Redo lung transplantation: a North American-European experience.

An international survey of redo lung transplantation was performed to identify the morbidity and mortality rates and factors correlating with increased or decreased survival after this procedure. Twenty institutions in North America and Europe participated, and the study cohort included 61 patients who underwent 63 redo lung transplantation operations. Patients undergoing a redo heart-lung transplantation were excluded. The indications for reoperation included obliterative bronchiolitis (32 patients), graft failure (14 patients), intractable airway problems (8 patients), severe acute lung rejection (5 patients), and miscellaneous complications (4 patients). Five types of retransplantation procedures were performed, including redo ipsilateral single lung transplantation (24 patients), redo contralateral single lung transplantation (11 patients), single lung transplantation after double lung or heart-lung transplantation (13 patients), redo double lung transplantation (8 patients), and double lung transplantation after a previous single lung transplantation (7 patients). Actuarial survival was 65%, 49%, 42%, 35%, and 32% at 1, 3, 6, 12, and 24 months, respectively; survival was significantly (p < 0.05) worse than that of first-time lung transplant recipients recorded in the International Society for Heart and Lung Transplantation Registry. Actuarial survival did not differ according to the original diagnosis of the recipients, the indication for reoperation, or the type of retransplantation procedure performed. Similarly, recipient cytomegalovirus status and ventilator status before reoperation did not affect postoperative survival. Trends toward an improved outcome were noted in patients who were ambulatory before reoperation and in those receiving an ABO identical, as opposed to ABO compatible, graft at reoperation. Life table and step-wise logistic regression analysis identified donor cytomegalovirus status at reoperation to be an important determinant of outcome, with significantly (p < 0.05) improved survival in the donor cytomegalovirus-negative group. Polymicrobial infection was the most common cause of death at all time intervals after reoperation. The presence of disseminated infection and established multiorgan failure was almost uniformly associated with a fatal outcome. We conclude that redo lung transplantation may be indicated only in well-selected patients with obliterative bronchiolitis, severe airway complications, or graft failure. Donor cytomegalovirus status at reoperation is an important predictor of survival. The presence of disseminated infection and established multiorgan failure should be contraindications to lung retransplantation.