Antiretroviral therapy to prevent HIV transmission in serodiscordant couples in China (2003-11) : A national observational cohort study

Background
On the basis of the results of the randomised clinical trial HPTN 052 and observational studies, WHO has recommended that antiretroviral therapy be offered to all HIV-infected individuals with uninfected partners of the opposite sex (serodiscordant couples) to reduce the risk of transmission. Whether or not such a public health approach is feasible and the outcomes are sustainable at a large scale and in a developing country setting has not previously been assessed.

Methods
In this retrospective observational cohort study, we included treated and treatment-naive HIV-positive individuals with HIV-negative partners of the opposite sex who had been added to the national HIV epidemiology and treatment databases between Jan 1, 2003 and Dec 31, 2011. We analysed the annual rate of HIV infection in HIV-negative partners during follow-up, stratified by treatment status of the index partner. Cox proportional hazards analyses were done to examine factors related to HIV transmission.

Findings
Based on data from 38 862 serodiscordant couples, with 101 295·1 person-years of follow-up for the seronegative partners, rates of HIV infection were 2·6 per 100 person-years (95% CI 2·4–2·8) among the 14 805 couples in the treatment-naive cohort (median baseline CD4 count for HIV-positive partners 441 cells per μl [IQR 314–590]) and 1·3 per 100 person-years (1·2–1·3) among the 24 057 couples in the treated cohort (median baseline CD4 count for HIV-positive partners 168 cells per μl [62–269]). We calculated a 26% relative reduction in HIV transmission (adjusted hazard ratio 0·74, 95% CI 0·65–0·84) in the treated cohort. The reduction in transmission was seen across almost all demographic subgroups and was significant in the first year (0·64, 0·54–0·76), and among couples in which the HIV-positive partner had been infected by blood or plasma transfusion (0·76, 0·59–0·99) or heterosexual intercourse (0·69, 0·56–0·84), but not among couples in which the HIV-positive partner was infected by injecting drugs (0·98, 0·71–1·36).

Interpretation
Antiretroviral therapy for HIV-positive individuals in serodiscordant couples reduced HIV transmission across China, which suggests that the treatment-as-prevention approach is a feasible public health prevention strategy on a national scale in a developing country context. The durability and generalisability of such protection, however, needs to be further studied.

Funding
Chinese Government's 12th Five-Year Plan, the National Natural Science Foundation of China, and the Canadian International Development Research Centre.