Review of scoliosis-specific exercise methods used to correct adolescent idiopathic scoliosis.

Only eight controlled studies had usable Cobb angle data, and many studies lacked baseline and ending angles with variability. The studies were too inconsistent to combine, so no reliable pooled estimate could be reported. The average PEDro score was 5.9, with scores from 3 to 9, and several studies lacked random allocation, concealed allocation, blinding, adequate follow-up, or intention-to-treat analysis. Four studies included bracing together with exercise, so the contribution of exercise alone could not be separated. Two studies could not be included in the quantitative analysis because data sets were incomplete, and adding them might have changed the findings. No study directly compared one scoliosis-specific exercise method with another. The review was limited to English-language articles and to Cobb angle outcomes, so it did not assess patient-reported outcomes.

Do not conclude that scoliosis-specific exercise reliably reduces adolescent spinal curve angles. The review could not combine the studies reliably, two confidence intervals crossed zero for SEAS versus standard exercise, and some exercise groups also received bracing.