Comparison of the Paley method using chronological age with use of skeletal maturity for predicting mature limb length in children.

BACKGROUND: Treating patients with congenital or acquired limb-length inequality requires accurate estimations of limb length at skeletal maturity. There is controversy over the best indicator of maturity to be used for limb-length calculations. Paley popularized the multiplier method, in which chronological age is used, which has the virtue of simplicity but does not account for the wide variance in timing of the adolescent growth spurt. The purpose of this study was to determine whether the use of chronological age or the level of skeletal maturity provides more accurate limb-length predictions.
METHODS: We identified patients with limb-length inequality, for whom scanograms had been obtained before and at maturity, and who had had no surgical procedures on their normal lower limb. Skeletal maturity was determined with use of the Greulich and Pyle atlas, Tanner-Whitehouse-3 method, and simplified stages described by Sanders et al. The length of the lower extremity was compared with the ultimate limb length and the actual multiplier (final limb length divided by current limb length) for each point in time. A linear model was used to determine the log-transformed multipliers for the level of skeletal maturity, and Paley's multipliers were used for chronological age. Residual standard errors were determined to compare the results of the methods. We also conducted piecewise linear regression on each of the methods and used the residual standard errors to rank their performance and cross-validated the results.
RESULTS: We identified twenty-four patients (twelve girls and twelve boys) who met the study criteria. Most subjects had had multiple scanograms along with skeletal age radiographs (average, 4.5) at different ages. When all ages are considered, the Paley method had the best overall performance, with residual standard errors that were typically =5 cm. However, the Paley method did not perform best for subjects at stage-2 skeletal maturity or above; in those cases, skeletal-maturity-based predictions had residual standard errors of <2 cm.
CONCLUSIONS: While the Paley method, which is based on chronological age, provides reasonable estimates of ultimate limb length for most patients, use of skeletal-maturity determinations appears to provide better predictions of mature limb length during adolescence.