OBJECTIVE: Hounsfield unit (HU) values measured using CT have been increasingly recognized to stand as a reliable corollary to dual-energy x-ray absorptiometry (DEXA) scores in evaluating bone mineral density. The authors examined the correlation between cervical HU values and DEXA T- and Z-scores and determined novel cervical HU thresholds for determining bone quality classification. METHODS: One hundred patients who underwent both cervical spine CT and DEXA, 85 patients who underwent both lumbar CT and DEXA, and 128 patients who underwent cervical and lumbar CT within 24 months at a single institution were included in this retrospective review. Two independent reviewers collected HU values from 3 cervical vertebral levels (C4–6) and 4 lumbar vertebral levels (L1–4), and the averaged values were used. Pearson’s correlation coefficient analysis was performed to compare the association of cervical HU values with lumbar HU values and T- and Z-scores. The mean cervical HU values for each DEXA classification were calculated and compared. Receiver operating characteristic (ROC) curves were created to determine the threshold and its sensitivity and specificity for diagnosis. RESULTS: Cervical (C4–6) HU values and average, lumbar, and femoral T- and Z-scores had significant correlations (0.436 > r > 0.274, all p < 0.01). A strong positive correlation between cervical and lumbar HU values was found (r = 0.79, p < 0.01). The average cervical HU value of healthy patients was 361.2 (95% CI 337.1–385.3); of osteopenic patients, 312.1 (95% CI 290.3–333.8); and of osteoporotic patients, 288.4 (95% CI 262.6–314.3). There was a significant difference between the cervical HU values of healthy and osteopenic patients (p = 0.0134) and between those of healthy and osteoporotic patients (p = 0.0304). The cervical HU value of 340.98 was 73.5% specific and 57.9% sensitive for diagnosing osteopenia with an area under the ROC (AUROC) curve of 0.655. The cervical HU value of 326.5 was 88.9% specific and 63.2% sensitive for diagnosing osteoporosis with an AUROC curve of 0.749. CONCLUSIONS: This is the second large-scale study and first with a patient population from the United States to show that HU values obtained using cervical CT were significantly correlated with bone quality based on DEXA T- and Z-scores and to establish a cervical HU threshold for determining bone quality classification. These results show that cervical HU values can and should be used to predict poor bone quality in surgical cervical spine patients.

Bone mineral density (BMD) is an essential indicator of bone strength and plays a crucial role in the clinical management of various spinal pathologies. Hounsfield units (HUs) calculated from computed tomography (CT) scans are a well-established, effective, and non-invasive method to determine bone density in the lumbar spine when juxtaposed to dual-energy X-ray absorptiometry (DEXA) scans, the gold standard for assessing trabecular bone density. Only recently have studies begun to investigate and establish HUs as a reliable and valid alternative for bone quality assessment in the cervical spine as well. In addition, multiple recent studies have identified cervical HUs as an accurate predictor of cage subsidence, an undesired complication of anterior cervical discectomy and fusion (ACDF) of anterior cervical corpectomy and fusion (ACCF) procedures. Subsidence involves migration of the spinal fusion cage into vertebral bodies, causing a loss of disk space, negatively altering spine alignment, and possibly necessitating further unwanted surgical intervention. Using the PRISMA-ScR checklist and the registered scoping review protocol (INPLASY2024100126), this review explores the current research on the use of cervical spine HU measurements as both a determinant of BMD and as a prognosticator of postoperative subsidence following cervical spine procedures (i.e., ACDFs and ACCFs) with the aim of improving clinical and surgical outcomes.

Background/Objectives: Dual-energy X-ray absorptiometry (DEXA), the gold standard for assessing bone mineral density (BMD), may yield inaccurate results in certain populations. This has prompted interest in alternative imaging methods, including the MRI-based cervical and lumbar vertebral bone quality (CVBQ and LVBQ) scores. The lumbar VBQ score is a validated MRI-based metric with excellent inter- and intra-rater reliability and established clinical utility in preoperative spine assessment, whereas the newer cervical VBQ (CVBQ) score has shown mixed results in early studies. This study investigates associations between a novel CVBQ score derived from MRI and established BMD metrics (Hounsfield unit (HU) measurements and DEXA values) to evaluate the comparative utility of these methods. Methods: A retrospective review was performed on patients who underwent cervical CT, DEXA, and non-contrast MRI of the cervical and lumbar spine between 2016 and 2022. BMD was assessed using DEXA T-scores, cervical HU measurements, and CVBQ and LVBQ scores derived from T1-weighted MRI sequences. Statistical comparisons between patients with and without osteoporosis were conducted using t-tests and Pearson correlation coefficients. Results: A total of 133 patients were included for CVBQ scoring and 127 for LVBQ. The CVBQ score showed poor correlation with both DEXA (r = −0.09, p < 0.001) and HU measurements (r = −0.34, p < 0.001), whereas a moderate correlation was found between LVBQ and DEXA scores (r = −0.34, p < 0.001). Conclusions: The LVBQ score demonstrates moderate correlation with DEXA and may serve as a reliable tool for preoperative BMD assessment. However, the CVBQ score showed weak correlation with both DEXA and HU, limiting its clinical utility in its current form. Further refinement of the CVBQ methodology is needed to enhance its accuracy and relevance for surgical risk assessment and postoperative outcome prediction.