Predicting the Development of Complications After Mitral Valve Repair Using Mathematical Analysis

Abstract

AIM. To study the possibilities of using risk factors, data from instrumental research methods identified at the preoperative stage as independent variables for predicting the development of complications after mitral valve surgery. MATERIAL AND METHODS. The study included 103 patients, 46 men and 57 women, who underwent surgical correction of mitral valve defect and were treated at the Cardiac Surgery Department of Chelyabinsk Regional Clinical Hospital № 1 in the period from 2014 to 2019. Among those patients, the presence of anemia at the preoperative stage (blood hemoglobin level less than 100 g/l) was detected in 13 (12,6 %) cases; obesity (BMI more than 30 kg/m2) — in 57 (55,3 %) cases; transfusion of blood and its components in anamnesis — in 14 (13,6 %) cases; type 2 diabetes mellitus — in 7 (6,8 %) cases; stroke in anamnesis — in 9 (8,7 %) cases; chronic Hepatitis C — in 6 (5,8 %) cases; HIV infection — in 4 (3,9 %) cases; antibiotic therapy before surgery was indicated in 21 (20,4 %) cases; edema of the lower extremities was detected in 40 (38,8 %) cases; an increase in body temperature before surgery — in 19 (18,4 %) cases; the presence of atrial fibrillation before surgery — in 63 (61,2 %) cases. The systematization of the source information was entered into a Microsoft Office Excel 2016 spreadsheet. Statistical analysis was carried out using the IBM SPSS Statistics v.26 program (developed by IBM Corporation). Binary logistic regression was used as a method of mathematical data analysis. Instrumental research methods: electrocardiography (ECG), echocardiography (ECHO-CG); coronary angiography was performed according to indications. RESULTS. The presence of risk factors such as anemia, obesity, as well as an increase in the size of the right atrium can be used as an independent variable to predict the development of complications. An increase in the diameter of the pulmonary artery trunk, and the presence of edema of the lower extremities reduces the risk of complications. The likelihood of developing complications after mitral valve surgery predicted by the logistic regression method was 53,4 %. The resulting model predicts the absence of complications with the likelihood of 82,5 %. The development of complications was correctly predicted in 93,3 % of cases. CONCLUSION. This model can be used as an additional tool in predicting the development of complications after mitral valve surgery.