NSAID use may be associated with increased VTE risk

There may be a statistically significant increased risk of venous thromboembolism (VTE) among users of nonsteroidal anti-inflammatory drugs (NSAIDs), a meta-analysis found.

Researchers conducted a systematic review and meta-analysis of 6 studies (1 cohort, 5 case-control) with 21,401 VTE events studies that reported odds ratios, relative risks, hazard ratios, or standardized incidence ratios for VTE among NSAID users compared with non-users. Results appeared online Sept. 24 in Rheumatology.

The pooled risk ratio of VTE in NSAID users was 1.80 (95% CI, 1.28 to 2.52; I², 95%) compared to non-users. The VTE risk among selective COX-2 inhibitor users (3 studies) was also significantly elevated, with a pooled risk ratio of 1.99 (95% CI, 1.44 to 2.75).

Because of the high heterogeneity, researchers conducted a jackknife sensitivity analysis by excluding 1 study at a time. The pooled risk ratios remained significantly elevated, ranging from 1.62 to 2.21, while the corresponding confidence intervals remained greater than 1.

The researchers suspect that differences in study design, definitions of NSAID exposure, and population were the main sources of heterogeneity. Other potential sources included the following.

• 3 studies were done in hospitalized subjects; 3 studies included both ambulatory and inpatient settings.
• 1 study included only patients with pulmonary embolism; the rest included patients with DVT and/or PE.
• 4 studies used a pharmacology-linked database; 2 used interviews.
• Controlling for confounders was done differently in different studies, from virtually no correction to control for a large number of confounders.

The researchers noted that, given the prevalence of NSAID use in the general population, their findings have important public health implications. They wrote, “Physicians should be aware of this association and NSAIDs should be prescribed with caution, especially in patients at high baseline risk of VTE.”