Ultrasound Determination of the Femoral Head-Neck Alpha Angle

The femoral head-neck alpha angle is used to quantify the degree of femoral head asphericity in patients suspected of cam-type femoroacetabular impingement. The measurement was first performed using magnetic resonance imaging and, more recently, three-dimensional computed tomography (CT). We set out to determine whether the alpha angle could be reliably measured using ultrasound. Patients were recruited from a cohort presenting for CT of the hip. Alpha angles were calculated following the departmental protocol by institutionally accredited radiographers. After the CT, patients were imaged with ultrasound and the alpha angle calculated from the ultrasound image by a sonographer blinded to the CT result. Statistical comparison of the two methods was performed with the Bland-Altman test using SPSS (version 21.0, Chicago, USA), and a p < 0.05 afforded significance. Twenty-eight patients were recruited. Eleven patients were bilateral examinations, providing 39 hips for analysis. There were 15 females and 13 males, with 21 right and 18 left hips examined. Average patient age (±standard deviation) was 40 y (±13.9 y). Mean (±standard deviation) measurements for CT and ultrasound were 62.5° (±14.2°) and 64.5° (±12.6°), respectively. The mean absolute difference between the two methods was 10.5° (95% confidence interval 6.9°-14.0°). Sensitivity of each individual ultrasound measurement was 91.3%. The specificity of ultrasound was 43.75%. The positive predictive value was 0.7, and the negative predictive value was 0.78. Overall accuracy of the ultrasound-derived alpha angle was calculated at 0.718. Ultrasound demonstrates good sensitivity and good negative predictive value in calculation of the femoral head-neck alpha angle compared with CT; however, specificity is low. Ultrasound measurement of the alpha angle can provide objective evidence of cam-type femoroacetabular impingement in symptomatic patients and can direct patients to more established imaging techniques where appropriate.