Evaluating the REST Scoring Method (Radiological Evaluation Score for Bone Tumors) as a Novel Radiological Tool in Distinguishing Benign and Malignant Bone Tumors: Accuracy Study at a Tertiary Referral Hospital

Abstract

Introduction: The misdiagnosis due to high mimicking of bone tumor malignancy and uneven access to advanced examinations in rural areas are yet several problems in bone tumor examinations. Not to mention, the high morbidity and mortality rates indicate obstacles to these cases. This study analyses the accuracy of X-rays processed using a new modality, Radiological Evaluation Score for Bone Tumors (REST) in primary bone tumor patients at Dr. Soetomo Regional Hospital Surabaya. Methods: This study was an analytical observational study that analyzed the correlation between each scoring criterion and its total score with histopathology (biopsy) results, calculated the cut-off value for benign and malignant type scoring, and diagnosis accuracy. The sample totaled 142 samples by a total sampling technique. Analysis calculation and variables were explained in the chapter. Discussion: Based upon the obtained p values, it showed that 1) there was no correlation between age, cortical erosion, and pathological fractures with histopathological results (p ≥ 0.05); 2) there were correlation between gender, characteristic, tumor content, lesion boundaries, transition zone, periosteal reaction, connective tissue mass, and the total scoring with histopathological results (p < 0.05); 3) the cut-off value was 3.5; 4) the diagnostic accuracy was 85.3%. The kappa value for interobserver reliability was 0.764. Conclusion: The correlation of REST scoring with the histopathology results has shown dominantly significant with considered good diagnostic accuracy. These results have demonstrated a promising future development for the utilization of X-ray as bone tumour diagnosis modality.