Observations on infantile coxa vara.

In 1907 Elmslie suggested that in some cases of congenital coxa vara without other abnormalities injury at birth might perhaps be one factor. He described accurately the cervical breach in infantile coxa vara with the typical metaphysial wedge included in the proximal fragment, and pointed out that there was often a clear history of injury. In 1913 he described thirty-four cases of the infantile or cervical type out of a total of seventy-seven cases of coxa vara. He believed that trauma, or gradual slipping, accounted for the pathological changes. His description of radiographs of three children aged five is of a downward displacement of the head of the femur, carrying with it the adjoining part of the base of the neck. He said that “ this displacement, whether occurring as a result of a sudden accident, or by a process of gradual slipping, exactly accounts for the appearances seen in the radiographs.” Fairbank (1928) accurately described the lesion under discussion and separated clearly the congenital group from the group which he called “ infantile or cervical coxa vara.” The characteristics of this second group were the later onset of signs and symptoms, the smaller amount of shortening and a characteristic radiological feature, which he described as “ the presence of a clear line through the neck of the femur other than the epiphysial line, the two lines forming an inverted V and enclosing between them a triangular piece of bone.” Fairbank could not accept trauma as the cause because the lesion was often bilateral, and he argued that a developmental error was the explanation. Babb, Ghormley and Chatterton (1949) agreed, and on the basis of fifteen cases concluded that the responsible developmental lesion was probably vascular and of congenital origin. Amstutz and Wilson (1962) admitted the confusion of names and entitled their paper “ dysgenesis of the proximal femur.’ ‘ They described the Fairbank type but called it “ congenital,” pointing out, however, that their seventeen cases of this type showed no other congenital disorders. The diagnosis was delayed until weight-bearing began and progressive deformity was invariable. They concluded that the responsible lesion was “ faulty maturation of the neck and irregular ossification.” These writers described coxa vara associated with a generalised bone disease, as had Fairbank (1928). Le Mesurier (1948) and Almond (1956). In Fairbank’s case this was cranio-cleido-dysostosis, and the coexistence led him to the conviction that as this disease was developmental, then so was infantile coxa vara. Morgan and Somerville (1960) described a congenital type due to