Infants

phology and pain in young adults with CP at skeletal maturity. Method: of a birth cohort of 367 children born in 1990–1992, 103 developed hip displacement (MP >30). Ninety-eight of these young adults attended a transition clinic where radiographs of the hips were obtained. Severity and frequency of pain were recorded using Likert scales. Motor function was classified by the GMFCS and hip morphology using the Melbourne Cerebral Palsy Hip Classification Scale (MCPHCS E&R). A Pearson correlation coefficient was used to describe the relationship between pain severity and frequency. Kruskall-Wallis tests of differences examined relationships between pain severity, GMFCS and MCPHCS. Non-parametric tests compared pain severity between those who had hip surveillance and those who did not. Results: There was a strong association between hip morphology, GMFCS and pain. The median (IQR) pain score for MCPHCS 1–4 was 2 (1.0–3.0) compared to 7 (6.0–8.0) for MCPHCS 5–6 (p<0.001). This association was also true for GMFCS: those at GMFCS I–III reported a median pain score of 2 (1.0–3.0), compared to 3 (2.0–5.0) for those at GMFCS IV–V (p<0.001). Hip surveillance was associated with improved hip morphology and less pain. Adolescents with worse hip morphology (MCPHCS 6–7) had fewer hip radiographs compared to those with hips graded 1–5 (p<0.001). The 14% of patients not under hip surveillance had a median pain score of 7 (5.0–8.0) compared to 2 (1.0–3.5) in the 86% patients under surveillance (p<0.001). Conclusion: Increasing GMFCS level, combined with limited or no hip surveillance, is associated with poor hip morphology and high pain levels. Conversely, adolescents at all GMFCS levels with access to hip surveillance and appropriately timed surgery had satisfactory hip morphology and less pain at skeletal maturity.