Reliability analysis for radiographic measurement of limb length discrepancy: full-length standing anteroposterior radiograph versus scanogram

Abstract

Patients with limb length discrepancy (LLD) often have associated angular deformities requiring a standing full-length radiograph of the lower limb in addition to a scanogram. The purpose of our study was to determine the intraobserver and interobserver reliability of measuring LLD with both techniques, using computed radiography. The LLD was measured on 70 supine scanograms and standing anteroposterior radiographs of the lower extremity by 5 blinded observers on 2 separate occasions. Intraclass correlation coefficient (ICC) and mean absolute difference (in millimeters) was calculated to assess intraobserver and interobserver reliability and found to be excellent for both radiographic techniques. Intraobserver ICC and mean absolute difference was 0.975 to 0.995 and 1.5 to 2.6 mm for scanogram and 0.939 to 0.996 and 1.5 to 4.6 mm for the standing radiograph, respectively. Repeated measurements for both radiographic studies were within 5 mm of the first measurement greater than 90% and within 10 mm greater than 95% of times. Interobserver ICC and mean absolute difference was 0.979 and 2.6 mm for scanogram and 0.968 and 3.0 mm for the standing radiograph. The reliability was excellent irrespective of age, sex, and underlying diagnosis other than Blount disease, which had good reliability. A standing anteroposterior radiograph of the lower extremity should be the imaging modality of choice when evaluating patients with limb length inequality who may have angular deformities because it allows a comprehensive evaluation of the extremity and is as reliable as a scanogram for measuring LLD. This approach may decrease the radiation exposure and financial burden involved in assessing patients with unequal limb lengths.