All statistical tests were performed using two-tailed p-values (P < 0.05) except for meta-regression where we considered P < 0.10 to detect potential heterogeneity among variables. Publication bias was assessed using Egger's method [37]. Analyses were conducted in stata (version 10; STATA Corporation, College Station, Texas, USA). The search strategy initially resulted in 5408 articles. We identified 418 for detailed

review. After reviewing the titles and abstracts in detail, we excluded 385 studies that were not relevant to CVD with HIV. Of 33 articles selected for potential eligibility, 10 were excluded as they were unrelated to our study question. We also searched conference proceedings of the Conference on Retroviruses and Opportunistic Infections (CROI) and International AIDS Society until 2010, and five out of 509 abstracts were selected [10, 11, 14, 15, 17]. A total of 23 studies were included, Protein Tyrosine Kinase inhibitor of which 21 were observational studies and two were randomized trials. Details of the search strategies and exclusion process are provided in Figure 1. Of the 23 studies included in our analysis, three were cross-sectional studies, two were case–control studies, 16 were cohort studies and two were randomized controlled trials. These studies recruited PLHIV and HIV-uninfected people with an average follow-up of 5 years. The studies varied greatly with respect to various ART combinations used as comparator. Three

studies recruited click here PLHIV who were not ART-experienced and HIV-uninfected people and compared the RR of CVD events. Three studies compared PLHIV treated with ART with HIV-uninfected Etofibrate people. Five studies compared PLHIV treated with ART with PLHIV without any treatment. Each of the identified studies was internally age-matched; the median age of the study populations was 40 (range 34–46) years. Table 1 gives the study characteristics in detail. Three identified studies reported the risk of CVD for PLHIV [19, 24, 27]. Lang et al. [19] compared 74958 PLHIV in France based on the France Hospital Database on HIV (FHDH) with uninfected people aged from 35 to 64 years. The estimated

age- and sex-standardized RR of MI was 1.50 (95% CI 1.3, 1.7). Obel et al. [24] reported the RR of IHD for 3953 PLHIV compared with 373 856 control subjects to be 1.39 (95% CI 0.82, 2.36) and 2.12 (95% CI 1.62, 2.76) for the pre-highly active antiretroviral therapy (HAART) and HAART eras, respectively. This study was based in Denmark and the study population consisted of adults older than 16 years of age; both HIV-infected subjects and control subjects were well matched in terms of distributions of age, sex, emigration, loss to follow-up and comorbidities. Another study, conducted in the USA by Triant et al. [27], compared 3851 PLHIV and 1 044 589 HIV-uninfected people and estimated the RR of acute MI to be 1.75 (95% CI 1.51, 2.02). This study compared PLHIV with the control group where the study populations were aged 18 years or older.