Background: Legg-Calvé-Perthes disease (LCPD) is a childhood hip disease characterized by osteonecrosis of the femoral head. Because severe deformity of the femoral head can cause secondary osteoarthritis in adulthood, progressive collapse should be prevented in children with a necrotic epiphysis. The prognosis of patients with LCPD generally worsens as the age at disease onset increases, and the appropriate treatment for late-onset LCPD remains unclear. Based on the limited effect of nonoperative treatment using a nonweightbearing brace, flexion varus osteotomy (FVO) was introduced in 2010 as an initial treatment for late-onset LCPD in place of brace treatment, which we used in our institution before that time., Questions/purposes: We asked, (1) Which treatment, FVO or a nonweightbearing brace, is associated with a lower likelihood of progressive femoral head collapse in children whose diagnosis of LCPD was made at the age of ≥ 8 years and who were followed for a minimum of 3 years after their intervention? (2) What proportion of patients in the brace group had surgery despite the treatment, and what percentage of children in the FVO group had a second operation to remove hardware and/or additional operations?, Methods: The initial treatment was applied in 181 patients with LCPD between 1995 and 2018 in our institution. Patients whose disease onset was at ≥ 8 years old (late-onset LCPD) with complete clinical and radiologic data were considered potentially eligible. In 2010, treatment for these patients changed from brace treatment to FVO for all patients. A total of 35% (42 of 121) of patients who were treated with a nonweightbearing brace between 1995 and 2009 and 40% (24 of 60) of patients who were treated with FVO between 2010 and 2018 were eligible. Among patients treated with a brace, 21% (nine of 42 patients) were excluded because of hospital transfer (three patients), short-term follow-up (three), the period from onset to the first visit was ≥ 7 months (two), and inability to use the brace because of mental incapacity (one patient). In patients treated with FVO, 12% (three of 24 patients) were excluded (two patients with a period from onset to the first visit ≥ 7 months and one with a comorbidity and multiple-epiphyseal dysplasia). Among the remaining patients, 79% (33 of 42 patients) were classified into the brace group and 88% (21 of 24 patients) were classified into the FVO group for analyses. There were no overlapping patients at the timepoint when the treatment strategy for late-onset LCPD changed. In the FVO group, subtrochanteric osteotomy with 35° to 40° of flexion and 15° to 20° of varus was performed using a locking compression plate for pediatric use. Patient demographics, radiographic parameters, and the assessment of femoral head deformity using the Stulberg classification were compared between the two groups. There was a greater proportion of boys than girls in both groups (brace: 88% and FVO: 86%), and there were no differences in the distribution of genders between the groups (p = 0.82). The right side was more frequently treated in the brace group, but there was no difference in laterality between the groups (brace: 58% right and FVO: 62% left; p = 0.16). There was no difference between groups in the median age at disease onset (9.0 years [range 8.0 to 12.5 years] in the brace group and 9.6 years [range 8.0 to 12.4 years] in the FVO group; p = 0.26). There was no difference between the groups in the period of treatment from onset (1.7 ± 1.9 months in the brace group and 1.5 ± 1.5 months in the FVO group; p = 0.73) or the follow-up period (6.7 ± 2.1 years in the brace group and 6.2 ± 2.1 years in the FVO group; p = 0.41). The LCPD stage at the first visit was assessed using the modified Waldenström classification. The intraobserver and interobserver values of the modified Waldenström classification, evaluated using kappa statistics, were excellent (kappa value 0.89 [95% CI 0.75 to 0.97]; p < 0.01) and good (kappa value 0.65 [95% CI 0.43 to 0.87]; p < 0.01). The radiographic degree of collapse at the maximum fragmentation stage was assessed using the lateral pillar classification. The intraobserver and interobserver reliabilities of the lateral pillar classification were excellent (kappa value 0.84 [95% CI 0.73 to 0.94]; p < 0.01) and excellent (kappa value 0.83 [95% CI 0.71 to 0.94]; p < 0.01). The degree of femoral head deformity at the most recent follow-up examination was compared between the groups in terms of the Stulberg classification, in which Classes I and II were classified as good and Classes III through V were classified as poor. The intraobserver and interobserver reliabilities of the Stulberg classification were good (kappa value 0.74 [95% CI 0.55 to 0.92]; p < 0.01) and good (kappa value 0.69 [95% CI 0.50 to 0.89]; p < 0.01). The evaluators were involved in the patients' clinical care as part of the treating team., Results: Good radiographic results (Stulberg Class I or II) were obtained more frequently in the FVO group (76% [16 of 21 patients]) than in the brace group (36% [12 of 33 patients]), with an odds ratio of 5.6 (95% CI 1.7 to 18.5; p < 0.01). In the brace group, a subsequent femoral varus osteotomy was performed in 18% (six of 33) of patients with progressive collapse and hinge abduction, and implant removal surgery was performed approximately 1 year after the first procedure. This traditional varus osteotomy was occasionally performed in patients who were considered for conversion from nonoperative treatment before 2009 because FVO had not yet been introduced. In the FVO group, all patients (n = 21) had a second procedure to remove the implant at a mean of 10.5 ± 1.2 months postoperatively. Additional procedures were performed in 24% (five of 21) of patients, including a second FVO for progressive collapse (one patient), guided growth for a limb length discrepancy (one patient), and flexion valgus osteotomy for coxa vara in patients with a limb length discrepancy (three patients)., Conclusion: Our historical control study found that FVO may increase the possibility of obtaining good radiographic results (Stulberg Class I or II) compared with brace treatment for patients with late-onset LCPD, although surgical interventions after the first and second implant removal procedures may be indicated. Surgeons can consider FVO if they encounter patients with late-onset LCPD, which is a challenging condition. A larger study with long-term follow-up is needed to confirm the efficacy of FVO., Level of Evidence: Level III, therapeutic study., Competing Interests: Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members. All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request., (Copyright © 2022 by the Association of Bone and Joint Surgeons.)