Effectiveness of Aerobic Exercise in Adults Living with HIV/AIDS: Systematic Review

Kelly O'Brien; Stephanie Nixon; Anne-Marie Tynan; Richard H. Glazier


Med Sci Sports Exerc. 2004;36(10) 

In This Article


We performed a systematic review and meta-analysis using methods of the Cochrane Collaboration.[3] We searched electronic databases for articles published between 1980 to November 2002 (MEDLINE, EMBASE, SCIENCE CITATION INDEX, AIDSLINE, CINAHL, HEALTHSTAR, PSYCHLIT, SOCIOFILE, SCI, SSCI, ERIC, DAI, and Cochrane Collaborative Review Group databases) using subject headings such as HIV, HIV infections, and exercise. We also reviewed abstracts from international and national AIDS conferences, searched reference lists from pertinent articles and books, made personal contact with authors, and hand searched targeted journals to identify potential studies for inclusion. All languages were included.

Titles and abstracts of all citations were reviewed independently by two reviewers to identify studies which met the following four inclusion criteria and included: 1) human participants who were HIV positive, 2) participants 18 yr of age or older, 3) an aerobic exercise intervention performed at least three times per week for at least 4 wk, and 4) a randomized comparison group. Two reviewers reviewed hard copies of an entire paper independently if one or both raters believed a study met eligibility criteria. Three of four possible reviewers examined full text to determine final inclusion. Disagreements were resolved through discussion and consensus.

Two reviewers (out of eight possible reviewers) abstracted relevant data from included studies onto standard data abstraction forms. Methodological quality of the studies was assessed using criteria developed by Jadad et al.[5] We also assessed whether the groups were similar at baseline. Our outcome measures included immunological/virological indicators (CD4 count, viral load), cardiopulmonary measures ( O2max), psychological measures, and adverse events including death.

We used RevMan (Version 4.1) software to perform statistical analyses. Where there were sufficient data available from the authors, and comparisons made practical sense, and in the absence of statistical heterogeneity (P < 0.05), meta-analyses were performed. For continuous variables, we used random effects models to calculate the weighted mean difference (WMD) and 95% confidence intervals. None of the outcomes were dichotomous variables.

Subgroups identified for separate analyses included: interval versus constant aerobic exercise and moderate versus heavy intensity aerobic exercise.

For the purposes of this review, we considered 50 cells·mm-3 to indicate a clinically important change in CD4 count, 0.5 log10 copies to indicate a clinically important change in viral load, and 2 mL·kg-1·min-1 to indicate a clinically important change in O2max. These values were based on extensive consultation with the clinical and research community, and are consistent with values used in previous literature.[8,17,18] We considered a P value of less than 0.05 as statistically significant.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.