Radiographic Assessment of Guided Growth: The Correlation Between Screw Divergence and Change in Anatomic Alignment.

BACKGROUND Assessment of changes in anatomic alignment following guided growth traditionally utilizes full-length standing radiographs which subjects patients to larger radiation doses than does a single anteroposterior radiograph of the knee. In an effort to minimize radiation exposure, the present study sought to determine whether changes in screw divergence (SD) of the 2-hole tension band plate used for hemiepiphysiodesis reliably predicts change in alignment. METHODS A retrospective review was conducted involving all patients with genu varum or genu valgum treated with hemiepiphysiodesis at a single institution. Preoperative anatomic alignment of the femur, using anatomic lateral distal femoral angle (aLDFA) and anatomic femoral-tibial angle (aTFA), and intraoperative divergence of hemiepiphysiodesis screws were compared with postoperative imaging. Linear regression analysis determined the relationship between changes in SD and changes in alignment, and multivariate regression analysis explored the relationship between the angular changes being measured and various demographic factors. RESULTS Linear regression analysis revealed that for every 1 degree change in SD there was a resultant 1.80 degrees of change in aTFA and 2.11 degrees of change in aLDFA. Change in aTFA is predicted by the equation: [INCREMENT]aTFA=0.41×|[INCREMENT]SD|+1.39. The change in aLDFA was predicted by the equation [INCREMENT]aLDFA=0.27×[INCREMENT]SD+1.84 with a R2 of 0.31. [INCREMENT]aTFA and [INCREMENT]SD had a correlation coefficient of 0.68 (95% confidence interval, 0.54-0.78.) [INCREMENT]aLDFA and [INCREMENT]SD had a correlation coefficient of 0.56 (95% confidence interval, 0.42-0.68). [INCREMENT]SD and sex were the only 2 independent predictors for [INCREMENT]aLDFA and [INCREMENT]aTFA as determined by multivariate regression analysis. CONCLUSION Change in coronal plane anatomic alignment in patients being treated for genu valgum or genu varum with hemiepiphysiodesis can be reasonably estimated by measuring the change in SD. Therefore, when following patients postoperatively, focal radiographic imaging of the knee can be utilized in lieu of standing full-length limb radiographs to limit radiation to the pelvis in this sensitive patient population. LEVEL OF EVIDENCE Level III-retrospective comparative study.