Opportunistic screening for osteoporosis using hydroxyapatite measurements of the vertebral by thorax dual-energy spectral CT in postmenopausal females

Abstract

The purpose of this study was to evaluate the feasibility of opportunistic screening for osteoporosis in postmenopausal females using the dual-energy CT(DECT)-derived hydroxyapatite (HAP) concentration and CT value of L1-vertebra. 239 consecutive postmenopausal female patients were enrolled and underwent both chest DECT and Dual energy X-ray absorptiometry (DXA). According to the T-score of the 1st lumbar vertebra on DXA, patients were divided into the osteoporosis group (T $$\le$$ ≤ − 2.5, n = 112) and non-osteoporosis group (T $$>$$ > − 2.5, n = 127). The HAP values of the 1st lumbar vertebra were measured from the coronal-view HAP(Fat)-based material decomposition(MD) images, and CT values were measured on the 75 keV monochromatic image. The cutoff values of using HAP and CT value for diagnosing osteoporosis were obtained by drawing receiver operating characteristic (ROC) curves. Both HAP and CT value of the 1st lumbar vertebra had moderate-high correlation with bone-mineral-density measurement on DXA (HAP, r = 0.614; CT value, r = 0.625; all p < 0.01). The area-under-the-curve (AUC), sensitivity, specificity, PPV and NPV for diagnosing osteoporosis was 0.754, 0.714, 0.693, 0.68 and 0.752 using HAP (cutoff value: 142.05 mg/cm3) and 0.766, 0.741, 0.7, 0.685 and 0.754 using CT value (cutoff value: 132HU), respectively. HAP measurements on HAP(Fat)-based MD images in DECT could provide reasonably accurate BMD quantification for diagnosing osteoporosis in postmenopausal females. DECT prescribed for lung cancer screening could also provide opportunistic screening for osteoporosis, extending the clinical application of DECT without additional radiation to patients.

Vertebral and Femoral Bone Mineral Density (BMD) Assessment with Dual-Energy CT versus DXA Scan in Postmenopausal Females

“…Some experimental and clinical studies indicate that the density value of the base substance can effectively represent the material content when the chosen base material pairs closely match the actual material. Similar to the findings of Deng et al, we found that the HAP concentration using the HAP-fat material pair was correlated with BMD, and the CT value (HU) was also measured on the DECT monochromatic images that are less susceptible to beam hardening artifacts [24].…”

“…DXA may introduce variability in measurements due to overlying soft tissues, vascular calcifications, intestinal contents, and degenerative bone changes, potentially influencing results. Similar to other studies in the literature, we chose to focus exclusively on the evaluation of the trabecular component to avoid the potential for a false increase in BMD values due to the causes mentioned above [24,25]. Additionally, trabecular bone is more susceptible to involution in osteopenic or osteoporotic processes.…”

Quantitative measurements of dual-energy CT parameters in the diagnosis of osteoporosis – a highly sensitive and specific technique: An observational study

“…Wesarg et al [19] investigated localized force measurements in 29 cadaveric specimens and found a strong correlation between DECT and DXA in quantifying local bone density. Another study involving 239 postmenopausal female patients [20] indicated a moderate-to-high correlation between Hounsfield attenuation profile (HAP) and CT values of the first lumbar vertebra with BMD obtained through DXA (HAP, R = 0.614; CT value, R = 0.625; all P < .01). Therefore, DECT can employ post-image processing techniques, such as material separation imaging and virtual monoenergetic imaging, to quantitatively assess vertebral bone density, facilitating the measurement of bone mass reduction and the diagnosis of OP.…”

Updates on the Applications of Spectral Computed Tomography for Musculoskeletal Imaging

“…Yet, many studies note that DXA scanning is an underutilized technique, with screening rates as low as 30% in Medicare patients [59]. Further, there are many limitations to traditional DXA scanning, as overlying tissue, bowel contents, and vascular calcifications can yield variations in BMD values [60]. Although quantitative CT (QCT) has also been used to evaluate BMD, this technique is costly, affected by marrow adipose tissue, and provides higher radiation-dose exposure than DXA scanning [61].…”

“…Yet, there is currently mixed data on the accuracy of DECT in the assessment of BMD. Certain studies state that DECTassociated HAP values moderately correlate with DXA images (r = 0.614) while others report linear correlations (r = 0.97) [60,63]. Yet, it is imperative to note that the latter study used a phantom setup and has yet to validate their methods in a clinical setting.…”