Does Early Proximal Femoral Varus Osteotomy Shorten the Duration of Fragmentation in Perthes Disease? Lessons From a Prospective Multicenter Cohort

Background:The prognosis of Legg-Calvé-Perthes disease (LCPD) is dependent upon several factors, with the length and severity of the fragmentation stage among the most important. Previous retrospectively collected data from a single center have suggested that early proximal femoral varus osteotomy (PFO) may shorten the length of fragmentation and allow 34% of patients to bypass fragmentation altogether resulting in less femoral head deformity. The purpose of this study was to validate these findings in a prospectively collected multicenter cohort.Methods:Patients with LCPD treated with early PFO (during Waldenström stage I) were prospectively followed with serial radiographs at 3-month intervals until a minimum of 2-year follow-up. Waldenström stages and lateral pillar class were determined by mode assessments from 3 pediatric orthopaedic surgeons. The duration of fragmentation was defined as the interval between the first radiographs demonstrating features of stage IIa and stage IIIa. "Complete" bypass was defined as the absence of stage IIa or IIb findings on sequential radiographs with no development of femoral head deformity or collapse. "Partial" bypass was defined as the absence of advanced features of fragmentation and femoral head collapse (stage IIb).Results:Forty-six patients (80% male individuals) with initial stage LCPD and a mean age of 8.2±1.2 years were identified. The weighted kappa statistics for Waldenström staging and lateral pillar classifications showed excellent (0.833) and substantial (0.707) agreement, respectively. Ninety-eight percent of patients (45/46) underwent some period of fragmentation lasting between 91 and 518 days; the median duration was 206 days (interquartile range, 181 to 280). One patient (2%) bypassed fragmentation completely; 8 patients (17%) demonstrated partial bypass. Patients who completely or partially bypassed fragmentation experienced significantly less severe lateral pillar collapse (P=0.016) and shorter fragmentation duration (P=0.001).Conclusions:In this prospective multicenter cohort, we found a lower rate of fragmentation bypass than previously reported. Nonetheless, our data support the previous contention that early PFO may shorten fragmentation and minimize collapse in LCPD compared with historical controls. Further study with larger cohorts and a more rigorous definition of what constitutes bypass is warranted to clarify the effect of early PFO on the reparative biology of LCPD.Level of Evidence:Level IV - therapeutic study.