Dual-Energy X-Ray Absorptiometry (DEXA) Scan Versus Computed Tomography for Bone Density Assessment

Rationale and objective Osteoporosis, a common non-pathological disease of bones, has been the cause of many disastrous consequences, in terms of physical, psychological, social, and economic loss. Therefore, it is crucial to diagnose it early for timely prevention and treatment of osteoporotic fractures. Dual-Energy X-Ray Absorptiometry (DEXA) is currently routinely used for determining bone mineral density. However, it has its limitations. Nowadays, CT technology has advanced so rapidly that the Hounsfield units (HU) values can be used in opportunistic screening for osteoporosis in patients during routine CT abdomen for other causes. Hence, there would be no need for additional study with DEXA and also reduce radiation exposure. The aim of our research is to determine whether there is a correlation between the bone mineral density and the T-score measured by DEXA and the HU values measured from the diagnostic CT images of L1-4 vertebrae. Also, to determine reference CT values that would help in screening the patients with osteoporosis. Materials and methods We conducted a retrospective study of 78 female patients who underwent CT lumbar spine, abdomen, and pelvis in our hospital between the years 2016-2020. We collected data of patients who performed DEXA and CT scans within an interval of up to two years. The final collected data was analyzed to find correlation values of HU with age group and with DEXA bone mineral density (BMD) and T-score using Pearson correlation coefficient. Results The mean of the 78 patients was 61.1 (range 37-88 years). Mean HU values decreased consistently with age, from 202.17 HU in the fifth decade to 71 HU in the ninth decade. Average L1-4 HU values ranged from 71 HU to 202.17 HU (mean with standard deviation), while their T-score ranged from -4.4 to +2.4 (mean was -1.7±1.41), and their BMD ranged from 0.62 to 1.465 g/cm2 (mean, 0.974±0.175 g/cm2). For each lumbar vertebra, the correlations of HU values with bone mineral density and T-score were calculated separately. For L1-4 vertebrae, the correlation coefficients (r2) for the HU value and T-score were 0.544, 0.600, 0.611, and 0.600, respectively. The correlation coefficients (r2) for the HU value and bone mineral density were 0.581, 0.623, 0.653,0.612, respectively. All the calculated correlations were significant (p<0.001). Therefore, it was concluded that there was a positive correlation between the HU values and the DEXA for the BMD and between the HU values and the T-score. Based on the WHO guidelines, the T-scores of the lumbar vertebrae were classified into three groups. The mean HU values for the subjects in the normal group were 174.05 (95% confidence interval, 153-194.49), in the osteopenia group were 120.45 HU (95% confidence interval, 106.98-133.91), and in the osteoporosis group were 115 HU (95% confidence interval, 104.60-125.40). The differences in the mean HU values between the groups were significant. Conclusion On analyzing the results of our study, we reached the conclusion that there is a positive correlation between the HU calculated from CT with automated exposure control and BMD calculated from the DEXA. Thus CT scans done for various reasons, for example, the abdomen, lumbar spine, etc. can provide us with information about the patient’s bone density as well. CT is a very popular, easily accessible, reproducible, and reliable tool for measuring HU values and thereby in the opportunistic screening of osteoporosis.