We have devised a combined pillar score (CPS) system, based on the lateral pillar (LP) and the posterior pillar (PP) classifications, together with the age at onset of Perthes’ disease, and examined its correlation with prognosis. The correlation coefficient of the Catterall classification, LP, PP, and CPS systems with the Stulberg system was 0.39, 0.52, 0.50, and 0.70, respectively. Overall 21 of the 22 hips (95.4%) with a CPS of 0 to 1 point had a good outcome and 12 of the 13 hips (92.3%) with a CPS of 3 points or more had a fair or poor outcome. None with a CPS of 2 points, had a poor outcome. The study shows that an accurate prediction of the prognosis is not possible with the LP classification alone for patients classified as belonging to group B (LP height 50% to 100% of contralateral height). The CPS system does allow accurate prediction of outcome.

We treated 120 children between the ages of 12 and 31 months with 137 developmental dislocations of the hip and reviewed them at a mean follow-up of 14 years.

We had used two-directional arthrography of all hips before reduction to evaluate the anterior, superior, and posterior portions of the limbus. Of the 137 hips, 64 had no interposed limbus in the AP view of the arthrogram, but 45 of these had an interposed anterior or posterior portion of the limbus. The hips with good stability and no interposed limbus in either AP or lateral arthrograms had excellent results by closed methods; in the other cases the results were less satisfactory.

Our findings suggest that hips suitable for management by closed reduction can be identified by two-directional arthrography. Hips shown to have an interposed limbus are best managed by open reduction.