Optimal starting point for antiretroviral HIV treatment in a town in Cameroon: a randomised controlled study

BACKGROUND: Optimal starting point for antiretroviral treatment (ART) has been uncertain. METHODS: Parallel group, single blind, randomised controlled study of adult HIV positive patients consulting at the Protestant Hospital, Ngaoundere, Cameroon in 2007-8. Simple randomisation of patients in WHO clinical stage 1-2 to start of ART early or deferred, i.e. when CD4 counts dropped below 350 versus 250 cells/mm(3), or when they reached clinical stage 3-4. Clinical follow-up every three months were offered for all patients. Main outcomes were clinical stage, CD4 differences and mortality. Of 424 consulting patients, most were excluded, mainly because they were already in WHO stage 3-4. Forty-four patients were randomised. RESULTS: In the ‘early’ group two patients died and five were lost to follow-up. In the ‘deferred’ group, six patients died and nine were lost to follow-up (Hazard ratio for death by early compared to deferred treatment 0.26, 95% confidence interval 0.05-1.29). Of the patients lost to follow-up, three patients in the ‘early’ group and four patients in the ‘deferred’ group were known to be alive when the study ended. Fourteen patients in the early group and 11 in the deferred group started ART. Twenty-two patients were evaluated clinically six to seven months after the study period was terminated. Except for one patient with AIDS, these were all still in clinical stage 1-2. CONCLUSIONS: In our small sample, relative risk for death did not differ significantly, but deferred treatment seemed to carry no increased survival or other clinical advantage. During the study period, other studies made WHO change its guidelines to conform to our early treatment. The tendency in our study lends support to this policy. TRIAL REGISTRATION: ISRCTN22114173 ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/1471-2458-14-828) contains supplementary material, which is available to authorized users.