Математическая модель прогнозирования риска аспиринорезистентности и сердечно-сосудистых осложнений у пациентов с ИБС после коронарного шунтирования

Abstract

Introduction. After on-pump coronary artery bypass grafting (CABG) surgery, the problem of prevention of bypass graft thrombosis, following the development of the phenomenon of induced aspirin resistance (Ar) remains unresolved. Aim. To study the clinical and laboratory signs associated with the development of Ar, as well as to assess the significance of this phenomenon in the risk of cardiovascular complications (CVC) in patients with coronary artery disease (CAD) which who underwent on-pump CABG surgery, using mathematical modeling. Materials and methods. A study included 260 men aged 45 to 70 years (58.2 ± 6.8 years) with a diagnosis of CAD (stable angina pectoris of II–IV functional class (FC) (234 people) and silent myocardial ischemia of II–IV FC (26 people)). All patients received conventional therapy, including beta-blockers, angiotensin-converting enzyme inhibitors, nitrates, statins and acetylsalicylic acid (ASA). Myocardial revascularization was performed by on-pump CABG. Blood sampling for the assessment of laboratory risk factors (RF) of Ar was carried out at three stages: 1) at the prehospital stage – before CABG; 2) at the early postoperative period – 2 days after CABG surgery; 3) in the long-term postoperative period – 1 year after CABG surgery. At baseline, during the preoperative period, all 260 patients were aspirin sensitive (AS). Systemic inflammatory response to surgical intervention was assessed by the level of interleukin-6 (IL-6) and highly sensitive C-reactive protein (HSRP) in the blood serum, and endotheliopathy was assessed by the level of endothelin-1 (ET-1). Platelet aggregation activity (PAA) was also studied. We classified as aspirin resistant (AR) patients, those individuals whose PAA did not reach target values while they were taking ASA at a dose of 100 mg/day, and decreased by less than 50% compared with the baseline parameters. To analyze the impact of various factors (predictors) on PAA, the linear stepwise regression analysis with inclusions was used. Binary logistic regression was used to assess the influence of Ar on the risk of developing CVC, and linear discriminant analysis – to assess the relationship of clinical and laboratory predictors with the risk of developing specific types of CVC. Recurrence of angina pectoris, myocardial infarction (MI), decompensation of heart failure (HF) and sudden cardiac death were considered as CVC. Results. Ar was detected in 122 people (46.9%) on the 2nd day after CABG. This cohort of patients constituted the main group. The comparison group included 138 people (53%) who remained aspirin sensitive (AS). A year after surgery (and taking ASA at a dose of 100 mg/day this year), the proportion of AR patients decreased to 62 people, amounting to 24% of AS patients at baseline in the preoperative period. In two groups of patients, we compared such indicators as IL-6, HSRP, ET-1, PAA at all three stages of the study, while the cardiopulmonary bypass time (CBT) and the number of coronary bypass grafts (NBGs) – immediately after the CABG. In the main group (AR patients), much higher levels of the studied parameters were revealed than in the comparison group (statistically significant) both on the 2nd day and 1 year after CABG. CVC were detected in 53 out of 260 people (20.4%) during one year follow-up. They were distributed as follows: recurrence of angina pectoris – 37 people (14.2%); MI – 7 people (2.7%); decompensation of HF – 5 people (1.9%); sudden cardiac death – 4 people (1.5%). A mathematical model was developed to predict the risk of CVC after CABG in order to improve the long-term prognosis after surgery. As a result of the regression analysis, we got a set of five predictors that are statistically significantly related to PAA: ET-1, IL-6, HSRP, CBT, NBGs. The regression equation, which allows predicting the value of PAA on the 2nd day after CABG, had the form: PAA = 34.40 + 0.45 · IL-6 + 0.92 · ET-1 + 0.25 · CBT + + 0.73 · HSRP + 0.13 · NGBS, where 34.40 is the absolute term of the equation. Thus, according to the calculated PAA value, it will be possible to concluded the presence of either As or Ar for a particular individual. When assessing the relationship of Ar with the risk of CVC in patients after on-pump CABG, using binary logistic regression, it was determined that PAA is a statistically significant prognostic factor for CVC (χ2 = 21.0; p = 0.013). Using discriminant analysis, a mathematical model has been created to evaluate the probability of developing certain types of CVC in patients after on-pump CABG. Conclusion. The grade of systemic inflammatory response to on-pump CABG affects the development of induced Ar. The persistence of a systemic inflammatory reaction is associated with the prolongation of Ar in the long-term postoperative period. The regression analysis revealed an important role of systemic inflammatory response markers and endotheliopathy (IL-6, HSRP, ET-1) in the development of the phenomenon of Ar in patients undergoing on-pump CABG. The mathematical model, created by utilizing discriminant analysis, makes it possible to predict with the accuracy of 87% the probability of developing CVC (recurrence of angina pectoris, MI, decompensation of HF, sudden cardiac death) in patients who underwent on-pump CABG. Significant risk factors for CVC include: CBT, the levels of PAA, IL-6, HSRP and ET-1. Correction of these factors’ parameters will improve the long-term prognosis in patients with CAD after CABG.