A population-based study of scoliosis among males diagnosed with a dystrophinopathy identified by the Muscular Dystrophy Surveillance, Tracking, and Research Network (MD STARnet).

ABSTRACT

INTRODUCTION/AIMS: Scoliosis is a common comorbidity among individuals diagnosed with a dystrophinopathy. We examined associations between clinical predictors and scoliosis in childhood-onset dystrophinopathy. METHODS: The progression and treatment of scoliosis were obtained from data collected by the US population-based Muscular Dystrophy Surveillance, Tracking, and Research Network. Associations between loss of independent ambulation (LoA) and corticosteroid use and scoliosis outcomes (ages at or exceeding Cobb angle thresholds [10°, 20°, 30°]; surgery) were estimated using Kaplan-Meier curve estimation and extended Cox regression modeling. RESULTS: We analyzed curvature data for 513 of 1054 individuals ascertained. Overall, approximately one-half had at least one radiograph and one-quarter had a curvature of at least 20°. The average maximum curvature was 25.0° (SD = 21.5°) among all individuals and 42.8° (SD = 18.8°) among those recommended for surgery. Higher adjusted hazards ratio of curvature (aHR(curvature) [95% confidence interval]) were found among individuals with LoA compared to those without LoA (aHR(10)  = 6.2 [4.4, 8.7], aHR(20)  = 15.3 [7.4, 31.7], aHR(30)  = 31.6 [7.7, 128.9]), among individuals who did not use corticosteroids compared to those who did (aHR(10)  = 1.2 [0.9, 1.7], aHR(20)  = 1.8 [1.1, 2.7], aHR(30)  = 2.3 [1.3, 4.0]), and among non-ambulatory individuals who used corticosteroids after LoA compared to those who did not (aHR(10)  = 1.8 [1.2, 2.8], aHR(20)  = 1.6 [1.0, 2.6], aHR(30)  = 3.6 [1.6, 7.9]). Scoliosis surgery among individuals with LoA who did not use corticosteroids was more than double compared to those who used (aHR = 2.3 [1.3, 4.2]). DISCUSSION: Our retrospective observational study suggests corticosteroids may delay spinal curvature progression and need for scoliosis surgery. Continuing corticosteroids after LoA also showed potential benefits of delaying curvature progression, additional studies are needed to confirm this finding or address the magnitude of benefit.