Background:  Solid organ transplantation is increasingly available to HIV-infected patients. Stable CD4⁺ T-cell counts and few opportunistic infections have been reported. Immunosuppression or drug interactions may make HIV RNA control difficult. Conversely, antiviral or immune modulating immunosuppressive effects may enhance control.

Methods:  Prospective, multi-site cohort study of HIV-infected liver and kidney recipients. Pre-transplant HIV RNA was undetectable (liver and kidney) or complete suppression was predicted (liver). ART and immunosuppression choices were individualized. We calculated incidence (95% CI) and median (IQR) peak HIV RNA and duration of detectability. Episodes were classified as single detectable RNA values that resolved (blip), occurred off ART, or resulted in ART change. Predictors of first detectable RNA were evaluated in univariate and multivariate Cox proportional hazards models.

Results:  Of 104 subjects, 38 episodes of detectable HIV RNA occurred in 27 (26%) (52 kidney and 52 liver) transplanted between November 2003 and September 2006. Only 8 (8%) subjects had >1 episode. The 1- and 2-year cumulative incidence was 27% (14%, 41%) and 36% (19%, 52%) for kidney, and 34% (19%, 50%) for liver recipients. Median peak HIV RNA was 291 copies/mL (76 to 2700) for kidney and 1800 copies/mL (454 to 14,500) for liver. Median duration of detection was 0 (0 to 129) days for kidney, and 12 (0 to 217) for liver. Blips occurred in only 9 subjects (33%) (7 kidney, 2 liver); 8 episodes (21%; 1 kidney, 7 liver) occurred while the subject was not on ART and 5 (13%, 2 kidney and 3 liver) resulted in ART change with subsequent suppression of HIV RNA. In univariate models, subjects not on ART (hazard ratio [HR] 4.2, CI 1.7 to 10.7) were at increased risk of detectable HIV RNA; increasing number of ART (HR 0.6, CI 0.5 to 0.8) and non-nucleoside reverse transcriptase inhibitor (NNRTI) -based regimens (HR 0.36, CI 0.14 to 0.96) were protective. Age, race, nadir, and most recent CD4⁺ T-cell count, transplanted organ, hepatitis C, baseline ART experience and HIV RNA, time to ART initiation post-transplant, total number of medications, immunosuppression type, organ function and hospitalization, rejection or infection in prior 30 days were not associated with detectable HIV RNA. In multivariate models, total number of ART adjusted for being off ART and NNRTI-based regimens remained significant (HR 0.4, CI 0.2 to 0.9).

Conclusions:  HIV RNA is well-controlled in liver and kidney transplant recipients despite complex drug interactions and multiple medications and co-morbidities. There are no unusual predictors of virologic breakthrough in this population.