基于决策曲线分析血清反式激活应答DNA结合蛋白-43、视锥蛋白样蛋白1、沉默信息调节因子3水平与脑卒中后抑郁的关系

Association of the serum levels of transactive response DNA-binding protein of 43 kDa, visinin-like protein 1, and silent information regulator 3 with post-stroke depression: A study based on decision curve analysis

摘要:

目的 基于决策曲线分析（DCA）评估血清反式激活应答DNA结合蛋白-43（TDP-43）、视锥蛋白样蛋白1（VILIP-1）、沉默信息调节因子3（SIRT3）水平与脑卒中后抑郁（PSD）的关系。方法 纳入2021年12月至2024年12月的脑卒中患者作为研究对象，共279例。根据患者出院后6个月是否发生PSD分为PSD组和非PSD组，根据PSD患者病情程度分为轻度组、中度组和重度组。采用酶联免疫吸附法检测血清TDP-43、VILIP-1、SIRT3水平。采用多因素Logistic回归分析确定影响PSD的因素；采用受试者操作特征（ROC）曲线及曲线下面积（AUC）分析血清TDP-43、VILIP-1、SIRT3水平对PSD的预测价值；采用DCA评估模型的临床应用价值。结果 PSD组脑卒中患者血清TDP-43、VILIP-1水平高于非PSD组（P<0.05），血清SIRT3水平低于非PSD组（P<0.05）；轻度组、中度组和重度组患者血清TDP-43、VILIP-1水平逐渐升高（P<0.05），血清SIRT3水平逐渐下降（P<0.05）。多因素Logistic分析表明，病灶最大径、TDP-43、VILIP-1是脑卒中患者发生PSD的危险因素（P<0.05），SIRT3是保护因素（P<0.05）。ROC曲线显示TDP-43、VILIP-1、SIRT3水平联合预测PSD患病的AUC值为0.942，显著大于各指标单独预测（P<0.05）。DCA显示，血清TDP-43、VILIP-1、SIRT3水平联合预测PSD在高风险阈值为0.08～0.95时，净获益率较高。结论 PSD患者血清TDP-43、VILIP-1水平升高，SIRT3水平降低，与病情相关，三者联合对PSD预测能力较高。

Abstract:

Objective To investigate the association of the serum levels of transactive response DNA-binding protein of 43 kDa （TDP-43）， visinin-like protein 1 （VILIP-1）， and silent information regulator 3 （SIRT3） with post-stroke depression （PSD） based on decision curve analysis （DCA）.Methods A total of 279 stroke patients from December 2021 to December 2024 were enrolled as subjects. According to the presence or absence of PSD at 6 months after discharge， the patients were divided into PSD group and non-PSD group， and according to the condition of PSD patients， they were divided into mild group， moderate group， and severe group. ELISA was used to measure the serum levels of TDP-43， VILIP-1， and SIRT3； the multivariate logistic regression analysis was used to determine the influencing factors for PSD； the receiver operating characteristic （ROC） curve and area under the curve were plotted to analyze the value of serum TDP-43， VILIP-1， and SIRT3 in predicting PSD； DCA was used to assess the clinical application value of the model.Results Compared with the non-PSD group， the PSD group had significantly higher serum levels of TDP-43 and VILIP-1 （P<0.05） and a significantly lower serum level of SIRT3 （P<0.05）. The mild group had the lowest serum levels of TDP-43 and VILIP-1， followed by the moderate group and the severe group （P<0.05）； the mild group had the highest serum level of SIRT3， followed by the moderate group and the severe group （P<0.05）. The multivariate logistic analysis showed that maximum lesion diameter， TDP-43， and VILIP-1 were risk factors for PSD in stroke patients （P<0.05）， while SIRT3 was a protective factor （P<0.05）. The ROC curve analysis showed that combined measurement of TDP-43， VILIP-1， and SIRT3 had an area under the AUC of 0.942 in predicting PSD， which was significantly higher than the AUC of each indicator alone （P<0.05）. DCA showed that combined measurement of serum TDP-43， VILIP-1， and SIRT3 had a relatively high net benefit rate for predicting PSD within the high-risk threshold of 0.08-0.95.Conclusions There are increases in the serum levels of TDP-43 and VILIP-1 and a reduction in the serum level of SIRT3 in PSD patients， which are associated with the severity of the disease. Combined measurement of these three indicators has relatively high performance in predicting PSD.