Role of protease inhibitors in preventing recurrent oral candidosis in patients with HIV-infection: A prospective case-control study

This study was conducted to evaluate the efficacy of highly active antiretr oviral therapy (HAART) in preventing recurrence of oral candidosis (OC) ass ociated with HIV. A prospective case-controlled observational study was per formed in an inner-city university-hospital HIV/AIDS clinic. Ninety-three H IV-positive study subjects with a history of recurrent OC were divided into two groups: protease inhibitors (PI)-treated patients (group 1, n = 30) an d non-PI-treated patients (group 2, n = 63). Study subjects were matched fo r sex, age, stage of HIV infection, and peripheral CD4(+) T-cell counts. Th e non-PI-treated group was further subdivided into the following three subg roups: HIV-positive study subjects treated with reverse transcriptase inhib itors (RTI; groups 2a and 2c) and HIV-positive study subjects not treated w ith RTIs (group 2b). Group 2c met the same inclusion criteria as group 2a h ad but was matched 6 months after the beginning of the study. We also asses sed in vitro peripheral blood mononuclear cells (PBMC) and their lymphoprol iferative response, as well as cutaneous delayed-type hypersensitivity (DTH ) response to Candida-associated antigens in a randomly selected sample of study subjects divided into those treated with PIs and those who were not. During a 1-year follow-up, OC was diagnosed in 2 (7%) PI-treated and 23 (36 %) non-PI-treated patients (p <.001). In addition to comparing findings in group 1 with those in group 2c, OC was detected in 14 (50%) non-PI-treated patients compared with no HAART-treated study subjects (p <.001). Only 41% of PI-treated study subjects had positive lymphoproliferative response in P BMCs and none was positive in terms of DTH to Candida antigens (p = not sig nificant versus non-PI-treated study subjects). While objectively demonstra ting a beneficial effect of HAART in preventing recurrence of OC infections , our findings suggest this effect cannot be not fully accounted for by rec onstitution of anti-Candida cell-mediated immunity, given that other mechan isms, even of a nonimmune nature, could have some effect.