Clinical Efficacy of Distal Tibial Hounsfield Unit for the Diagnosis of Osteoporosis in Ankle Fractures

Abstract

Category: Trauma; Ankle Introduction/Purpose: Ankle fracture is one of the most common fractures over the age of 50 years. However, there is a lack of research on osteoporotic ankle fractures compared to other osteoporotic fractures. Recently, some studies have been published that Hounsfield Unit(HU) measuring bone attenuation from CT scan is used for assessing the bone density of not only the hip and spine but also extremities such as the distal radius and proximal humerus. This study aims to evaluate the correlation between HU of distal tibia without ankle fracture and DEXA-BMD and determine the cut-off value of HU for osteoporosis diagnosis. Additionally, this study determines whether the HU of the distal tibia obtained from the preoperative CT scan of ankle fractures helps evaluate patients with osteoporotic ankle fractures. Methods: From June 2016 to August 2020, We retrospectively studied foot and ankle traumatic patients who are postmenopausal women or men over 50 years. They conducted a foot or ankle CT scan containing the axial images of the distal tibia and DEXA- BMD within three months. Patients with open fractures with severe displacement or arthritic change were excluded. The mean HU was obtained from two sequential axial images of the distal tibial metaphysis's largest region of interest (ROI), more than 1cm proximal to the tibial plafond. We divided the patients into Group A without ankle fractures and Group B with distal tibia or fibula fractures in the ankle. The correlation analysis was done between the distal tibial HU and DEXA-BMD. We obtained the cut-off value of the distal tibial HU for osteoporosis using receiver operating characteristic(ROC) curves and compared the results between the two groups. Results: Group A was 38 cases(mean age 69.9), and Group B was 59 (mean age 71.0). There were no statistical differences in patients' characteristics, the mean distal tibial HU, bone density, T-score, and rate of osteoporosis in BMD. All HU values in both groups showed excellent inter- and intra-observer reliability(All intraclass coefficients>0.9). Both groups had significant correlations between the distal tibial HU and DEXA-BMD, especially in the femur neck and L3 spine(r=0.706, 0.622, p< 0.001 in Group A, r=0.625,0.499, p< 0.001 in Group B). The cut-off value for osteoporosis of the distal tibial HU was 153.15(sensitivity 73.9%, specificity 86.7%; area under the curve(AUC) 0.841, 95% CI 0.716-0.965, p< 0.001) in Group A. The cut-off value of Group B was 167.35(sensitivity 66.7%, specificity 75.9%; AUC 0.737, 95% CI 0.610-0.865, p=0.002). Conclusion: The distal tibial HU values on the axial images on the CT showed a significant correlation with the central bone of DEXA-BMD. The distal tibial HU value showed excellent reliability with or without fracture. HU value can be easily obtained using a preoperative ankle CT scan without additional cost & radiation exposure to patients with ankle fractures. Therefore, it can help treat osteoporotic ankle fractures by screening for osteoporosis and deciding the fixation method of surgery.