Dislocation rates in posterior total hip arthroplasty (THA) range between 2% and 5%, but long-term course (recurrence of dislocation or revision surgery) is not known, most series having short follow-up or small populations. We therefore conducted a retrospective study on a large series, to determine long-term rates of recurrence and surgical revision and recurrence risk factors.

Long-term follow-up of a large cohort of THA dislocations enables recurrence rate and factors to be determined.

Five hundred and nine cases of THA dislocation were admitted to our center between 1994 and 2008. A hundred and twenty seven incomplete files and 150 patients who had received their THA elsewhere were excluded, leaving 232 patients: 150 female, 82 male. Mean age at THA (163 primary, 69 revision) was 63 years (range, 15–90 years), and 65 years (range 20–90 years) at first dislocation, with a mean interval to dislocation of 25 months. Minimum follow-up was 8 years up to 2016, or 1 year taking account of deaths (111 deaths). There were 46 anterior, 185 posterior and 1 multidirectional dislocations. The following potential recurrence factors were assessed: gander, age, body-mass index (BMI), etiology, surgical history, bearing diameter and type, component fixation means, dislocation direction, and time to dislocation.

A hundred and thirty three of the 232 patients (57%) showed at least 1 recurrence, at a mean 38 months (range, 0.5-252 months); 78 experienced a second and 32 a third recurrence. Ninety-nine (43%) had only 1 dislocation, without recurrence, but 17 of these (17%) underwent reoperation for other causes. The reoperation rate was 17/232 (7%) excluding recurrent instability, and 84/232 (36%) for instability. Fourty-eight months after the first dislocation, 84/133 cases of recurrence (63%) had been reoperated on: 16 complete replacements, 18 bearing replacements, 42 dual mobility cups, one large diameter cup, seven Lefèvre retentive cups. The rate of revision surgery for instability was high, at 84/232 (36%), and higher again in relation to recurrence (84/133: 63%). Only posterior dislocation emerged as a factor for recurrence (HR=1.774, 95% CI [1.020–3.083]), the other tested factors showing no correlation.14 of the 84 revision surgeries for instability (16.6%) were followed by recurrence, without identifiable risk factors.

The recurrence rate was 57%, with posterior dislocation as the only risk factor. The rate of revision surgery for recurrence was 84/232 (36%), with 14/84 revision procedures (16.6%) followed by further recurrence.

IV, retrospective, without control group.