The Results of Epiphysiodesis for Treatment of Limb Length Discrepancy

BACKGROUND: Epiphysiodesis is a pediatric surgical procedure often used to correct a 2-5 cm limb length discrepancy (LLD) by surgically arresting physeal growth of the longer limb. OBJECTIVE: The purpose of the study is to compare three models commonly used to calculate expected LLD in patients who have undergone an epiphysiodesis to determine which most accurately predict the patient outcomes. METHODS: Out of 469 patients who have undergone an epiphysiodesis for LLD at the Texas Scottish Rite Hospital for Children between1991-2011, 84 patients were selected for the study. The inclusion criteria were availability of three pre-operative x-rays and skeletal age data before the surgery, with at least a six months interval between scans. Each patient must have been followed to skeletal maturity with limb length measurements at that time and must not have suffered growth arrest of traumatic etiology or had postoperative complications. Using radiographic limb measurements, we compared the accuracy of the growth remaining methods by assessing predicted limb lengths and predicted LLD with White-Menelaus, Moseley/Rotterdam, and Green-Anderson methods or their variants. Where applicable, the predictions were made with and without accounting for growth inhibition rate (GIR), which is a quantification of the retarded rate of growth in the shorter leg. The differences across methods were assessed with one-way repeated measures ANOVA and reliability measures using intraclass correlation coefficients (ICC). A p-value < 0.05 indicated statistical significance. RESULTS: Analysis comparing the mean absolute difference between the predicted and actual outcomes for each method demonstrates a similar efficacy. CONCLUSION: The White-Menelaus, Moseley/Rotterdam, and Green-Anderson methods of prediction are nearly indistinguishable, with the White-Menelaus being slightly more accurate. Modifying the existing models to include GIR did not improve the accuracy of the predictions. We advocate that the White-Menelaus be the preferred method as it is as accurate as the others and, from the clinical perspective, has the advantage of simplicity: requiring only one measurement and the patient's chronological age, as opposed to skeletal age. It also does not require the physician to create graphs or consult growth charts.