Abstract
Objectives
To compare the performance of areal Hounsfield units (aHUs), volumetric Hounsfield units (vHUs), and volumetric bone mineral density (vBMD) by quantitative CT (QCT) in discriminating vertebral fractures (VFs) risk.
Methods
We retrospectively included CT scans of the lumbar spine 101 VFs cases (60 women, mean age: 64 ± 4 years; 41 men, mean age: 73 ± 10 years) and sex- and age-matched 101 control subjects (60 women, mean age: 64 ± 4 years; 41 men, mean age: 72 ± 7 years). In order to assess the discriminatory capability of aHU, vHU, and vBMD measurements at the L1 and L2 levels in identifying VFs, we conducted binary logistic regression and receiver operating characteristic (ROC) curve analyses in men and women. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated.
Results
In both men and women with and without VFs, aHU, vHU, and vBMD were highly correlated with each other (r2 from 0.832 to 0.957, all P < .001). There was a statistically significant difference in aHU, vHU, and vBMD between subjects with and without VFs (P < .001). When age, gender, and BMI were taken into account as covariances and adjusted simultaneously, odds ratios (ORs) for aHU, vHU, and vBMD values, which represent the risk of VFs, were significant (P < .001). Compared with aHU and vHU, vBMD was more strongly associated with VF risk (vBMD: OR, 6.29; 95% CI, 3.83-10.35 vs vHU: OR, 3.64; 95% CI, 2.43-5.46 vs aHU: OR, 2.56; 95% CI, 1.79-3.67). In both men and women, further, vBMD had higher values for AUC, sensitivity, specificity, PPV, and NPV compared to vHU, with vHU in turn surpassing aHU. The area under the receiver operating characteristic curve (AUC) for discriminating VFs using the average aHU, vHU, and vBMD of 2 vertebrae was 0.72, 0.77, and 0.87 in men and 0.76, 0.79, and 0.86 in women. In both men and women, there exist statistically significant differences in the AUC when employing the 3 measurements—namely, aHU, vHU, and vBMD—to discriminate fractures (P < .05).
Conclusions
The QCT-measured vBMD is more associated with acute VFs than vHU and aHU values of the lumbar spine. Although the use of vHU and aHU values for the diagnosis of osteoporosis and discriminating fracture risk is limited to scanner- and imaging protocol-specific, they have great potential for opportunistic osteoporosis screening, particularly vHU.
Advances in knowledge
The novelty of this study presents a comparison of the VF discriminative capabilities among aHU, vHU, and vBMD. The vHU values introduced in this study demonstrate a greater capacity to discriminate fractures compared to aHU, presenting an improved clinical choice. Although its discriminatory capability is slightly lower than that of vBMD, it is more convenient to measure and does not require specialized software.
Keywords: volumetric bone mineral density, vertebral fractures, Hounsfield units, quantitative computer tomography
Introduction
With aging, lumbar vertebral fractures (VFs), which represent 42.4%-75.2% of all spinal fractures in the elderly, have become a prevalent clinical illness associated with pain, disability, and mortality.1–4 VFs will more than triple by 2035, reaching 5.99 (95% CI, 5.44-6.55) million and costing $25.43 (95% CI, 23.92-26.95) billion by 2050.5 Thus, it is critical to identify individuals at high risk of VFs and start interventions such as anti-osteoporosis medication as soon as possible.
Low bone mineral density (BMD) is a recognized indicator of high VF risk, and dual-energy X-ray absorptiometry (DXA) is the most widely used method for evaluating BMD in vivo. However, areal BMD (aBMD) measured by DXA is 2-dimensional and susceptible to artificial elevation due to degenerative changes and osteophytes.6 In recent studies,7–9 the vertebral Hounsfield units (HUs) value measured from clinical computed tomography (CT) has been recognized as a surrogate of bone strength and also recommended to detect osteoporosis. This method utilizes the clinical routine CT scans obtained for various unrelated indications, such as low-dose CT, lumbar CT, and abdominal CT, requiring no additional cost or radiation. However, CT HU values were introduced to calibrate water equivalent materials across different scanners. The mass absorption coefficient of bone significantly differs from that of water, and therefore the absorption of bone may vary among scanners. Moreover, the HU value of lumbar vertebrae depends on X-ray field inhomogeneity, which varies with table height. Therefore, compared to the recognized volumetric BMD (vBMD) measured by quantitative CT (QCT), the use of HU values for the diagnosis of osteoporosis and predicting fracture risk is scanner- and imaging protocol-specific.
To date, most studies using area HU (aHU) values for diagnosis were based on DXA measurements.10–12 Some reports recently also demonstrated the discrimination capability of aHU values in VF risk.13–15 However, these previous studies have analysed only single-slice based oval regions of interest (ROI) to assess the HU of vertebrae.8,16,17 Due to the 3-dimensional bone structure and density distribution of the vertebra, measuring only a single ROI cannot represent the whole bone strength and may result in questionable measurement reliability. The association of the volumetric HU (vHU) values with VFs is still unclear. Furthermore, the relationship and the comparative assessment of the predictive capabilities for VFs among aHU, vHU, and QCT-based vBMD have not been reported in previous studies. For this reason, the purpose of this study is to explore the abilities to discriminate VFs among these 3 methods.
Methods
Subjects
This study was approved by the institutional review board of our hospital. The informed consent was waived because this was a retrospective study. We reviewed the cases and retrospectively included 101 patients diagnosed with acute lumbar VFs in our hospital from March 2014 to June 2022. Inclusion criteria were (1) patients over the age of 50 years and (2) those who underwent spine CT with a QCT phantom in our hospital within 1 month after the fracture occurred. The exclusion criteria were (1) patients who had fractures at each vertebra and (2) patients after fracture surgery.
One hundred and one age- and sex-matched dwelling-community adults without VFs served as controls. These control participants were from China Action on Spine and Hip Status (CASH) study, which determines the prevalence of osteoporotic fracture, osteoporosis, and osteoarthritis in an older Chinese population using QCT and/or dual energy-ray absorptiometry (DXA).18
CT acquisition
CT scans were performed in a supine position at full inspiration using a Toshiba CT scanner (Aquilion Prime, Toshiba Medical Systems, Japan). Scanning parameters were as follows: slice thickness, 1.0 mm; beam pitch, 1.0; tube voltage, 120 kV; tube current, 250 mA. All CT examinations were performed without the use of intravenous contrast. Scanners were calibrated daily to ensure accurate CT attenuation numbers for the vertebrae. A calibration phantom (Mindways Software Inc., Austin, TX, USA) was placed under the patient's lumbar spine during the scan.
QCT volumetric BMD measurements
For QCT vBMD (mg/cm3) of the L1-2 vertebral bodies measurements, all CT images were transferred to a QCT Pro workstation (Mindways, Software, Inc.), which could generate and semiautomatically analyse the volumes of interest (VOIs) (Figure 1). The VOIs were with 9 mm height and the cross-sectional cylinder was drawn to be as large as possible without intersecting the vertebral cortex and without including dense bone islands, venous plexus, or focal lesions. However, it needed to be reset if an obvious identification error was observed. In addition, if the L1 or L2 vertebral bodies were fractured, they were replaced by an adjacent vertebral body.
Figure 1.
Bone mineral density (BMD) measurements of L1 and L2 by using quantitative CT (QCT) Pro (A) and the corresponding volume of interest (VOI) Hounsfield unit (HU) values by using 3D slicer software (B).
CT Hounsfield unit measurements
All CT images were initially stored in an electronic Picture Archiving and Communication System (PACS) as DICOM files and assessed using a 3D slicer, version 5.0.3. All vBMD and HU (vHU and aHU) measurements were performed by 1 observer. The vBMD and vHU values were measured simultaneously to make sure the consistency of the VOI location, VOI size, and measured vertebrae of each individual between 2 methods (Figure 1). The volumetric reconstruction and analyses of the ROIs were performed semi-automatically using the 3D slicer software to calculate the average vHU value of each vertebra. In the process of calculating aHU values, ROIs were positioned at the mid-axial level of every vertebral body. The size of ROI delineation and the calculation of aHU values remain consistent with the above-mentioned method.
Statistical analyses
SPSS 26.0 (IBM Crop., Armonk, NY, USA), MedCalc 20.0, GraphPad Prism 8.0, and PASS 15.0 were used for the statistical analyses. This study was designed as a matched case-control study, with a calculated sample size of N1 = N2 = 95 for both the control and case groups, resulting in a total required sample size of 190. The baseline sociodemographic and clinical characteristics are described as the mean ± standard deviation (SD) for continuous variables. A Student’s t-test was used to compare the demographic differences between the VFs and control groups in men and women, respectively. Binomial logistic regression models with unadjusted and adjusted gender, BMI, and age were used to calculate odds ratios (ORs) of fracture per standard deviation increase in aHU, vHU, and vBMD. We assessed the diagnostic performance of aHU, vHU, and vBMD of sensitivity, specificity, PPV, and negative predictive value (NPV). Subsequently, aHU, vHU, and vBMD combining age and BMI were constructed model and compared with each other using the area under the receiver operating characteristic (ROC) curves (AUCs). The statistical significance of differences among these AUCs was evaluated using DeLong's test. P < .05 was regarded as statistically significant.
Results
Baseline characteristics
Table 1 shows the baseline characteristics of the study population. A total of 202 subjects were eligible for further analysis. There were 82 males (41 VFs, age 73.37 ± 9.86 years; 41 non-VFs, age 72.49 ± 6.50 years) and 120 women (60 VFs, age 63.9 ± 4.21 years; 60 non-VFs, age 63.57 ± 4.23 years). No statistical difference in height, and weight was found between VFs and the control group. Significant differences were observed in HU (aHU and vHU) and vBMD values between VFs and controls (P < .001).
Table 1.
Demographic characteristics and bone density.
| Characteristic | All (n = 202) | Men (n = 82) |
Women (n = 120) |
||
|---|---|---|---|---|---|
| Control group (n = 41) | Fracture group (n = 41) | Control group (n = 60) | Fracture group (n = 60) | ||
| Age (y) | 67.47 ± 7.65 | 72.49 ± 6.50 | 73.37 ± 9.86 | 63.57 ± 4.23 | 63.9 ± 4.12 |
| Height (cm) | 159.70 ± 26.78 | 170.56 ± 6.03 | 167.27 ± 27.32 | 154.03 ± 29.57 | 152.78 ± 28.87 |
| Weight (kg) | 63.47 ± 14.67 | 72.88 ± 13.96 | 66.27 ± 13.08 | 60.18 ± 15.01 | 58.43 ± 12.47 |
| BMI (kg/m2) | 23.64 ± 4.87 | 24.91 ± 3.59 | 22.64 ± 4.40* | 23.71 ± 5.79 | 23.38 ± 4.86 |
| aHU mean value (L1 + L2) (mg/cm3) | 100.73 ± 40.91 | 104.52 ± 27.53 | 78.57 ± 32.44** | 127.18 ± 39.11 | 86.83 ± 41.04** |
| vHU mean value (L1 + L2) (HU) | 99.09 ± 39.48 | 109.23 ± 26.68 | 76.91 ± 33.84** | 124.81 ± 36.30 | 81.61 ± 36.55** |
| vBMD mean BMD (L1 + L2) (HU) | 78.63 ± 35.57 | 94.21 ± 23.33 | 56.33 ± 25.23** | 104.29 ± 32.56 | 57.57 ± 28.54** |
Abbreviations: aHU = areal Hounsfield units; vBMD = volumetric bone mineral density; vHU = volumetric Hounsfield unit.
P-value < .05; statistically, differences were compared with the control group stratified by sex..
P < .001; statistically, differences were compared with the control group stratified by sex..
Correlation of Hounsfield units and QCT
Figures 2 and 3 show the correlation coefficients among aHU, vHU, and QCT-measured vBMD measured by QCT in men and women. Among both males and females, there is a strong correlation among the 3 measurements, with the highest correlation observed between vHU and vBMD in both the fracture groups (men: r2 = 0.943, P < .001; women: r2 = 0.957, P < .001) and control groups (men: r2 = .921, P < .001; women: r2 = 0.938, P < .001).
Figure 2.
The correlation among areal Hounsfield unit (aHU), volumetric Hounsfield unit (vHU), and quantitative CT (QCT)-measured volumetric bone mineral density (vBMD) in the men control and fracture groups.
Figure 3.
The correlation among areal Hounsfield unit (aHU), volumetric Hounsfield unit (vHU), and quantitative CT (QCT)-measured volumetric bone mineral density (vBMD) in the women control and fracture groups.
Discriminatory ability of HU and QCT measurements for vertebral fracture risk
Table 2 presents the association of each SD increase in vBMD and HU (aHU and vHU) variables with prevalent VF. There was a significant association of higher vBMD and HU (aHU and vHU) with a lower prevalence of VF in unadjusted and adjusted models. In all the models, the OR values of vBMD (OR, 6.29; 95% CI, 3.83-10.35) were higher than those of vHU (OR, 3.64; 95% CI, 2.43-6.46), with vHU exhibiting a superiority over aHU (OR, 2.56; 95% CI, 1.79-3.67).
Table 2.
Unadjusted and adjusted odd ratios of VFs per one SD increase of aHU, vHU, and QCT-measured BMD analysed by logistic regression analysis.
| P | OR (95% CI) | |
|---|---|---|
| Unadjusted | ||
| aHU | <.001 | 2.411 (1.712-3.396) |
| vHU | <.001 | 3.523 (2.357-5.277) |
| vBMD | <.001 | 6.115 (3.754-9.960) |
| Adjusted | ||
| Sex | .496 | 0.771 (0.364-1.630) |
| Age | .843 | 0.995 (0.949-1.044) |
| BMI | .034 | 1.070 (1.005-1.140) |
| aHU | <.001 | 2.564 (1.790-3.674) |
| Model 1 | Prob(fracture) = −3.463 −0.261sex −0.005age + 0.068BMI +0.942 aHU | |
| Sex | .370 | 0.693 (0.311-1.546) |
| Age | .670 | 0.989 (0.940-1.040) |
| BMI | .100 | 1.056 (0.989-1.129) |
| vHU | <.001 | 3.640 (2.428-5.457) |
| Model 2 | Prob(fracture) = −3.568 −0.367sex −0.011age +0.055BMI +1.292vHU | |
| Sex | .595 | 0.789 (0.329-1.890) |
| Age | .764 | 0.992 (0.940-1.047) |
| BMI | .100 | 1.061 (0.989-1.138) |
| vBMD | <.001 | 6.293 (3.826-10.351) |
| Model 3 | Prob(fracture) = −4.694 −0.237sex −0.008age +0.059BMI +1.839vBMD | |
Abbreviations: aHU = areal Hounsfield units; QCT = quantitative CT; vBMD = volumetric bone mineral density; vHU = volumetric Hounsfield unit.
Table 3 summarizes the sensitivity, specificity, PPV, NPV, and AUC for 3 variables. The AUC for distinguishing VFs using vBMD was 0.87 (95% CI, 0.77-0.93) in men and 0.86 (95% CI, 0.78-0.91) in women, which were higher compared to the AUC values obtained from aHU and vHU. Statistically significant differences in the AUC values exist among these 3 measurements (P < .05). For vHU, an optimal threshold value of 80.69 resulted in a sensitivity of 85.4% with a specificity of 61% in men. Similarly, in women, a threshold value of 86.42 led to 83.3% sensitivity and 61.7% specificity.
Table 3.
Calculated area under the ROC curves and the diagnostic performance of aHU, vHU, and vBMD with optimal thresholds for detecting fracture.
| Diagnosis | AUC (95% CI) | Optimal thresholda | Sensitivity, % (n/N) | Specificity, % (n/N) | PPV, % (n/N) | NPV, % (n/N) |
|---|---|---|---|---|---|---|
| Men (n = 82) | ||||||
| aHU | 0.72 (0.61-0.82) | 90.46 | 73.20 (30/41) | 63.40 (26/41) | 66.67 (30/45) | 70.29 (26/37) |
| vHU | 0.77 (0.67-0.86) | 80.69 | 85.40 (35/41) | 61.00 (25/41) | 68.65 (35/51) | 80.69 (25/31) |
| vBMD | 0.87 (0.77-0.93) | 62.91 | 87.80 (36/41) | 68.30 (28/41) | 73.47 (36/49) | 84.84 (28/33) |
| Women (n = 120) | ||||||
| aHU | 0.76 (0.67-0.83) | 91.64 | 81.70 (49/60) | 61.70 (37/60) | 68.05 (49/72) | 77.08 (37/48) |
| vHU | 0.79 (0.71-0.86) | 86.42 | 83.30 (50/60) | 61.70 (37/60) | 68.50 (50/73) | 78.70 (37/47) |
| vBMD | 0.86 (0.78-0.91) | 60.42 | 90.00 (54/60) | 58.30 (35/60) | 68.34 (54/79) | 85.36 (35/41) |
Abbreviations: aHU = areal Hounsfield units; AUC = area under the ROC curve; BMD = bone mineral density; CI = confidence interval; NPV = negative predictive value; PPV = positive predictive value; vBMD = volumetric bone mineral density; vHU = volumetric Hounsfield unit.
Optimal threshold is the threshold where the largest proportion of patients is correctly classified.
Figure 4 shows the ROC curve and AUC for aHU, vHU, and vBMD used in detecting VFs in adjusting and unadjusting age and BMI. After adjusting for age and BMI, there is no statistically significant difference in the AUC between aHU and vHU only in men (P = .051), while statistically significant differences exist in the remaining cases (P < .05).
Figure 4.
Receiver operating characteristic curves of different models for discrimination of vertebral fractures (VFs) in men and women.
Discussion
In this study, we first compared the relevance of aHU, vHU values, and QCT-measured vBMD of L1 and L2 vertebrae for VFs discrimination. Our study shows that vBMD is more associated with acute VFs than vHU and aHU of the lumbar spine. Compared to the contribution of vBMD to VFs discrimination, vHU and aHU measurements had a lower discriminatory ability but still had the potential to be used in opportunistic osteoporosis screening, especially vHU.
Previous studies have reported a good correlation between vBMD based on QCT and DXA-measured aBMD.12,19,20 Only a few studies have investigated the correlation between the HU of lumbar spine and the QCT data.21–24 Interestingly, most of these studies evaluated HU values obtained from only a single oval ROI for each vertebra. The extent to which vHU can effectively discriminate VFs remains unclear, and there is no prior study of a comparative assessment of the VFs discrimination ability, as used in this study, among aHU values, vHU values, and vBMD derived from QCT. Compared to aHU and DXA-measured aBMD, vHU and QCT-measured vBMD circumvent degenerative changes of the spine, and can better reflect trabecular bone microstructure.
Our findings are similar to those of Zou et al,13 who discovered that the case patients had a much lower average of 2 vertebrae than the control group in aHU value and in QCT-measured vBMD. Compared to previous studies, the overall HU (aHU and vHU) value and vBMD of this study are low, which may be explained by the large number of lumbar spine fractures we included, as demonstrated by Kim et al.25
Our findings showed that HU (aHU and vHU) value has an excellent correlation with QCT and diagnostic performance in both men and women. In this retrospective study, trabecular vBMD assessed by QCT showed that a high association with the risk of incident VFs, followed by vHU and then aHU values (Table 2), which is similar to previous research.19,26,27 In addition, Leonhardt et al20 also demonstrated that osteoporotic trabecular vBMD of lumbar vertebrae assessed by QCT was associated with an increased risk of VFs, following for better risk assessment (OR, 1.03; 95% CI, 1.01-1.06, P = .005). The large difference between our results and theirs could be attributed to the fact that they recruited patients with a single fracture and suspected osteoporosis.
According to Lee et al,28 structural measures of the spine are best suited to predict osteoporotic fracture status in the spine (AUCs, 0.62-0.75). In our study, before and after including BMI and age in the models, the AUC values to identify patients with VFs were all greater than 0.75, indicating that HU (aHU and vHU) value and vBMD were good markers for detecting the high-risk population. Again, trabecular vBMD performed also better (male: AUC = 0.89; female: AUC = 0.86) than vHU value (male: AUC = 0.83; female: AUC = 0.79) and aHU (male: AUC = 0.79; female: AUC = 0.76) value of the lumbar spine in VF discrimination.
Opportunistic screening for osteoporosis becomes increasingly popular.29,30 In contrast to numerous studies of opportunistic screening, where HU values in thoracic or lumbar vertebra were reported,27,31,32 we used the phantom calibration to obtain lumbar trabecular vBMD. The benefits of conventional QCT can be appreciated without its disadvantages of additional radiation and costs compared to DXA. Moreover, within this study, we introduced the HU value of lumbar spine was based on volumetric measurements, which reduced the error caused by the changes of bone microstructure. Then, we compared aHU, vHU, and vBMD to evaluate their capacities in discerning fractures. This comparative analysis has the potential to provide clinical practitioners with valuable reference options. Finally, we controlled for the effects of gender, BMI and age on the results and included relevant models in predicting VFs risk.
There are several limitations to this study. Our retrospective inclusion of subjects who underwent CT scans may have resulted in a selection bias, including subjects with clinical indications. Second, while the fractures group was age- and sex-matched, the subjects with fractures more likely to have second fracture or above-average spinal degeneration, causing some bias in our results. Third, we did not perform a categorical analysis of the subjects' fracture severity. Finally, this was a single-centre study conducted on a limited number of patients.
In conclusion, the QCT-measured vBMD is more associated with acute VFs than vHU and aHU values of the lumbar spine. Although the use of HU (aHU and vHU) values for the diagnosis of osteoporosis and predicting fracture risk is limited to scanner- and imaging protocol-specific, vHU values have great potential for opportunistic osteoporosis screening, especially vHU.
Contributor Information
Fengyun Zhou, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Wenshuang Zhang, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Jian Geng, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Yandong Liu, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Yi Yuan, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Kangkang Ma, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Zitong Cheng, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Pengju Huang, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Xiaoguang Cheng, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Ling Wang, Department of Radiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China; JST sarcopenia Research Centre, National Center for Orthopaedics, Beijing Research Institute of Traumatology and Orthopaedics, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Yajun Liu, JST sarcopenia Research Centre, National Center for Orthopaedics, Beijing Research Institute of Traumatology and Orthopaedics, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China; Department of Spine Surgery, Beijing Jishuitan Hospital, Capital Medical University, Beijing 100035, China.
Author contributions
All authors contributed to the study’s conception and design. Material preparation, data collection, and analysis were performed by Yajun Liu, Ling Wang, Xiaoguang Cheng, Fengyun Zhou, Wenshuang Zhang, Jian Geng, Yandong Liu, Yi Yuan, Kangkang Ma, Zitong Cheng and Pengju Huang. The first draft of the manuscript was written by Fengyun Zhou, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
This work was supported by the Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (code: ZYLX202107), National Key R&D Program of China (2021YFC2501703), National Natural Science Foundation of China (grant no. 82371956), National Natural Science Foundation of China (grant no. 82371957), Beijing Municipal Public Welfare Development and Reform Pilot Project for Medical Research Institutes (JYY2023-8).
Conflicts of interest
The authors declared no potential conflicts of interest concerning the research, authorship, and/or publication of this article.
References
- 1. Bigdon SF, Saldarriaga Y, Oswald KAC, et al. Epidemiologic analysis of 8000 acute vertebral fractures: evolution of treatment and complications at 10-year follow-up. J Orthop Surg Res. 2022;17(1):270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. den Ouden LP, Smits AJ, Stadhouder A, et al. Epidemiology of spinal fractures in a level one trauma center in the Netherlands: a 10 years review. Spine (Phila Pa 1976). 2019;44(10):732-739. [DOI] [PubMed] [Google Scholar]
- 3. Robinson WA, Carlson BC, Poppendeck H, et al. Osteoporosis-related vertebral fragility fractures: a review and analysis of the American Orthopaedic Association’s Own the Bone Database. Spine (Phila Pa 1976). 2020;45(8):E430-E438. [DOI] [PubMed] [Google Scholar]
- 4. Skjødt MK, Nicolaes J, Smith CD, et al. Fracture risk in men and women with vertebral fractures identified opportunistically on routine CT scans and not treated for osteoporosis: an observational cohort study. JBMR Plus. 2023;7(5):e10736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Si L, Winzenberg TM, Jiang Q. et al. Projection of osteoporosis-related fractures and costs in China: 2010–2050. Osteoporos Int. 2015;26(7):1929-1937. [DOI] [PubMed] [Google Scholar]
- 6. Muraki S, Yamamoto S, Ishibashi H, et al. Impact of degenerative spinal diseases on bone mineral density of the lumbar spine in elderly women. Osteoporos Int. 2004;15(9):724-728. [DOI] [PubMed] [Google Scholar]
- 7. Pickhardt PJ, Lee LJ, del Rio AM, et al. Simultaneous screening for osteoporosis at CT colonography: bone mineral density assessment using MDCT attenuation techniques compared with the DXA reference standard. J Bone Miner Res. 2011;26(9):2194-2203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pickhardt PJ, Pooler BD, Lauder T, et al. Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann Intern Med. 2013;158(8):588-595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hendrickson NR, Pickhardt PJ, Del Rio AM, et al. Bone mineral density T-scores derived from CT attenuation numbers (Hounsfield units): clinical utility and correlation with dual-energy X-ray absorptiometry. Iowa Orthop J. 2018;38:25-31. [PMC free article] [PubMed] [Google Scholar]
- 10. Davidson S, Vecellio A, Flagstad I, et al. Discrepancy between DXA and CT-based assessment of spine bone mineral density. Spine Deform. 2023;11(3):677-683. [DOI] [PubMed] [Google Scholar]
- 11. Marinova M, Edon B, Wolter K, et al. Use of routine thoracic and abdominal computed tomography scans for assessing bone mineral density and detecting osteoporosis. Curr Med Res Opin. 2015;31(10):1871-1881. [DOI] [PubMed] [Google Scholar]
- 12. Paggiosi MA, Debono M, Walsh JS, et al. Quantitative computed tomography discriminates between postmenopausal women with low spine bone mineral density with vertebral fractures and those with low spine bone mineral density only: the SHATTER study. Osteoporos Int. 2020;31(4):667-675. [DOI] [PubMed] [Google Scholar]
- 13. Zou D, Ye K, Tian Y, et al. Characteristics of vertebral CT Hounsfield units in elderly patients with acute vertebral fragility fractures. Eur Spine J. 2020;29(5):1092-1097. [DOI] [PubMed] [Google Scholar]
- 14. Zou D, Li W, Deng C, et al. The use of CT Hounsfield unit values to identify the undiagnosed spinal osteoporosis in patients with lumbar degenerative diseases. Eur Spine J. 2019;28(8):1758-1766. [DOI] [PubMed] [Google Scholar]
- 15. Yang G, Wang H, Wu Z, et al. Prediction of osteoporosis and osteopenia by routine computed tomography of the lumbar spine in different regions of interest. J Orthop Surg Res. 2022;17(1):454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Perrier-Cornet J, Omorou AY, Fauny M, et al. Opportunistic screening for osteoporosis using thoraco-abdomino-pelvic CT-scan assessing the vertebral density in rheumatoid arthritis patients. Osteoporos Int. 2019;30(6):1215-1222. [DOI] [PubMed] [Google Scholar]
- 17. Engelke K, Chaudry O, Bartenschlager S.. Opportunistic screening techniques for analysis of CT scans. Curr Osteoporos Rep. 2023;21(1):65-76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wang L, Yin L, Yang M, et al. Muscle density is an independent risk factor of second hip fracture: a prospective cohort study. J Cachexia Sarcopenia Muscle. 2022;13(3):1927-1937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Löffler MT, Jacob A, Valentinitsch A, et al. Improved prediction of incident vertebral fractures using opportunistic QCT compared to DXA. Eur Radiol. 2019;29(9):4980-4989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Leonhardt Y, May P, Gordijenko O, et al. Opportunistic QCT bone mineral density measurements predicting osteoporotic fractures: a use case in a prospective clinical cohort. Front Endocrinol (Lausanne). 2020;11:586352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Buenger F, Eckardt N, Sakr Y, et al. Correlation of bone density values of quantitative computed tomography and Hounsfield units measured in native computed tomography in 902 vertebral bodies. World Neurosurg. 2021;151:e599-e606. [DOI] [PubMed] [Google Scholar]
- 22. Baum T, Müller D, Dobritz M, et al. Converted lumbar BMD values derived from sagittal reformations of contrast-enhanced MDCT predict incidental osteoporotic vertebral fractures. Calcif Tissue Int. 2012;90(6):481-487. [DOI] [PubMed] [Google Scholar]
- 23. Mueller DK, Kutscherenko A, Bartel H, et al. Phantom-less QCT BMD system as screening tool for osteoporosis without additional radiation. Eur J Radiol. 2011;79(3):375-381. [DOI] [PubMed] [Google Scholar]
- 24. Bauer JS, Henning TD, Müeller D, et al. Volumetric quantitative CT of the spine and hip derived from contrast-enhanced MDCT: conversion factors. AJR Am J Roentgenol. 2007;188(5):1294-1301. [DOI] [PubMed] [Google Scholar]
- 25. Kim YW, Kim JH, Yoon SH, et al. Vertebral bone attenuation on low-dose chest CT: quantitative volumetric analysis for bone fragility assessment. Osteoporos Int. 2017;28(1):329-338. [DOI] [PubMed] [Google Scholar]
- 26. Dieckmeyer M, Löffler MT, El Husseini M, et al. Level-specific volumetric BMD threshold values for the prediction of incident vertebral fractures using opportunistic QCT: a case–control study. Front Endocrinol (Lausanne). 2022;13:882163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Johannesdottir F, Allaire B, Kopperdahl DL, et al. Bone density and strength from thoracic and lumbar CT scans both predict incident vertebral fractures independently of fracture location. Osteoporos Int. 2021;32(2):261-269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lee DC, Campbell PP, Gilsanz V, et al. Contribution of the vertebral posterior elements in anterior-posterior DXA spine scans in young subjects. J Bone Mineral Res. 2009;24(8):1398-1403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Engelke K, Lang T, Khosla S, et al. Clinical use of quantitative computed tomography-based advanced techniques in the management of osteoporosis in adults: the 2015 ISCD official positions-Part III. J Clin Densitom. 2015;18(3):393-407. [DOI] [PubMed] [Google Scholar]
- 30. Engelke K. Quantitative computed tomography-current status and new developments. J Clin Densitom. 2017;20(3):309-321. [DOI] [PubMed] [Google Scholar]
- 31. Alacreu E, Moratal D, Arana E.. Opportunistic screening for osteoporosis by routine CT in southern Europe. Osteoporos Int. 2017;28(3):983-990. [DOI] [PubMed] [Google Scholar]
- 32. Marinova M, Edon B, Wolter K, et al. Use of routine thoracic and abdominal computed tomography scans for assessing bone mineral density and detecting osteoporosis. Curr Med Res Opin. 2015;31(10):1871-1881. [DOI] [PubMed] [Google Scholar]