The effect of hip reconstructive surgery on health related quality of life in non‐ambulatory children with cerebral palsy

identification of hip dysplasia and early treatment through a hip surveillance program of regular hip radiographs can substantially improve a child’s long-term function and comfort, even preventing hip dislocation in the population. The migration percentage (MP), or the percent of the ossified femoral head that is outside of the lateral margin of the acetabulum, is the most accepted measurement for quantifying progressive hip displacement in patients with cerebral palsy undergoing routine hip surveillance. An MP of 30% is considered a threshold for referral to an orthopaedic surgeon. Traditional measurements of the MP using plain radiographs have now been supplanted by reliable measurements on digital Picture Archiving and Communication System (PACS). The touch-screen capability and embedded cameras of smartphones may allow practitioners to obtain this hip surveillance measurement without reliance on digital PACS infrastructure. The objective of this study was to establish the validity of a smartphone-based measurement of MP. Study Design: Instrument Development and Validation Study. Study Participants & Setting: Convenience sample of pelvic radiographs of children with cerebral palsy undergoing routine hip surveillance at a tertiary referral children’s hospital. Materials/Methods: We designed a custom smartphone app called “HipScreen” that contains a tool for rapid measurement of MP using the phone’s touchscreen and camera (Figure 1). Twenty-one pelvic radiographs with varying levels of MP were selected for blinded and random analysis by two experienced pediatric orthopaedic surgeons. Inter-class correlation coeffecients were used to determine the reliability of the measurements using the HipScreen app. The sensitivity and specificity of the HipScreen app were also calculated when referenced against the reference standard of digital PACS. Results: The measurement of MP using the HipScreen app showed very strong interand intra-observer reliability (Figure 2). When an MP of 30% or greater on digital PACS is considered a “positive” case, the HipScreen App had a sensitivity of 98%, specificity of 72%, positive predictive value of 93%, and negative predictive value of 93% (Figure 2). There was a broad distribution of MP represented among the 42 hips, with 24 hips (57%) having an MP between 20–49%, a critical zone for discriminating whether a child needs orthopaedic referral (Figure 3). Conclusions/Significance: Measurement of MP using a custom smartphone app is accurate and valid relative to the PACS standard when used by experienced clinicians. As a screening tool, the HipScreen app shows a low false-negative rate with high sensitivity. Future studies will investigate whether less experienced users can measure MP with acceptable accuracy using this smartphone tool.