Risk factors for heterotopic ossification in primary total hip arthroplasty

Abstract

Heterotopic ossification (HO) is a recognised complication of total hip arthroplasty (THA). This study aims to correlate demographics, surgical approach and type of arthroplasty to the incidence of HO in an attempt to quantify patient risk. A total of 920 primary THAs in 893 patients were performed between 2006 and 2008 in a single arthroplasty centre. Radiographic evaluation was conducted and all cases of HO were classified using the Brooker classification. Age, sex, arthroplasty type and surgical approach were all considered as variables. Arthroplasty type was classified into four groups; total cemented (TC), total uncemented (TU), hybrid (H) and reverse hybrid (RH). Two-level logistic regression analysis was conducted. The overall incidence of HO was 24%. Male sex [OR=3.57, 95% CI (1.79-7.10); p=0.001], lateral approach [OR=2.47, (95% CI 1.23-4.95); p=0.001] and total cemented implants [OR=3.14, (95%CI 1.37-7.23); p=0.007] were significantly associated with HO. The intra-class correlation coefficient was 0.52 [95% CI (0.21, 0.81); p=0.004], demonstrating that patients with previous HO to one THA were very likely to suffer HO in subsequent THA. The results demonstrate very large effects for sex, surgical approach, and implant type on HO incidence. This raises a three arm hypotheses that reaming of the femoral canal for the insertion of cemented implants contaminates the surgical field with bone marrow increasing the risk of HO, whereas modern cementless implants generally employ impaction broaching. In addition surgical insult to the hip abductors during exposure, particularly in males due to higher muscle mass, may also predispose to HO.