Spica magnetic resonance imaging for determination of abduction angle: initial results and reproducibility assessment

PURPOSE: Spica magnectic resonance imaging (MRI) is an established technique for postoperative determination of hip reduction in patients treated for developmental dysplasia of the hip (DDH). A hip abduction angle >55° is considered excessive and has been associated with epiphyseal osteonecrosis. Our purpose was to establish objective criteria for measuring hip abduction angles on MRI after hip reduction and spica casting in patients with DDH, and evaluate reproducibility and reliability of angle measurement using these criteria. METHODS: Forty patients with DDH at our institution who underwent spica MRI after hip reduction between 3 April 2008 and 3 March 2015 were identified. Hip abduction angles were measured on proton density axial images as follows. A transverse line was drawn connecting the posterior ischial tuberosities. A second line was drawn medially along the distal femoral diaphysis, and the angle between these two lines was measured; this value was subtracted from 90°, yielding the degree of abduction from midline. Measurements were independently performed by three faculty radiologists, one orthopedist, and one radiology resident. Inter-reader and intra-reader reliability was assessed using intraclass correlation (ICC), with 0 representing no agreement and 1 representing perfect agreement. RESULTS: For inter-reader reliability, the ICC of the five physicians was 0.89 (95 % CI 0.84–0.92). For intra-reader reliability, the ICC of the five physicians ranged from 0.90−0.97 (95 % CI 0.85–0.98). The mean standard deviation of hip abduction angle measurement among readers was 3.6°. CONCLUSION: The proposed hip abduction angle measurement criteria for spica MRI are both reproducible and easy to perform. The high ICC and low standard deviation of independently evaluated hip abduction angles indicates high reproducibility of measurement. This applies to both inter- and intra-reader reliability.