Internal fixation of intertrochanteric fractures using the percutaneous compression plate: the event of cutting out and its prediction by the tip-apex-distance

Abstract

In a recent study, the analyzed percutaneous compression plate (PCCP) data identified the tip-apex-distance (TAD) as the most valuable risk factor for the reoperation rate and in particular for the event of cutting out of the cephalic screws after internal fixation of intertrochanteric fractures. However, it is unknown which potential distance between the tip of the screws and the apex of the femoral head is most valuable in predicting the event of cutting out. As the PCCP device has two cephalic screws, there are at least three different distances which can be measured, including the distance between the tip of the lower screw and the apex of the femoral head, the distance between the center of the line between the two cephalic screws and the apex, and finally between the tip of the upper screw and the apex. Our study was motivated by the question which measurement can provide the best results for the prediction of the event of cutting out of the cephalic screws. The digitalized postoperative X-ray films of 174 patients with 174 PCCP implants were analyzed, and we obtained three different measurements for every patient. The primary outcome was the proportion of those patients with a cutting out of the screws. A univariate logistic regression was performed to assess the impact of the TAD on the event of cutting out. To describe the degree of consistency among measurements, we calculated an intra-class correlation coefficient (ICC) and the corresponding 95 % confidence limits. For each of the different measurements, we performed a receiver operating characteristic (ROC) curve analysis and calculated the area under the curve (AUC) and the corresponding 95 % confidence limits. In order to identify the most valuable distance measure, we added pairwise comparisons of the AUCs. Out of 174 patients, six patients (3.4 %) were reoperated due to cutting out of the cephalic screws. In the univariate logistic regression model, the p value was 0.0012 (OR 0.874; 95 % CI 0.806 to 0.948). The ROC curve analysis showed similar results for the different measurements; however, the highest AUC value was obtained for the center position. In conclusion, all three measurements provide similar results and are suitable to predict the risk of cutting out of the cephalic screws.