Objective: The aim of this study was to assess the prevalence of cytomegalovirus (CMV) viremia in HIV-positive patients starting antiretroviral therapy (ART) and to evaluate its impact on clinical outcomes. Design: A retrospective analysis of four clinical trials (INSIGHT FIRST, SMART, START, and ANRS REFLATE TB). Methods: Stored plasma samples from participants were used to measure CMV viremia at baseline prior to initiating ART and at visits through 1 year of follow-up after ART initiation. CMV viremia was measured centrally using a quantitative PCR assay. Within FIRST, associations of CMV viremia at baseline and through 8 months of ART were examined with a composite clinical outcome of AIDS, serious non-AIDS events, or death using Cox proportional hazards regression. Results: Samples from a total of 3176 participants, 1169 from FIRST, 137 from ANRS REFLATE TB, 54 from SMART, and 1816 from START were available with baseline CMV viremia prevalence of 17, 26, 0, and 1%, respectively. Pooled across trials, baseline CMV viremia was associated with low CD4₊T-cell counts and high HIV RNA levels. In FIRST, CMV viremia was detected in only 5% of participants between baseline and month 8. After adjustment for CD4₊T-cell count and HIV RNA levels, hazard ratios for risk of clinical outcomes was 1.15 (0.86-1.54) and 2.58 (1.68-3.98) in FIRST participants with baseline and follow-up CMV viremia, respectively. Conclusion: Baseline CMV viremia in HIV-positive patients starting ART is associated with advanced infection and only persistent CMV viremia after ART initiation is associated with a higher risk of morbidity and mortality.