Latent class growth analysis predicts long term pain and function trajectories in total hip arthroplasty: a study of 605 consecutive patients

Abstract

Purpose: To characterize groups of subjects according to their trajectory of hip pain and function over one to five years post total hip arthroplasty (THA). Methods: Patients from one centre who underwent primary THA (N=605) between 2006 and 2008. The Harris Hip Score (HHS) was collected pre-operatively and annually post-operatively. Latent Class Growth Analysis (LCGA) was used to classify groups of subjects according to their trajectory of hip pain and function over 1-5 years post-surgery. Results: Results: LCGA identified a class of patients with persistent moderate to marked hip pain (9.0%), (Figure 1). Predictors (OR, 95% CI) of this pain trajectory class membership were pre-surgery Short Form 12 (SF12) Mental Component Summary (MCS) per 15 points (0.49, 0.32 - 0.76) and Physical Component Summary (PCS) per 15 points (0.21, 0.07 - 0.61), Charlson Comorbidity Index (2.4, 1.25 - 4.73) and Socio-economic status (0.85, 0.77 - 0.95). LCGA identified a class of patients with persistent poor function (16.2%), (Figure 2). Predictors (OR, 95% CI) of this function trajectory class membership were female sex (2.2, 1.3 - 4.0), advancing age per 10 years (1.6, 1.2 - 2.1), pre-surgery SF12 MCS per 15 points (0.65, 0.45 - 0.94) and PCS per 15 points (0.28, 0.12 - 0.70), Charlson Comorbidity Index ≥ two (4.0, 1.8 - 9.0), lateral vs posterolateral approach (1.9, 1.1 - 3.3), and ASA score >2 (2.6, 1.5 - 4.6). Conclusions: Modifiable predictors of poor response to THA included baseline co-morbidity, physical and mental well-being, socio-economic status and lateral vs posterior approach. This provides useful information for clinicians in terms of informing patients of the expected course of longer term outcomes of THA and for developing of risk prediction algorithms that identify patients in whom there is a high likelihood of poor surgical response.