Adherence to antiretroviral therapy and CD4 T-cell count responses among HIV-infected injection drug users

OBJECTIVE: To evaluate the time to CD4 cell count response (> or = 50 cells/mm3) among patients initiating highly active antiretroviral therapy (HAART) with and without a history of injection drug use, and to examine the potential role of non-adherence to HAART on differential CD4 responses. METHODS: Population-based analysis of treatment-naive patients initiating HAART during the period 1 August 1996 to 31 July 2000 and who were followed until 31 March 2002. Patients were stratified based on 95% adherence and history of injection drug use, and Kaplan-Meier methods and Cox regression were used to evaluate CD4 response rates and factors associated with CD4 responses. RESULTS: Overall, the CD4 cell count response rate was slower among injection drug users in Kaplan-Meier analyses (log-rank: P<0.05). However, no differences existed when the analyses were restricted to adherent patients (log-rank: P=0.349). Similarly, the differences in the time to CD4 cell count response observed in univariate Cox regression analyses for patients with a history of injection drug use [relative hazard: 0.85 (95% CI: 0.75-0.97)] diminished after adjustment for adherence [adjusted relative hazard: 1.02 (95% CI: 0.89-1.16)]. CONCLUSION: These data demonstrate the importance of adherence on CD4 cell count responses and highlight the need for interventions to improve antiretroviral adherence among injection drug user.