The 1-year survival rate approaches 80% for patients

Lung Transplantation for Respiratory Failure Resulting From Systemic Disease Frank A. Pigula, MD, Bartley P. Griffith, MD, Marco A. Zenati, MD, James H. Dauber, MD, Samuel A. Yousem, MD, and Robert J. Keenan, MD Division of Cardiothoracic Surgery, Presbyterian University Hospital, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Background. Lung transplantation for pulmonary failure resulting from systemic disease is controversial. We reviewed our transplant experience in patients with sarcoidosis, scleroderma, lymphangioleiomyomatosis, and graft-versus-host disease. Methods. This retrospective review examined the outcome of 23 patients who underwent pulmonary transplantation for these systemic diseases. Group 1 included 15 patients with pulmonary hypertension who underwent transplantation (9 for sarcoidosis, 6 for scleroderma), and group 2 included 8 patients with normal pulmonary artery pressures who underwent transplantation (5 for lymphangioleiomyomatosis, 3 for graft-versus-host disease). The incidences of infection and rejection, pulmonary function, and survival were measured and compared with those of patients who underwent transplantation for isolated pulmonary disease. Results. Although there were no differences in the rate of infection between patients who underwent transplantation for systemic versus isolated disease, patients with pulmonary hypertension who underwent transplantation for systemic disease had significantly lower rates of rejection. Four patients with sarcoidosis and 2 with lymphangioleiomyomatosis demonstrated recurrence in the allograft. Survival was similar between patients who underwent transplantation for systemic versus isolated disease. Conclusions. Patients with respiratory failure resulting from these systemic diseases can undergo transplantation with outcomes comparable to those obtained in patients who undergo transplantation for isolated pulmonary disease. (Ann Thorac Surg 1997;64:1630 4) 1997 by The Society of Thoracic Surgeons The 1-year survival rate approaches 80% for patients who receive lung transplants for isolated pulmonary diseases such as emphysema, pulmonary fibrosis, and pulmonary hypertension [1]. Transplantation for patients who have pulmonary failure secondary to systemic diseases remains controversial. These unconventional indications include pulmonary involvement resulting from sarcoidosis, scleroderma, lymphangioleiomyomatosis (LAM), and graft-versus-host disease (GVHD). This retrospective review examines the outcome of lung transplantation as a treatment option for these patients, and compares it to that of patients who undergo transplantation for conventional, isolated pulmonary pathology. As of December 1996, 468 lung transplantations have been performed at the University of Pittsburgh. Twentythree (5%) have been performed for unconventional indications: 9 for sarcoidosis, 6 for scleroderma, 5 for LAM, and 3 for GVHD. Although these patients represent a small percentage of all those evaluated for lung transplantation, they frequently are referred to transplant centers for consideration. Candidates for transplantation were less than 60 years old and were functionally limited (New York Heart Association class III and IV) by Accepted for publication June 26, 1997. Address reprint requests to Dr Keenan, Division of Cardiothoracic Surgery, University of Pittsburgh Medical Center, 200 Lothrop St, Suite c-700, Pittsburgh, PA 15213. respiratory symptoms. Candidates had to be free of evidence of significant involvement of extrapulmonary organ systems. Pretransplantation evaluation included spirometry, arterial blood gases, exercise oxygen titration, 6-minute walk, cardiac echocardiography, quantitative left and right perfusion lung scans, and right heart catheterization with measurement of pulmonary artery pressures. Pulmonary hypertension (defined as a mean pulmonary artery pressure 30 mm Hg) is a common physiology affecting patients being considered for lung transplantation. Accordingly, the patients in our review were divided into two groups. Group 1 contained 15 patients (9 with sarcoidosis, 6 with scleroderma) who had pulmonary hypertension, and group 2 contained 8 patients (5 with LAM, 3 with GVHD) who had normal pulmonary artery pressures. Twelve single-lung transplantations, 2 heart-lung transplantations, and 1 double-lung transplantation were performed in group 1. Eight single-lung transplantations and 1 double-lung transplantation were performed in group 2, including a second single-lung transplantation performed 18 days postoperatively because of primary graft failure. A contemporaneous group of patients undergoing single-lung transplantation for isolated pulmonary disease was identified. Group 1C consisted of 17 patients with pulmonary hypertension (10 with primary pulmonary 1997 by The Society of Thoracic Surgeons 0003-4975/97/$17.00 Published by Elsevier Science Inc PII S0003-4975(97)00930-2

Ann Thorac Surg PIGULA ET AL 1997;64:1630 4 LUNG TRANSPLANTATION IN SYSTEMIC DISEASE 1631 hypertension, 7 with pulmonary fibrosis), and group 2C included 66 patients without pulmonary hypertension who underwent transplantation for emphysema. The outcomes, including survival, rates of rejection and infection, and pulmonary function, of patients who underwent transplantation for systemic versus isolated disease were compared. Material and Methods Donor Selection and Organ Preservation Details of the procurement procedure have been described elsewhere [2]. Briefly, lungs were considered suitable for transplantation if the donor was less than 55 years old and had no history of pulmonary disease, a normal chest radiograph, adequate gas exchange (ie, arterial oxygen tension 400 mm Hg on an inspired oxygen fraction of 1.0) and negative antibody titers for hepatitis B and C and human immunodeficiency virus. Recipients were matched with donors by body size and ABO blood group. Operation Patients were anesthetized and interfaced with appropriate hemodynamic monitoring devices, including arterial and pulmonary artery catheters. The side transplanted was determined by quantitative ventilation and perfusion of each lung as determined by preoperative evaluation, previous thoracotomy or pleurodesis, and donor availability. Cardiopulmonary bypass was used when intraoperative hypoxemia or hemodynamic instability could not be managed pharmacologically. Telescoping bronchial anastomoses usually were performed. The completed atrial and pulmonary artery anastomoses were examined with intraoperative transesophageal ultrasound for technical complications after weaning from cardiopulmonary bypass and before chest closure. Immunosuppression Postoperative immunosuppression usually was accomplished with triple-drug therapy including azathioprine, corticosteroids, and either cyclosporine or tacrolimus. The cyclosporine dose was titrated to maintain a whole blood level by the Abbott TDx method of 700 to 1,000 ng/dl, and the tacrolimus dose was titrated to maintain a level of 20 to 25 ng/dl. When a patient remained hemodynamically unstable after transplantation or had evidence of significant renal dysfunction early in the postoperative period, induction therapy with antithymocyte globulin was administered for 5 days, instead of cyclosporine or tacrolimus. The dose of azathioprine was titrated to maintain a white blood cell count greater than 5,000/ L. Patients generally were maintained on lowdose corticosteroids (prednisone, 10 to 20 mg/d) after operation. Postoperative Care Transplant recipients were evaluated every 3 months during the first year after transplantation and whenever infection or rejection was suspected. Each evaluation consisted of an interval history, physical examination, chest radiograph, spirometry, arterial blood gases, and bronchoscopy with bronchoalveolar lavage and transbronchial biopsy. Hematologic evaluation included a complete blood count and blood urea nitrogen, creatinine, electrolyte, and cyclosporine or tacrolimus levels. Statistical Analysis Linearized rates of rejection and infection, presented as treated episodes per 100 days, were compared using the two-tailed t-test. Survival comparisons were performed using the log-rank test and Kaplan-Meier survival curves. Forced expiratory capacity in 1 second and forced vital capacity were analyzed using repeated measures of variance. All data are presented as means standard deviation. A p value of less than 0.05 was considered statistically significant. Results Twenty-three patients underwent transplantation for respiratory failure resulting from systemic disease. Group 1 included 15 patients with pulmonary hypertension (9 with sarcoidosis, 6 with scleroderma). Group 2 included 8 patients (5 with LAM, 3 with GVHD) with normal pulmonary artery pressures. Although the average length of follow-up for both groups 1 and 2 tended to be longer than for groups 1C and 2C, this was not statistically different (Table 1). Twenty single-lung transplantations were performed in 19 patients (group 1 12, group 2 8). One patient (with LAM) underwent a second transplantation 18 days postoperatively because of primary graft failure and multiple pulmonary emboli. Two heart-lung and 1 double-lung transplantations were performed in group 1 patients, and 1 double-lung transplantation was performed in a patient with GVHD. In the early postoperative period, mean pulmonary artery pressures in group 1 dropped from 43 17 to 27 11 mm Hg (p 0.05) and were equivalent to those in group 2 (23 4mmHg) (Table 2). Six (50%) of 12 patients in group 1 who underwent single-lung transplantation required cardiopulmonary bypass, compared with no patients in group 2. Mortality The actuarial survival rates for group 1 were 53% (8/15) at 1 year and 47% (7/15) at 2 years. This was similar to group 1C, which had respective rates of 63% (12/19) and 53% (10/19) (Fig 1). The causes of death in group 1 included infection in 6 patients, hemorrhage after transbronchial biopsy in 1 patient, and postoperative multisystem organ failure in 1 patient. The survival rates for group 2 were 75% (6/8) at 1 year and 63% (5/8) at 2 years. Again, these rates were similar to those of group 2C (79% [52/66] and 65% [43/66], respectively) (Fig 2). The cause of death in group 2 was infection in all 3 cases. Four (44%) of the 9 patients who underwent transplantation for sarcoidosis died within 1 year (at 1 week, 2

1632 PIGULA ET AL Ann Thorac Surg LUNG TRANSPLANTATION IN SYSTEMIC DISEASE 1997;64:1630 4 Table 1. Results Parameter Group 1 (n 15) Group 1C (n 17) Group 2 (n 8) Group 2C (n 65) p Value Follow-up (mo) 18.9 15.6 15.4 12.9 21.5 13.1 14.2 10.9 NS Acute cellular rejection (episodes/100 d) 0.58 0.68 a 1.75 1.21 a 0.49 0.44 1.58 1.85 0.05 Obliterative bronchiolitis 13% (2/15) 18% (3/17) 25% (2/8) 18% (12/66) NS Infections (episodes/100 d) 1.7 3.5 0.9 1.5 0.3 0.6 0.53 0.76 NS Bacterial (%) 60 31 29 46 Viral (%) 22 43 14 32 Fungal (%) 18 26 57 22 a p 0.05, group 1 versus group 1C. NS not significant. months, 8 months, and 11 months, respectively). The early death occurred in a patient who received a heartlung transplant early in our transplant experience (1984) and died of multisystem organ failure. The survivors (5/9) remain alive at 6, 21, 23, 34, and 44 months, respectively, after transplantation. Five of 6 patients who underwent transplantation for scleroderma are alive 6 to 60 months after transplantation. One death occurred at 22 days as a result of bacterial pneumonia and pancreatitis. The 3 patients who underwent transplantation for GVHD after bone marrow transplantation are alive at 14, 19, and 40 months, respectively, after transplantation. Three of the 5 patients who underwent transplantation for LAM have died of infection at 2, 25, and 30 months, respectively, after transplantation. Two survivors remain alive at 8 and 34 months, respectively, after transplantation. Infection Six deaths in group 1 were related to infection (40%), as were 3 in group 2 (38%). Episodes of infection were Table 2. Clinical Characteristics of Patients With (Group 1) and Without (Group 2) Pulmonary Hypertension Who Underwent Transplantation for Systemic Disease Characteristic Group 1 Group 2 Sex Male 5 1 Female 10 7 Age (y) 40 6 34 11 Mean pulmonary artery pressure (mm Hg) Before transplantation 43 17 23 4 After transplantation 27 11 24 3 Type of disease Sarcoidosis 9... Scleroderma 6... Lymphangioleiomyomatosis 5 Graft-versus-host disease 3 Type of transplant Single-lung 12 8 Double-lung 1 1 Heart-lung 2 0 linearized to the number of episodes per 100 days and presented as means SD. Although group 1 tended to have more frequent infections than group 2, this did not reach statistical significance (Table 1). Invasive pulmonary aspergillosis was identified in 5 patients in group 1 (4 with sarcoidosis, 1 with scleroderma), and 1 died of a Nocardia brain abscess 11 months after transplantation. Invasive pulmonary Aspergillus with brain abscess was identified at autopsy in 2 patients with LAM in group 2. Three patients in group 1 had gastrointestinal cytomegalovirus infection, and 1 patient in group 2 had cytomegalovirus pneumonia. When the frequency of infection in patients with pulmonary hypertension who underwent transplantation for isolated pulmonary disease (group 1C) was compared with that in patients who underwent transplantation for systemic diseases (group 1), no differences were found (Table 1). Likewise, the frequency of infection was similar among the two groups of patients without pulmonary hypertension (groups 2 and 2C) who underwent transplantation. The relative frequencies of bacterial, viral, and fungal infections are shown in Table 1. Although group 2 had a Fig 1. Kaplan-Meier survival curves for patients with sarcoidosis and scleroderma (group 1, solid line) versus control patients with pulmonary hypertension (group 1C, broken line). One-year (53% and 63%, respectively) and 2-year (47% and 53%, respectively) survival rates were similar in the two groups (p 0.05, log-rank test).

Ann Thorac Surg PIGULA ET AL 1997;64:1630 4 LUNG TRANSPLANTATION IN SYSTEMIC DISEASE 1633 Recurrence Six patients had biopsy-proven recurrence of the primary disease. Four patients with sarcoidosis had granulomas identified in the allograft by transbronchial biopsy. Two patients with LAM had recurrent disease in the allograft identified at autopsy, at 2 months and 30 months, respectively, after transplantation. These patients died of infectious causes (herpes pneumonia and disseminated aspergillosis, respectively). In no case was recurrence suspected clinically, nor did it appear to contribute significantly to mortality. Fig 2. Kaplan-Meier survival curves for patients with LAM and GVHD, (group 2, solid line) versus control patients with emphysema (group 2C, broken line). One-year (75% and 79%, respectively) and 2-year (63% and 65%, respectively) survival rates were similar in the two groups (p 0.05, log-rank test). preponderance of fungal infections and a lower percentage of viral infections, no clear etiologic pattern is apparent. Rejection Patients underwent routine biopsy every 3 months during the first year and whenever rejection was suspected on clinical grounds. Sixty-six percent (10/15) of the patients in group 1 were receiving steroids before transplantation, as were 50% (4/8) of the patients in group 2. Rejection was treated on the basis of tissue obtained by transbronchial biopsy and clinical evaluation. Linearized rates of treated acute cellular rejection per 100 days were similar between the two groups (Table 1). Compared with the patients in group 1C, those in group 1 had a significantly lower frequency of rejection. The patients in group 2 had a lower frequency of rejection than those in group 2C, but this did not reach statistical significance. The incidence of obliterative bronchiolitis, determined on the basis of biopsy results, was similar for all groups (Table 1). Pulmonary Function Early pulmonary function tests (3 months) showed similar results between groups 1 and 1C and groups 2 and 2C (Table 3). Pulmonary function measured at 12 months also was similar and showed no intergroup differences. In general, there tended to be slight, clinically insignificant improvement in most parameters over time. Comment Lung transplantation for the treatment of respiratory failure stemming from systemic diseases remains controversial. The fact that mortality rates approach 20% for patients on the waiting list for lung transplantation emphasizes the need for information regarding the allocation of scarce transplant resources [3]. Survival analysis has shown no differences in 1- and 2-year survival rates between patients who undergo transplantation for the systemic diseases considered here and those who undergo transplantation for isolated pulmonary disease. Clearly, patients without pulmonary hypertension tend to do better, as has been suggested by previous studies [4]. Survival results for all groups are comparable to those obtained in other series [4, 5]. Although there appears to be equivalent survival between groups of patients who undergo transplantation for systemic and isolated diseases, a consideration unique to those with systemic diseases is the possibility of recurrence. The primary systemic disease is not cured by transplantation, and the allograft may be vulnerable to recurrent damage. Evidence of recurrent disease has been demonstrated in 4 (44%) of 9 patients with sarcoidosis and 2 (40%) of 5 patients with LAM. Although this is a concern, we were unable to discern any clinical effect of this phenomenon. Lung transplantation for the treatment of end-stage sarcoidosis specifically was reviewed by Johnson and colleagues in 1993 [6]. Survival, pulmonary function, and the incidence of bronchiolitis obliterans were similar to those of contemporary patients undergoing transplantation for other indications. These investigators found that, although recurrence was documented in 4 patients, no clinical effect of the recurrence could be discerned. Recurrent LAM in pulmonary allografts has been described previously by O Brien and associates [7], and Table 3. Pulmonary Function Test Results a Time of Test FVC (% predicted) FEV 1 (% predicted) 3 Months after transplantation Group 1 (n 11) 60.7 13.3 63.6 14.8 Group 1C (n 16) 69.3 16.5 72.8 23 Group 2 (n 8) 60.9 15.1 55.3 16.9 Group 2C (n 54) 63.2 16.5 55.1 14.1 12 Months after transplantation Group 1 (n 9) 65.7 15.6 60.2 14.7 Group 1C (n 11) 73.6 16.3 68 28.1 Group 2 (n 7) 64 16.7 56.3 9.2 Group 2C (n 40) 69.6 15.5 56.9 13.6 a All p values were nonsignificant. FEV 1 forced expiratory volume in 1 second; capacity. FVC forced vital

1634 PIGULA ET AL Ann Thorac Surg LUNG TRANSPLANTATION IN SYSTEMIC DISEASE 1997;64:1630 4 there is evidence that it stems from cells of donor origin [8]. Recurrence in our 2 patients was noted at autopsy after death resulting from infectious causes, and appears to be unrelated to their clinical outcome. Anecdotal reports of transplantation for scleroderma have emerged [9], but the 5 patients presented here comprise one of the largest series. Although this disease often is responsive to medical treatment, refractory disease with pulmonary involvement carries a 7-year mortality rate of 50% [10]. Transplantation is considered the final treatment option. Graft-versus-host disease is a systemic disease, but pulmonary manifestations often predominate. Pulmonary fibrosis occurs in nearly 20% of these patients, with the mortality rate approaching 50%. In 1992, Calhoon and co-workers [11] reported the first successful lung transplant for GVHD. The 3 patients reported here are alive and have marked improvement in their pulmonary function and minimal infection- and rejection-related complications. Our use of treated episodes of infection or rejection per 100 days reflects the clinical considerations present in the care of transplant recipients. Perceived threats to the allograft or patient prompted a thorough search for infection or rejection. Therapy then was directed at the most likely cause. There were no demonstrable differences in the frequency of infection between patients who underwent transplantation for systemic versus isolated pulmonary disease. In addition, no differences were seen between patients with and without pulmonary hypertension who underwent transplantation. The reason for the lower incidence of rejection in patients with pulmonary hypertension who underwent transplantation for systemic diseases is unclear. Although 56% of the patients in group 1 were treated with steroids before transplantation, so were 50% of the patients in group 1C. The inherent involvement of the immune system in these systemic diseases may contribute. Conversely, patients in group 1 tended to have more episodes of treated infection, suggesting an underlying immunosuppression from the disease itself. Despite the lack of differences in measured parameters, lower levels of induced immunosuppression in these patients should be considered. Patients who undergo transplantation for these systemic diseases have acceptable morbidity and satisfactory survival. However, although these are systemic diseases, pulmonary manifestations have been their predominant clinical feature. This experience, although limited, suggests that lung transplantation need not necessarily be denied to patients with respiratory failure resulting from these systemic diseases. Rather, morbidity and mortality seem to be a reflection of the presence of pulmonary hypertension before transplantation. Clearly, the concern regarding extrapulmonary involvement in systemic disease is important. In considering the transplant candidacy of patients with systemic diseases, an exhaustive pretransplant evaluation directed at collateral organ systems is required. In the absence of significant extrapulmonary organ dysfunction in an otherwise acceptable candidate, lung transplantation can be offered to patients with these systemic diseases. These patients can undergo transplantation with expectations of results comparable to those obtained in patients who undergo transplantation for more traditional, isolated pulmonary diseases. References 1. Al Kattan K, Tadjkarimi S, Cox A, et al. Evaluation of the long-term results of single lung versus heart lung transplantation for emphysema. J Heart Lung Transplant 1995;14: 824 31. 2. Sundaresan S, Trachiotis G, Aoe M, Patterson A, Cooper J. Donor lung procurement: assessment and operative technique. Ann Thorac Surg 1993;56:1409 13. 3. Harper AM, Baker AS. The UNOS OPTN waiting list: 1988 1995. Clinical Transplants 1995:69 84. 4. Bando K, Keenan RJ, Paradis IL, et al. Impact of pulmonary hypertension on outcome after single lung transplantation. Ann Thorac Surg 1994;58:1336 42. 5. Bando K, Paradis IL, Keenan RJ, et al. Comparison of outcomes after single and bilateral lung transplantation for obstructive lung disease. J Heart Lung Transplant 1995;14: 692 8. 6. Johnson BA, Duncan SR, Ohori NP, et al. Recurrence of sarcoidosis in pulmonary allograft recipients. Am Rev Respir Dis 1993;148:1373 7. 7. O Brien JD, Lium JH, Parosa JF, et al. Lymphangiomyomatosis recurrence in the allograft after single lung transplantation. Am J Respir Crit Care Med 1995;151:2033 66. 8. Nine JS, Yousem SA, Paradis IL, et al. Lymphangioleiomyomatosis: recurrence after lung transplantation. J Heart Lung Transplant 1994;13:714 9. 9. Levine SM, Anzueto A, Peters JI, et al. Single lung transplantation in patients with systemic disease. Chest 1994;105: 837 41. 10. Arroliga AC, Podell DN, Matthay RA. Pulmonary manifestations of scleroderma. J Thorac Imaging 1992;7:30 45. 11. Calhoon JH, Levine SM, Anzueto A, Bryan CL, Trinkle JK. Lung transplantation in a patient with a prior bone marrow transplant. Chest 1992;102:948.