Youdan Tu, Yujia Kong, Yanyu Zhu, Suyang Wu, Mingzhe Du, Hao Jiang, Ming Wang, Yanbo Peng, Yonghong Zhang, Tan Xu, Chongke Zhong
Elevated plasma FABP4 levels were independently associated with PSD after ischemic stroke, suggesting that FABP4 may serve as a potential prognostic biomarker to improve risk stratification for PSD.
PURPOSE: Fatty acid-binding protein 4 (FABP4) has been reported to be involved in cerebral ischemia injury and neuropsychiatric disorders by modulating lipid metabolism and neuroinflammatory responses. But its effect on post-stroke depression (PSD) remains unclear. This study aimed to prospectively investigate the correlation between plasma FABP4 levels and PSD in a multicenter cohort study.
METHODS: This study included 611 ischemic stroke patients from the Chinese Acute Ischemic Stroke Antihypertensive Trial. The 24-item Hamilton Depression Rating Scale was used to assess depression at 3 months after stroke, with PSD defined as a score of ≥8. Logistic regression models were performed to estimate the relationships between plasma FABP4 and the risk of PSD. A random forest regression model was used to incorporate multiple biomarkers and evaluate their predictive importance for PSD.
RESULTS: The highest tertile of FABP4 levels was positively associated with an increased risk of PSD (multivariable-adjusted odds ratio, 1.80; 95% CI, 1.10-2.92) compared with the lowest tertile. Adding FABP4 to the conventional model improved the risk reclassification for PSD (net reclassification improvement: 27.6%, p < 0.001; integrated discrimination improvement: 1.1%, p = 0.013). The random forest regression model showed that FABP4 ranked among the top 5 most important predictors for PSD, with a predictive importance higher than high-sensitivity C-reactive protein, growth differentiation factor 15, homocysteine and other biomarkers.
CONCLUSION: Elevated plasma FABP4 levels were independently associated with PSD after ischemic stroke, suggesting that FABP4 may serve as a potential prognostic biomarker to improve risk stratification for PSD.