If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Assessment of bone density using the 1.5 T or 3.0 T MRI-based vertebral bone quality score in older patients undergoing spine surgery: does field strength matter?
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Department of Radiology and Image, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Corresponding author. Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), Chairman of the Department of Spine Surgery, No. 1 Jiazi Rd, Shunde district, Foshan city, Guangdong Province, China. Tel.: (86) 18620866697.
Department of Spine Surgery, Shunde Hospital, Southern Medical University (The First People's Hospital of Shunde Foshan), No. 1 Jiazi Rd, Shunde district, Foshan, Guangdong, China
Recently published studies have revealed a correlation between MRI-based vertebral bone quality (VBQ) score and bone mineral density (BMD) measured using dual X-ray absorptiometry (DXA) or quantitative computed tomography (QCT). However, no studies have determined if differences in field strength (1.5 vs. 3.0 T) could affect the comparability of the VBQ score among different individuals.
PURPOSE
To compare the VBQ score obtained from 1.5 T and 3.0 T MRI (VBQ1.5T vs. VBQ3.0T) in patients undergoing spine surgery and assess the predictive performance of VBQ for osteoporosis and osteoporotic vertebral fracture (VCF).
DESIGN
A nested case‒control study based on an ongoing prospective cohort study of patients undergoing spine surgery.
PATIENT SAMPLE
All older patients (men aged >60 years and postmenopausal women) with available DXA, QCT and MR images within 1 month were included.
OUTCOME MEASURES
VBQ score, DXA T-score, and QCT derived vBMD.
METHODS
The osteoporotic classifications recommended by the World Health Organization and American College of Radiology were used to categorize the DXA T-score and QCT-derived BMD, respectively. For each patient, the VBQ score was calculated using T1-weighted MR images. Correlation analysis between VBQ and DXA/QCT was performed. Receiver operating characteristic (ROC) curve analysis, including determination of the area under the curve (AUC), was performed to assess the predictive performance of VBQ for osteoporosis.
Results
A total of 452 patients (98 men aged >60 years and 354 postmenopausal women) were included in the analysis. Across different BMD categories, the correlation coefficients between the VBQ score and BMD ranged from -0.211 to -0.511, and the VBQ1.5T score and QCT BMD demonstrated the strongest correlation. The VBQ score was a significant classifier of osteoporosis detected by either DXA or QCT, with VBQ1.5T showing the highest discriminative power for QCT-osteoporosis (AUC=0.744, 95% CI=0.685–0.803). In ROC analysis, the VBQ1.5T threshold values ranged from 3.705 to 3.835 with a sensitivity between 48% and 55.6% and a specificity between 70.8% and 74.8%, while the VBQ3.0T threshold values ranged from 2.59 to 2.605 with a sensitivity between 57.6% and 67.1% and a specificity between 67.8% and 69.7%.
CONCLUSIONS
VBQ1.5T exhibited better discriminability between patients with and without osteoporosis than VBQ3.0T. Considering the non-negligible difference in osteoporosis diagnosis threshold values between the VBQ1.5T and VBQ3.0T scores, it is essential to clearly distinguish the magnetic field strength when assessing the VBQ score.
Osteoporosis: recent recommendations and positions of the American Society for Bone and Mineral Research and the International Society for Clinical Densitometry.
J Bone Joint Surg Am.2021; 103 (Epub 2021/02/16): 741-747
]. Dual X-ray absorptiometry (DXA), as an established measure for evaluating the BMD, is widely used in scientific research and clinical practice, while quantitative computed tomography (QCT) is becoming gradually accepted for lumbar BMD measurement due to its ability to assess the volumetric BMD (vBMD) three dimensionally [
]. However, these methods inevitably expose patients to excessive ionizing radiation.
Because it is noninvasive, does not utilize ionizing radiation, and demonstrably excellent soft-tissue contrast, magnetic resonance imaging (MRI) plays a pivotal role in the field of spinal disorders and is currently widely used [
MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis.
AJNR Am J Neuroradiol.2015; 36 (Epub 2015/09/12): 2394-2399
]. Considering the availability of MR images, MRI-based vertebral bone quality (VBQ) was first applied to bone density in 68 operative spine patients, revealing an accuracy of 81% in predicting osteopenia/osteoporosis [
]. It has been confirmed that VBQ is correlated with the lumbar BMD measured by DXA or QCT and demonstrates high predictive value for vertebral fractures in patients with spinal metastasis or osteoporosis, making it a promising tool for radiation-free osteoporosis screening [
]. However, different studies published on the topic have yielded highly variable results, with accuracies in predicting the presence of osteopenia/osteoporosis ranging from 67% to 89%, an overall sensitivity ranging from 58% to 84.7%, and a specificity ranging from 40.6% to 90% [
Bone quality in patients with osteoporosis undergoing lumbar fusion surgery: analysis of the MRI-based vertebral bone quality score and the bone microstructure derived from microcomputed tomography.
In previous studies regarding VBQ score, the reference standard for the diagnosis of osteoporosis was either DXA or QCT. Although discordant lumbar BMD measurements are obtained between DXA and QCT, few studies have assessed whether the different diagnostic criteria for osteoporosis may affect the VBQ score [
Discordance in lumbar bone mineral density measurements by quantitative computed tomography and dual-energy X-ray absorptiometry in postmenopausal women: a prospective comparative study.
]. In addition, differences in MRI systems, including vendor, field strength (1.5/3.0 T), coil setup, and sequence type, could influence the quality of the acquired images of the musculoskeletal tissue and may affect the comparability of the VBQ score among individuals [
Measuring and reporting of vertebral endplate bone marrow lesions as seen on MRI (Modic changes): recommendations from the ISSLS Degenerative Spinal Phenotypes Group.
]. To the best of our knowledge, prior studies on VBQ have not addressed these concerns. There is a need to reconsider the prognostic value of VBQ under different reference standards and field strengths.
Therefore, this study aimed to (1) visually compare the VBQ score obtained from 1.5 T and 3.0 T MRI (VBQ1.5T vs. VBQ3.0T) in patients undergoing spine surgery (2), examine the correlation among the DXA T-score, QCT-derived vBMD and VBQ score, and (3) assess the predictive performance of VBQ for osteoporosis and osteoporotic vertebral fracture (VCF).
Materials and methods
Study design and population cohort
The study was approved by the local institutional review board and followed the tenets of the Declaration of Helsinki. This nested case‒control study is based on an ongoing prospective cohort study of patients undergoing spine surgery. The cohort was established for the purpose of assessing spinal QCT-derived vBMD and DXA-derived areal BMD and evaluating the predictive performance of VCF in patients undergoing spine surgery; some of the DXA and QCT data have been previously partially reported. We retrospectively collected the data from elderly patients (men aged >60 or postmenopausal women) who visited the spine surgery department of a single tertiary medical center and were about to undergo surgery between September 2020 and November 2022. The collected information included age, sex, body mass index (BMI), medical history, and clinical diagnosis. All patients with available DXA, QCT, and MR images within 1 month were included. The exclusion criteria were as follows: (1) history of lumbar instrumentation surgery; (2) spinal metastases; (3) spinal infection; (4) inadequate image quality; and (5) inability to measure the lumbar vBMD or VBQ. We collected data from 471 patients, 452 of whom were included in the final analysis (Fig. 1).
Lumbar spine (L1–L4) DXA data were acquired with a GE Lunar DXA scanner (GE Lunar Prodigy and DPX Brovo DXA scanners, GE Healthcare, WI, USA) according to the manufacturer's standard protocol. CT datasets were obtained using an Aquilion 64-row CT scanner (Toshiba Medical Systems, Tokyo, Japan) and translated to a QCT workstation (Mindways QCT Pro, Mindways Software, Austin, TX, USA) for all vBMD measurements. Details regarding the calibration methods and specific operation protocol have been previously described [
Discordance in lumbar bone mineral density measurements by quantitative computed tomography and dual-energy X-ray absorptiometry in postmenopausal women: a prospective comparative study.
]. The osteoporosis diagnostic categories from the World Health Organization were used to categorize the areal BMD measured with DXA: normal, T score≤-2.5; osteopenia; -2.5 <T score <-1; and osteoporosis, T score≥-1. The standard 2018 American College of Radiology classifications were used to categorize the QCT-derived vBMD: normal, vBMD>120 mg/cm3; osteopenia, 80 mg/cm3≤vBMD≤120 mg/cm3; and osteoporosis, vBMD<80 mg/cm3. For convenience, the osteoporosis diagnoses obtained from DXA and QCT were referred to as DXA-osteoporosis and QCT-osteoporosis, respectively.`
MRI-based VBQ score
One senior musculoskeletal radiologist (HC) trained two spine surgeons (WL and HY) on the VBQ measurement technique using the picture archiving and communication system. Both 1.5 T and 3.0 T MRI were performed using three scanners: an Achieva 1.5 T (Philips Achieva, Philips Medical Systems, Best, The Netherlands), a Skyra 3.0 T (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany), and a Lumina 3.0 T (MAGNETOM Lumina, Siemens Healthcare, Erlangen, Germany). According to the technique previously described by Ehresman et al. [
], the region of interest (ROI) tool was utilized to calculate the average signal intensity (SI) in the medulla portion on the mid‐sagittal position of the L1–L4 vertebral body as well as the average signal cerebral spinal fluid (CSF) posterior to the L1–L3 vertebral body. The VBQ score was calculated as the quotient of the mean of the average SI across L1–L4 and the mean of the average CSF SI across L1–L3. All SIs were obtained from contrast-free T1-weighted MR images (Fig. 2).
Fig. 2Representative image of regions of interest used to calculate the VBQ score. Left, T1-weighted MR image; Right, T2-weighted MR image; VBQ, vertebral bone quality; Avg, average signal intensity; Dev, standard deviation.
If the ROI within the vertebral body was not visible due to abnormalities such as modic change, severe scoliosis, vertebral hemangiomas, or compression fracture, the remaining vertebral body was used to calculate the VBQ score. There were no restrictions as to what MRI manufacturer or dominant magnetic field was used, and the VBQ scores obtained from 1.5 T and 3.0 T MRI were recorded as VBQ1.5T and VBQ3.0T, respectively. A stratified analysis was performed between VBQ1.5T and VBQ3.0T and between their predictive performance for osteoporosis or VCF.
Statistical analysis
Continuous variables are expressed as the mean±standard deviation, and categorical variables are expressed as the frequency and percentage. The Kolmogorov–Smirnov test was conducted to test the normality of the distributions of different variables. Continuous variables conforming to a normal distribution were compared using Student t test, while skewed data were tested by the Mann-Whitney U test. One-way analysis of variance (ANOVA) was used for multiple-group analysis. Correlations among DXA, QCT, and VBQ were demonstrated by scatter plot and quantified with Pearson's correlation. Following univariate logistic regression, odds ratios (ORs) and their corresponding 95% confidence intervals (CIs) were calculated by multivariate logistic regression. The diagnostic performance of the VBQ score for osteoporosis and osteoporotic fractures was evaluated by receiver operating characteristic curve (ROC) analysis, and the area under the curve (AUC) was calculated. Youden's index was used to determine the VBQ score threshold value with the best sensitivity and specificity. All analyses were performed using the SPSS statistics software package (version 26.0, SPSS Inc., Chicago, IL, USA). Statistical significance was assigned as p<.05.
Results
A total of 452 patients (98 men and 354 women) with available DXA, QCT, and MRI data were included in the analysis, with a mean age of 68.9±9.9 years and a mean BMI of 23.8±3.7 kg/m2. In the case cohort, the most predominant reason for spine surgery was vertebral compression fracture (40.9%), followed by lumbar disc herniation (37.2%). The characteristics of the study population are presented in Table 1.
Table 1Baseline characteristics of study participants
All patients were divided into two groups on the basis of the magnetic field used: 1.5 T MRI group (N=273) and 3.0 T MRI group (N=179). No significant differences were observed in age, sex ratio, BMI, primary diagnosis, or osteoporosis category between the two groups. With respect to MRI-derived parameters, both the mean vertebral body SI and VBQ score were significantly higher in the 1.5 T MRI group than in the 3.0 T MRI group, while the mean CSF SI was higher in the 3.0 T MRI group than in the 1.5 T MRI group. The proportion of patients with osteoporosis was not significantly different between the two groups (Table 1). A comparison of VBQ scores among the different MRI devices is visualized in Fig. 3.
Fig. 3Distribution of the vertebral bone quality (VBQ) score classified by different MRI scanners (Achieva 1.5T, Skyra 3.0T, and Lumina 3.0T). The box plot shows the distribution of the VBQ scores as the median (line), interquartile range (top and bottom of the box), and 1.5 times the interquartile range (whiskers). The circles represent the individual patient scores, and those outside the whiskers represent outliers. Different letters indicate statistically significant differences (p<.05)
Fig. 4 shows a plot of the DXA T-score and QCT-vBMD versus the MRI-based VBQ score for different scanner types and magnetic field strengths. The correlation coefficient between the DXA T-score and the VBQ score ranged from -0.211 to -0.383. QCT-vBMD was found to be more strongly correlated with the VBQ score (correlation coefficient, range -0.387 to -0.511) than the DXA T-score, and the strongest correlation was observed between the VBQ1.5T score and QCT-vBMD.
Fig. 4Correlation analysis of MRI-based VBQ score versus DXA T-score (Left, y-axis) and QCT-vBMD (Right) for different MRI scanner types and manufacturers. For the DXA T-score, the Pearson correlation coefficients were -0.211 (red, Achieva 1.5 T, p<.001), -0.383 (green, Skyra 3.0 T, p<.001), and -0.231 (purple, Lumina 3.0 T, p=.204). For the QCT-vBMD, the Pearson correlation coefficients were -0.511 (red, Achieva 1.5 T, p<.001), -0.428 (green, Skyra 3.0 T, p<.001), and -0.387 (purple, Lumina 3.0 T, p=.029).
Association between VBQ score and osteoporosis/VCF
A total of 237 patients (52.4%) were diagnosed with osteoporosis by DXA based on the World Health Organization criteria, whereas 280 (61.9%) were considered to have osteoporosis by QCT based on the American College of Radiology classification. Regardless of the classification system used, patients with osteoporosis had significantly higher VBQ1.5T and VBQ3.0T scores than those without osteoporosis (Table 2).
Table 2Comparison of VBQ scores between the different groups
The risk of both DXA-osteoporosis and QCT-osteoporosis was significantly associated with the VBQ1.5T score, the latter demonstrating an unadjusted OR per SD decrease of 2.762 (95% CI=1.776–4.296) and 5.469 (95% CI=3.184–9.396), respectively. This association remained statistically significant after adjustment for age, sex, BMI, serum uric acid, creatinine clearance, alkaline phosphatase, and comorbidities, which resulted in reductions in the ORs to 1.946 (95% CI=1.153–3.283) and 4.046 (95% CI=2.133–7.673), respectively. For those who underwent 3.0 T MRI examination, the VBQ3.0T score was a risk factor for osteoporosis, with an unadjusted OR of 2.205 (95% CI=1.257–3.868) for DXA-osteoporosis and 4.008 (95% CI=2.021–7.952) for QCT-osteoporosis. After adjustment, the VBQ3.0T score remained a risk factor for QCT-osteoporosis (adjusted OR=2.744, 95% CI = 2.154–6.523) but not for DXA-osteoporosis (OR=0.776, 95% CI=0.334–1.806) (Table 3).
Table 3Logistic regression analysis of risk factors associated with osteoporosis detected by DXA/QCT
In the ROC curve analysis, the VBQ score was a significant classifier of osteoporosis as detected by either DXA or QCT; the highest discriminative power was achieved by VBQ1.5T in predicting QCT-osteoporosis (AUC=0.744, 95% CI=0.685–0.803) (Fig. 5, Left). At the maximum Youden's index, the threshold values of VBQ1.5T ranged from 3.705 to 3.835 with a sensitivity between 48% and 55.6% and a specificity between 70.8% and 74.8%, while the VBQ3.0T threshold values ranged from 2.59 to 2.605 with a sensitivity between 57.6% and 67.1% and a specificity between 67.8% and 69.7% (Table 4).
Fig. 5ROC curve analysis was performed to assess the discriminability between patients with and without osteoporosis (Left) and with and without corresponding VCFs (Right). (Left) -QCT indicates osteoporosis detected by QCT; -DXA indicates osteoporosis detected by DXA.
A total of 195 patients suffered a prevalent VCF of at least one anatomic level. Both the VBO1.5T (4.18 vs. 3.83, p<.001) and VBQ3.0T scores (2.82 vs. 2.65, p=.046) were significantly higher in patients with VCF than in those without (Table 2). In discriminating between patients with and without VF, QCT-vBMD (AUC=0.794, 95% CI=0.753–0.835) demonstrated better diagnostic performance than the DXA T-score (AUC=0.754, 95% CI=0.71–0.8), the VBQ1.5T score (AUC=0.662. 95% CI=0.596–0.729) and the VBQ3.0T score (AUC=0.591, 95% CI=0.507–0.675) (Fig. 5, Right).
Discussion
The present study compared the VBQ1.5T and VBQ3.0T scores at the L1–L4 levels in postmenopausal women and older men. Overall, the VBQ score obtained from 1.5 T MRI was higher than that from 3.0 T MRI. One unexpected oddity was that the mean SI at the L1–L4 vertebral bodies measured by 1.5 T MRI was higher than that measured by 3.0 T MRI, while the mean CSF SI measured by 1.5 T MRI was lower than that measured by 3.0 T MRI. This is maybe due to differences intrinsic to MR images from different field strength [
], but this has not been demonstrated in the literature. Differences in the VBQ score for T1, T2, or short tau inversion recovery sequences have been previously reported, and physiologic variables, such as hyperlipidemia, may also influence the VBQ score by impacting bone composition [
Combination of vertebral bone quality scores from different magnetic resonance imaging sequences improves prognostic value for the estimation of osteoporosis.
]. To the best of our knowledge, this study is the first to evaluate the effect of magnetic field strength on the VBQ score.
Based on the scatter plot and Pearson's correlation, the correlation between the VBQ score and BMD ranged from 0.211 to 0.511 depending on the selected parameters. Overall, higher correlations were found for the VBQ score with QCT-vBMD than with the DXA T-score. Unlike the DXA T-score, both the MRI-based VBQ score and QCT-derived vBMD are evaluated based on 3D imaging techniques. The former is calculated as the average SI in the medulla portion on the mid-sagittal position at the L1–L4 vertebral body, while the latter involves outlining an elliptical ROI on the interior space of each vertebral body on a CT image and calculating its vBMD. This finding is to be expected, given the more considerable overlap in the ROI selection between the MRI-based VBQ score and QCT than between the VBQ score and DXA. Previous studies also confirmed a weak to moderate correlation between the VBQ score and T-score/vBMD [
Bone quality in patients with osteoporosis undergoing lumbar fusion surgery: analysis of the MRI-based vertebral bone quality score and the bone microstructure derived from microcomputed tomography.
]. In our study, the strongest correlation was observed between the VBQ1.5T score and QCT-vBMD (r=-0.511), which was higher than previously reported in series using both 1.5 T and 3.0 T MRI scanners from multiple brands or unknown MR system (r ranging from -0.27 to -0.358) [
Combination of vertebral bone quality scores from different magnetic resonance imaging sequences improves prognostic value for the estimation of osteoporosis.
]. Although previous investigations into the impact of the MR system have been performed and revealed no significant difference between the mean VBQ scores obtained from different vendors, including Philips, GE, Toshiba, and Siemens, the limited sample size (N=68) may have resulted in insufficient statistical power for detecting an association [
]. Furthermore, no details of the dominant magnetic field were provided for their study, while our results showed that the VBQ score seems to be influenced by field strength rather than MRI manufacturer.
Most previous work evaluating the correlation between lumbar BMD and VBQ score focused on discriminability for the presence of low BMD (normal vs. osteopenia/osteoporosis) [
Bone quality in patients with osteoporosis undergoing lumbar fusion surgery: analysis of the MRI-based vertebral bone quality score and the bone microstructure derived from microcomputed tomography.
] and did not investigate the performance in identifying patients with and without osteoporosis. However, lumbar osteoporosis rather than osteopenia has been shown to be a risk factor for cage subsidence and revision surgery following lateral lumbar interbody fusion [
]. In patients with osteoporosis, implant loosening following pedicle screw fixation due to loss of fixation at the screw-bone interface is one of the most common instrumentation-related complications [
], while adequate preoperative planning, including strengthening the screw-bone interface with cement augmentation and cortical bone trajectory screws, may contribute to a reduction in such adverse events [
Cortical trajectory fixation versus traditional pedicle-screw fixation in the treatment of lumbar degenerative patients with osteoporosis: a prospective randomized controlled trial.
]. On these grounds, osteoporosis screening but not osteopenia screening in patients who undergo spine surgery is more likely to be of clinical significance. To our knowledge, only one prior report, by Kim et al. [
], assessed whether the VBQ score can differentiate between patients with and without osteoporosis. However, Kim used a QCT T-score ≤-2.5 as a diagnostic criterion, which likely resulted in overestimating the number of osteoporosis patients, leading to an inaccurate analysis regarding the performance of the VBQ score in identifying patients with the disease [
Discordance in lumbar bone mineral density measurements by quantitative computed tomography and dual-energy X-ray absorptiometry in postmenopausal women: a prospective comparative study.
Clinical use of quantitative computed tomography and peripheral quantitative computed tomography in the management of osteoporosis in adults: the 2007 ISCD Official Positions.
Considering the paucity of evidence-based data regarding comparisons between the VBQ1.5T and VBQ3.0T scores, the focus of this study was to evaluate the diagnostic accuracy of different VBQ scores in predicting osteoporosis and the corresponding VCF. In this regard, we observed that every 0.1-unit increase in the VBQ1.5T score was associated with a 9.5% increase in the risk of DXA-osteoporosis and a 30.5% increase in the risk of QCT-osteoporosis, while for each 0.1-unit increase in VBQ3.0T score, the only risk increase, 17.4%, was observed for QCT-osteoporosis, indicating that the VBQ1.5T score seems to have a stronger association with lumbar osteoporosis than the VBQ3.0T score based on multivariate logistic regression analyses. Regarding the specific threshold value at which the VBQ score can differentiate between patients with and without osteoporosis, according to the DXA category, we observed a value of 3.705 for the VBQ1.5T score, with an AUC of 0.62, and 2.605 for the VBQ3.0T, with an AUC of 0.628, which is slightly lower than the previously reported AUC of 0.7 in distinguishing DXA-osteoporosis from healthy bone [
]. Another study reported a VBQ threshold value of 3.12, which is between the above two threshold values, understandable given the VBQ scores obtained from 1.5 T and 3.0 T MRI [
]. Furthermore, in that study, the threshold value was calculated in the discrimination between normal bone and osteopenic/osteoporotic bone, whereas we compared patients with and without osteoporosis in our study; this might explain the higher AUC (0.829) in the previous study. In discriminating QCT categories, the threshold value for the VBQ1.5Tscore was 3.835, with an AUC of 0.744, and that for the VBQ3.0T score was 2.59, with an AUC of 0.703, showing an overall improvement in osteoporosis diagnostic capability than for different DXA categories. Previous studies have also reported that the VBQ score shows moderate diagnostic ability in differentiating patients with normal BMD versus osteopenia/osteoporosis based on QCT, with an AUC ranging from 0.67 to 0.713 and VBQ threshold values ranging from 2.18 to 2.388 [
Bone quality in patients with osteoporosis undergoing lumbar fusion surgery: analysis of the MRI-based vertebral bone quality score and the bone microstructure derived from microcomputed tomography.
Combination of vertebral bone quality scores from different magnetic resonance imaging sequences improves prognostic value for the estimation of osteoporosis.
], which largely supports the results of the present study. However, considering the difference between our study and previous studies regarding the BMD category, we did not make a direct comparison of the results from the ROC analysis.
Most spine surgeons do not focus on the dominant magnetic field of MRI, which could be a potential contributor to the obvious discrepancies with the previous study regarding VBQ assessments. Our results first revealed a larger gap between osteoporosis diagnosis thresholds for the VBQ1.5T and VBQ3.0T scores (3.705–3.835 vs. 2.58–2.605). For the images obtained from the same MRI scanners, the choice of osteoporosis categorization (DXA vs QCT) did not seem to affect the threshold value for the VBQ score. In clinical practice, identifying the magnetic field of an MRI scanner is arguably more important than distinguishing the two methods of assessing BMD when utilizing the VBQ score to evaluate lumbar BMD.
The present study also has several limitations. First, the study population is highly homogeneous, including only patients scheduled for spine surgery. This might limit the generalizability to healthy individuals and other patient groups. Second, a clinical diagnosis of osteoporosis does not consider the presence of fragility fractures, meaning some patients should have been diagnosed with osteoporosis rather than osteopenia. Third, only 3 MRI scanners from different manufacturers were used for VBQ assessment, and whether our findings can be generalized to other MR manufacturers requires further research. Finally, despite the considerable differences between VBQ1.5T and VBQ3.0T, the rationale for these findings cannot be explained based on the present data, and a further validation study is necessary.
In conclusion, the VBQ1.5T score was generally higher than the VBQ3.0T score and was moderately correlated with QCT-derived vBMD. The VBQ1.5T score exhibited better discriminability between patients with and without osteoporosis than the VBQ3.0T score. Considering the non-negligible difference in osteoporosis diagnosis threshold values between the VBQ1.5T and VBQ3.0T scores, it is essential to clearly identify the magnetic field strength when assessing the VBQ score.
Ethical review committee statement
This study was performed in accordance with the ethical standards in the 1964 Declaration of Helsinki. This study was carried out in accordance with relevant regulations of the US Health Insurance Portability and Accountability Act (HIPAA). Details that might disclose the identity of the subjects under study have been omitted. Ethical approval was obtained from the Institutional Review Board of Shunde Hospital of Southern Medical University.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request (Zhiyun Wang, E-amil: [email protected]).
Declaration of Competing Interest
None of the authors have a conflict of interest to disclose.
Acknowledgments
We thank the patients and the patients’ guardians and families for their support. Without their participation this report of course could not have been possible; we are very thankful. The author (ZW) has received funding from Guangdong Basic and Applied Basic Research Foundation (2021A1515011508). Another author (WL) has received funding from the Scientific Research Start Plan of Shunde Hospital, Southern Medical University (SRSP2021007).
References
Dimar J
Bisson EF
Dhall S
Harrop JS
Hoh DJ
Mohamed B
et al.
Congress of neurological surgeons systematic review and evidence-based guidelines for perioperative spine: preoperative osteoporosis assessment.
Osteoporosis: recent recommendations and positions of the American Society for Bone and Mineral Research and the International Society for Clinical Densitometry.
J Bone Joint Surg Am.2021; 103 (Epub 2021/02/16): 741-747
MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis.
AJNR Am J Neuroradiol.2015; 36 (Epub 2015/09/12): 2394-2399
Bone quality in patients with osteoporosis undergoing lumbar fusion surgery: analysis of the MRI-based vertebral bone quality score and the bone microstructure derived from microcomputed tomography.
Discordance in lumbar bone mineral density measurements by quantitative computed tomography and dual-energy X-ray absorptiometry in postmenopausal women: a prospective comparative study.
Measuring and reporting of vertebral endplate bone marrow lesions as seen on MRI (Modic changes): recommendations from the ISSLS Degenerative Spinal Phenotypes Group.
Combination of vertebral bone quality scores from different magnetic resonance imaging sequences improves prognostic value for the estimation of osteoporosis.
Cortical trajectory fixation versus traditional pedicle-screw fixation in the treatment of lumbar degenerative patients with osteoporosis: a prospective randomized controlled trial.
Clinical use of quantitative computed tomography and peripheral quantitative computed tomography in the management of osteoporosis in adults: the 2007 ISCD Official Positions.
Author disclosures: WL: Grant: Scientific Research Start Plan of Shunde Hospital, Southern Medical University (B). CH: Nothing to disclose. FX: Nothing to disclose. TC: Nothing to disclose. GZ: Nothing to disclose. HY: Nothing to disclose. HC: Nothing to disclose. ZW: Grant: Guangdong Basic and Applied Basic Research Foundation (B, Paid directly to institution).
Continuous variables were expressed as means and standard deviation; Categorical variables represent counts and frequencies.
00152-3&id=gr1.jpg){kind=link}
00152-3&id=gr2.jpg){kind=link}
00152-3&id=gr3.jpg){kind=link}
00152-3&id=gr4.jpg){kind=link}
00152-3&id=gr5.jpg){kind=link}