The effect of NSAIDs on postfracture bone healing: a meta-analysis of randomized controlled trials

To determine whether nonsteroidal anti-inflammatory drugs (NSAIDs) have an adverse effect on bone healing by evaluating all available human randomized controlled trials (RCTs) on this subject. DATA SOURCES: A comprehensive search of electronic databases (PubMed, MEDLINE, and Cross-References) until October 2018 comparing the occurrence of nonunion in patients who received NSAIDs to the control group through RCTs. STUDY SELECTION: Inclusion criteria were English-only studies, and the type of studies was restricted to RCTs. DATA EXTRACTION: Two authors independently extracted data from the selected studies, and the data collected were compared to verify agreement. DATA SYNTHESIS: Nonunion was the main outcome evaluated in each study. Regression analysis was used to estimate the relative risk comparing the duration and the type of NSAIDs by calculating the odds ratio (OR) for dichotomous variables. Studies were weighed by the inverse of the variance of the outcome, and a fixed-effects model was used for all analyses. CONCLUSIONS: Six RCTs (609 patients) were included. The risk of nonunion was higher in the patients who were given NSAIDs after the fracture with an OR of 3.47. However, once the studies were categorized into the duration of treatment with NSAIDs, those who received NSAIDs for a short period (<2 weeks) did not show any significant risk of nonunion compared to those who received NSAIDs for a long period (>4 weeks). Indomethacin was associated with a significant higher nonunion rate and OR ranging from 1.66 to 9.03 compared with other NSAIDs that did not show a significant nonunion risk.