Table 2

‚ÄÉHuman studies on the effect of non-steroidal anti-inflammatory drugs (NSAIDS) on bone healing

ReferenceYearPopulationInterventionResultQuality
Adolphson et al231993Double blind randomised study of postmenopausal women with displaced Colles fracturesPiroxicam v placeboNo significant delay in healingA
Butcher and Marsh251996Retrospective analysis of 94 patients with tibial fracturesAny NSAIDsMore likely to have delayed healing with NSAIDsB
Glassman et al301997Retrospective analysis of 288 spinal surgery patientsPostoperative intramuscular ketorolacSignificantly increased risk of non-union after NSAIDsB
Khan261997Retrospective analysis of patients with clavicle fracturesAny NSAIDIncreased rate of non-union with NSAIDsB
Alho et al.271999Survey of 165 elderly patients with femoral neck fracturesStandard treatmentsNo mention of NSAIDsB
Perlman and Thordarson291999Retrospective analysis of 67 ankle fusionsNoneNon-union associated with open trauma, smoking, alcohol, diabetes, and illicit drug use; no mention of NSAIDsB
Giannoudis et al22000Retrospective analysis of 32 patients with non-union of the femoral diaphysis (67 controls)Any, mainly diclofenac and ibuprofenSignificant association between NSAIDs and non-unionB
Van Staa et al92000Retrospective cohort of 500 000 patients taking NSAIDs in primary care with matched controlsAny NSAIDNot supportive of overall effect on bone metabolismB
No differences in fracture risk between regular and irregular NSAID users
Karladani et al282001100 consecutive tibial shaft fracturesNoneIncreased rates of non-union with high energy trauma and with open fractures; no mention of NSAIDsB
Reuben et al142001Retrospective analysis of 106 spinal surgery patientsRofecoxibNo significantly increased risk of non-union compared with studies without NSAIDsB
Clough32002Case report of femoral neck stress fractureNot clear from articleGiven NSAIDs before non-unionB
Burd et al242003112 patients at risk of heterotopic ossification after open reduction and external fixation of acetabular fractureProspectively randomised to radiation therapy v indomethacin v no prophylaxisPatients who received indomethacin had a greater risk of non-union of the long bones than those who either received radiotherapy or no prophylaxisA