# Application of pedicle bone quality scoring and Hounsfield unit in predicting the risk of cage subsidence after single-segment lumbar fusion surgery

## TL;DR

This study shows that combining pedicle bone quality scores and Hounsfield units improves the prediction of cage subsidence after lumbar fusion surgery.

## Contribution

A novel predictive model combining pedicle bone quality scoring and Hounsfield units for assessing cage subsidence risk is proposed and validated.

## Key findings

- Hounsfield units and pedicle bone quality scores are independent predictors of cage subsidence.
- The combined model outperformed individual predictors with an AUC of 0.831.
- The combined model achieved a sensitivity of 76.7% and specificity of 79.0%.

## Abstract

This study aimed to investigate the clinical value of combining pedicle bone quality scoring and Hounsfield units (HUs) in predicting the risk of cage subsidence following single-segment lumbar fusion surgery.

We conducted a retrospective analysis of clinical data from 160 patients who underwent single-segment lumbar fusion surgery at Panyu District Traditional Chinese Medicine Hospital between January 2017 and May 2023. Clinical data comparisons and multivariate logistic regression analyses were performed using SPSS 27.0. Receiver operating characteristic (ROC) curves were generated using MedCalc 23 to evaluate diagnostic efficacy.

Comparative analysis of clinical data revealed statistically significant differences between the two groups in terms of age, hypertension, HUs, vertebral body quality (VBQ) scores, and pedicle bone quality (PBQ) scores (P < 0.05). The multivariate logistic regression analysis indicated that HUs (OR = 0.98, 95% CI: 0.97–0.99; P = 0.018) and PBQ scores (OR = 3.99, 95% CI: 1.63–9.79; P = 0.002) are independent predictors of cage subsidence in patients undergoing single-segment lumbar interbody fusion. ROC analysis demonstrated that the area under the curve (AUC) for HUs was 0.781 (95% CI: 0.709–0.843), with an optimal threshold of 95, yielding a maximum Youden index of 0.460, corresponding to a diagnostic sensitivity of 60.0% and specificity of 86.0%. The AUC for PBQ scores was 0.772 (95% CI: 0.699–0.835), with an optimal threshold of 3.088, yielding a maximum Youden index of 0.480, corresponding to a diagnostic sensitivity of 85.0% and specificity of 63.0%. Furthermore, the predictive model constructed by combining HUs and PBQ scores achieved an AUC of 0.831 (95% CI: 0.763–0.885), with a maximum Youden index of 0.557, resulting in a diagnostic sensitivity of 76.7% and specificity of 79.0%. DeLong’s test results indicated that the combined diagnostic model outperformed the individual use of HUs and PBQ scores (P-values of 0.049 and 0.014, respectively).

Pedicle bone quality scoring and HUs provide significant reference value in diagnosing cage subsidence following single-segment lumbar interbody fusion. The predictive model constructed through the combined assessment of these two factors demonstrated superior diagnostic efficacy.