Left ventricular function outcome after coronary artery bypass grafting, King Abdullah Medical City (KAMC)- single-center experience

Background Coronary artery bypass grafting is known to be associated with better outcome in ischemic heart disease patients with low ejection fraction. We aim to demonstrate the effect of coronary artery bypass grafting (CABG) on left ventricle (LV) systolic function and to identify the predictors that adversely lead to postoperative poor outcome. Result This is a cross-sectional prospective study; we included 110 patients with left ventricular ejection fraction (LVEF) < 50% who underwent CABG with a mean age of 56.1 ± 12.2 years old. Those patients were classified into two groups: group I, 76 (69%) patients with LVEF > 35%, and group II, 34 (31%) patients with LVEF < 35%. Our results as regards demographic and clinical data revealed that group II patients had a significantly higher prevalence of diabetes mellitus (DM) and Euro SCORE II compared to group I patients (p = 0.05 and < 0.001 respectively); otherwise, all other clinical predictors did not differ between the two studied groups. There was a significant improvement in LVEF post-surgery (p = 0.05) in both groups with observed no significant difference recorded for in-hospital mortality rate among patients with different groups. DM, significant diastolic dysfunction, and insertion of IABP are predictors of in-hospital mortality of the patients (p = 0.001, 0.03 and < 0.001, respectively) Conclusion We concluded that there is a significant improvement of LV systolic function after CABG and hence better survival rate. DM, significant diastolic dysfunction, and perioperative insertion of IABP are predictors of mortality after cardiac surgery. Special care should be provided to such patients to improve their outcome


Background
Despite improvements in medical therapy and surgical techniques, management of patients with left ventricular (LV) dysfunction and coronary artery disease undergoing cardiac surgery is still challenging [1,2]. Coronary artery bypass grafting (CABG) has appeared to be superior to medical therapy alone for patients with low ejection fraction (EF), representing a significant clinical improvement and long-term survival [3][4][5][6][7]. For these patients, CABG is associated with higher postoperative morbidity and mortality rates compared with patients with normal EF [5][6][7][8][9].
The studies investigating early postoperative changes have yielded conflicting results. Some have found ventricular improvement within weeks postoperatively [10,11], while other studies have detected no change [12,13] or a worsening of ventricular function [14].
Therefore, recognition of the predictors that increase the patients risk for a worse outcome plays a crucial role in the clinical decision-making process [15].
The aim of this study was to assess the effect of cardiac surgery on LV systolic function in patients with abnormal preoperative systolic function and to identify the predictors that may affect the outcome in patients undergoing CABG.

Methods
It is a cross-sectional, prospective single-center study that included 110 patients who underwent CABG at the cardiac center-King Abdullah Medical City (KAMC)during 2016 and 2017.

Inclusion criteria
Patients operated for elective isolated CABG or CABG and mitral valve (MV) repair at KAMC.

Exclusion criteria
Patients with normal left ventricular ejection fraction (LVEF), severe pulmonary hypertension, cardiogenic shock, atrial fibrillation, combined CABG with other valve intervention, and those who did not have postoperative echocardiography follow-up. Also, patients with very low EF < 20% were excluded as they were rejected by our multidisciplinary team for surgery.

Surgical principals
The surgical procedure was performed via median sternotomy with aorto-caval (single/bicaval). Cardioprotection was performed through antegrade mixed with retrograde or direct into the vein after distal anastomosis and topical hypothermia to achieve complete protection of the myocardium. The surgical procedure was done with cardioplegia every 20 min, wean off bypass. Trans-esophageal echocardiography was done for assessment of wall motion and ventricular function after revascularization.

Echocardiography
All patients had a baseline echocardiogram before and after surgery. Subsequently, study participants were monitored repeatedly throughout the first few months after CABG, in specific regard to their LV function.
Preoperative transthoracic echocardiography (TTE) was collected as well as TTE performed after surgery. Echocardiography was performed with a vivid 7 ultrasound system. Parasternal and apical views were obtained. Ejection fraction is commonly measured by echocardiography, in which the volumes of the heart's chambers are measured during the cardiac cycle. Ejection fraction can then be obtained by dividing the volume ejected by the heart (stroke volume) by the volume of the filled heart (end-diastolic volume) [16]. Severe LV systolic dysfunction was defined in our cohort study as EF < 35%. Improvement in LVEF was defined as > 5% absolute increase in LVEF in comparison to the preoperative echocardiogram. Consequently, LVEF that decreased by > 5% compared to the preoperative echocardiogram was categorized as worsened. All other postoperative EF measurements within ± 5% of the preoperative values were categorized as unchanged [17]. Other echocardiographic parameters (diastolic function, mitral regurgitation (MR), right atrial (RA) size, right ventricular (RV) size, LV size, left atrial (LA size), right ventricular systolic function (RVSP) and RV function) all were assessed. RV function was evaluated by means of tricuspid annular plane systolic excursion (TAPSE) and tissue Doppler imaging (RV S').

Statistical analysis
Statistical analysis was performed by the use of the SPSS software package (SPSS Inc.; Chicago, IL), version 21.0. Data are presented as mean ± SD or as median and range according to the type of distribution of each variable. Chisquare test was used to compare the existence of ventricular dysfunction pre-and postoperatively (logistic DM, HTN, smoking, dyslipidemia, CKD, MI, BNP, LVEF, LV size, TAPSE, RV size). Linear regression analysis was performed. For all analyses, a p value < 0.05 was considered significant and not significant if it is > 0.05.

Results
Out of 264 patients undergoing coronary artery bypass grafting at the cardiac center-King Abdullah Medical City (KAMC)-in the study period, 110 had preoperative LVEF < 50%. Of these, 76 (69%) patients had LVEF > 35% (group I), 34 (31%) had LVEF < 35% (group II), and all were included in the study. Baseline characteristics, comorbidities, type of operation, and intraoperative management of the study population are reported in Table 1.
Among our patients, mortality was 5.4% and was consistent with preoperative predictions (mean Euro SCORE).
Preoperative characteristics of the two studied groups were statistically similar except for the significantly higher prevalence of DM and Euro SCORE II in group II. Operative characteristics were statistically similar between both groups (Table 2).
Early 30-day in-hospital mortality was 5.2% in group I and 5.8% in group II, with a statistically insignificant difference. In Table 3, early mortality was observed in one patient (3%) of group II and it was due to cardiogenic shock. The causes of death were cardiogenic shock (3 patients in group I and one patient in group II), multi-organ failure (one patient in group I) and respiratory insufficiency (one patient in group II). There were insignificant differences in postoperative complications, except the significantly higher perioperative insertion of IABP and inotropic support in group II (Table 3 and Fig. 1).

Change of LV function after surgery
Most of our patients had significant improvement of LVEF > 5% in the early postoperative period as the mean LVEF of the whole cohort increased significantly from 29.76 ± 4.868 before surgery to 33.53 ± 9.655 in the first week after surgery. Postoperative improvement in LVEF was also observed in subgroups of patients with a different type of cardiac surgery ( Table 4).

Predictors of adverse outcome after CABG
DM, significant diastolic dysfunction, and insertion of IABP all were found to be predictors of adverse outcome and mortality in the studied patients (p = 0.00, 0.03 and < 0.001 successively). Insertion of IABP was found as an independent predictor of mortality in different groups (Table 5).

Discussion
The identification of LVEF after CABG in patients with ischemic cardiomyopathy is considered as a crucial factor to predict the outcome. Improved function and survival following surgical revascularization have been shown in patients with LV dysfunction which is predominantly due to hibernating or stunned myocardium [17]. Our study focused on the effect of CABG on LVEF in patients with baseline impaired systolic function. In this study, a significant improvement in LV systolic function was observed with CABG. This supports the hypothesis that surgical revascularization and restoration of the blood flow to ischemic myocardium preserve already viable and functioning myocardial muscle against later infarction, recruit the hibernating myocardium, and reduce LV remodeling and ischemic burden which all impact LV function recovery.
Because of the lack of routine echocardiographic examination postoperatively, few studies identified the changes in LV systolic function after CABG. Similarly, a recent study was conducted and concluded that successful results of CABG in patients with EF < 50% can be achieved by careful selection of patients and management [18]. Also, another study found that a significant improvement was observed in LV systolic function in patients with preoperative systolic dysfunction [19]. Many factors contributing to the outcome of patients with baseline LV systolic dysfunction after CABG include perioperative care, severity of preoperative LV systolic dysfunction, surgical skills, complete revascularization, type of myocardial protection, cardiac anesthesia management, emergency cardiac facilities, and postoperative intensive care monitoring and management. However, Koene et al. [2] conducted a similar study and concluded that CABG is associated with worse outcome on LV systolic function [20]. A decrease in LV systolic function with CABG surgery could be  explained by many factors: significant intraoperative global ischemia which adversely affects the LV function [21] or from myocardial stunning [22] or early postoperative graft failure [23]. In contrast also to our findings, a small study did not find a significant change in LVEF immediately post CABG [24].
Our study demonstrates that in the early mortality, 5.4% is acceptable and in agreement with the current published data [25,26]. Similarly, Elassy et al. reported higher but insignificant mortality rate in patient with LVEF < 35% (5.6% vs 2.4%) [27]. This is reflecting a marked improvement in the outcome of ischemic cardiomyopathy patients with CABG in highly qualified centers.

Predictors of adverse outcome with CABG
In this study, we classified our patients into two groups with and without severe LV systolic dysfunction and investigated all the predictors that could lead to poor outcome with CABG.
Type II DM was a significant predictor of adverse outcome of our patients.
Although CABG is considered the best revascularization strategy in diabetic patients, a significantly higher rate of mortality was continually observed in patients with type II DM compared to patients without type II DM [28,29]. Reasons that have been suggested for that are more aggressive disease with advanced stage of DM in these patients, adverse effects of insulin therapy, hyperinflammatory, and hormonal hyperactivation response [30,31].
The use of intra-or postoperative IABP was a significant predictor of in-hospital mortality in all patients and when EF was < 50%. Although the use IABP is important to support failing circulation during CABG, it is well known as a risk factor for mortality [32,33]. It can be explained by that patients who are receiving IABP are already at high risk of increased mortality because of unstable hemodynamic status and its complications (stroke, paraplegia, limb ischemia, infection, and hemolysis) that all have an impact on mortality after CAGB [34,35].
We also demonstrated advanced diastolic dysfunction as a predictor of adverse outcome and mortality in patients with LV dysfunction. Diastolic dysfunction has been reported to be an independent risk predictor of postoperative heart failure, atrial fibrillation, and cardiac death in different studies [36,37].
Finally, a limited number of patients are included due to the nature of a single center.
Some patients also did not have echocardiography follow-up, hence excluded from our study. Follow-up echocardiography was done few months only post CABG, and thus, our results could not be correlated with longterm outcome after surgery. Like other studies evaluating adverse outcome after CABG in patients with reduced LVEF, data regarding patency of grafts were not applicable in our study. The results of this study are encouraging, and it needs corroboration in multicenter larger population with longer follow-up.

Conclusion
This study confirmed that there is a remarkable improvement of LV systolic function after coronary artery bypass grafting and that reflects the high benefit of CABG in patients with reduced EF. Diabetes mellitus, high Euro SCORE, advanced diastolic dysfunction, and insertion of IABP were significant predictors of adverse outcome. So, identification of patients with those risk predictors could provide complementary prognostic information and help to maximize the care, monitoring, and close follow-up to improve their expected poor outcome. More investigation is required for similar data in other tertiary centers to provide multicenter results and hence generalize our conclusion.