Are normal hips being labeled as pathologic? A CT-based method for defining normal acetabular coverage.
Academic Article
Overview
abstract
BACKGROUND: Plain radiographic measures of the acetabulum may fail to accurately define coverage or pathomorphology such as impingement or dysplasia. CT scans might provide more precise measurements for overcoverage and undercoverage. However, a well-defined method for such CT-based measurements and normative data regarding CT-based acetabular coverage is lacking. QUESTIONS/PURPOSES: The purposes of the study were (1) to develop a method for evaluation of percent coverage of the femoral head by the acetabulum; and (2) to define normative data using a cohort of asymptomatic patient hip and pelvic CT scans and evaluate the variability in acetabular version for asymptomatic patients with normal lateral coverage (lateral center-edge angle [LCEA] 20°-40°) that has previously been defined as abnormal based on radiographic parameters. METHODS: Two-hundred thirty-seven patients (474 hips) with hip CT scans obtained for reasons other than hip-related pain were evaluated. The scans were obtained from a hospital database of patients who underwent CT evaluation of abdominal trauma or pain. In addition, hips with obvious dysplasia (LCEA < 20°) or profunda (LCE > 40°) were excluded resulting in a final cohort of 222 patients (409 hips [115 men, 107 women]) with CT scans and a mean age of 25 ± 3 years. CT scan alignment was corrected along the horizontal and vertical axis and percent acetabular coverage around the clockface (3 o'clock = anterior), and regional (anterior, superior, posterior) and global surface area coverage was determined. Percent coverage laterally was correlated with the LCEA and the presence and prevalence of cranial retroversion (crossover sign) and a positive posterior wall sign were determined. RESULTS: The mean regional percent femoral head surface area coverage for the asymptomatic cohort was 40% ± 2% anteriorly, 61% ± 3% superiorly, and 48% ± 3% posteriorly. Mean global coverage of the femoral head was 40% ± 2%. The local coverage anteriorly (3 o'clock) was 38% ± 3%, laterally (12 o'clock) was 67% ± 2%, and posteriorly (9 o'clock) was 52% ± 3%. The mean lateral coverage represented a mean LCEA of 31° (± 1 SD). Fifteen percent of hips demonstrated cranial retroversion that would correlate with a crossover sign, and 30% had < 50% posterior coverage that would correlate with a positive posterior wall sign on an anteroposterior pelvis radiograph. In addition, male hips had a higher prevalence of a crossover sign (19%; 95% confidence interval [CI], 14%-25% versus 11%; 95% CI, 7%-16%; p = 0.03) and posterior wall sign (46%; 95% CI. 39%-53% versus 13%; 95% CI, 9%-19%; p < 0.001) compared with women. A positive crossover sign or posterior wall sign was present for 113 male hips (53%; 95% CI, 46%-60%) compared with 39 female hips (20%; 95% CI, 15%-26%; p < 0.001). CONCLUSIONS: This study provides normative coverage data and a reproducible method for evaluating acetabular coverage. Cranial acetabular retroversion (crossover sign) and a positive posterior wall sign were frequent findings in a young asymptomatic cohort and might be a normal variant rather than pathologic in a significant number of cases.
