Establishing Age- and Sex-Specific Norms for Pediatric Return-to-Sports Physical Performance Testing.

Abstract

Background:Graft tears and contralateral anterior cruciate ligament (ACL) tears are common in pediatric athletes after ACL reconstruction. Use of objective return-to-sports (RTS) criteria, in particular physical performance tests (PPTs), is believed to reduce the incidence of secondary injury; however, pediatric norms for these tests are unknown.Purpose:To establish a proof of concept for the creation of age- and sex-based norms for commonly used RTS PPTs in healthy pediatric athletes, allowing the creation of growth curves for clinical referencing.Study Design:Cross-sectional study; Level of evidence, 3.Methods:A total of 100 healthy people who were between the ages of 6 and 18 years and involved in organized sports were enrolled, with even distributions of age and sex. All participants underwent 9 common RTS PPTs: stork test, stork test on Bosu, single-leg squat, single-leg squat on Bosu, clockwise and counterclockwise quadrant hops, single-leg hop for distance, 6-m timed hop, and triple crossover hop for distance. Mean performance across limbs was calculated for each individual. Chronological age, height, weight, sex, and self-reported Pubertal Maturational Observational Scale (PMOS) score were recorded. Univariable and multivariable models were created for each PPT, assessing the importance of the recorded descriptive variables. Quantile regression was used to create growth curves for each PPT.Results:The cohort was 52% female, and the mean ± standard deviation age was 11.7 ± 3.6 years. PMOS was highly correlated with age (r = 0.86) and was excluded from the regressions. In univariable regression, age, height, and weight were strong predictors of performance for all PPTs, whereas sex was a predictor of performance on the single-leg and triple crossover hops for distance (with males outperforming females). Height and weight were excluded from multivariable regression because of multicollinearity with age. Multivariable regression showed predictive patterns for age and sex that were identical to those shown in the univariable analysis. Given ceiling effects, quantile regression for the stork tests was not possible, but quantile regression growth curves were successfully created for the 7 remaining PPTs.Conclusion:Chronological age and sex accurately predicted performance on common RTS PPTs in pediatric patients. The growth curves presented herein could assist clinicians with benchmarking pediatric patients postoperatively against a healthy athletic cohort.