Reliability of radiographic assessment of distraction osteogenesis site.

Abstract

The presence of three of four continuous cortices on anteroposterior and lateral radiographs has been stated to be an indication of bony healing of distraction osteogenesis. In this study the authors assessed the level of agreement of radiographic assessment of bony union. Forty-two lower extremity radiographs of consolidating distraction gaps were reviewed by nine examiners on two occasions for number of cortices and whether fixator removal was indicated. For number of cortices, the kappa coefficients for interobserver and intraobserver variability were 0.127 and 0.290. For fixator removal, the interobserver and intraobserver coefficients were 0.352 and 0.461. Variation in assessment of number of cortices was slightly better than chance, indicating an inadequate measure of healing. The clinicians used radiographic criteria other than three of four cortices for fixator removal. The decision to remove an external fixation device based on radiographic assessment alone resulted in intraobserver and interobserver variability moderately above chance.