BACKGROUND: Co-infection with Mycobacterium tuberculosis remains a leading cause of morbidity and mortality among HIV infected individuals especially in developing countries. Early initiation of cART in these patients when CD4+ T cell count is less than 200cells/mm(3) has reduced disease progression and mortality. However for patients with higher CD4+ T cell counts greater than 350cells/mm(3) evidence is conflicting. In this study we seek to evaluate the effectiveness of cART in reducing mortality among TB-HIV co-infected patients with CD4â+âT cells above 350cells/mm(3) at the time of TB diagnosis. METHOD: In a retrospective cohort study we analyzed 337 HIV-TB co-infected patients with CD4+ T cells above 350cells/mm(3) at baseline who were diagnosed between 2006 and 2012 in the southern province of Zambia. The primary outcome was all-cause mortality. We estimated the effect of cART by comparing survival according to cART and controlling for differential loss to follow-up. RESULTS: Of the 257 patients on cART, 22 died (9Â %) and 20 (8Â %) were lost to follow-up; of 80 patients not on cART, 20 died (25Â %) and 19 (24Â %) were lost to follow-up. Patients treated with cART had better survival compared to those not treated (Pâ<â0âÂ·â0001, log-rank test). In a proportional hazard regression adjusting for Cotrimoxazole, the risk of death was reduced by 78Â % with cART (95Â % CI: 0âÂ·â47, 0âÂ·â91). In a propensity score analysis, the effect of cART was still beneficial. CONCLUSION: In patients with HIV-associated TB and CD4+ T cells above 350cells/mm(3), treatment with cART reduced mortality for up to 4Â years as compared to no cART and was associated with better retention in care.