Significance of peak height velocity as a predictive factor for curve progression in patients with idiopathic scoliosis

BACKGROUND Much attention has been paid to peak height velocity (PHV) as a possible predictor of curve progression in patients with idiopathic scoliosis (IS). The aim of this study was to analyze the relationship between the magnitude of the Cobb angle at PHV and scoliosis progression, defined as having surgery prior to skeletal maturity in female patients with IS. METHODS A retrospective review identified 56 skeletally immature female IS patients who were followed until maturity. The mean age and the mean pubertal status at the initial visit were 10 years and 24 months before menarche respectively, with a follow-up period of 5 years. They were divided into two groups: non-surgery group (NS) and surgery group (S), depending on their treatment method in use at the final follow-up visit. Surgery group was defined as an ultimately having surgery due to Cobb angle greater than 45 degrees prior to skeletal maturity regardless of conservative management. Height measurements were recorded at each visit; height velocity was calculated as the height change, in cm, divided by the time interval, in years. The PHV, chronological age at PHV (APHV), height at PHV (HPHV), and final height (FH) were determined for each group. In patients with Cobb angle greater than 30 degrees, the corrected height was calculated by Kono formula and corrected height velocity values were provided. The sensitivity, specificity, and area under the curve (AUC) of the receiver-operating -characteristic (ROC) analysis were calculated to predict spinal curve progression for various Cobb-angle cutoff values at PHV. RESULTS The corrected PHV had a mean value of 8.5 and 8.9 cm/year in the NS-group and S-group, respectively. The APHV was 11.9 and 11 years, the corrected HPHV was 152.9, and 149.3 cm, and the corrected FH was 159.9 and 159.3 cm, respectively. When a Cobb angle of 31.5 degrees was at PHV, ROC analysis revealed 78% sensitivity, 82% specificity, and an AUC of 0.93, acceptable values for curve progression in patients with IS. CONCLUSIONS These findings indicate that 31.5 degrees of spinal curvature when patients are at PHV is a significant predictive indicator for progression of the curve to a magnitude requiring surgery. We suggest that the curve-progression risk assessment in patients with IS should include PHV, along with measures of skeletal and non-skeletal maturities.