Treatment initiation with zidovudine-containing potent antiretroviral therapy impairs CD4 cell count recovery but not clinical efficacy

Objective: Zidovudine-containing antiretroviral therapy has been associated with a lower rise in absolute CD4 cell counts in several randomized trials. We examined the predictive factors for this phenomenon and assessed its impact on clinical progression during treatment in a large patient cohort. Design: An analysis of data from the Swiss HIV Cohort Study. Methods: All 2177 treatment-naive adults who began potent antiretroviral therapy (ART) between September 1995 and September 2004 were included. Exclusion criteria were previous ART and treatment duration of less than 3 months. Follow-up was censored in the case of a treatment switch or stop. Results: A total of 1312 patients initiated zidovudine-containing ART and 865 started ART without zidovudine. Except for slightly higher absolute CD4 cell counts in the zidovudine group, prognostic characteristics at baseline and viral suppression during treatment did not differ. During an observation time of 2343 and 1486 patient-years, the CD4 cell count increased by a median of 221 versus 286 cells/mu l at 2 years and 290 versus 379 cells/mu l at 4 years in the zidovudine versus no zidovudine group; however, the rise in the percentage of CD4 cells was similar in both groups. The zidovudine group had a significantly slower rise in total lymphocytes and haemoglobin. In multivariable Cox models, the hazard for new HIV-associated clinical events was not affected by zidovudine-containing ART. Conclusion: Over 4 years, zidovudine led to a smaller increase in absolute, but not percentage, CD4 cell counts. The effect can be explained as a slower rise in total lymphocytes and has no impact on clinical efficacy. (C) 2007 Lippincott Williams & Wilkins.