冠状动脉旁路移植术后新发心房颤动的血浆预测因子:倾向性评分匹配研究

目的:研究术前血浆白细胞介素-1(interleukin-1,IL-1)、IL-6、肿瘤坏死因子-α(tumor necrosis factor-α,TNF-α)、同型半胱氨酸(homocysteine,Hcy)、内皮素-1(endothelin-1,ET-1)水平与冠状动脉旁路移植(coronary artery bypass grafting,CABG)术后新发心房颤动(简称房颤,atrial fibrillation,AF)之间的相关性。方法:选取2017年1月1日至12月30日在北京大学人民医院心脏中心行择期单纯CABG的患者148人,采集患者术前24 h内空腹静脉血,用酶联免疫吸附法检测血浆IL-1、IL-6、TNF-α、Hcy、ET-1的含量,其中术后新发房颤患者39人,以术后是否新发房颤为标准将患者分为房颤组和非房颤组,进行1 ∶1倾向性评分匹配后两组各38人。将匹配后两组患者的5个指标分别进行配对t检验,如果不符合正态分布则进行Wilcoxon符号秩和检验,然后将各指标进行条件Logistic回归分析,探究术前各指标血浆水平与术后新发房颤之间的相关性。结果:经1 ∶1倾向性匹配后,两组均衡可比,匹配后房颤组的IL-1、IL-6、TNF-α、Hcy血浆水平均大于非房颤组[(0.867±0.589) ng/L vs. (0.742±0.262) ng/L,21.55 (6.50, 209.90) ng/L vs. 17.95 (3.60, 86.70) ng/L,20.30 (5.70, 361.00) ng/L vs. 21.50 (7.50, 251.80) ng/L,(0.29±0.11) μmol/L vs. (0.27±0.09) μmol/L],但两组差异均无统计学意义(P=0.165,P=0.891,P=0.817,P=0.285),经条件Logistic回归分析后,上述4个变量均不是CABG术后新发房颤的预测因子。倾向性匹配后,房颤组和非房颤组的ET-1分别为(25.80±6.20) ng/L、(29.10±8.54) ng/L,患者术前较低的血浆ET-1水平与CABG术后新发房颤有统计学相关性(P=0.003),条件Logistic回归分析也显示术前血浆ET-1水平与CABG术后新发房颤有相关性(P=0.039,调整后OR=0.637,95%CI: 0.415~0.977)。结论:CABG术后新发房颤患者的术前血浆IL-1、IL-6、TNF-α及Hcy的水平均高于术后无房颤的患者,但差异无统计学意义,术前血浆低ET-1水平与CABG术后新发房颤有关。

Preoperative plasma predictive factors of new-onset atrial fibrillation after coronary artery bypass graft surgery:A propensity score matching study

Objective: To study the relationship between preoperative plasma interleukin-1 (IL-1), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), homocysteine (Hcy), endothelin-1 (ET-1) levels and new-onset atrial fibrillation (AF) after coronary artery bypass grafting (CABG). Methods: In the study, 148 patients who underwent isolated CABG in Peking University People’s Hospital from January 1, 2017 to December 30, 2017 were enrolled, of whom 39 had new-onset AF. The fasting venous blood was collected within 24 hours before the surgery. The preoperative plasma IL-1, IL-6, TNF-α, Hcy, ET-1 levels were detected by enzyme-linked immunosorbent assay (ELISA). The patients were divided into AF group and non-AF group according to whether new-onset AF occurred after operation. After 1 ∶1 propensity score matching (PSM), 38 people were in each group. The paired sample t-tests were performed on the five factors’ concentrations of the matched AF group and the non-AF group respectively. If the concentration values did not conform to the normal distribution, the Wilcoxon signed rank sum test was performed. Conditional Logistic regression analysis was performed on the concentrations of the five indicators to explore the correlation between preoperative plasma concentrations of IL-1, IL-6, TNF-α, Hcy, ET-1 and postoperative new-onset AF after CABG. Results: After a 1 ∶1 propensity score matching, the AF group was comparable to the non-AF group. The concentrations of IL-1, IL-6, TNF-α, and Hcy in the AF group were higher than those in the non-AF group [(0.867±0.589) ng/L vs. (0.742±0.262) ng/L, 21.55 (6.50, 209.90) ng/L vs. 17.95 (3.60, 86.70) ng/L, 20.30 (5.70, 361.00) ng/L vs. 21.50 (7.50, 251.80) ng/L, (0.29±0.11) μmol/L vs. (0.27±0.09) μmol/L], but the differences were not statistically significant (P=0.165, P=0.891, P=0.817, P=0.285). After the conditional Logistic regression analysis, the above four variables were not predictors of new-onset AF after CABG. The concentrations of ET-1 in the matched AF group and non-AF group were (25.80±6.20) ng/L and (29.10±8.54) ng/L, respectively. The correlation between preoperative low plasma ET-1 concentration and the new-onset AF after CABG were statistically significant (P=0.003). After conditional Logistic regression analysis, preoperative plasma ET-1 concentration was correlated with postoperative new-onset AF after CABG (P=0.039, adjusted OR=0.637, 95%CI: 0.415-0.977). Conclusion: The levels of preoperative plasma IL-1, IL-6, TNF-α and Hcy in the patients with new-onset AF after CABG were higher than those in the patients without AF, but the difference was not statistically significant. Preoperative plasma low ET-1 concentration was statistically associated with new-onset AF after CABG.