A novel CT soft tissue window-based visual scoring system for rapid assessment of bone density.

Abstract

<h4>Purpose</h4>To develop a rapid CT soft tissue window-based visual scoring system, the vertebral low-attenuation area (VLAA) score, for initial assessment of bone density, and to evaluate its correlation with established densitometric measures.<h4>Methods</h4>Each cervical vertebra (C3-C7) was scored based on the presence and continuity of low-attenuation areas (0-3 points per vertebral body), and a composite VLAA score was calculated. Correlations between VLAA scores and T-scores, Hounsfield unit (HU) values, and vertebral bone quality (VBQ) scores were assessed using Pearson's correlation. Multiple linear regression was performed to evaluate the independent predictive value of the VLAA score for the lowest T-score. Receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) were used to assess the performance of the VLAA score in predicting bone density and titanium mesh cage (TMC) subsidence.<h4>Results</h4>Cohort A included 292 patients. VLAA scores correlated significantly with T-scores (r = - 0.647 to - 0.676), HU values (r = - 0.823), and VBQ scores (r = 0.492). Multiple regression models confirmed the independent predictive role of the VLAA score for the lowest T-score (p < 0.001). The optimal VLAA cutoff for diagnosing osteopenia was 8 (sensitivity = 0.88, specificity = 0.63, AUC = 0.83), and for osteoporosis was 11 (sensitivity = 0.71, specificity = 0.88, AUC = 0.89). Cohort B included 41 patients. Regression analysis revealed a significant correlation between VLAA scores and TMC subsidence (OR = 1.78, 95% CI: 1.20-2.65), with an AUC of 0.73 (95% CI: 0.57-0.89).<h4>Conclusions</h4>The VLAA score provides a rapid visual assessment of bone density, demonstrating significant correlations with T-scores, HU values, and VBQ scores, and reliably identifying osteopenia and osteoporosis diagnosed by DXA. It also shows potential for predicting postoperative TMC subsidence.