Predictors of spine deformity progression in adolescent idiopathic scoliosis: A systematic review with meta-analysis

AIM: To evaluate published data on the predictors of progressive adolescent idiopathic scoliosis (AIS) in order to evaluate their efficacy and level of evidence. METHODS: Selection criteria: (1) study design: randomized controlled clinical trials, prospective cohort studies and case series, retrospective comparative and none comparative studies; (2) participants: adolescents with AIS aged from 10 to 20 years; and (3) treatment: observation, bracing, and other. Search method: Ovid MEDLINE, Embase, the Cochrane Library, PubMed and patent data bases. All years through August 2014 were included. Data were collected that showed an association between the studied characteristics and the progression of AIS or the severity of the spine deformity. Odds ratio (OR), sensitivity, specificity, positive and negative predictive values were also collected. A meta-analysis was performed to evaluate the pooled OR and predictive values, if more than 1 study presented a result. The GRADE approach was applied to evaluate the level of evidence. RESULTS: The review included 25 studies. All studies showed statistically significant or borderline association between severity or progression of AIS with the following characteristics: (1) An increase of the Cobb angle or axial rotation during brace treatment; (2) decrease of the rib-vertebral angle at the apical level of the convex side during brace treatment; (3) initial Cobb angle severity (> 25 degrees); (4) osteopenia; (5) patient age < 13 years at diagnosis; (6) premenarche status; (7) skeletal immaturity; (8) thoracic deformity; (9) brain stem vestibular dysfunction; and (10) multiple indices combining radiographic, demographic, and physiologic characteristics. Single nucleotide polymorphisms of the following genes: (1) calmodulin 1; (2) estrogen receptor 1; (3) tryptophan hydroxylase 1; (3) insulin-like growth factor 1; (5) neurotrophin 3; (6) interleukin-17 receptor C; (7) melatonin receptor 1B, and (8) ScoliScore test. Other predictors included: (1) impairment of melatonin signaling in osteoblasts and peripheral blood mononuclear cells (PBMC); (2) G-protein signaling dysfunction in PBMC; and (3) the level of platelet calmodulin. However, predictive values of all these findings were limited, and the levels of evidence were low. The pooled result of brace treatment outcomes demonstrated that around 27% of patents with AIS experienced exacerbation of the spine deformity during or after brace treatment, and 15% required surgical correction. However, the level of evidence is also low due to the limitations of the included studies. CONCLUSION: This review did not reveal any methods for the prediction of progression in AIS that could be recommended for clinical use as diagnostic criteria.