Predictive value of the product term BRI × carotid plaque thickness for stroke and transient ischemic attack: a...

Carotid plaque thickness and BRI are each associated with an increased risk of stroke. However, the value of their interaction in predicting stroke remains unclear. This study aimed to investigate the predictive performance of maximum carotid plaque thickness, BRI, and their interaction for the occurrence of stroke or TIA.

In this prospective cohort study, 230 elderly Chinese adults were enrolled. Baseline measurements included maximum carotid plaque thickness and BRI, and an interaction term was calculated. Participants were followed for 1 year, during which the incidence of stroke or TIA was recorded. Multivariable logistic regression was used to assess the predictive value of each variable. Receiver operating characteristic curve analysis with 95% confidence intervals was conducted to determine the area under the curve (AUC) for model performance, and internal validation using bootstrap resampling (B = 1,000) was performed to correct for potential optimism.

Both maximum plaque thickness (3.305 ± 0.515 mm vs. 2.245 ± 0.820 mm, p < 0.001) and BRI (4.872 ± 1.240 vs. 3.751 ± 0.916, p < 0.001) were significantly higher in the stroke group than in the non-stroke group. Logistic regression analysis showed that maximum plaque thickness (Full multivariable adjustment: OR = 3.619, 95%CI: 1.781-7.355, p = 0.00038) and BRI (Full multivariable adjustment: OR = 3.116, 95% CI: 1.784-5.444, p = 0.00006) were both independent predictors. ROC analysis revealed that the interaction term yielded the highest AUC (0.9192, 95% CI: 0.8772-0.9612), compared with maximum plaque thickness (0.8819, 95% CI: 0.8353-0.9285) and BRI (0.7632, 95% CI: 0.6266-0.8997). Statistical comparisons indicated that the interaction model significantly outperformed BRI, while its advantage over maximum plaque thickness was numerically higher but did not reach statistical significance, likely due to the limited number of events. After bootstrap correction (B = 1,000), the optimism-corrected AUC of the interaction model was 0.897 (95% CI: 0.788-0.954).

Both maximum carotid plaque thickness and BRI independently predict the risk of stroke and TIA after adjusting for confounders. Their interaction further improves predictive performance. Combined assessment of these indicators may optimize early stroke risk stratification and warrants further validation in clinical practice.