Although generally considered successful, total hip arthroplasty (THA) can yield suboptimal results in terms of pain and function in up to one forth of patients suffering from hip osteoarthritis (OA). A thorough understanding of the determinants of patient-reported pain and function following THA can help plan interventions directed at improving surgical results. Previously published systematic reviews do not permit to confidently identify the determinants of THA outcomes because of unsatisfactory methodological designs. Therefore, we aimed to answer: (1) which preoperative factors are most consistently associated with postoperative patient-reported pain and function up to 2 years following primary unilateral THA for hip OA. Medline, Pubmed, Embase and CINAHL were screened from their respective inception dates until April 2015 using a combination of keywords and MESH terms. Criteria for inclusion were: (1) participants with primary unilateral THA for hip OA followed for at least 3 months with a maximal follow-up of 2 years; (2) validated disease-specific patient-reported outcome measures assessing pain and/or disability; (3) identification of determinants obtained via multivariate analyses. Methodological quality was assessed using a modified version of the methodology checklist for prognostic studies. Twenty-two manuscripts were included. Mean score of the methodological quality was 81.0±10.3% (66.7% to 100%). Among socioeconomic determinants, a lower educational level was significantly related to worse pain and function (3 out of 3 studies evaluating the relationship). Clinical determinants of poor outcomes included preoperative levels of pain and physical function (9 out of 12 studies), higher body mass index (6 out of 10 studies), presence/greater level of comorbidities (7 out of 8), worse general health (4 out of 4 studies) and lower radiographic OA severity (3 out of 4 studies). Study heterogeneity limited the pooled assessment of the strength of associations between the preoperative variables and THA outcomes. Studies with moderate-to-high methodological quality allowed to identify 6 preoperative variables consistently associated with medium term pain and function following THA. This knowledge may assist the management of patients at risk of poor results. Further research is required to clarify the force of associations between determinants and THA outcomes.

Level II. Systematic review of cohort studies.