Predicting Growth Quantitatively Through Proximal Tibia Radiograph Markers.

Background: The creation of accurate markers for skeletal maturity has been of significant interest to orthopaedic surgeons. They guide the management of diverse disorders such as adolescent idiopathic scoliosis, leg length discrepancy, cruciate ligament injuries, and slipped capital femoral epiphysis. Multiple systems have been described to predict growth using radiographic skeletal markers; however, no such system has yet been developed for the proximal tibia. The purpose of this study was to establish quantitative radiographic parameters within the proximal tibia that can be used to assess degree of skeletal maturity.
Methods: From the Bolton Brush collection, 94 children, consisting of 49 girls and 4 boys between the ages of 3 and 18 years old, were followed annually throughout growth with serial radiographs and physical examinations. Final height at maturity was used to calculate the growth remaining at each visit. Multiple measurements for each knee radiograph were performed and correlated with the percentage of growth remaining. Tibial epiphysis width, tibial metaphysis width, and height of the lateral tibial epiphysis were measured on each film and the composite ratios between each of these sets of variables along with their respective accuracy and reliability were calculated. Single and multiple linear regression models were constructed to determine accuracy of prediction. Interobserver and intraobserver studies were performed with 4 investigators ranging from medical student to senior attending and calculated using the intraclass correlation coefficient. All 4 examiners measured all of the subjects and the ratios created were averaged.
Results: Tibial epiphysis width, tibial metaphysis width, and height of the lateral tibial epiphysis were all found to be strongly correlated with growth remaining with R values ranging from 0.57 to 0.84. In addition, all 3 ratios were found to be reliable with intraobserver and interobserver intraclass correlation coefficients ranging from 0.92 to 0.94 and 0.80 to 0.94, respectively. A multiple linear regression model demonstrated that combining these 3 ratios allows for a predictive R value of 0.917, showing that these ratios when combined were highly predictive of growth remaining. All findings were independent of sex (P=0.996).
Conclusions: We describe 3 measurements that can easily be obtained on an anteroposterior radiograph of the knee. We demonstrate that ratios of these variables can be measured reliably and correlate closely with remaining growth, independent of sex. Together, we believe that these factors will improve the accuracy of determining growth from lower extremity radiographs that include the proximal tibia.
Clinical Relevance: This study provides a new quantitative technique to evaluate growth in the lower extremity, which can inform a range of conditions including adolescent idiopathic scoliosis, leg length discrepancy, cruciate ligament injury, and slipped capital femoral epiphyses.