Tracheobronchial healing after lung and heart-lung transplantation

Eur J Cardio-thorac

Surg (1993) 7:453-456

Tracheobronchial healing after lung and heart-lung transplantation A critical review of 64 anastomoses R. Giudicelli, P. Thomas, G. Massard, M. Reynaud, P. Fuentes, M. Noirclerc and the Joint Marseille-MontrCal Lung Transplant Program Department 09, France

of Thoracic Surgery, Hbpital Sainte Marguerite,

270, boulevard

de Sainte Marguerite.

BP 29, F-l 3274 Marseille - Ccdex

Abstract. The authors report on an analysis concerning the healing of tracheobronchial anastomoses after lung and heart-lung transplantation. The present study includes 64 anastomoses selected from a total of 80. Sixteen were excluded because of early postoperative death; none of these deaths was related to an airway complication. Bronchial healing was assessed with bronchoscopic follow-up; the aspect of the suture line was classified according to the grades of Couraud. The initial reference was the examination at 2 weeks postoperatively, which was compared to subsequent follow-ups. At the initial assessment, 42 anastomoses were grade I, 4 were grade II, and 18 were grade III. The subsequent anatomic result was satisfactory for 52 sutures (81%). The complications observed in the remaining patients were malacia in 2, stenosis treated with a stenting device in 4 and dehiscence in 6. The duration of ischemia and postoperative mechanical respiratory support, as well as the proximal or distal location of the anastomosis appeared to be of significant prognostic value. [Eur J Cardio-thorac Surg (1993) 7:453-4561 Key words: Lung transplantation

~ Bronchial healing - Airway complications

Trachea1 or bronchial healing after lung and heart-lung transplantation is still a source of major concern because of the prevalence of anastomotic complications. These are in part explained by the fact that lung transplantation is the only graft procedure in which restoration of the systematic arterial blood supply towards the transplanted organ is not addressed. Airway complications either damage the functional results (stenosis, malacia), or may lead to life-threatening mediastinitis (dehiscence). Several techniques have been developed to decrease the air-way ischemia by either indirect or direct revascularization; airway complications have not entirely disappeared with any of them. The purpose of the present study was to assess which are the factors, beside tracheobronchial ischemia, contributing to the occurrence of airway complications.

Received for publication: Accepted for publication: Correspondence

March 17. 1992 April 21. 1993

to: Dr. Roger Giudicelli

Material and methods Patient population, operative procedure and immunosuppressive therap_v From November 1987 through December 1991,45 transplantations were performed in 43 patients, 2 having undergone retransplantation. There were 23 men and 20 women with a mean age of 25 years (range: 6-61 years). Double lung transplantation (DLT) was performed in 37 patients: 15 had enbloc replacement with trachea-totrachea (n = 2) or bronchus-to-bronchus anastomosis (n = 13). and 22 bilateral sequential lung graft. Six patients underwent single lung transplantation (SLT) (1 re-transplantation), and 2 underwent heart-lung transplantation (HLT) (1 retransplantation). A total of 80 airway anastomoses was performed. The early postoperative deaths (first month) of 9 patients necessitated the exclusion of 16 anastomoses for reasons of insufficient follow-up; none of these deaths was related to an airway complication. The present study includes 64 anastomses: 4 were tracheal-to-tracheal and 60 were bronchial-to-bronchial. Initially we performed the anastomosis in a proximal location, close to the carina (24 sutures). but currently the site of anastomosis is distal, close to the lobar branches (36 sutures). All anastomoses were performed using a running suture with an absorbable 3-O monofilament. A pediculized omental wrap was performed only in the first 3 cases, whereas the following 61 anastomoses were covered with the peribronchial mediastinal tissues. The immunosuppressive regimen included a 3-day course of antithymocyte globulin. starting at the time of transplantation; cy-

454 closporine, maintaining a whole blood level of 200 to 400 ug/ml; azathioprine (2 mg/kg per day), according to white blood cell count (up to 5 /mm3); and methylprednisolone sodium succcinate (8 mg/kg on day I,4 mg/kg on day 2,0.6 mg/kg on day 3) gradually reduced daily to 0.25 mg/kg per day after the first month. Rejection was treated with a 3-day course of intravenous methylprednisolone (1 mg).

Endoscopic assessment The quality of bronchial healing was appreciated on endoscopy that was performed routinely on the day of transplantation, weekly during the first month, and monthly thereafter. The endoscopic aspect was classified according to the bronchoscopic grades of Couraud [3]. The examination undertaken at the end of the second postoperative week was considered as the initial reference for subsequent comparisons.

Statistical analysis The means were compared with unpaired Student’s l-test, and frequencies with the X-square test. Correlation was assessed with linear regression analysis. Statistical significance was accepted for any value of P below the 0.05 level.

Results Overall results

The duration of follow-up was 1 to 2 months for 12 anastomoses, 3 to 6 months for 10, and more than 6 months for 42 (mean: 20 months, medium: 18.6 months).

Table 1. Tracheobronchial of Couraud [3])

healing: endoscopic

assessment

(grades

Beyond the 1st month

At 2 weeks postoperatively

Favorable

Unfavorable 4 stenosis

Grade I:

42 (66%)

38

Grade II:

4 (6%)

3

1 malacia

11

1 malacia 6 leakages

Grade III: 18 (28%)

Grade I: muco-mucosai healing; grade II: peritracheobronchial healing; grade III: partial or complete necrosis of variable depth. Unfuvarnble: outcome complicated by stenosis, malacia or dehiscence

Table 2. Tracheobronchial

healing with respect to duration

Storage

The bronchoscopy at the end of the second week disclosed 42 muco-mucosal first-intention healings (66%) which were classified grade I. A delayed muco-mucosal healing following peritracheobronchial scarring corresponding to grade II was noted in 4 (6%), and a grade III pattern characterized by focal necrosis was observed in 18 (28%). No grade IV (ulcerations or necrosis in areas distant to the anastomosis) was seen. Bronchoscopic follow-up showed that the healing at 2 weeks postoperatively is not permanent and may change subsequently. Among 42 grade I anastomoses, 4 ultimately developed a bronchial stenosis. On the other hand, a patient with grade III changes observed at the end of the second week may have a favorable outcome without any sequelae, as we noted in 11 cases (Table 1). Beyond the first month, 52 anastomoses had a favorable outcome @I%), and 12 were complicated (19%). Two sutures presented with a stable malacia. Despite a grade I pattern 2 weeks postoperatively, four sutures developed stenosis which required permanent endobronchial stenting. Six anastomoses classified grade III, 2 weeks postoperatively experienced subsequent dehiscence. Two leaks were partial: one patient with muco-mucosal dehiscence required permanent stenting with a Gianturco-device and is alive 33 months postoperatively; one patient had temporary insertion of a Dumon-device for a month resulting in a satisfactory calibration, still 15 months postoperatively. The remaining four had major dehiscence that resulted in the deaths of three patients (one having had dehiscence of both his bronchial sutures) as a final consequence of mediastinitis and overwhelming sepsis.

Parameters affecting bronchial healing

Early rejection occurred in 14 patients (22 anastomoses). Among them, three anastomotic stenoses were present (13.6%). On the other hand, the incidence of anastomotic complications for the 22 patients (42 anastomoses) who did not experience early rejection was 21.4% (g/42). The difference was not significant (x2 = 0.6; 0.3 < P I 0.5). Healing of tracheobronchial anastomoses was related to the duration of cold ischemia (Table 2); the longer the storage time, the greater was the incidence of anastomotic complications. All 4 grade I anastomoses that developed stenosis had an ischemic time of more than 4 hours (5 to 7 hours), whereas the 15 of grade I with conservation

of cold ischemia Storage > 4 hours

< 4 hours

Number

Favorable

Unfavorable

Number

Favorable

Unfavorable

Grade I

15

15

0

21

23

4

Grade II

2

2

0

2

1

1

Grade III

5

2

3

13

9

4

Unfavorable:

outcome complicated

by stenosis, malacia or dehiscence

455 Table 3. Tracheobronchial plant procedure Type of transplantation procedure

healing with respect to the type of trans-

Endoscopic grade at 2 weeks postoperatively

Outcome beyond the 1st month Favorable

Double lung en bloc

Grade I: Grade II: Grade Ill:

19 1 4

15 1 1

Sequential double lung

Grade I: Grade II: Grade III:

18 2 12

18 1 9

Single lung


Unfavorable: outcome cence

4 0 0

complicated

0

60

120

180

240

300

360

420

460

540

Storage time (min) Fig. 1. Correlation between storage time and postoperative chanical ventilation. The regression line is outlined

4 0 0

me-

by stenosis, malacia or dehis-

Table 4. Tracheobronchial healing with respect mechanical respiratory support Respiratory support

Number

Favorable outcome

I 5 days

Gradel: 16 Grade II: 2 Grade 111: 1

16 1

5 days < RS I 15 days

Grade I: IO Grade 11: 0 Grade Ill: 1

8 0 1

> 15 days


16 2 16

Unfavorable

to postoperative

Unfavorable outcome

1

14 2 9

RS = mechanical respiratory support Unfavorable: outcome complicated by stenosis, malacia or dehiscence

times of less than 4 hours had perfect anatomic results. However, the overall stenosis rates were 13.6% and 21.4%, respectively, and the difference did not achieve statistical significance (x2 = 0.6; 0.3 < P I 0.5). The proximal or distal location of the bronchial anastomosis was an important factor (Table 3). The four grade I patterns that developed stenosis were in proximal anastomoses. The incidence of anastomotic complications was 11.1% in patients with distal and 29.2% in patients with proximal suture lines. This difference was close to being significant (x2 = 3.2; 0.05 < P I 0.1). The influence of the duration of post-operative assisted ventilation is shown in Table 4. The four grade I anastomoses developing stenosis had ventilatory support for more than 5 days; all dehiscences complicating grade III sutures occurred in patients ventilated for more than 15 days. The mean duration of post-operative assisted ventilation was 16.3 days ( + 17.2; 5-74) in uncomplicated sutures and 31.2 days ( ) 19.5; 5 -60) in complicated sutures. This increase was statistically significant (It I = 2.64; P I 0.01). Regression analysis confirmed a significant cor-

relation between the storage time and the duration of postoperative mechanical ventilation (R = 0.39; P _< 0.005) (Fig. 1).

Discussion When no bronchial arterial revascularization procedure is performed, the donors’ trachea, carina or bronchus depends for its vascular supply on the retrograde flux originating from the low-pressured pulmonary artery [9, IO]. This precarious situation partly explains the prevalence of airway complications after lung and heart-lung transplantations [3, 13, 161. Besides, various factors may affect the healing of the airway suture. In fact, the tracheobronchial risk varies largely with respect to the type of transplantation procedure. The tracheal anastomosis of heart-lung replacements appears to be protected by the anastomotic network between coronary arteries and bronchial arteries. By contrast, the tracheal anastomosis performed in en bloc DLT is at the highest risk of ischemia [13]. In order to lower the airway ischemia, a proximal bi-bronchial anastomosis, performed at a 2-ring distance from the carina, has been proposed, leaving the recipient’s carina with its original vascularization [9, 111. For the same reasons, Bisson and Cooper have introduced the technique of bilateral sequential DLT [l, 51; the bronchial sutures are located at a level comparable to SLT. The risk of donor bronchial ischemia is certainly lower with distal anastomoses since a shorter bronchial segment depends on the retrograde revascularization originating from the pulmonary artery [9, lo]. The prognosis of these distal sutures has been reported to be similar to that of SLT [2], and the results of our small series tended towards the same conclusion. The other reason why several authors, like ourselves, currently prefer bilateral sequential lung transplantation is the considerable reduction of intraoperative bleeding achieved by avoiding cardiopulmonary bypass [l, 51. Tracheobronchial healing may be impeded by other factors. Chronologically, the first group of factors is re-

456

lated to the donor lung: type and quality of storage and duration of cold ischemia. The second group depends on postoperative outcome and management: duration of postoperative respiratory supply, early severe acute rejection and infection. In our series, we were not able to demonstrate an increased risk related to early rejection episodes occurring during the first 2 months postoperatively. On the other hand, location of the anastomosis, duration of ischemia and duration of postoperative positive pressure ventilation were important factors determining the outcome of our patients. A long period of ischemia and prolonged postoperative artificial ventilation were associated with a poor prognosis for bronchial healing; of course, both factors are interdependent, the second being a consequence of the first. Another factor which must be considered is the experience of the transplanting team. It appears that with our increasing experience, the prevalence of complications lessened: among the 24 anastomoses performed in 1991, a single minor complication occurred. It has been claimed that immunosuppressive therapy does not negatively influence adequate healing, despite the use of steroids [3, 12, 141. Furthermore, steroids may have a positive effect on airway healing [17]. The prevalence and severity of airway complications have led to the development of several preventive measures. The Toronto group has demonstrated experimentally the possibility of bronchial neo-vascularization with a pedicled omental wrap [8] and has successfully transposed this technique to human transplantation. Other kinds of wraps have similarly been shown to be useful [7, 15, 191. Whatever kind of wrap is used, the neovascularization takes several days to become established and function efficiently. The problem of bronchial ischemia during the immediate postoperative period however, remains unresolved. The theoretically ideal solution would be a revascularization procedure of the bronchial arteries. Such techniques have been developed in experimental surgery and appeared to improve bronchial healing reliably [6, 141.A reappraisal of human bronchial circulation by means of cadaver dissection [18] has led to the first clinical applications, which seem to provide promising results [4]. One might object, however, that these procedures prolong the operative procedure, require cardiopulmonary bypass, and probably prolong cold ischemia, which is perhaps one of the most important factors. We conclude that the ischemia of the donor lung’s airway is the central problem. In addition, tracheobronchial healing is affected by several factors the most important of which are the location of the anastomosis and the duration of both ischemia and postoperative ventilation. The ideal solution would be a direct revascularization of the bronchial arteries. However, the recent advent of bilateral sequential transplantation with distal bronchial anastomoses seems to reduce the incidence and severity of airway complications to an acceptable level. The further advantages of the technical simplicity of this procedure and the possible avoidance of cardiopulmonary bypass make the latter the technique of choice at this point.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

IO.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Bisson A, Bonnette P (1992) A new technique for double lung transplantation: bilateral single lung transplantation. J Thorac Cardiovasc Surg 103: 40-46 Calhoon JH, Grover L, Gibbons J, Bryan L, Levine M, Bailey R, Nichols L, Lum C, Trinkle K (1991) Single lung tranplantation. J Thorac Cardiovasc Surg 101:816-825 Couraud L, Martigne C, Velly JF, Gallon P, Dromer C, Baudet E (1991) La cicatrisation trachea-bronchique chez le transplant6 pulmonaire ou cardio-pulmonaire. Etude anatomique de 40 anastomoses. Ann Chir Thorac Cardiovasc 45: 641-648 Couraud L, Baudet E, Martigne C, Roques X, Velly JF, Laborde N, Dubrez J, Clerc F, Dromer C, Vallieres E (1992) Bronchial revascularization in double-lung transplantation: a series of 8 patients. Ann Thorac Surg 53: 88-94 Kaiser LR, Pasque MK, Trulock EP, Low DE, Dresler CM, Cooper JD (1991) Bilateral sequential lung transplantation: the procedure of choice for double-lung replacement. Ann Thorac Surg 52: 438-446 Laks H, Louie HW, Haas GS, Drinkwater DC, Lewis W, Permut LC, Stein DG, Chang P (1991) New technique of vascularization of the trachea and bronchus for lung transplantation. J Heart Lung Transplant 10:280-287 Legal YM, Chittal SM, Wright ES (1985) Early bronchial revascularisation with an intercostal pedicle graft following canine lung autotransplantation. Can J Surg 28: 518-522 Lima 0, Goldberg M, Peters WJ (1982) Bronchial omentopexy in canine lung transplantation. J Thorac Cardiovasc Surg 83:418-421 Metras D, Noirclerc M, Vaillant A, Brunet C, Kreitmann B (1990) Double lung transplant: the role of bilateral bronchial suture. Transplant Proc 22: 1477-1478 Mills NL, Boyd AD, Gheranpong C (1970) The significance of bronchial circulation in lung tranplantation. J Thorac Cardiovast Surg 60: 866-878 Noiclerc M, Metras D, Vaillant A, Camboulives J, Couvely JP, Pannetier A, Garbi 0, Chazalette JP, Goudard A, Carcassonne M (1989) Technique chirurgicale de la transplantation bipulmonaire. Lyon Chir 85: 247-252 Novick RJ, Menkise AH, McKenzie FN, Reid KR, Pflugfelder PW, Kostuk WJ, Ahmad D (1991) The safety of low-dose prednisone before and immediately after heart-lung tranplantation. Ann Thorac Surg 51: 642-645 Patterson AG, Todd TR, Cooper JD (1990) Airway complications after double lung transplantation. J Thorac Cardiovasc Surg 99: 14-21 Pinsker KL, Veith FJ, Kamholz SL, Montefusco C, Emeson P, Hagstrom JW (1984) Influence of bronchial circulation and corticosteroid therapy on bronchial anastomotic healing. J Thorac Cardiovasc Surg 87: 439-444 Riquet M, Bonette P, Carnot F (1989) Anatomic study of the pericardial fatty fringes in the dog. Revascularization of the bronchial suture in auto-transplantation of the left lung. Surg Radio1 Anat 11: 252-253 Schlfers HJ, Haydock DA, Fracs CB, Cooper JD (1991) The prevalence and management of bronchial anastomotic complications in lung transplantations. J Thorac Cardiovasc Surg 101: 1044-1052 Schafers HJ, Haverich A, Wagner TO, Wahlers T, Alken A, Borst HG (1992) Decreased incidence of bronchial complications following lung transplantation. Eur J Cardio-thorac Surg 6: 174-178 Schreinemakers HJ, Weder W, Miyoshi S, Harper BD, Shimokawa S, Egan TM, McKnight R, Cooper JD (1990) Direct revascularization of bronchial arteries for lung transplantation: an anatomical study. Ann Thorac Surg 49: 614-619 Turrentine MW, Kesler KA, Wright CD, McEwen KE, Faught PR, Miller ME, Mahomed Y, King H, Brown JW (1990) Effects of omental, intercostal and internal mammary artery pedicle wraps on bronchial healing. Ann Thorac Surg 49: 574-579