Introduction: The natural history of idiopathic scoliosis is not well understood. Previous reports focused on characteristics of curve progression pre-defined at 5–6 degrees. However, the absolute curve magnitude at skeletal maturity is more predictive of long-term curve behavior rather than progression of defined magnitude over shorter periods of growth. It is generally agreed that curves < 30 degrees are unlikely to progress after skeletal maturity. Hence, defining factors that influence curve progression to an absolute magnitude of ≥30 degrees at skeletal maturity significantly aids clinical decision-making.

Methods: Of 279 patients with idiopathic scoliosis detected by school screening of 72,699 adolescents, 186 fulfilled the study criteria and were followed up to skeletal maturity. Initial age, gender, pubertal status and initial curve magnitude were used as predictive factors for curve progression to ≥30 degrees at skeletal maturity. Uni and multivariate, logistic regression and receiver operating characteristic (ROC) analysis was performed.

Results: Curve magnitude at first presentation was the most important predictive factor for curve progression to ≥30 degrees at skeletal maturity. An initial curve of 25 degrees had the best ROC of 0.8 with a positive predictive value of 68% and a negative predictive value of 92% for progression to ≥30 degrees at skeletal maturity. The highest risk was a pre-pubertal female < 12 years of age with a Cobb of ≥25 degrees at presentation; with an 82% chance of progression to a Cobb of ≥30 degrees. Probability of progression to ≥30 degrees was defined by 1/(1 + exp (−z)). [z = −3.709 + 0.931(Gender) + 0.825(Puberty) + 3.314(Cobb) + 0.171(Age)].

Conclusions: Initial curve magnitude is the most important independent predictor of long-term curve progression past skeletal maturity. An initial Cobb of 25 degrees is an important threshold. Combined with other factors, we identify patient profiles with high or low risk for progression.

Study Design: Prospective cohort study

Objective: To determine the rate of curve progression and factors related to curve progression in untreated adolescent idiopathic scoliosis in a prospective cohort study within a national school screening program.

Methods: Over 140,000 school children are screened annually in Singapore for common health conditions, including scoliosis. In 1996–97, a randomized sample in four age groups consisting of 72,699 children was enrolled in a scoliosis prevalence study, 263 were found to have spinal curvatures of 10 degrees or more. After exclusions, 250 children were followed up over a five year period up to 2001–02. Basic demographic data, age at menarche or break of voice, scoliometer reading, curve type, Cobb angle, curve rotation, and Risser grade were recorded. Curve progression was correlated to individual factors such as age, sex, puberty, curve type and magnitude as well as combinations of factors.

Results: Overall, 28% of the 250 curves progressed. Age at diagnosis, sex, pre-menarche status, and curve magnitude were statistically correlated to curve progression. Taking curve magnitude and age together, 53% of 11–12 year-olds with curves 20 degrees or more progressed compared to 10% of 13–14 year-olds with curves less than 20 degrees. 56% of children with curves 20 degrees or more and Risser grades 0–2 progressed, compared to 17% with curves less than 20 degrees and Risser grades 3–5. Combining curve magnitude, age, sex, and puberty together, a pre-pubertal female under 13 years old with a curve of 25 degrees or more has a 70% chance of curve progression. In comparison, a post-pubertal female older than 13 years of age and a curve of less than 25 degrees has only a 10% chance of progression. Curve progression in adolescent idiopathic scoliosis has been reported to vary from 5.2% to 56%, with the lower rates being found in school screening studies. Nachemson et al (1982) reported that 10–12 year old girls with untreated scoliosis of 20–29 degrees had a 60% risk of curve progression. Lonstein and Carlson (1984) reported progression in 23.2% of untreated children and that curve magnitude, skeletal immaturity, and curve pattern were associated with progression.

Conclusions: Our findings are similar, with pre-pubertal females under the age of 13 years old and with large curves at diagnosis having the greatest risk of progression.