Posterior sloping angle of the capital femoral physis: interobserver and intraobserver reliability testing and predictor of bilaterality

J Pediatr Orthop. 2007 Oct-Nov;27(7):801-4. doi: 10.1097/bpo.0b013e31814256af.

Abstract

The management of the contralateral hip after unilateral slipped capital femoral epiphysis (SCFE) is controversial. The aims of this study were to assess the interobserver and intraobserver reliability of the posterior sloping angle (PSA) as described by Barrios et al (J Pediatr Orthop. 2005;25:445-449) and whether it can be used as a predictor of need for prophylactic pinning of the contralateral hip. The PSA was measured by 4 different surgeons on 2 separated occasions in 47 unaffected contralateral hips after unilateral SCFE. The interobserver intraclass correlation coefficient (ICC) was 0.83 (excellent) for the first set of measurements and 0.74 (fair to good) for the second set of measurements. Measurement of the intraobserver ICC revealed an excellent agreement (ICC > 0.75) in 3 of the 4 surgeons and fair to good agreement in 1 of the surgeons (ICC between 0.4 and 0.74). Comparison of the PSA measurements of the first author (M.Z.) with a control group of 20 hips revealed that the PSA was significantly higher (P < 0.05) in the group of patients that eventually had a contralateral slip when compared with the PSA of patients in the study group who did not slip and with normal hips in the control group. Patients with contralateral SCFE that did not slip had a significantly higher PSA (P < 0.05) when compared with normal individuals. A PSA of 14.5 degrees or more is recommended for prophylactic pinning of the contralateral hip. The number needed to treat (NNT) at this cut off to avoid 1 future slip was calculated to be 1.9.

MeSH terms

  • Adolescent
  • Child
  • Disease Progression
  • Epiphyses, Slipped / diagnostic imaging*
  • Female
  • Femur / diagnostic imaging*
  • Humans
  • Male
  • Observer Variation
  • Predictive Value of Tests
  • ROC Curve
  • Radiography
  • Reproducibility of Results
  • Statistics, Nonparametric