections were treated with aerosolized amphotericin B, oral itraconazole, or both. Further continuation of therapy was directed by patient response and appearance of the anastomosis by bronchoscopic visualization.

1577

Classification of Airway Anastomosis Complications We classified our airway anastomosis complications into five categories based upon bronchoscopic evaluation. Airway anastomosis complications were categorized as stenosis, bronchomalacia, exophytic granulation tissue, dehiscence, and anastomotic infections. This classification system allowed a systematic approach to management of these distinct problems. Most patients had a combination of airway anastomosis problems and required treatments directed at each particular entity.

Technique of Bronchodilation and Stent Placement Metal stent placement was preceded by bronchoscopy, which was used to inspect the bronchus, clear secretions, and, in conjunction with fluoroscopy, identify the site of the lesion and the origin of the upper lobe bronchus. Bronchoscopy was also used to guide the placement of a guidewire across the lesion into a distal bronchus. To avoid stent placement across the origin of the upper lobe bronchus, the upper lobe was marked by contrast injection employing nonionic contrast, which permitted a temporary opacification. With experience, this approach was modified to marking the upper lobe bronchus by placing a separate guidewire into the upper lobe. With further experience, it became clear that crossing of the upper lobe bronchus origin by a Gianturco stent (Cook Inc., Bloomington, IN) would not result in any clinical problems. At that time, specific marking of the upper lobe bronchus during stent placement was discontinued and Gianturco stent placement was instead optimized for treatment of the stenotic or bronchomalacic segment without consideration for the upper lobe origin. Metal stent placement was accomplished under fluoroscopic guidance. The stent was placed into the area where it was required and then either released in the case of the self-expanding Gianturco stents or Wallstents (Schneider Co, Minneapolis, MN) or expanded with a balloon in the case of the Palmaz stent (Johnson & Johnson Co, Warren, NJ). Stents were fluoroscopically and bronchoscopically inspected and then redilated if necessary to “force” the stent struts against the wall of the bronchus. This promoted anchoring of the stent to the wall of the bronchus in the short term and in the long term encouraged coverage of the metal struts of the stent with bronchial epithelium. Malpositioned stents were removed through a large rigid bronchoscope using a 9F rigid grasper. When a Gianturco stent strut is grasped and pulled into the bronchoscope, it will usually break, which then allows unfolding and straightening of the stent. The stent can then be pulled through the scope with surprising ease, usually in a single piece. Palmaz stents collapse readily as they are pulled into the bronchoscope and are then removed with minimal difficulty. Wallstents have angulated struts that appose to the bronchial wall, making removal extremely difficult.

1578

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

Ann Thorac Surg 1997;63:1576 – 83

Table 2. Demographics, Airway Complication, and Time to Treatment Patient No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Primary Disease

Age (y)

Sex

Tx Type

Complication

POD

a1-Antitrypsin Lymphangiomatosis Lymphocytic pneumonitis a1-Antitrypsin Pulmonary HTN a1-Antitrypsin Cystic fibrosis IPF a1-Antitrypsin COPD PPH COPD COPD Histiocytosis X Cystic fibrosis COPD IPF PPH

48 37 18 30 19 45 33 55 56 59 38 54 53 44 11 56 60 53

F F F M F M M M M M F M M F F F M F

RSL RSL LSL LSL RSL BSL BSL RSL LSL RSL BSL LSL LSL RSL BSL RSL RSL BSL

ST/BM ST/GT DH DH ST GT/IN ST BM/ST/IN ST/BM BM Bilateral ST/BM/IN ST/IN GT/IN ST/GT GT/ST DH/IN ST/GT GT/IN

1,245 108 14 7 60 68 120 119 190 453 85 64 103 80 79 35 60 30

BM 5 bronchomalacia; BSL 5 bilateral sequential lung transplant; GT 5 granulation tissue; HTN 5 hypertension; IN 5 infection; POD 5 postoperative day; PPH 5 primary pulmonary hypertension;

Results Patients and Airway Evaluation One hundred twenty-seven airway anastomoses were evaluated in 102 patients (77 SLs and 25 BSLs). A total of 19 airway anastomotic complications developed in 18 patients for an overall complication rate of 15.0%. There were 9 men and 9 women with a mean age at time of transplantation of 40.4 years (range, 18 to 60 years). Patient airway complications by anastomotic technique and type of transplant are listed in Table 1. A total of 30 airway anastomoses were performed using the omental wrap technique (22 SLs and 4 BSLs), with 7 complications noted for an incidence of 23.3%. Using this technique, complications developed in 5 of 22 SL airway anastomoses (22.7%) and 2 of 8 BSL airway anastomoses (25%). In comparison, in those patients with the telescoping technique, 12 complications developed in 97 airway anastomoses at risk (12.4%): in 7 of 55 SL airway anastomoses (12.7%) and 5 of 42 BSL airway anastomoses (12%). Changing our technique from omental wrapping to the telescoping anastomosis reduced the incidence of complications by nearly 50%, but did not reach the level of statistical significance (p 5 0.15 by Fischer’s exact test). Follow-up for all patients included in this study was 23.2 6 17.6 months. Mean follow-up for patients undergoing BSL transplantation was 19.6 6 14.8 months versus 24.4 6 18.4 months for SL transplant recipients.

Timing of Airway Complications and Patient Survival Patient demographics, airway anastomotic complication type, and interval to treatment of complications is illustrated in Table 2. Interval to the occurrence of an airway

COPD 5 chronic obstructive pulmonary disease; DH 5 dehiscence; IPF 5 idiopathic pulmonary fibrosis; LSL 5 left single-lung transplant; RSL 5 right single-lung transplant; ST 5 stenosis; Tx 5 transplant.

anastomotic complication requiring treatment intervention had a mean of 162.3 6 287.6 days, ranging from 7 to 1245 days after transplantation for all patients. Those patients with the omental wrap technique had complications at 231.7 6 448.8 days (range, 7 to 1,245 days) compared with those with the telescoping technique at 118.1 6 119.3 days (range, 35 to 190 days). Actuarial survival (Table 3) was not different for patients in whom airway anastomotic complications developed compared with those in whom they did not (p 5 1.0 by x2 analysis).

Organ Ischemic Time and Acute Rejection Allograft ischemia was similar among patients in whom airway anastomotic complications developed regardless of anastomotic technique used. Those patients undergoing omental wrapping in whom airway anastomotic complications developed had a mean graft ischemic interval of 290.7 6 102.3 minutes versus 290.0 6 86.7 minutes for telescoped anastomoses. Allograft ischemia was not significantly different for any patients undergoing pulmonary transplantation, with an average organ ischemia time of 260 6 70.7 minutes for SL transplants (median,

Table 3. Actuarial Survival of Patients With Versus Without Airway Complications Time After Transplantation 6 Months 1 Year 2 Years

No Airway Complications

Airway Complications

0.96 0.90 0.81

0.94 0.94 0.84

Ann Thorac Surg 1997;63:1576 – 83

Fig 1. Buckling of the airway anastomosis noted immediately after transplantation due to excessive telescoping of the cartilaginous portion of the bronchus.

245 minutes; range, 97 to 397 minutes) and 358 6 143 minutes for BSL transplants (median, 326 minutes; range, 187 to 708 minutes; calculated from the second of two lungs transplanted). Thus, allograft ischemia time was not significantly different for wrapping versus telescoping techniques for those patients in whom airway anastomotic complications developed and did not appear to be a critical etiologic factor responsible for the increased airway anastomotic complication rate observed in the omental wrapping group of patients. The incidence of acute rejection before the development of airway anastomotic complications was not significantly different from that of patients without complications.

Management of Stenosis Perianastomotic stenoses were managed with balloon bronchodilation in our early experience. This proved to be a difficult problem in several patients, with multiple

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

1579

Fig 2. Airway anastomosis buckling, as noted in Figure 1, now 2 weeks after transplantation showing beginning of stenosis.

dilation procedures required. Improved results were noted with the use of stents. Early experience with silicone stents was associated with a high incidence of dislodgment from coughing, requiring metal stent placement in 2 patients. A total of 10 patients with 11 airway anastomoses were diagnosed with bronchial anastomotic stenosis. Stenosis occurred on the right in 7 airway anastomoses and on the left in 4. Eight patients underwent bronchodilation, with a mean of 3.9 dilation procedures per airway anastomotic stenosis. Those patients not responding to repeated dilation procedures underwent stent placement. Patients were considered candidates for metal stent placement if they had failed attempts at laser recanalization, silicone stent placement, or repeat balloon dilation; and as a result of the failed therapy had either decreased lung function or repeated infections severe enough to require hospitalization. Six patients (7 airway anastomoses) eventually underwent metal stent placement. Of

Fig 3. Multiple views of bronchomalacia during inspiration and expiration.

1580

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

Ann Thorac Surg 1997;63:1576 – 83

Management of Bronchomalacia

Fig 4. Pulmonary function tests (forced expiratory volume in 1 second [FEV1] and forced vital capacity [FVC]) before and after stenting of the airway anastomosis for stenosis or bronchomalacia. (ns 5 not significant.)

these 7 stenotic bronchi, 3 had associated areas of perianastomotic bronchomalacia. Four bronchi with pure stenoses failed repeated balloon bronchodilation and were subsequently treated with 1 Wallstent and 3 Gianturco stents. Three stenotic bronchi and 3 bronchi with a combination of stenosis and bronchomalacia were treated with Gianturco stents. Two of 6 stents dislodged during or immediately after placement, necessitating removal and replacement. All 6 successful placements employed tandem stents, 4 of which were smaller in the more peripheral portion of the stent, which decreased migration toward the trachea. Stent placement across the right upper lobe origin has occurred in 3 patients but has been tolerated without problems. Two patients with right bronchial anastomotic stenoses also had development of right bronchus intermedius stenosis. Both were successfully treated with repeated bronchodilation and simultaneous dual balloon inflation in the bronchus intermedius and main bronchus. The technique of telescoping the airway anastomosis in single-lung transplantation must be performed carefully to avoid “buckling” of the anastomosis (Figs 1, 2). Over-telescoping the airway anastomosis at the cartilaginous-mucosal junction has led to infolding of the cartilaginous portion of the bronchus, resulting in progressive stenosis. This technical problem is currently avoided by performing an end-to-end anastomosis of the cartilaginous portion of the bronchus with interrupted figure-of-8 monofilament sutures. Gianturco stent fracture has occurred in 2 of 8 stents, resulting in airway compromise. In 1 case the fractured struts were removed and the lumen remained patent. In the other case, the single fractured strut was removed and the lumen was then held in the open position with a Wallstent.

A total of 5 patients were determined to have some degree of bronchomalacia, based on bronchoscopic examination, in combination with other airway problems (Fig 3). All 5 patients (5 airways, 3 with associated stenoses) with clinically significant bronchomalacia underwent metal stenting and demonstrated clinical improvement after successful stent placement. The 2 patients with pure bronchomalacia were treated primarily with Palmaz stent placement. Both stents collapsed due to the forces of coughing, resulting in airway compromise. Balloon reexpansion was unsuccessful in maintaining long-term patency. Both airway anastomoses were eventually successfully treated with placement of a Gianturco stent inside the balloon-reexpanded Palmaz stent. One patient required a Wallstent inside the Gianturco and Palmaz stents after the struts of the Gianturco stent fractured, resulting in recurrence of airway collapse with exhalation. Stenting of airways for bronchomalacia or stenosis resulted in a significant improvement in pulmonary function tests (Fig 4) and in degree of dyspnea. Eight patients had 9 stents placed, with a mean forced expiratory volume in 1 second of 1.05 6 0.11 L before stent placement and 1.58 6 0.13 L after stent placement (p 5 0.04), and a forced vital capacity of 2.38 6 0.19 L before stent placement and 2.72 6 0.15 L after stent placement (p 5 not significant).

Management of Exophytic Granulation Tissue Exophytic granulation tissue was treated with either rigid bronchoscopic debridement or yttrium-aluminum garnet laser debridement. Granulation tissue was differentiated from necrotic, ischemic, or infected mucosa based on clinical evidence and visual inspection. Tissue samples obtained from bronchoscopic debridement were routinely cultured. Five patients underwent debridement, including 3 with yttrium-aluminum garnet laser, 1 with rigid bronchoscopic and subsequent repeated yttriumaluminum garnet laser treatments, and 1 with rigid bronchoscopic debridement alone. Four of the 5 patients had a successful result after one debridement procedure, and did not have recurrence during long-term follow-up. Debridement was well tolerated, with no major complications noted.

Management of Dehiscence Airway anastomotic dehiscence occurred in 3 patients, 2 of whom underwent the omental wrapping bronchial anastomotic technique and 1 a telescoping anastomosis. Allograft ischemia for these patients was 202, 373, and 333 minutes, respectively, and occurred after right SL in 1 and left SL transplantation in 2 patients. In 1 patient a significant bronchial air leak developed requiring reexploration 7 days after transplant for revision. Another patient showed substantial mucosal sloughing with partial or impending dehiscence at the perianastomotic region that was treated locally with autologous plateletderived wound healing factor, as previously reported [12]. An additional patient with a less severe degree of

Ann Thorac Surg 1997;63:1576 – 83

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

Table 4. Bronchial Anastomotic Infections Infecting Organisms Rhizopus sp Aspergillus fumigatus

Aspergillus fumigatus

Group D enterococcus, Staphylococcus aureus Pseudomonas aeruginosa, Staphylococcus aureus Candida sp Polymicrobial

a

Airway Complication

Treatment

Granulation tissue Antifungal agents, debridement Stenosis Antifungal agents, debridement, stenting Granulation tissue Debridement, antifungal agents Stenosis Antibiotics, dilation/stenting Granulation tissue, Antibiotics, bronchomalacia stenting Granulation tissue Antifungal agents, debridement Dehiscence Antibiotics, antifungal agents, debridement, attempted airway revisiona

Revision technically impossible.

mucosal sloughing was treated in a similar fashion and improved without the development of dehiscence. Application of wound healing factors was successful in both patients, with a normal-appearing airway 2 years after treatment in one patient. In 1 patient a partial dehiscence developed 35 days after transplantation that was associated with an anastomotic fungal infection. Airway revision and debridement via thoracotomy was attempted but was technically impossible due to extensive adhesions. This patient died 48 hours later of disseminated fungal sepsis.

Management of Airway Anastomotic Infections Seven patients with airway complications experienced anastomotic infections, including 4 with fungal infec-

1581

tions, 2 with bacterial infections, and 1 with both bacterial and fungal microorganisms. Infections were classified as significant based on the need for debridement of the airway and administration of antimicrobial therapy. These airways underwent early biopsy and careful debridement for removal of necrotic tissue and culture. Bacterial infections were polymicrobial, with grampositive and gram-negative bacterial species, including Staphylococcus aureus (n 5 3) and Pseudomonas aeruginosa (n 5 2). Fungal infections were predominantly with Aspergillus fumigatus. Treatment of infections consisted of appropriate antimicrobial and antifungal therapy in addition to debridement of the airway anastomosis. Antifungal therapy consisted of aerosolized amphotericin B and oral itraconazole. Some patients required a course of intravenous amphotericin B due to the severity of infection and limited response to aerosolized treatments. A summary of infecting organisms, concurrent or subsequent airway complication, and treatment is shown in Table 4.

Algorithm for Evaluation and Treatment of Airway Anastomotic Complications Lung transplant recipients underwent periodic bronchoscopic evaluation of the airway anastomosis at scheduled time periods after transplantation. In addition, immediate bronchoscopic evaluation of the airway anastomosis was performed when indicated clinically. An algorithm summarizing treatment interventions after diagnosis of airway anastomotic complications is shown in Figure 5.

Comment Approaches to airway complications have undergone marked evolution as transplant centers have become more experienced in the techniques of lung transplantation. Early lung transplant centers reported an airway anastomotic complication rate as high as 80% [1]. Changes in surgical technique and increased experience Fig 5. Algorithm for bronchoscopic diagnosis and treatment of airway complications.

1582

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

in lung transplantation have dramatically reduced the complication rate. Despite these improvements, airway complications remain a significant source of morbidity and occasional mortality. The pathophysiologic events leading to airway complications remain incompletely understood. Donor organ ischemia is thought to be a significant contributor to the development of complications, including stenosis and bronchomalacia. The bronchus has a limited collateral blood supply, which increases the risk of necrosis and dehiscence. Innovative techniques have been developed to protect the bronchial anastomosis and to hasten neovascular ingrowth. Wrapping the bronchial anastomosis with omentum [7] or other vascularized pedicles of tissue such as intercostal muscle and internal mammary artery [8] has demonstrated reestablishment of collateral circulation to the donor bronchus. Modifying the airway anastomotic technique by reducing the length of the donor bronchus has been shown to minimize the degree of ischemia and promote bronchial healing [6]. Invaginating the donor bronchus into that of the recipient bronchus in a telescoping fashion supplies vascularity to the donor bronchus [9, 10]. Direct revascularization of donor bronchial arteries has been attempted both in experimental [4] and clinical settings [5]; however, results are preliminary and bronchial revascularization is technically demanding. Airway stenosis has been a significant problem in a number of patients after lung transplantation. Stenosis has been managed with silicone rubber stents [13, 14]. Recently, however, expandable metallic stents have been used due to their improved mucociliary clearance and potential for overgrowth with respiratory mucosa [15, 16] without need for later removal. The placement of metallic stents has been associated with an immediate marked improvement in forced expiratory volume [15]. Airway stenoses have also reportedly been managed surgically, with sleeve resection of stenotic segments, bilobectomy, and retransplantation as interventions, all successful in removing existing stenoses [17]. These, however, are dramatic interventions that are rarely indicated. Airway anastomoses with bronchomalacia are also best managed with stents. It has been advised to insert stents early after ablation of the stricture when anastomotic stenosis is complicated by bronchomalacia [18]. Use of expandable wire stents in our experience has resulted in no significant morbidity and no mortality. The combined use of bronchodilation with stenting has been successful in most of our patients with airway stenosis. We advocate this approach and believe that frequent and early bronchoscopic examination of the anastomotic sites will allow for treatment intervention before progressive stenosis becomes unmanageable by conservative measures. Wire expandable stents cause less impairment in the clearance of airway secretions, are resistant to migration, and have the advantage of low internal to external diameter ratio. Although adjustment of these stents can be difficult, we have been able to balloon-dilate collapsed or stenotic stents and place new stents within the lumen of existing stents. None of these stent manipulations has

Ann Thorac Surg 1997;63:1576 – 83

resulted in any adverse sequelae, including no episodes of hemorrhage or infection. Stent fracture has occurred on several occasions, but has also been treated with either stent removal or placement of new stents within the internal lumen of the fractured stent. Newer techniques in the manufacture of these stents has produced a stronger metal alloy, potentially reducing the incidence of stent fracture. In contrast, use of silicone stents is associated with encrustation, which may serve as a nidus for infection, and dislodgment or migration, necessitating stent replacement or repositioning [19]. Modifications in bronchial anastomotic technique have also improved the incidence of airway complications. Dramatic improvements have been made by modifying bronchial anastomotic techniques from wrapping procedures to the telescoping technique [20]. Bilateral sequential lung transplants have also been reported to result in a higher complication rate compared with SL transplantation. In our series we noted a high complication rate (25%) with the use of omental wrapping compared with telescoping each anastomosis (9.5%) in BSL transplantation. However, the difference between these groups was not statistically significant due to small numbers in the omental wrap group. In addition, comparison is difficult due to our early experience in performing lung transplantation and the improvements made over time in donor organ preservation, immunosuppression, and patient management. Furthermore, the use of the omental wrap technique may result in abdominal complications postoperatively [21], none of which have occurred after telescoping anastomoses. From these observations, we believe that changing the anastomotic technique has contributed substantially to a decrease in airway complications. Bronchial dehiscence remains a disastrous complication in the posttransplantation period. Most cases occur early after transplantation, are extremely difficult to treat, and are associated with high mortality. The initiating events leading to dehiscence most likely are due to inadequate savascularization at the bronchial anastomosis. One patient in this series underwent successful operative revision of the airway anastomosis due to an unresolving air leak 1 week after transplantation. Bronchial dehiscence has rarely been successfully managed surgically [22]. The advent of wrapping procedures and, later, telescoping anastomotic techniques have provided increased tissue coverage at the anastomotic site. This has undoubtedly led to a decreased incidence of bronchomediastinal fistulas in patients in whom dehiscence has developed. Management of partial dehiscence may also be attempted with the local application of growth factors [12]. Steroids have also been implicated in impaired healing of the bronchial anastomosis [23]. Early experience with clinical lung transplantation suggested that avoiding steroids in the first postoperative week was advantageous. The avoidance of steroids combined with the use of bronchial omentopexy in canine models demonstrated a marked decrease in the incidence of airway complications [24, 25]. With the introduction of cyclosporine,

Ann Thorac Surg 1997;63:1576 – 83

steroid dosages were lowered, and, by using cyclosporine in combination with azathioprine, adequate immunosuppression could be achieved with bronchial healing. However, current modifications in preservation, operative, and immunosuppression techniques have improved bronchial anastomosis healing and steroids are now given early after transplantation [9, 20]. Anastomotic obstruction may also be caused by exophytic granulation tissue. This can be managed with rigid bronchoscopic debridement or yttrium-aluminum garnet laser phototherapy. Extreme caution must be taken due to the possibility of creating a bronchomediastinal fistula. This is particularly the case the first 3 months after transplantation. Debridement must be only of excess granulation tissue to prevent perforation. The use of these procedures in this series of patients has been well tolerated without development of complications. Airway anastomotic infections may coexist with or predispose to latter complications. Twelve of 14 patients requiring stent placement for stenosis after single, bilateral single, and heart-lung transplantation in one report had infections complicating the airway problem, with Aspergillus fumigatus and Pseudomonas aeruginosa as the most commonly isolated microorganisms [15]. Of the 11 cases of stenosis in the present series, 3 were complicated by anastomotic infections. Careful debridement of necrotic mucosal tissue is indicated as well as the administration of appropriate intravenous and inhalational antimicrobial agents. In summary, airway anastomotic complications after lung transplantation remain a difficult and challenging problem but can be successfully managed with a comprehensive multimodality approach. Our experience with these complications has led to early and more frequent bronchoscopic visualization of all airway anastomoses. Early recognition and treatment are paramount to decrease morbidity and mortality.

References 1. Wildevuur CRH, Benfield JR. A review of 23 human lung transplantation. Ann Thorac Surg 1970;9:489 –515. 2. Siegelman SS, Hagstrom JWC, Koerner SK, Veith FJ. Restoration of bronchial artery circulation after canine lung allotransplantation. J Thorac Cardiovasc Surg 1977;73:792–5. 3. Barman SA, Ardell JL, Parker JC, Perry ML, Taylor AE. Pulmonary and systemic blood flow contributions to upper airway in canine lung. Am J Physiol 1988;255:H1130 –5. 4. Mills NL, Boyd AD, Gheranpong C. The significance of bronchial circulation in lung transplantation. J Thorac Cardiovasc Surg 1970;60:866–78. 5. Couraud L, Baudet E, Nashef SAM, et al. Lung transplantation with bronchial revascularisation. Eur J Cardiothorac Surg 1992;6:490–5. 6. Pinsker KL, Koerner SK, Kamholz SL, Hagstrom JWC, Veith FJ. Effect of donor bronchial length on healing: a canine

KSHETTRY ET AL AIRWAY COMPLICATIONS AFTER TRANSPLANTATION

7.

8. 9. 10. 11. 12. 13. 14. 15. 16.

17. 18. 19.

20. 21. 22.

23.

24.

25.

1583

model to evaluate bronchial anastomotic problems in lung transplantation. J Thorac Cardiovasc Surg 1979;77:669–73. Morgan E, Lima O, Goldberg M, Ferdman A, Luk SK, Cooper JD. Successful revascularization of totally ischemic bronchial autograft with omental pedicle flaps in dogs. J Thorac Cardiovasc Surg 1982;84:204–10. Fell SC, Mollenkopf FP, Montefusco CM, et al. Revascularization of ischemic bronchial anastomoses by an intercostal pedicle flap. J Thorac Cardiovasc Surg 1985;90:172– 8. Calhoon JH, Grover FL, Gibbons WJ, et al. Single lung transplantation. Alternate indications and technique. J Thorac Cardiovasc Surg 1991;101:816–25. Veith FJ, Richards K. Improved technique for canine lung transplantation. Ann Surg 1970;171:553– 8. Bolman RM, Shumway SJ, Estrin JA, Hertz MI. Lung and heart-lung transplantation. Evolution and new applications. Ann Surg 1991;214:456–70. Hertz MI, Harmon KR, Knighton DR, et al. Combined laser phototherapy and growth factor treatment of bronchial obstruction after lung transplantation. Chest 1991;100:1717–9. Cooper JD, Pearson FG, Patterson GA, et al. Use of silicone stents in the management of airway problems. Ann Thorac Surg 1989;47:371– 8. Griffith BP, Magee MJ, Gonzalez IF, et al. Anastomotic pitfalls in lung transplantation. J Thorac Cardiovasc Surg 1994;107:743–54. Higgins R, McNeil K, Dennis C, et al. Airway stenoses after lung transplantation: management with expanding metallic stents. J Heart Lung Transplant 1994;13:774– 8. Nashef SAM, Dromer C, Velly J-F, Labrousse L, Couraud L. Expanding wire stents in benign tracheobronchial disease: indications and complications. Ann Thorac Surg 1992;54: 937– 40. Scha¨fers H-J, Scha¨fer CM, Zink C, Haverich A, Borst HG. Surgical treatment of airway complications after lung transplantation. J Thorac Cardiovasc Surg 1994;107:1476– 80. Shennib H, Massard G. Airway complications in lung transplantation. Ann Thorac Surg 1994;57:506–11. Sonett JR, Keenan RJ, Ferson PF, Griffith BP, Landreneau RJ. Endobronchial management of benign, malignant, and lung transplantation airway stenoses. Ann Thorac Surg 1995;59: 1417–22. DeHoyos AL, Patterson GA, Maurer JR, Ramirez JL, Miller JD, Winton TL. Pulmonary transplantation. Early and late results. J Thorac Cardiovasc Surg 1992;103:295–306. Smith PC, Slaughter MS, Petty MG, Shumway SJ, Kshettry VR, Bolman RM. Abdominal complications after lung transplantation. J Heart Lung Transplant 1995;14:44–51. Kirk AJB, Conacher ID, Corris PA, Ashcroft T, Dark JH. Successful surgical management of bronchial dehiscence after single-lung transplantation. Ann Thorac Surg 1990;49: 147–9. Lima O, Cooper JD, Peters WJ, et al. Effects of methylprednisolone and azathioprine on bronchial healing following lung autotransplantation. J Thorac Cardiovasc Surg 1981;82: 211–5. Goldberg M, Lima O, Morgan E, et al. A comparison between cyclosporin A and methylprednisolone plus azathioprine on bronchial healing following canine lung allotransplantation. J Thorac Cardiovasc Surg 1983;85:821– 6. Lima O, Goldberg M, Peters WJ, et al. Bronchial omentopexy in canine lung transplantation. J Thorac Cardiovasc Surg 1982;83:418–21.