1 Introduction

Osteoporosis is a systemic skeletal disorder characterized by reduced bone mineral density (BMD) and deterioration of trabecular and cortical microarchitecture, both of which contribute to increased bone fragility and a heightened risk of fracture. Vertebral compression fractures are the most common osteoporotic fractures and are strongly associated with chronic pain, progressive deformity, functional decline, and reduced quality of life. Despite their clinical importance, a substantial proportion of vertebral fractures remain undiagnosed or are misattributed to nonspecific back pain, particularly in the early stages of disease.¹

Dual-energy X-ray absorptiometry (DEXA) remains the standard modality for assessing BMD; however, several well-recognized limitations reduce its diagnostic accuracy. Degenerative spinal changes, aortic calcifications, and metallic instrumentation can artifactually elevate lumbar BMD values, leading to underestimation of osteoporosis severity.²,³ Many patients with normal or mildly reduced BMD still experience osteoporotic vertebral fractures.²

Computed tomography (CT)-derived Hounsfield unit (HU) measurements have emerged as an opportunistic, three-dimensional surrogate for trabecular bone density, with demonstrated value in predicting pedicle screw loosening and mechanical failure following spinal instrumentation.⁶ However, HU predominantly reflects mineral content and provides limited insight into marrow composition or microstructural deterioration—key elements of true bone quality.

Recently, the vertebral bone quality (VBQ) score, derived from T1-weighted magnetic resonance imaging (MRI), has been introduced as a non-invasive and radiation-free biomarker of trabecular bone quality. VBQ quantifies marrow fat infiltration by calculating the ratio of vertebral trabecular signal intensity to cerebrospinal fluid (CSF) signal intensity.⁴,⁵ Prior studies have shown that VBQ correlates with DEXA T-scores, differentiates normal from osteopenic and osteoporotic bone, and predicts vertebral compression fractures with diagnostic performance comparable to or exceeding that of DEXA.⁴,⁶ VBQ has also demonstrated high reproducibility across scanners and MRI field strengths.

Furthermore, VBQ has gained increasing interest in the surgical context. Elevated VBQ scores are strongly associated with pedicle screw loosening, instrumentation failure, and adjacent-level fractures after lumbar fusion.⁷,⁸ Taken together, VBQ and HU provide complementary insights into skeletal integrity: HU reflects bone density (“quantity”), whereas VBQ reflects trabecular marrow composition and microarchitectural quality (“quality”). Their combined use may enhance preoperative risk stratification and guide surgical planning in patients with suspected osteoporosis.

Teriparatide (PTH 1-34) is an established anabolic agent that stimulates osteoblast activity, enhances collagen cross-linking, reduces marrow adiposity, and improves trabecular microarchitecture.³ While the effects of PTH on DEXA-derived BMD and CT-based HU have been documented, its influence on MRI-based qualitative metrics—particularly VBQ—remains largely unknown.

Therefore, the purpose of this study was to evaluate perioperative changes in MRI-derived VBQ scores among patients undergoing single-level lumbar fusion and to compare these changes between individuals who received PTH therapy and untreated controls. A secondary aim was to relate VBQ changes to HU and BMD measurements, thereby clarifying the relative contributions of bone density and bone quality to postoperative skeletal integrity in the osteoporotic spine.

2 Materials and methods

2.5

2.5 MRI-based VBQ score assessment

All imaging measurements were performed by observers blinded to clinical information and treatment details. Preoperative and 12-month postoperative lumbar MRI scans were obtained using a 3-T system. VBQ scores were calculated from mid-sagittal T1-weighted sequences according to previously validated methods. A circular region of interest (ROI) was manually placed within the central trabecular marrow of three vertebral bodies (T12–L2), carefully avoiding cortical bone, endplates, and basivertebral venous structures (Fig. 1). These levels were selected to minimize the influence of the instrumented segment.

Close

Fig. 1 Calculation of the vertebral bone quality (VBQ) score using T1-weighted sagittal MRI. Regions of interest (ROIs) were manually placed in the central trabecular regions of three consecutive vertebral bodies (T12–L2), carefully avoiding the endplates and basivertebral veins. The signal intensities (SIs) of the vertebral ROIs were averaged and divided by the SI of cerebrospinal fluid (CSF) measured at the L3 level. The VBQ score was calculated using the following formula: VBQ score = mean SI (T12–L2)/SI (CSF).

Export to PPT

The signal intensities (SIs) of the three vertebral ROIs were averaged and divided by the SI of CSF measured at the L3 level to compute the VBQ score:

VBQ score = mean SI (T12–L2 trabecular bone)/SI (L3 CSF)

This approach has been validated in previous studies as a reproducible and reliable MRI-based surrogate marker of vertebral bone quality.⁴,⁵

2.6

2.6 CT-based HU assessment

Axial CT images of the T12–L2 vertebrae were acquired preoperatively and at 12 months postoperatively using a 1-mm slice thickness. All images were reformatted in axial planes parallel to the vertebral endplates. For HU measurements, an elliptical region of interest (ROI) was placed centrally within the cancellous bone, carefully excluding cortical margins and vascular channels to avoid volume-averaging artifacts.

For each vertebral body, HU values were obtained from three non-consecutive axial slices—immediately below the superior endplate, at the mid-vertebral level, and immediately above the inferior endplate. The HU values from the three slices were averaged to yield the mean HU value for each vertebra, and these values were subsequently averaged across T12–L2 (Fig. 2).

Close

Fig. 2 Measurement of vertebral cancellous bone density using axial CT. Axial CT images were reformatted parallel to the vertebral endplates. For each level (T12–L2), Hounsfield units (HU) were obtained from three axial slices—just below the superior endplate, at mid-body, and just above the inferior endplate—using an elliptical ROI placed centrally within the cancellous bone while excluding cortical margins. HU values from the three slices were averaged to determine the mean HU for each vertebra.

Export to PPT

HU measurements served as a quantitative index of trabecular bone density and have been shown in previous studies to correlate with DEXA T-scores and the risk of pedicle screw loosening.³,⁶

3 Result

4 Discussion

This study is the first to demonstrate that perioperative PTH therapy significantly improves MRI-based VBQ, providing direct evidence that PTH induces measurable changes in vertebral marrow composition. By integrating VBQ—a qualitative, microarchitecture-sensitive MRI biomarker—with traditional bone density metrics, this study introduces a novel imaging framework for evaluating bone quality and treatment response in spinal fusion candidates.

Our results demonstrated that perioperative PTH administration significantly improved vertebral bone quality in patients undergoing lumbar spinal fusion, as indicated by reductions in VBQ scores. Because elevated VBQ scores reflect increased marrow fat infiltration and deterioration of trabecular structure, the observed decrease suggests a qualitative enhancement of cancellous bone. These findings support the hypothesis that PTH improves trabecular bone integrity and suggest that VBQ serves as a sensitive, non-invasive imaging marker capable of monitoring both therapeutic response and perioperative bone health.

VBQ, derived from T1-weighted MRI, has gained recognition as an imaging surrogate for trabecular microarchitecture. It reflects the degree of bone marrow fat accumulation—a phenomenon that increases with aging and osteoporosis.⁵,¹³,¹⁴,¹⁵,¹⁶ Prior studies have linked high VBQ scores to osteoporotic vertebral fractures, pedicle screw loosening, and diminished biomechanical strength.⁴,⁶-⁸ In the present study, patients treated with PTH showed a significant reduction in VBQ, supporting the notion that anabolic therapy can favorably alter marrow composition and improve bone quality at the microstructural level.

The relationship between ΔVBQ and postoperative BMD aligns with earlier evidence demonstrating an inverse correlation between marrow fat content and bone mineral density.¹³,¹⁷ Although HU values measured by CT are validated as quantitative markers of bone density correlated with DEXA T-scores and mechanical strength,¹⁸,¹⁹ HU primarily captures mineral content and does not reflect alterations in marrow composition or trabecular microstructure. The greater sensitivity of VBQ compared with HU in detecting early postoperative changes may therefore highlight the unique value of VBQ as a qualitative biomarker responsive to anabolic intervention.

Importantly, VBQ and HU provide complementary information regarding vertebral integrity, reflecting distinct biological aspects of bone health. HU values, derived from CT attenuation, quantify mineral density and correlate strongly with DEXA T-scores and vertebral mechanical strength. In contrast, VBQ detects qualitative alterations in trabecular microarchitecture by capturing marrow fat infiltration on T1-weighted MRI—a hallmark of deteriorated bone quality that cannot be evaluated using density-based techniques. Whereas HU represents the “quantity” of bone, VBQ reflects its “quality,” including microstructural deterioration and marrow compositional shifts. The ability to evaluate both dimensions provides a more comprehensive assessment of vertebral strength than either parameter alone. In this context, the significant improvement in VBQ observed with PTH therapy underscores the unique sensitivity of MRI-based bone quality metrics to anabolic, microarchitecture-level changes that may precede measurable increases in mineral density.

The biological plausibility of these findings is supported by the known anabolic mechanisms of PTH, which stimulates osteoblast activity, enhances collagen cross-linking, suppresses marrow adipogenesis, and promotes new bone formation.,^3,9–11 Prior clinical studies have shown that PTH increases spinal BMD and reduces fracture risk more effectively than antiresorptive therapies in selected patients.¹⁰ In our cohort, both PTH usage and treatment duration emerged as independent predictors of ΔVBQ on regression analysis, underscoring the direct effect of pharmacologic intervention on vertebral bone quality. The absence of significant associations between VBQ improvement and demographic variables such as age or sex further supports this interpretation.

These findings carry important clinical implications. Patients with osteoporosis undergoing spinal fusion are at increased risk of pseudarthrosis, pedicle screw loosening, and implant failure.¹² Because many patients undergoing lumbar MRI do not routinely receive preoperative DEXA screening,²⁰ VBQ scoring offers an accessible, radiation-free adjunct for evaluating vertebral bone quality using routinely acquired MRI. Preoperative identification of patients with high VBQ scores may facilitate risk stratification and guide selection of bone-enhancing therapy or augmented fixation strategies.⁶,⁷ Moreover, the demonstrated sensitivity of VBQ to PTH-induced qualitative improvements suggests its potential as a monitoring tool during osteoporosis treatment in the perioperative period.

Several limitations should be acknowledged. First, the retrospective single-center design entails inherent risks of selection and information bias and precludes causal inference, despite inclusion of consecutive patients and multivariable adjustment. Second, although a post hoc power analysis indicated adequate statistical power for the primary correlation analyses (n = 102), subgroup analyses may still be underpowered to detect smaller effects and should therefore be regarded as exploratory. Third, VBQ was measured at non-instrumented thoracolumbar levels (T12–L2) to avoid MRI artefact; thus, the findings primarily reflect global vertebral bone quality and may not be fully generalizable to the instrumented index level (L4–5). Fourth, although imaging protocols were standardized and inter- and intra-rater reliability was excellent, manual ROI placement may limit reproducibility across centers. Fifth, heterogeneous use of antiresorptive agents precluded stratified analyses or direct comparisons with anabolic therapy. Sixth, this study focused on imaging biomarkers and did not evaluate long-term clinical outcomes, such as fusion status, implant-related complications, reoperation, or patient-reported outcomes, nor was histological validation of marrow fat performed; therefore, the prognostic value of VBQ for surgical success remains uncertain. Previous reports have shown that preoperative VBQ influences JOABPEQ and SF-36 outcomes five years after lumbar fusion,²¹ underscoring the importance of long-term follow-up. Finally, the modest correlations and R² values reflect the multifactorial nature of vertebral strength, indicating that VBQ should be considered an adjunctive, rather than definitive, imaging biomarker. Future prospective multicenter studies are warranted.

5 Conclusion

In this retrospective cohort study, we demonstrated that perioperative PTH administration significantly improves vertebral bone quality, as evidenced by reductions in VBQ scores. These findings may be consistent with mechanisms involving reduced marrow adiposity, as suggested by prior experimental and imaging studies; however, marrow adiposity was not directly assessed in the present study. that PTH favorably modifies marrow composition and may enhance overall bone health in the fusion environment. VBQ represents a non-invasive, radiation-free, and cost-effective imaging biomarker for evaluating vertebral bone quality and may be particularly valuable when DEXA is unavailable or unreliable. Importantly, this study is the first to show that VBQ can detect qualitative improvements in vertebral bone resulting from teriparatide therapy, highlighting the potential of MRI-based bone quality assessment as a sensitive tool for monitoring anabolic treatment effects in the perioperative management of osteoporotic spine patients. Incorporating VBQ assessment into preoperative evaluation may help identify patients with compromised bone quality and optimize osteoporosis management strategies. Future multicenter and prospective studies are needed to validate VBQ as a standard tool for diagnosis, monitoring, and prediction of long-term fusion and instrumentation outcomes.

Guardian/patient's consent

The study protocol for this retrospective and noninvasive study was approved by the institutional review board of our hospital. Informed consent was obtained in the form of an opt-out, and the need to obtain informed consent from individual patients was waived by the ethics committee of our institution.

Ethical statement

This study was conducted following the ethical principles of the Declaration of Helsinki, and the study protocol for this retrospective and noninvasive study was approved by the institutional review board of Oita University. Informed consent was obtained in the form of an opt-out, and the need to obtain informed consent from individual patients was waived by the ethics committee of Oita University.

Credit author statement

Tetsutaro Abe: Analyze and Write the original draft.

Masashi Miyazaki: Conceptualization, Methodology, and Review & editing.

Noriaki Sako: Validation, Data curation.

Nobuhiro Kaku: Supervision.

All authors have read and approved the final version to be published.

Funding

None.

Funding statement

This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.