Background/Purpose: Preliminary investigation in this cohort supported an association between hip morphology and symptoms, however, the influence of race, gender, age, BMI and radiographic hip OA (rHOA) remains unclear.

Methods: Data are from the baseline visit of the Johnston County OA Project. Hip symptoms were assessed by: 1) patient-reported groin pain (yes/no) and 2) hip pain on internal rotation (yes/no).  RHOA was defined as a Kellgren Lawrence grade of 2 or more. Standardized software (OxMorf, Oxford, UK) was used to assess hip morphology. We focused on 5 key hip morphologies (Table). Joint-based logistic regression models, stratified by sex, and employing generalized estimating equations to account for intra-person correlation, were used to estimate Odds Ratios (OR) associations between morphologic measures and symptomatic outcomes, adjusting for race, age, BMI, and side (labeled OR¹), while a second model considered the influence of baseline rHOA (labeled OR²); interactions between rHOA and morphologic measures were considered significant at a p value of <0.1, in which case stratified analyses were performed.

Results: The analysis sample included 4530 hips from 2274 individuals, mean age 64 years, mean BMI 29kg/m², 30% Black, 39% male. Approximately 20% of hips had rHOA. Among women, significant associations were identified between hip symptom outcomes and increasing alpha angle, Gosvig ratio, triangular index height, and protrusio acetabula independent of race, age, and BMI.  However, after adjusting for rHOA, the associations with pain on internal rotation were no longer statistically significant, and those with groin pain were seen only for hips with rHOA. Additionally, protrusio acetabula was protective for groin pain in this model (Table, left). For men, a similar statistically significant association between hip symptom outcomes and increasing alpha angle, Gosvig ratio, triangular index height, and lateral center edge angle (CEA) was noted.  After adjustment for rHOA, these associations were attenuated, although significant associations remained between pain on internal rotation and increasing Gosvig ratio, triangular index height, and for CEA <=25 degrees (versus referent 25-40 degrees). For the groin pain outcome, a statistically significant association with alpha angle was present only in those with rHOA. In this model, a low CEA (<=25 degrees) was associated with greater odds of groin pain, while a high CEA (>40 degrees) was protective (Table, right).

Conclusion: Hip pain on internal rotation and groin pain were more likely with certain hip morphologies in men and women, however this association was highly dependent on baseline rHOA status.  Pincer-type morphologies (CEA>40 degrees and protrusio) appeared protective for groin pain when coupled with rHOA.   One hypothesis is that the added depth and acetabular coverage could increase stability and reduce extreme range of motion, resulting in reduced symptoms.