睡眠状况与中青年急性心肌梗死风险的关联：基于匹兹堡睡眠质量指数量表的病例–对照研究

Association between Sleep Status and Acute Myocardial Infarction Risk in Young and Middle-Aged Adults: A Case-Control Study Based on the Pittsburgh Sleep Quality Index

摘要: 目的：探讨睡眠状况与中青年急性心肌梗死(Acute Myocardial Infarction, AMI)风险的关系，探讨匹兹堡睡眠质量指数量表(Pittsburgh Sleep Quality Index, PSQI)总分与AMI发病的独立关联及其关联特征。方法：采用病例–对照研究设计，纳入中青年急性心肌梗死患者99例同时选取同期于安徽医科大学第二附属医院就诊的中青年非心源性胸痛病人，心肌酶学阴性且冠状动脉影像学检查证实冠状动脉阴性者68例作为对照组。采用PSQI评估睡眠状况。比较两组PSQI总分差异，并通过多因素Logistic回归分析PSQI总分与AMI风险的独立关联。采用散点图与线性拟合线探讨PSQI总分与AMI风险的线性关系。进一步进行主成分分析(PCA)和带L1惩罚的Logistic回归(LASSO)筛选关键睡眠维度。结果：总共纳入167例研究对象，其中AMI组99例，对照组68例。AMI组匹兹堡睡眠质量指数(PSQI)总分显著高于对照组[11.0 (9.0, 13.0) vs. 5.0 (4.0, 6.0), P < 0.001]，且97.0%的AMI患者存在不同程度的睡眠问题，而对照组仅36.8%存在轻度睡眠问题。单因素Logistic回归显示，BMI、高血压病史、总胆固醇、低密度脂蛋白、糖化血红蛋白、冠心病家族史及PSQI总分与AMI发生显著相关(P < 0.05)。多因素Logistic回归逐步调整人口学、传统心血管及代谢性危险因素后，PSQI总分仍为AMI风险的独立预测因子(OR = 2.284, 95% CI: 1.776~3.187, P < 0.001)。散点图分析显示PSQI与AMI风险呈持续递增的线性剂量–反应关系(非线性检验P = 0.63)。主成分分析(PCA)及LASSO回归结果表明，“睡眠障碍”“入睡时间”“睡眠时间”和“日间功能障碍”是与AMI风险关联最密切的关键睡眠维度。结论：中青年AMI患者睡眠质量显著下降，PSQI总分为AMI风险的独立预测因子。PSQI与AMI风险呈线性剂量–反应关系，睡眠障碍、入睡困难及日间功能障碍可能在发病中起重要作用。改善睡眠质量可能是预防中青年AMI的重要干预方向。

Abstract: Objective: To explore the relationship between sleep status and the risk of acute myocardial infarction (AMI) in young and middle-aged adults, and to investigate the independent association between the total score of the Pittsburgh Sleep Quality Index (PSQI) and AMI onset, as well as the characteristics of this association. Methods: A case-control study was conducted, enrolling 99 young and middle-aged AMI patients. Meanwhile, 68 young and middle-aged patients with non-cardiac chest pain, negative myocardial enzymes, and confirmed negative coronary arteries by coronary imaging, who visited The Second Affiliated Hospital of Anhui Medical University during the same period, were selected as the control group. Sleep status was assessed using the PSQI. The difference in PSQI total scores between the two groups was compared, and the independent association between the PSQI total score and AMI risk was analyzed using multivariate logistic regression. Scatter plots with linear fit lines were used to explore the linear relationship between the PSQI total score and AMI risk. Furthermore, principal component analysis (PCA) and L1-penalized logistic regression (LASSO) were performed to identify key sleep dimensions. Results: A total of 167 subjects were included, comprising 99 in the AMI group and 68 in the control group. The PSQI total score was significantly higher in the AMI group than in the control group [11.0 (9.0, 13.0) vs. 5.0 (4.0, 6.0), P < 0.001]. Furthermore, 97.0% of AMI patients had varying degrees of sleep problems, while only 36.8% in the control group had mild sleep problems. Univariate logistic regression showed that BMI, history of hypertension, total cholesterol, low-density lipoprotein, glycated hemoglobin, family history of coronary heart disease, and PSQI total score were significantly associated with AMI occurrence (P < 0.05). After stepwise adjustment for demographic, traditional cardiovascular, and metabolic risk factors in multivariate logistic regression, the PSQI total score remained an independent predictor of AMI risk (OR = 2.284, 95% CI: 1.776~3.187, P < 0.001). Scatter plot analysis indicated a continuous, increasing linear dose-response relationship between the PSQI score and AMI risk (P for nonlinearity = 0.63). Results from PCA and LASSO regression indicated that “sleep disturbance”, “sleep latency”, “sleep duration”, and “daytime dysfunction” were the key sleep dimensions most closely associated with AMI risk. Conclusion: Sleep quality is significantly impaired in young and middle-aged AMI patients. The PSQI total score is an independent predictor of AMI risk. A linear dose-response relationship exists between the PSQI score and AMI risk, with sleep disturbance, difficulty falling asleep, and daytime dysfunction potentially playing important roles in the pathogenesis. Improving sleep quality may be an important intervention direction for preventing AMI in young and middle-aged adults.