PICO 1.1 – impact of NSAIDs on non-union
| Author | Non-union definition | Groups | N | Male, n (%) | Age, year, mean (SD) | Non-union, n (%) | Lost to follow-up, n (%) | Study conclusions |
| Hunter22 | Persistent fracture at 24 weeks | ASA No ASA | 152 354 | 66 (43) 163 (46) | 44 (NA) 42 (NA) | 3 (3) 2 (1) | 54 (34) 154 (41) | ASA for 6 to 8 weeks postoperatively does not influence time to union |
| Moed21 | Not defined | Indomethacin No indomethacin | 16 19 | NA NA | 32 (NA) 32 (NA) | 0 (0) 0 (0) | NA NA | No problems with fracture healing noted |
| Buchheit18 | Not defined | NSAID Acute NSAID Chronic NSAID No NSAID | 2525 2180 527 306 579 | NA NA | NA NA | 227 (8.3) 183 (8.4) 44 (7.7) 15 022 (4.9) | 0% at 12 months (only complete cases included) | Numerous medications are associated with non-union including antibiotics, anticoagulants, bisphosphonates, opioids and NSAIDs. Non-union depends on fracture location. |
| Sagi15 | Lack of bridging callus and visible fracture line | Indomethacin 3 days 1 week 6 weeks Placebo | 72 24 25 23 26 | (70%) (76%) (80%) (67%) | 41 (NA) 34 (NA) 42 (NA) 46 (NA) | 18 (25) 6 (35) 4 (24) 8 (62) 4 (19) | 29% 32% 43% 19% | Indomethacin is not indicated in the prevention of HO after acetabular fracture. One week on indomethacin may be safe to prevent HO without an increase in non-union. |
| McDonald14 | Clinical healing (walking) and radiographic healing (blinded review) | Ketorolac No ketorolac | 64 64 | 23 (36) 30 (57) | 48 (15) 46 (17) | 0 (0) 0 (0) | One, unclear which group | Ketorolac reduced post-op pain and OMEs but unclear effect on healing due to study power. |
| Burd13 | Not defined | Indomethacin No indomethacin | 38 74 | NA NA | 40 (NA) 38 (NA) | 11 (29) 5 (7) | NA NA | Avoid NSAIDs for analgesia or anti-inflammatory purposes during healing of fractures. |
| Zura12 | Coded non-union or code for electrical bone stimulation | NSAID No NSAID | 23 847 285 384 | NA NA | NA NA | 661 (2.8) 14 588 (5.1) | NA NA | No impact of NSAIDs on healing; fracture healing is a diverse process and non-union can result from many risk factors. |
| Tucker6 | Coded non-union | NSAID No NSAID | 15 119 2 570 | NA NA | NA NA | 1 179 (7.8) 392 (10.4) | NA NA | Exposure to NSAIDs at any point during the 90-day postoperative period was associated with fracture non-union (no opioid adjustment) |
| George4 | Non-union diagnosis code or a procedure to treat non-union | NSAID No NSAID | 25 001 279 720 | 11 229 (45) 103 003 (37) | 49 (17) 60 (19) | 456 (1.8) 4 118 (1.5) | All had 1- year follow-up | Filling a prescription for a non-selective NSAID after fracture was not associated with an increased risk of non-union in the subsequent year. Both COX-2-inhibitor and opioid prescription fills after fracture were associated with a greater risk of non-union. |