OBJECTIVETo investigate the preventive effects and differences of NSAIDs combined with radiotherapy, NSAIDs and radiotherapy for heterotopic ossification(HO) after total hip arthroplasty(THA).

METHODSFrom February 2015 to July 2016, 168 hips undergoing primary THA were divided into group A, B and C, and 163 patients were followed up (54 cases and 54 hips in group A, 55 cases and 55 hips in group B, 54 cases and 54 hips in group C). Among group A, 5 hips were primary osteoarthritis, 37 hips were secondary osteoarthritis due to avascular necrosis of the femoral head, 12 hips were secondary osteoarthritis due to acetabular dysplasia. Patients in group A received oral celecoxib (0.2 g, 2 times a day) for 2 weeks after operation. Among group B, 6 hips were primary osteoarthritis, 32 hips were secondary osteoarthritis due to avascular necrosis of the femoral head, 17 hips were secondary osteoarthritis due to acetabular dysplasia, all of which in group B were treated with preoperative single 7 Gy radiotherapy. Among group C, 5 hips were primary osteoarthritis, 35 hips were secondary osteoarthritis due to avascular necrosis of the femoral head, 14 hips were secondary osteoarthritis due to acetabular dysplasia. Patients in group C were treated with preoperative radiotherapy and celecoxib after operation. The side effects of gastrointestinal reactions were observed after operation, and the heterotopic ossification was evaluated by pelvic anterior and posterior X-ray (Brooker grading).

RESULTSThe mean clinical and radiological follow-up was 21 months(12 to 30 months). In group A, 54 hips were followed up with 7 hips with heterotopic ossification, including 5 hips of Brooker I and 2 hips of Brooker II. In group B, 55 hips were successfully followed up, with 8 hips of heterotopic ossification occurred, including 6 hips of Brooker I, 2 hips of Brooker II. In group C, 54 hips were successfully followed up, with 5 hips of heterotopic ossification occurred , including 4 hips of Brooker I, 1 hip of Brooker II. There was no significant difference in efficacy among 3 groups (²=0.743, 0.690) by chi-square test. The prevalence of side effects were as following: in group A, there were 6 hips with side effects;in group B, there were 6 hips with side effects;in group C, there were 7 hips with side effects. There was also no significant difference in side effects among 3 groups (²=0.135, 0.935).

CONCLUSIONSThe combined-therapy group has lower prevalence of HO than the NSAIDs group or radiotherapy group, but the statistical difference between them is not significant. NSAIDs is still the first choice to prevent HO after THA.