Can We Discriminate Symptomatic Hip Patients From Asymptomatic Volunteers Based on Anatomic Predictors? A 3-Dimensional Magnetic Resonance Study on Cam, Pincer, and Spinopelvic Parameters

Abstract

Background: Given the high prevalence of patients with hip deformities and no ongoing hip dysfunction, understanding the anatomic factors predicting the symptomatic state is critical. One such variable is how the spinopelvic parameters (SPPs) may interplay with hip anatomic factors. Hypothesis/Purpose: SPPs and femoral- and acetabular-specific parameters may predict which patients will become symptomatic. The purpose was to determine which anatomic characteristics with specific cutoffs were associated with hip symptom development and how these parameters relate to each other. Study Design: Cohort study (Diagnosis); Level of evidence, 2. Methods: 548 participants were designated either symptomatic patients (n = 176, scheduled for surgery with hip pain and/or functional limitation) or asymptomatic volunteers (n = 372, no pain) and underwent 3-dimensional magnetic resonance imaging. Multiple femoral (α angle, Ω angle, neck angle, torsion), acetabular (version, coverage), and spinopelvic (pelvic tilt, sacral slope [SS], pelvic incidence) parameters were measured semiautomatically. Normative values, optimal differentiating thresholds, and a logistic regression analysis were computed. Results: Symptomatic patients had larger cam deformities (defined by increased Ω angle and α angle), smaller acetabular coverage, and larger pelvic incidence and SS angles compared with the asymptomatic volunteers. Discriminant receiver operating characteristic analysis confirmed that radial 2-o’clock α angle (threshold 58°-60°, sensitivity 75%-60%, specificity 80%-84%; area under the curve [AUC] = 0.831), Ω angle (threshold 43°, sensitivity 72%, specificity 70%; AUC = 0.830), acetabular inclination (threshold 6°, sensitivity 65%, specificity 70%; AUC = 0.709), and SS (threshold 44°, sensitivity 72%, specificity 75%; AUC = 0.801) ( P < .005) were the best parameters to classify participants. When parameters were entered into a logistic regression, significant positive predictors for the symptomatic patients were achieved for SS, acetabular inclination, Ω angle, and α angle at 2-o’clock, correctly classifying 85% of cases (model sensitivity 72%, specificity 91%; AUC = 0.919). Conclusion: Complex dynamic interplay exists between the hip and SPPs. A cam deformity, acetabular undercoverage, and increased SPP angles are predictive of a hip symptomatic state. SPPs were significant to discriminate between participants and were important in combination with other hip deformities. Symptomatic patients can be effectively differentiated from asymptomatic volunteers based on predictive anatomic factors.