2026 年第 3 期 第 21 卷

急性脑梗死伴脑微出血患者血清过氧化物酶体增殖物激活受体γ辅助激活因子1α和可溶性CD40配体水平与认知障碍的关系及其预测价值

Correlation between serum peroxisome proliferator-activated receptor γ coactivator 1α and soluble CD40 ligand levels with cognitive impairment and their predictive value in patients with acute cerebral infarction complicated with cerebral microbleeds

作者：宋晓征1何佳1唐一植2

英文作者：Song Xiaozheng1 He Jia1 Tang Yizhi2

单位：1三六三医院神经内科，成都610041；2三六三医院超声医学中心，成都610041

英文单位：1Department of Neurology 363 Hospital Chengdu 610041 China; 2Ultrasound Medical Center 363 Hospital Chengdu 610041 China

关键词：急性脑梗死；脑微出血；认知障碍；过氧化物酶体增殖物激活受体γ辅助激活因子1α；可溶性CD40配体

英文关键词：Acutecerebralinfarction;Cerebralmicrobleeds;Cognitiveimpairment;Peroxisomeproliferator-activatedreceptorγcoactivator1α;SolubleCD40ligand

• 摘要：

• 目的 探讨急性脑梗死（ACI）伴脑微出血患者血清过氧化物酶体增殖物激活受体γ辅助激活因子1α（PGC-1α）、可溶性CD40配体（sCD40L）水平与认知障碍的关系及其预测价值。方法 选取2021年6月至2025年3月三六三医院神经内科收治的291例ACI伴脑微出血患者作为研究对象，根据是否合并认知障碍将患者分为认知障碍组（124例）与非认知障碍组（167例）。采用酶联免疫吸附试验法检测血清PGC-1α、sCD40L水平。多因素Logistic回归分析ACI伴脑微出血患者发生认知障碍的因素，受试者工作特征（ROC）曲线分析基于PGC-1α、sCD40L所构建的联合预测模型对ACI伴脑微出血患者发生认知障碍的预测价值。结果 认知障碍组血清PGC-1α水平低于非认知障碍组［（0.58±0.12）μg/L比（0.85±0.10）μg/L］，sCD40L水平高于非认知障碍组［（330±59）ng/L比（226±21）ng/L］（均P＜0.001）。多因素Logistic回归分析结果显示，高美国国立卫生研究院卒中量表（NIHSS）评分、高水平sCD40L是ACI伴脑微出血患者认知障碍的危险因素，高水平PGC-1α是保护因素(均P＜0.05)。以上述多因素回归所得概率风险模型构建ACI伴脑微出血患者认知障碍的联合预测评估模型：Logit［P/(1-P)］=-0.166+0.141×NIHSS评分-0.536×PGC-1α+0.716×sCD40L，以模型值为预测效应量，行ROC分析，结果显示联合预测模型曲线下面积为0.828(95%置信区间：0.673～0.988)，敏感度为84.7%，特异度为82.0%。结论 ACI伴脑微出血患者血清PGC-1α水平降低，sCD40L水平增高与认知障碍有关，基于PGC-1α、sCD40L所构建的ACI伴脑微出血患者认知障碍的联合预测模型能较好地评估认知障碍风险。

• Objective To investigate the correlation between serum peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) and soluble CD40 ligand (sCD40L) levels with cognitive impairment and their predictive value in patients with acute cerebral infarction (ACI) complicated with cerebral microbleeds. Methods A total of 291 patients with ACI complicated with cerebral microbleeds admitted to the Department of Neurology, 363 Hospital from June 2021 to March 2025 were enrolled as research subjects. According to the presence or absence of cognitive impairment, the patients were divided into the cognitive impairment group (124 cases) and non-cognitive impairment group (167 cases). Serum levels of PGC-1α and sCD40L were detected by enzyme-linked immunosorbent assay. Multivariate Logistic regression analysis was used to identify the influencing factors for cognitive impairment in patients with ACI complicated with cerebral microbleeds. Receiver operating characteristic (ROC) curve was used to analyze the predictive value of the combined prediction model constructed based on PGC-1α and sCD40L for cognitive impairment in patients with ACI complicated with cerebral microbleeds. Results The serum PGC-1α level in the cognitive impairment group was significantly lower than that in the non-cognitive impairment group ［(0.58±0.12)μg/L vs (0.85±0.10)μg/L］, while the serum sCD40L level was significantly higher than that in the non-cognitive impairment group ［(330±59)ng/L vs (226±21)ng/L］(both P＜0.001). Multivariate Logistic regression analysis showed that high National Institutes of Health Stroke Scale (NIHSS) score and high sCD40L level were risk factors for cognitive impairment in patients with ACI complicated with cerebral microbleeds, while high PGC-1α level was a protective factor (all P＜0.05). A combined prediction model for cognitive impairment in patients with ACI complicated with cerebral microbleeds was constructed based on the probability risk model derived from the above multivariate regression: Logit［P/(1-P)］=-0.166+0.141×NIHSS score-0.536×PGC-1α+0.716×sCD40L. ROC analysis with the model value as the predictive effect size showed that the area under the curve of the combined prediction model was 0.828 (95% confidence interval: 0.673-0.988), with a sensitivity of 84.7% and specificity of 82.0%.  Conclusion Decreased serum PGC-1α level and increased sCD40L level are associated with cognitive impairment in patients with ACI complicated with cerebral microbleeds. The combined prediction model constructed based on PGC-1α and sCD40L can well evaluate the risk of cognitive impairment in these patients.