Establishment Of Diagnostic Model For Acute Pulmonary Embolism Based On Electrocardiogram Combined With D-dimer

Objective:To investigate the diagnostic effect of multiple ECG findings combined with D-dimer on acute pulmonary embolism.Methods:Patients with suspected acute pulmonary embolism admitted to the hospital from January 2015 to December 2019 were obtained through the electronic medical record system of the First Affiliated Hospital of Xinjiang Medical University.Pulmonary embolism was definitely diagnosed by pulmonary CT arteriography(CTPA).The patients were divided into pulmonary embolism control group and pulmonary embolism group for risk stratification.The acute pulmonary embolism group was divided into high-risk group,intermediate-risk group and low-risk group according to the original and simplified pulmonary embolism severity index,severity classification of pulmonary embolism and early(hospitalization or 30-day)death risk in the 2019 European Guidelines for Pulmonary Embolism.Clinical data analysis,laboratory data analysis and ECG analysis were performed among multiple groups to clarify the changes of each clinical data and ECG with the increase of pulmonary embolism risk stratification between each group.A new combined predictor was established by logistic regression analysis combined with multiple ECG findings and D-dimer and other indicators for high-risk group and control group,intermediate-risk group and control group,and low-risk group and control group.The area under ROC curve(AUC)of the new combined predictor was compared with each original single indicator to determine the optimal cut-off value of the combined predictor and calculate its sensitivity,specificity,positive predictive value,and negative predictive value.Results: A total of 168 patients were enrolled in this study,including 77 males and 91 females,and their mean age was calculated to be(66.08 ± 9.4)years.There were 108 patients in the pulmonary embolism group(33 in the high-risk group,44 in the intermediate-risk group,and 31 in the low-risk group)and 60 patients in the control group.Statistical analysis showed that right bundle branch block,a VR ST segment elevation,V1-V6 QTc interval difference and D-dimer were significantly correlated with predicting pulmonary embolism.The AUC of combined predictor 1 of right bundle branch block,a VR ST segment elevation,V1-V6 QTc interval difference,and D-dimer in predicting pulmonary embolism was higher than that of single index alone.When the optimal cut-off value of combined predictor 1 was 0.2,the sensitivity was 1 and the specificity was 0.9 for the diagnosis of pulmonary embolism.Statistical analysis showed that increased a VR R wave and V1-V6 QTc interval difference were significantly correlated with predicting pulmonary embolism.The area under the curve of the predictor fitted with the above indicators for predicting intermediate-risk pulmonary embolism was smaller than that of the prediction of V1-V6 QTc interval difference alone,so the combined factor could not be used to determine intermediate-risk pulmonary embolism.Using the presence or absence of acute low-risk pulmonary embolism as a binary outcome variable,it was shown that only the V1-V6 QTc interval difference had a greater correlation with low-risk pulmonary embolism,and the sensitivity of V1-V6 QTc interval difference alone in the diagnosis of acute low-risk pulmonary embolism was 0.935 and the specificity was 0.633.Conclusion:With the increase of risk stratification of pulmonary embolism,ECG changes and D-dimer levels related to pulmonary embolism gradually increased.Electrocardiogram changes combined with D-dimer are better for the diagnosis of pulmonary embolism.