106.
Poor agreement between radiographs and clinical examinations in
children with genu valgus
Nina Hardgrib Madsen, Martin Gottliebsen, Ole Rahbek, Michel Bach
Hellfritzsch, Bjarne Møller Madsen
Department of Children`s Orthopaedics Aarhus University Hospital, Denmark;
Department of Children`s Orthopaedics, Aarhus University Hospital,
Denmark; Department of Children`s Orthopaedics, Aarhus University
Hospital, Denmark; Department of Radiology, Aarhus University Hospital,
Denmark; Department of Children`s Orthopaedics, Aarhus University
Hospital, Denmark
Background:
Genu valgus is a condition characterized by a lateral shift of the
knee mechanical axis. The deformity can be characterized using clinical
examinations and long hip-knee-ankle (HKA) x-rays. It is unclear how these
investigations correlate with each other
Purpose / Aim of Study:
To a) examine the interobserver and intraobserver
reliabilities in clinical measurements and b) examine the agreement between
clinical and radiographic measurements in children
Materials and Methods:
a) 11children had Intermalleolar (IM) double
measurements performed, both standing and lying down, by two different
examiners to determine interobserver and intraobserver reliabilities for IM
measurements. Intra Class Correlations (ICC) coefficients were calculated. b)
49
children referred with genu valgus deformity. Criteria for inclusion: IM
distance measured and a standardized AP radiograph of the lower extremities
taken. We calculated Spearman’s rank correlation coefficient to study the
reliability between radiographs and clinical examinations
Findings / Results:
The overall ICCs for intraobserver reliability for standing
and lying clinical IM measurements was 0.84–0.98. We examined the
reliability between standing and lying examinations within observers with
ICCs ranging from 0.53–0.94. When comparing standing and lying
examinations between observers we found ICCs between 0.84-0.99.
Spearman’s rho for comparison between clinical IM distance measurements
and HKA angles was found to be 0.19 indicating no relationship
Conclusions:
We report good reliability with ICC coefficients >0.80 for
clinical examinations of IM distance except for the reliability between standing
and lying examinations. We found no convincing correlation between clinical
examinations and radiographic angles, so it seems useful to consider if both
measurements should be used to follow children with genu valgus
