Containment surgery and radiographic outcomes in children with Waldenström stage III LCPD

Photo by Sakarie Mustafe Hidig / Unsplash

Frontiers in Medicine Published April 30, 2026 Medically reviewed May 1, 2026 DOI ↗ By Dr. Sofia Müller, MD · Lifespan & Whole-Person Care

Key Takeaway

Note that containment surgery showed a trend toward favorable Stulberg outcomes in stage III LCPD, but results were not statistically significant.

This retrospective cohort study evaluated 58 children diagnosed with Waldenström stage III Legg-Calvé-Perthes disease (LCPD). The study compared a surgical group (n=35) undergoing containment surgery to a non-surgical group (n=23). Follow-up lasted until skeletal maturity, with a mean duration of 11.94 ± 5.31 years in the surgical group and 7.15 ± 2.68 years in the non-surgical group.

Regarding femoral head morphology via the modified Stulberg classification, the surgical group showed 34.3% Class I, 51.4% Class II, and 14.3% Class III. The non-surgical group showed 1t7.4% Class I, 43.5% Class II, and 39.1% Class III. Although an OR of 3.857 was noted, the difference in the proportion of acceptable outcomes did not reach statistical significance (P = 0.057).

Secondary outcomes regarding acetabular development showed significant differences. The modified Tönnis angle was significantly smaller in the surgical group (6.12° ± 6.27°) compared to the non-surgical group (11.63° ± 5.73°) with P = 0.002. Additionally, the acetabular head index (AHI) was higher in the surgical group (86.98% ± 8.99%) than in the non-surgical group (75.51% ± 6.70%).

Safety and adverse event data were not reported. The study reports an association between containment surgery and radiographic outcomes, but the primary outcome did not reach statistical significance.

Study Details

Study typeCohort

EvidenceLevel 3

PublishedApr 2026

View Original Abstract ↓

BackgroundThis study aimed to evaluate the long-term radiographic outcomes of containment surgery in children with Legg–Calvé–Perthes disease (LCPD) during the reossification stage.MethodsThis retrospective study included 58 children diagnosed with Waldenström stage III LCPD. Patients were divided into a non-surgical group (n = 23, mean age 8.56 ± 1.74 years) and a surgical group (n = 35, mean age 8.28 ± 1.29 years). All patients were monitored with serial anteroposterior pelvic and frog-leg lateral radiographs. At the final follow-up, femoral head morphology was assessed using the modified Stulberg classification, while acetabular development was evaluated using the modified Tönnis angle and the acetabular head index (AHI).ResultsAll 58 patients were followed until skeletal maturity. The mean follow-up duration was 7.15 ± 2.68 years in the non-surgical group and 11.94 ± 5.31 years in the surgical group. At final follow-up, in the surgical group (n = 35), the simplified three-group modified Stulberg classification showed 34.3% Class I (Excellent; Stulberg I–II), 51.4% Class II (Good; Stulberg III), and 14.3% Class III (Poor; Stulberg IV–V). In the non-surgical group (n = 23), the corresponding proportions were 17.4%, 43.5%, and 39.1%, respectively. Although a trend toward more favorable outcomes was observed in the surgical group, the difference in the proportion of acceptable outcomes did not reach statistically significant (P = 0.057, OR = 3.857). However, the surgical group exhibited a significantly smaller modified Tönnis angle (6.12° ± 6.27° vs. 11.63° ± 5.73°, P = 0.002) and a higher AHI (86.98% ± 8.99% vs. 75.51% ± 6.70%, P