血清sST2、FGF-23、GDF-15对阵发性房颤的预测价值探讨

Prognostic Value of Serum sST2, FGF-23 and GDF-15 in Paroxysmal Atrial Fibrillation

摘要: 目的：检测阵发性心房颤动(Paroxysmal Atrial Fibrillation, PAF)患者与非心房颤动(Atrial Fibrillation, AF)患者血清可溶性生长刺激基因表达蛋白2 (Soluble Growth Stimulating Gene Expression Protein 2, sST2)、成纤维细胞生长因子-23 (Fibroblast Gowth Factor-23, FGF-23)、人生长分化因子15 (Growth differentiation factor 15, GDF-15)水平，探讨此三个因子对PAF的预测价值。方法：选取符合入排标准的PAF患者61例作为观察组，非AF患者61例作为对照组。应用酶联吸附免疫实验法(Enzyme linked immunosorbent assay, ELISA)测定两组患者血清sST2、FGF-23、GDF-15的浓度。比较两组患者的临床资料、血液学检测指标。采用Spearman相关性分析sST2、FGF-23、GDF-15与炎症指标及其与超声心动图各参数的相关性。采用单因素及多因素logistic回归分析PAF发生的相关因素。结果：与非AF组比较，PAF组血清sST2、FGF-23、GDF-15水平均明显升高(P < 0.05)。Spearman相关性分析显示，血清sST2与FGF-23呈正相关(r = 0.219, P = 0.015)，sST2与GDF-15呈正相关(r = 0.211, P = 0.020)，FGF-23与GDF-15呈正相关(r = 0.198, P = 0.028)。sST2与中性粒细胞呈正相关(r = 0.268, P = 0.003)、与NLR呈正相关(r = 0.265, P = 0.003)，与白细胞、淋巴细胞无显著相关性；FGF-23、GDF-15与炎症指标均无显著相关性。sST2、FGF-23、GDF-15均与超声心动图各参数无显著相关性。单因素logistic回归分析显示，年龄、心率、舒张压、左房前后径、右房长径、右房横径、sST2、FGF-23、GDF-15、谷丙转氨酶、总胆红素、直接胆红素、肌酐、肾小球滤过率、高密度脂蛋白胆固醇(P < 0.05)是PAF的危险因素。多因素logistic回归分析显示，年龄(OR = 1.197, 95%CI 1.066~1.343, P = 0.002)、FGF-23 (OR = 1.006, 95%CI 1.002~1.011, P = 0.009)、GDF-15 (OR = 1.003, 95%CI 1.001~1.005, P < 0.001)、高密度脂蛋白胆固醇(OR = 19.136, 95%CI 1.050~348.601, P = 0.046)是PAF的独立危险因素。结论：1) PAF组血清sST2、FGF-23、GDF-15水平较非AF组明显升高，多因素logistic回归分析显示FGF-23、GDF-15是PAF的独立危险因素。2) 血清sST2、FGF-23、GDF-15浓度两两呈正相关性，sST2与中性粒细胞、NLR呈正相关性，提示炎症反应参与了PAF的发生发展。3) 血清sST2、FGF-23、GDF-15水平均与超声心动图参数无显著相关性，其与房颤发生的相关性可能独立于心脏结构改变之外。

Abstract: Objective: To detect the levels of serum sST2, FGF-23 and GDF-15 in patients with PAF and without AF, and investigate the predictive value of these three factors on PAF. Methods: 61 patients with PAF who met the inclusion criteria were selected as the observation group, and 61 patients without AF were selected as the control group. Serum levels of sST2, FGF-23 and GDF-15 were determined by ELISA. The clinical data and hematological indicators of the two groups were compared. Spearman correlation analysis was used to analyze the correlation between sST2, FGF-23, GDF-15 and inflammatory indicators and echocardiographic parameters. The factors related to the occurrence of PAF were analyzed by univariate and multivariate logistic regression. Results: Compared with non-AF group, serum levels of sST2, FGF-23 and GDF-15 in PAF group were significantly increased (P < 0.05). Spearman correlation analysis showed that serum sST2 was positively correlated with FGF-23 (r = 0.219, P = 0.015), and sST2 was positively correlated with GDF-15 (r = 0.211, P = 0.020). FGF-23 was positively correlated with GDF-15 (r = 0.198, P = 0.028). sST2 was positively correlated with neutrophils (r = 0.268, P = 0.003) and NLR (r = 0.265, P = 0.003), but had no significant correlation with leukocytes and lymphocytes. There was no significant correlation between FGF-23, GDF-15 and inflammatory indexes. sST2, FGF-23 and GDF-15 were not significantly correlated with echocardiographic parameters. Univariate logistic regression analysis showed that age, heart rate, diastolic blood pressure, left anteroposterior atrial diameter, right atrial long diameter, right atrial transverse diameter, sST2, FGF-23, GFF-15, alanine aminotransferase, total bilirubin, direct bilirubin, creatinine, glomerular filtration rate, and high density lipoprotein cholesterol were risk factors for PAF (P < 0.05). Multivariate logistic regression analysis showed that Age (OR = 1.197, 95%CI 1.066~1.343, P = 0.002), FGF-23 (OR = 1.006, 95%CI 1.002~1.011, P = 0.009), GDF-15 (OR = 1.003, 95%CI 1.001~1.005, P < 0.001), high density lipoprotein cholesterol (OR = 19.136, 95%CI 1.050~348.601, P = 0.046) were independent risk factors for PAF. Conclusion: 1) Serum levels of sST2, FGF-23 and GDF-15 in PAF group were significantly higher than those in non-AF group, and multiple logistic regression analysis showed that FGF-23 and GDF-15 were independent risk factors for PAF. 2) Serum sST2, FGF-23 and GDF-15 concentrations were positively correlated, and sST2 was positively correlated with neutrophils and NLR, suggesting that inflammation was involved in the occurrence and development of PAF. 3) Serum sST2, FGF-23 and GDF-15 levels were not significantly correlated with echocardiographic parameters, and their correlation with atrial fibrillation may be independent of cardiac structural changes.