Publications

Background
Hip migration is a common deformity in children with cerebral palsy (CP). Although children with CP often
have anatomically normal hips at birth, progressive spasticity can lead to hip subluxation, dislocation, and
acetabular dysplasia over time. Early recognition of hip migration is crucial because timely intervention can

significantly change the prognosis and reduce morbidity. The aim of this study was to evaluate the inter-
and intra-observer reliability of radiographic hip migration measurements in children with CP.

Materials and methods
We conducted a prospective observational study from 2020 to 2022 to determine inter- and intra-observer
variation in the diagnosis of hip migration among children with CP. Eligible participants were recruited from
the clinic and followed up for a period of two years. They underwent serial supine AP pelvic radiographs at
six-month intervals. The radiographic parameters evaluated were Reimer’s migration percentage (MP) and
the acetabular index (AI). Four observers with varying levels of clinical experience independently assessed
the radiographs. Inter- and intra-observer agreement was analyzed using intraclass correlation coefficients
(ICCs) and standard statistical methods.
Results
Eighteen children (mean age 12.2 years, range 2-18 years, SD 7.72) with spastic CP were included in the
study. To ensure consistency and avoid inter-hip variability in the same patient, the hip showing greater
displacement served as the unit of analysis. Based on the Gross Motor Function Classification System, nine
were classified as level IV, seven as level III, and two as level V. The MP demonstrated consistently high
inter- and intra-observer reliability, with ICCs and 95% CIs of 0.999 (0.988-1.000) at baseline and 0.999
(0.999-1.000) at endline. Similarly, the AI also showed excellent inter- and intra-observer reliability, with
ICCs and 95% CIs of 0.992 (0.984-0.997) at baseline and 0.998 (0.997-0.999) at endline. These findings
indicate strong reproducibility for both measurements across observers and time points. MP analysis showed
a significant increase from baseline to endline (Wilcoxon Z = -3.724, p = 0.0002), indicating progression of
hip migration. AI also increased significantly over time (paired t = -3.944, p = 0.0010), reflecting the
progression of acetabular dysplasia. Both represent secondary outcomes of the study.
Conclusions

This study demonstrates that both MP and AI provide comparable and clinically acceptable levels of inter-
and intra-observer reliability, with a slight advantage for MP. These results underscore the value of MP as a

primary surveillance metric while reaffirming AI as a reliable adjunct for evaluating acetabular morphology
in children with CP. By contributing additional evidence on the reproducibility of these measurements, our
study supports their continued use within structured hip surveillance programs. Further multicenter,
prospective studies are needed to validate these findings in broader populations and to strengthen the
foundation for standardized, universally applicable surveillance protocols.