TLS-VBQ: a novel MRI-based bone quality score for opportunistic osteoporosis screening in spinal surgery candidates

Opportunistic screening for osteoporosis remains underutilized in spinal surgery candidates, despite the critical impact of bone quality on postoperative outcomes. While the L1-4 vertebral bone quality (VBQ) score derived from MRI has been proposed as an alternative to dual-energy X-ray absorptiometry (DXA), its reliability is limited in the presence of vertebral fractures or instrumentation. We introduce the TLS-VBQ score, a novel MRI-based metric that samples T11, T12, L3, and S1, to overcome these limitations.

In this retrospective multicenter study, 218 patients undergoing spinal surgery with preoperative T1-weighted MRI and DXA were included. The TLS-VBQ score was calculated using signal intensities from T11, T12, L3, and S1, normalized to cerebrospinal fluid at L3. Diagnostic performance was evaluated using receiver operating characteristic (ROC) analysis, and robustness was tested by simulating single-level vertebral exclusions.

The TLS-VBQ score demonstrated a stronger correlation with DXA T-scores (r = - 0.772, p < 0.001) than the conventional L1-4 VBQ (r = - 0.708, p < 0.001). The area under the curve (AUC) for TLS-VBQ was 0.854, significantly higher than that of L1-4 VBQ (0.825, DeLong p < 0.05). TLS-VBQ maintained an AUC > 0.85 even with simulated exclusion of T12 or S1, whereas L1-4 VBQ dropped below 0.80 when L1 or L4 was excluded. Inter-reader reliability was also higher for TLS-VBQ (ICC = 0.914).

The TLS-VBQ score is a robust and reproducible MRI-based tool for assessing bone quality in spinal surgery candidates. It outperforms the conventional L1-4 VBQ in diagnostic accuracy, resilience to vertebral-level exclusions, and inter-observer consistency. TLS-VBQ offers a practical, non-invasive method for opportunistic osteoporosis screening, facilitating early intervention without additional scans or radiation.