This is a phase II, open label, randomized-controlled pilot study designed to study both the efficacy and safety of salsalate in decreasing endothelial cell dysfunction, systemic inflammation, and insulin resistance in HIV-infected adults. The investigators hypothesis is that salsalate will reduce inflammation and therefore endothelial cell activation and insulin resistance. The sample size will be 40, with an equal number of people being randomized to one of two groups. The first arm will be randomized to salsalate therapy. The second arm will act as a control group. The study duration will be 13 weeks.

Inclusion Criteria:

1. 18 years of age or older
2. HIV-infected
3. Evidence of durable virologic suppression, i.e., must have HIV-1 RNA < 400 copies/ml at study entry and for at least 12 weeks prior to entry
4. On a stable antiretroviral (ARV) regimen, i.e., on the same ARV for at least 12 weeks prior to study entry
5. No intention to stop or modify ARV regimen during the study period

Exclusion Criteria:

1. Current pregnancy or breast feeding, or women of child bearing age who refuse or are unable to use appropriate methods for contraception during the study period
2. Any of the following conditions: diabetes (2 fasting glucose levels > 126 mg/dL or confirmed random glucose level > 200), creatinine clearance < 50, aspirin-sensitive asthma, COPD, history of bleeding gastric or duodenal ulcer, hepatic dysfunction, active hepatitis B or C, and any active infectious or inflammatory condition
3. Need for regular use of any of the following medications: salsalate, aspirin, non-steroidal antiinflammatories (NSAIDS), corticosteroids, warfarin or other anticoagulation therapy, phenytoin, valproic acid, carbonic anhydrase inhibitors, lithium, probenecid or sulfinpyrazone
4. Consumption of alcohol on a daily basis
5. Active use of illicit drugs
6. Unable to attend follow-up appointments
7. Allergy to any salicylic acid-containing medication or salsalate
8. AST or ALT > 2 upper limit of normal (ULN) within 6 months prior to study entry

Lack of placebo in the control group precluded blinding of the study investigators. Adherence was measured utilizing pill counts which assumes that pills not returned were taken by the participant. Sample size too small to detect a small effect.