Osteonecrosis of the femoral head (ONFH) classification systems are based on the size, volume, and location of necrotic lesions. Often-but not always-ONFH results in femoral head collapse. Because acetabular coverage is associated with mechanical stress on the femoral head, it might also be associated with femoral head collapse in patients with ONFH. However, the association between acetabular coverage and femoral head collapse in these patients has not been established. (1) Is femoral head collapse associated with acetabular coverage or pelvic incidence (PI) in patients with ONFH? (2) Are established predictors of femoral head collapse in ONFH classification systems associated with acetabular coverage? Between 2008 and 2018, we evaluated 343 hips in 218 patients with ONFH. We considered all patients with ONFH except for those with a traumatic etiology, a history of surgical treatment before collapse, or those with collapse at initial presentation as potentially eligible for this study. Of those, 101 hips with ONFH (50% [50] were in males with a mean age of 44 ± 15 years) met our inclusion criteria. These patients were subsequently divided into two groups: those with femoral head collapse within 12 months (collapse group, 35 hips) and those without femoral head collapse (noncollapse group, 66 hips). No differences in patient demographics were observed between the two groups. CT images were used to measure the PI and acetabular coverage in three planes: the lateral center-edge angle (LCEA) in the coronal plane, the anterior and posterior center-edge angle in the sagittal plane, and the anterior and posterior acetabular sector angle in the axial plane; in addition, the difference between these parameters was investigated between the groups. The thresholds for femoral head collapse in the parameters that showed differences were investigated. Necrotic location and size were evaluated using the Japanese Investigation Committee (JIC) classification and the Steinberg grade classification, respectively. We examined the relationship between these parameters and classifications. The mean LCEA was slightly greater in the noncollapse group than in the collapse group (32° ± 6° versus 28° ± 7°; mean difference 4° [95% CI 1.15° to 6.46°]; p = 0.005); the clinical importance of this small difference is uncertain. There were no differences in PI between the two groups. After accounting for sex, age, BMI, and etiology as confounding factors, as well as acetabular coverage parameters and PI, we found a lower LCEA to be independently associated with increased odds of collapse, although the effect size is small and of questionable importance (OR 1.18 [95% CI 1.06 to 1.33]; p = 0.001). The threshold of LCEA for femoral head collapse was 28° (sensitivity = 0.79, specificity = 0.60, area under the curve = 0.73). The percentage of patients with an LCEA less than 28° was larger in JIC Type C1 (OR 6.52 [95% CI 1.64 to 43.83]; p = 0.006) and C2 (OR 9.84 [95% CI 2.34 to 68.38]; p = 0.001) than in patients with both Type A and Type B. The acetabular coverage data for the excluded patients did not differ from those of the patients included in the analysis. Our findings suggest that acetabular coverage appears to have little, if any, association with the likelihood of collapse in patients with ONFH. We found a small association between a lower LCEA and a higher odds of collapse, but the effect size may not be clinically important. Factors other than acetabular coverage need to be considered, and if our findings are verified by other investigators, osteotomy is unlikely to have a protective role. As the patients in our study were fairly homogeneous in terms of ethnicity and BMI, these factors need to be further investigated to determine whether they are associated with femoral head collapse in ONFH. Level III, prognostic study.
Read full abstract