Active CMV infection before lung transplantation: risk factors and clinical implications

Cytomegalovirus (CMV) infection is a major cause of morbidity following lung transplantation, but active CMV infection has not been described before transplantation. Since 1990, we have screened all lung-transplant recipients for CMV infection with viral urine cultures on the day of transplantation. We retrospectively reviewed the medical records of all 102 lung-allograft recipients transplanted between March 1990 and September 1998. Patients with positive urine cultures for CMV were compared to culture negative patients for age, gender, pretransplant diagnosis, time from diagnosis to transplantation, CMV serostatus, use of pretransplant immunosuppression, T-lymphocyte subsets, and presence of fever. Posttransplant outcomes assessed were duration of intubation and hospitalization, acute rejection, frequency of CMV disease, duration of Nashville rabbit antithymocyte serum or globulin (N-RATS/G) and ganciclovir, and survival. Five (5%) of 102 patients had positive urine cultures for CMV; none had symptoms of CMV infection. All 5 had idiopathic pulmonary fibrosis (IPF) (5/5 vs 27/97; p = 0.002). The age, gender, and CMV serostatus of these patients did not differ from the 97 patients in the culture negative group. Four (80%) of the 5 patients with positive cultures were receiving treatment with azathioprine or cyclophosphamide vs only 18 (19%) of the 97 patients with negative cultures (p = 0.007), and all 5 (100%) were receiving steroids compared to 50 (52%) of 97 patients with negative cultures (p = 0.06). Culture-positive IPF patients, when compared with the 27 culture-negative IPF patients, did not differ in any demographic variable or in the use of immunosuppression, but culture-positive patients were more likely to have a CD4/CD8 T-cell subset ratio <1.0 (p = 0.02). Following transplantation, 3 (60%) of 5 IPF patients with positive CMV cultures developed CMV disease compared to 3 (11%) of 27 IPF patients with negative cultures (p = 0.03). Patients with positive cultures also received more days of parenteral antiviral therapy (mean 44 +/- 11 days vs 16 +/- 10 days; p < 0.001). Utilizing pretransplant screening, we have discovered that 16% of patients with IPF had active CMV infection, which was associated with both alterations in their T-cell subsets and a greater risk for CMV disease after transplantation. This occurrence of occult CMV infection in patients with IPF has not been previously recognized, and has important implications.