[Correlation between serum growth differentiation factor 11 level and severity of coronary artery disease in patients with acute myocardial infarction]

Objective: To investigate the correlation between serum growth differentiation factor 11 (GDF11) level and coronary artery lesions in patients with ST-segment elevation myocardial infarction (STEMI), and the predictive efficacy of nomogram risk prediction model based on GDF11 combined with traditional risk factors on the occurrence of STEMI. Methods: This study was a retrospective cross-sectional study. Patients hospitalized in the Department of Cardiology of the 904th Hospital of Joint Logistic Support Force of People's Liberation Army of China from 2016 to 2018 were selected and divided into control group and STEMI group. The demographic data, blood lipid level, laboratory indicators of blood and GDF11 level were collected. Logistic regression analysis screened out independent correlated factors for the occurrence of STEMI. Spearman correlation analysis clarified the correlation of each indicator with the SYNTAX or Gensini scores. A nomogram risk prediction model for the risk of STEMI occurrence and the receiver operating characteristic curve was used to compare the prediction efficiency of each model. Results: A total of 367 patients were enrolled, divided into control group (n=172) and STEMI group (n=195), age (66.5±11.8), male 222 (60.49%). The serum GDF11 level of STEMI group was significantly lower than that of the control group (36.20 (16.60, 70.75) μg/L vs. 85.00 (53.93, 117.10) μg/L, P<0.001). The results of multivariate logistic regression analysis showed serum GDF11(OR=0.98, 95%CI: 0.97-0.99) and traditional independent risk factors such as smoking, diabetes, C-reactive protein, homocysteine, lipoprotein (a) and apolipoprotein A1/B were independent correlate factors for the occurrence of STEMI (P<0.05). Spearman correlation analysis showed that serum GDF11 was negatively correlated with SYNTAX score and Gensini score (P<0.05). The nomogram model constructed by serum GDF11 combined with traditional independent risk factors (AUC=0.85, 95%CI: 0.81-0.89) had better predictive value for the occurrence of STEMI than the traditional nomogram model constructed by independent risk factors(AUC=0.80, 95%CI:0.75-0.84) or serum GDF11 (AUC=0.76, 95%CI: 0.72-0.81), all P<0.01. Conclusions: Serum GDF11 is an independent correlate factor in the occurrence of STEMI and is negatively correlated with the severity of coronary artery lesions in patients with STEMI. The nomogram model constructed based on GDF11 combined with traditional risk factors can be a good predictor for the occurrence of STEMI.