INTRODUCTION

After pelvic osteotomy for the treatment of symptomatic hip dysplasia, the longevity of the hip joint can be compromised by acetabular overcorrection. This iatrogenic pincer-type deformity is considered to be one of the major risk factors for persistent pain and progressing osteoarthritis. There is evidence that acetabula in the borderline range, defined by a lateral center edge angle (LCEA) between 18° and 25°, are more delicate to be orientated physiologically. The aim of this study was to assess the quality of acetabular orientation by triple pelvic osteotomy (TPO), established by Tönnis and Kalchschmidt, especially with respect to acetabular overcorrection.

MATERIALS AND METHODS

A retrospective examination on 368 consecutive hips treated with TPOs was conducted. On the preoperative pelvic radiograph and the radiographic control 5 days after surgery, LCEA, acetabular index (AI), and anterior (AWI) and posterior wall index (PWI) were measured. According to the above-mentioned definition, the hips were divided into a borderline (n = 196) and a dysplastic (n = 172) group. Acetabular overcorrection was defined as when LCEA exceeded 35°, AI was below 0° and AWI exceeded 0.60, postoperatively. The postoperative occurrence of a relevant femoroacetabular impingement was correlated to these thresholds. Statistics comprised a priori power analysis, correlation analyses and receiver operating characteristics (ROC).

RESULTS

In the borderline group, in 64 hips (32.7%), LCEA and AI indicated lateral overcorrection. In the dysplastic group, in 14 hips (8.1%), solely AI indicated overcorrection. In none of the hips, relevant anterior overcorrection was detected since AWI never exceeded 0.60. Chi-square test demonstrated a significant correlation between the occurrence of a postoperative femoroacetabular impingement and LCEA exceeding 35°, as well as AI below 0° (p < 0.001, resp.). Bravais-Pearson's analysis showed a significant correlation between the pre- and postoperative values of all parameters in the borderline and the dysplasia group (p < 0.001). Thus, ROC analysis could be performed and provided preoperative cutoff values for LCEA (23°) and AI (12.5°), hinting at postoperative overcorrection.

CONCLUSION

The comparison of radiographic parameters after TPO showed a considerably greater percentage of laterally overcorrected acetabula in the borderline hips than in the dysplastic hips. According to the wall indices, anterior overcorrection was not observed. ROC analysis anticipated unfavorable lateral overcorrection when preoperative LCEA was above 23° and AI below 12.5°. These findings should sensitize the surgeon to the delicate acetabular correction in borderline dysplastic hips.