Purpose: Non-steroidal anti-inflammatory drugs (NSAIDs) are powerful analgesics, frequently used for post-operative pain control. However, concerns regarding the potential deleterious effects of NSAIDs on bone healing have compelled many physicians to avoid NSAIDs in patients with fractures, osteotomies, and fusions. The purpose of this study was to systematically review and meta-analyze the best clinical evidence regarding the effects of NSAIDs on bone healing.

Method: We performed a literature search for studies of fracture, osteotomy or fusion patients with NSAID exposure, and non-union as an outcome. Data on study design, patient characteristics and risk estimates were extracted. Pooled effect estimates were calculated. Study inclusion results were checked for evidence of publication bias. Metaregressions were performed to assess the impact of age, smoking, and study quality on reported risk of non-union.

Results: Seven spine fusion and four long-bone fracture studies were included. A significant association between lower quality studies and higher reported odds ratios for non-union was identified. When only higher quality studies were considered, seven spine fusion studies were analyzed, and no statistically significant association between NSAID exposure and non-union was identified (OR=2.2, 95%CI:0.8, 6.3). No statistically significant association was found in sub-analysis of patients exposed to high dose IV/IM ketorolac (OR=2.0, 95%CI:0.4, 11.1), low dose IV/IM ketorolac (OR=1.2 95%CI:0.3, 4.5), or standard oral NSAIDs (OR=7.1, 95%CI:0.1, 520). In sub-analysis of the four most clinically relevant studies of adult spine fusion patients with well defined peri-operative NSAID exposure, no statistically significant association was found between NSAID exposure and risk of non-union (OR=0.8 95%CI:0.4, 1.4).

Conclusion: Studies on NSAID exposure in long-bone healing settings were of lesser quality than studies in the spine fusion setting. Within the spine literature we could not demonstrate any increased risk of non-union with NSAID exposure. Randomized controlled trials (and meta-analyses of such trials) on the impact of standard NSAID and COX-2 inhibitor exposure in spine and long-bone fracture, fusion and osteotomy populations are warranted to confirm or refute the findings of this meta-analysis of observational studies.