We investigate a new measurement method for bowleg deformity in children to reduce intra-observer and inter-observer error and to assess change in the angles measured with respect to the rotation of the lower leg.

Fifty-one children younger than 3 years old with bow leg were analyzed. To assess the intra-observer and inter-observer error of the Metaphyseal-Diaphyseal Angle, Intermetaphyseal-metaphyseal angle and Tibio-Femoral angle measurements, supine view antero-posterior radiographs of the lower legs were taken at the neutral rotation, at 10 ° and at 20 ° internal rotation positions. Also, the effect of the rotation of lower legs upon the measured angle was assessed by comparing and analyzing the angle values with respect to rotations of the lower leg.

TInter-observer error showed statistically significant difference (p<0.05). All measured angles showed statistically significant differences with respect to the rotation of the lower legs. The measurement variability introduced by rotation was significantly greater than the intra-observer and inter-observer errors. The Intermetaphyseal-metaphyseal angle measurement had the lowest standard error and the highest correlation coefficient.

The Intermetaphyseal-metaphyseal angle measurement seems a highly reliable method, which could be clinically utilized. Also, since the rotation of the lower legs influences the metaphyseal-diaphyseal angle, the clinician should consider this effect when studying a simple radiograph for bowleg deformity.