OF-score for osteoporotic thoracolumbar fractures : which parameter is decisive for the therapy decision? : A prospective multicentric cohort study

Abstract

Introduction

Various treatment strategies for osteoporotic vertebral compression fractures (OVCF) have been proposed, with the OF-score being a promising tool for therapeutic decision-making. The aim of this study is to determine the significance of the individual components of the OF-score for the therapy recommendation. Furthermore, score modifications will be sought that lead to improved agreement between the score recommendation and clinically successful therapy. Methods

Data from the prospective multicenter study “Clinical evaluation of the OF-score for therapy planning and treatment recommendation for osteoporotic fractures of the thoracolumbar spine” (EOFTT) with 518 patients (128 male, 390 female), including 344 surgically and 174 conservatively treated, were used. Binary discriminant analysis and logistic regression models were applied to analyze the predictive power of OF-score variables for the treatment decision. ROC analysis determined the predictive value of VAS pain, with thresholds defined using the Youden index. Results

A total of 508 patients were analyzed. Binary discriminant analysis showed an explanatory power of 79.7%, and logistic regression showed 80.4%. The most powerful variables were VAS pain (0.809), mobilization (−0.405), and OF classification (0.302). ROC analysis identified a VAS threshold of 5.5 for surgical treatment (AUC = 0.811, p < 0.001). A VAS threshold of ≥ 5 yielded the highest accuracy (71.5%), while thresholds ≥ 4 had the highest sensitivity (71.9%) and ≥ 6 the highest specificity (82.0%). Conclusions

A modified OF-score with a VAS pain threshold of ≥ 5 improves accuracy and balances specificity and sensitivity. Despite this adjustment, the OF-score’s predictive power of 80% remains underutilized. Further research could enhance the clinical utility of the modified score and explore the potential of other variables.