Solid organ transplantations have been performed successfully in selected HIV-positive patients with highly active antiretrovirus therapy (HAART). However, some of the medications in the HAART regimen require metabolism via the cytochrome P4503A, the same enzyme complex responsible for clearance of the calcineurin inhibitors cyclosporine and tacrolimus. Several case reports have described significant interactions between the agents used in HAART and immunosuppressive drugs. The goal of this report is to examine the extent of potential drug interactions between antiretroviral agents and tacrolimus after liver and kidney transplantation. Seven liver transplant (LTx) patients (M = 6, F = 1) and four kidney transplant (KTx) patients (M = 4) infected with HIV underwent surgery between September 1997 and January 2001. Initial immunosuppression consisted of tacrolimus and steroids for LTx patients or tacrolimus, steroids, and mycophenolate mofetil for KTx recipients. Their current baseline immunosuppression and HAART regimen were examined retrospectively. Of the seven liver recipients, one (case 4) died 2 weeks after LTx and never received HAART therapy posttransplantation. The remaining six patients were placed on a regimen consisting of two nucleoside reverse transcriptase inhibitors (NRTI) and one protease inhibitor (PI) (nelfinavir in 5, indinavir in 1) based on known viral sensitivities or history of a previous clinical response. Kidney recipients received NRTI and non-nucleoside reverse transcriptase inhibitors (NNRTI). The mean dose of tacrolimus in liver recipients was 0.6 mg/d, with mean trough concentration of 9.7 mg/mL. Compared with historic controls (liver transplant patients not on HAART), the average tacrolimus dose was 16-fold lower in patients on HAART. In contrast to liver recipients, HIV-positive kidney recipients not on PI therapy required a mean tacrolimus dose of 9.5 mg/d to maintain a mean trough concentration of 9.6 ng/mL. Of the two protease inhibitors used, nelfinavir seems to have a more profound effect than indinavir. When patients on nelfinavir alone (n = 5) were compared with a control group not on antiretroviral therapy, the need for a tacrolimus dose was 38 times lower (mean dose, 0.26 mg/d). Profound drug interactions between PI and tacrolimus have been observed requiring up to 50-fold reductions in dosage. This effect seems to be most pronounced with the use of nelfinavir as opposed to indinavir, although further experience is required to confirm this observation. In contrast, HAART using NRTI and NNRTI without the use of PI, as shown in kidney recipients, produces less significant effects on tacrolimus metabolism. Great caution and frequent drug level monitoring are necessary when HAART is introduced or withdrawn in HIV-positive recipients of organ transplants.