CT Hounsfield Unit Measurements Around the Hip and Their Utility for Opportunistic Screening: A Systematic Review
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Background: Computed tomography scans (CTs) are routinely obtained before hip preservation surgery or to evaluate pelvic and/or hip fractures. Analyzing bone mineral density (BMD) using CT Hounsfield units (HU) before surgical intervention may improve surgical management and outcomes. This systematic review summarizes the current literature on reliability and placement of regions of interest (ROIs) within the hip, pelvis, and sacrum, and their correlation with local and systemic BMD. Methods: A systematic review in PubMed, Scopus, and Web of Science was performed. Twenty-four studies reported intraclass correlation coefficients of HU measurements and/or compared the HU value obtained from CT and t-score of DXA for predicting regional BMD within the hip, pelvis, and sacrum. Results: Twenty-four studies were included, enrolling 7,035 patients. The HU measurement was most frequently made at the femoral neck (n = 10). Femur and pelvis intrarater and interrater reliability was moderate to excellent. Moderate to strong correlations between femoral head and femoral neck CT HU and DXA, and poor to moderate correlations between the pelvis CT HU and DXA were found. Conclusion: A variety of ROIs have been described around the pelvis and proximal femur all with reasonable reliability for the measurement of CT HU. The strongest correlations between CT HU and DXA were reported between HU of the femoral neck and DXA of the hip/femur suggesting this ROI may be used for opportunistic BMD scre
Abstract
Background: Computed tomography scans (CTs) are routinely obtained before hip preservation surgery or to evaluate pelvic and/or hip fractures. Analyzing bone mineral density (BMD) using CT Hounsfield units (HU) before surgical intervention may improve surgical management and outcomes. This systematic review summarizes the current literature on reliability and placement of regions of interest (ROIs) within the hip, pelvis, and sacrum, and their correlation with local and systemic BMD. Methods: A systematic review in PubMed, Scopus, and Web of Science was performed. Twenty-four studies reported intraclass correlation coefficients of HU measurements and/or compared the HU value obtained from CT and t-score of DXA for predicting regional BMD within the hip, pelvis, and sacrum. Results: Twenty-four studies were included, enrolling 7,035 patients. The HU measurement was most frequently made at the femoral neck (n = 10). Femur and pelvis intrarater and interrater reliability was moderate to excellent. Moderate to strong correlations between femoral head and femoral neck CT HU and DXA, and poor to moderate correlations between the pelvis CT HU and DXA were found. Conclusion: A variety of ROIs have been described around the pelvis and proximal femur all with reasonable reliability for the measurement of CT HU. The strongest correlations between CT HU and DXA were reported between HU of the femoral neck and DXA of the hip/femur suggesting this ROI may be used for opportunistic BMD screening using preoperative hip/pelvis CT scans. Level of Evidence: III.
