Utility of the vertebral bone quality score to predict complications after spine surgery: a systematic review

OBJECTIVE The MRI-based vertebral bone quality (VBQ) score has emerged as a safe, convenient alternative to dual energy x-ray absorptiometry (DEXA) and CT for preoperative bone health assessment, which correlates with the outcomes of spine surgery. In this study, the authors aimed to systematically review the literature characterizing the utility of the VBQ score in predicting postoperative complications to inform operative planning and patient management. METHODS This systematic review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration no. CRD42024542755) and adhered to PRISMA guidelines. PubMed, Embase, and Cochrane databases were searched for all original research published between January 1, 2020, and May 20, 2024, that had analyzed the VBQ score as a predictor of postoperative complications following spine surgery including pedicle screw loosening (PSL), cage subsidence (CS), adjacent segment disease (ASD), proximal junctional kyphosis (PJK) or proximal junctional failure (PJF), distal junctional kyphosis (DJK) or distal junctional failure (DJF), vertebral refracture, and need for reoperation. Applications of the VBQ score and its derivatives were characterized across methods, demographics, and outcomes. The Newcastle-Ottawa Scale was used for study quality assessment. RESULTS Twenty-seven studies comprising 4068 patients, 60.5% of whom were female, with a mean age of 58.5 ± 17.0 years were eligible for study inclusion. Sixteen studies used the traditional VBQ score calculation (L1–4/L3); alternatives included C3–6/C2 (5 studies), C3–6/C5 (1 study), L1–5/L3 (1 study), L4–5/L3 (1 study), and S1/L3 (3 studies). The VBQ score significantly predicted PSL (ranges: thresholds 2.9–3.175, areas under the curve [AUCs] 0.72–0.77, ORs 1.02–5.778), CS (thresholds 2.68–4.10, AUCs 0.785–0.99, ORs 1.513–23.158), ASD (thresholds 2.91–2.95, AUCs 0.934–0.963, ORs 1.509–1.601), PJK or PJF (thresholds 2.715–3.205, AUCs 0.721–0.943, ORs 1.745–26.49), DJK or DJF (threshold 2.66, AUC 0.935, OR 1.46), refracture (combined T1- and T2-weighted VBQ nomogram threshold 0.73, and ratio of adjacent to injured vertebral levels VBQ score < 1.4; AUCs 0.753–0.838, ORs 0.32–2.239), and reoperation (threshold 2.6 to > 3, AUCs 0.702–0.808, ORs 1.569–2.096). CONCLUSIONS The data suggested that the VBQ score serves as a safe, convenient measure to predict complications after spine surgery. The lowest thresholds reported across all complications were 2.66 in the cervical spine and 2.6 in the lumbar spine, which may serve as rough cutoffs for prompting further patient testing. However, methodological heterogeneity limits guideline development. Future research with consistent methodology is necessary. Systematic review registration no.: CRD42024542755 ( www.crd.york.ac.uk/prospero )