A study of femoral torsion in children.

Techniques for measuring upper femoral torsion in children have been progressively developed by Dunn (1952), Ryder and Crane (1953), Dunlap, Shands, Hollister, Gaul and Streit (1953), Magilligan (1956) and Shands and Steele (1958). The projected neck-shaft angle of inclination and the projected angle of femoral torsion (Fig. 1) may be used to plot the true torsion on a table based on a mathematically developed geometrical relationship (Dunn 1952, Ryder and Crane 1953, Dunlap et a!. (1953), Magilligan 1956, Shands and Steele 1958). The essential features of the radiographic positioning include an antero-posterior view for inclination taken in neutral rotation and a view for torsion with hip and knee each flexed to a right angle and with the hip abducted to a fixed and known angle between 10 degrees and 30 degrees. For this study the technique of Ryder and Crane (1953) has been used, and for normal values the figures of Shands and Steele (1958) which vary little from those of Ryder and Crane (1953) and Crane (1959) have been averaged with the latter (Table I). Only figures at least 10 degrees over or under the normal were recorded as abnormal torsion. In the past ten years 1,320 children presenting consecutively were examined personally to determine the incidence and sequelae of toeing-in and toeing-out when walking. In the