Background: As part of the Safety Of non-Steroidal antiinflammatory drugs (SOS) project funded by the European Commission, a systematic review of randomized clinical trials (RCTs) was performed to select the most relevant RCTs in terms of safety of NSAIDs.
Objectives: To calculate the incidence of myocardial infarction (MI), ischemic stroke (IS), hemorrhagic stroke (HS) and heart failure (HF) in NSAIDs users during RCTs.
Methods: Medline, Scopus, ISI web of Science, and Cochrane database of Systematic reviews were searched for RCTs published between 1983 and 2008 for the 10 most sold NSAIDs in Europe. RCTs with no treatment arm over 100 patients-years (P-Ys) were excluded. The incidence was expressed in incidence per 1,000 P-Ys.
Results: Among 1,222 potentially relevant references, 51 RCTs fulfilled inclusion criteria. For MI, celecoxib incidence was 5.5 per 1,000 P-Ys (95% CI, 4.2-6.8), diclofenac 4.4 (3.6-5.1), etoricoxib 4.2 (3.4-4.9), ibuprofen 2.8 (1.4-4.2), indomethacin 8.7 (0.2-48.2), meloxicam 0.0 (0.0-24.1), and naproxen 2.5 (1.6-3.4). For IS, celecoxib incidence was 1.9 per 1,000 P-Ys (95% CI, 0.5-4.8), diclofenac 2.1 (1.6-2.6), etoricoxib 2.0 (1.5-2.5), ibuprofen 2.1 (0.9-3.3), indomethacin 0.0 (0.0-31.9), meloxicam 0.0 (0.0-24.1), and naproxen 3.0 (1.6-4.4). For HS, celecoxib incidence was 2.0 per 1,000 P-Ys (95% CI, 1.1-3.0), diclofenac 0.7 (0.0-4.0), ibuprofen 0.5 (0.1-1.5), indomethacin 0 (0.0-31.9), meloxicam 0.0 (0.0-24.1), and naproxen 0.2 (0.0-1.1). For HF, celecoxib incidence was 2.7 per 1,000 P-Ys (95% CI, 1.8-3.6), diclofenac 2.4 (1.3-3.4), etoricoxib 0.6 (0.1-1.2), ibuprofen 4.2 (2.5-5.9), indomethacin 0 (0.0-31.9), meloxicam 0.0 (0.0-7.2), and naproxen 4.1 (2.7-5.5).
Conclusions: Potential differences in the incidence of the studied CV events were observed between the NSAIDs for which sufficent long term data from RCTs are available.
