Patients

An anonymized database was created from the hospital registry (from 2014 to 2021) to search for patients who underwent isolated coronary artery bypass grafting procedures. Further selection criteria included preoperative EF < 40% (heart failure with reduced EF) and procedural use of either cold blood or del Nido cardioplegia. Patients who required additional procedures or had retrograde cardioplegia delivery were excluded from the study.

Research Ethics Board Consent

This study is a retrospective analysis of a fully anonymized institutional dataset. No further interventions were made to subjects, and all data were readily available. As such, following the National Code for Clinical Trials (National Code on Clinical Researches, 2011), research ethics board consent is not mandatory for the quantitative part of the study. Furthermore, all admitted patients consented to medical data processing for scientific purposes before admission to the hospital.

Surgical Treatment of Coronary Artery Disease - Qualification and Strategy

Before the procedure, coronary angiography was carefully evaluated. The qualification process throughout the entire period was based on the actual European Society for Cardiology/European Association for Cardiothoracic Surgery Clinical Guidelines^[5,6]. Our institution has a Heart Center structure, so the heart-team assessment was performed whenever indicated.

Our facility routinely uses the left internal mammary artery (LIMA) for left anterior descending grafting. Other coronary arteries are revascularized using the saphenous vein. However, in patients younger than 50 years old, total arterial revascularization is the preferred method.

Anesthesia and Preparation for Coronary Artery Bypass Grafting

Combined general anesthesia was used for every patient. Each patient was intubated, and a central venous port was introduced, preferably via the right jugular vein. A Foley catheter was placed into the bladder. At the end of LIMA harvesting, unfractionated heparin was administered at a calculated dose adjusted to the patient’s body mass to achieve an activated clotting time (ACT) of at least 480 seconds. After completing the procedure, anticoagulation was reversed with protamine sulfate (1.5 mg/100 units of heparin).

Surgical Procedure

Median sternotomy was used for all cases. LIMA was harvested using low-energy coagulation. The bridging harvesting technique was used for saphenous vein harvest, and the standard surgical technique was used for radial artery harvest. After achieving an adequate ACT, an arterial cannula was inserted into the ascending aorta, and a double-staged venous cannula was inserted into the right atrium. Cardiopulmonary bypass was initiated. After placing the cross-clamp and administering cardioplegia (following the specific protocol below), distal anastomoses were performed using 7.0 or 8.0 sutures. The cross-clamp was released, and the return of rhythm was observed. Defibrillation was performed whenever necessary. Proximal anastomoses were performed on the ascending aorta using 6.0 sutures with the vessel partially clamped. Subsequently, weaning from extracorporeal circulation was done, and the cannulas were removed. After reversing anticoagulation, chest tubes were inserted into the left pleura and pericardium, and the sternum was closed with surgical wires. The subcutaneous tissue and skin were then sutured.

Cardioplegia Protocol

Antegrade cardioplegia was administered to most of the patients at our institution. Retrograde cardioplegia was rarely used and reserved for selected cases. The cold blood cardioplegia used in our protocol contained 500 mL of crystalloid, consisting of 435 mL Plasma-Lyte A, 20 mL mannitol 15%, 20 mL NaHCO₃ 8.4%, and 25 mL KCl 2 mEq/mL. The crystalloid was mixed with autologous blood in a ratio of 4:1 (blood:crystalloid). The electrolyte concentration and administration protocol are provided in Table 1.

Table 1 - Cardioprotection protocol comparison.

The crystalloid component of the del Nido cardioplegia was prepared using calculated doses to provide a solution identical to the originally described protocol (1000 mL Plasma-Lyte A, 16.3 mL mannitol 20%, 4 mL MgSO4 50%, 13 mL NaHCO₃ 8.4%, 13 mL KCl 2 mEq/mL, and 13 mL lidocaine 1%). The crystalloid was mixed with autologous blood in a ratio of 4:1 (crystalloid:blood). The electrolyte concentration and administration protocol are provided in Table 1. A dose of 20-40 mg furosemide was given unless intraoperative hemofiltration was used to manage high volume given in a single dose.

Propensity Matching

To improve the precision of the study, propensity matching was performed. The matched parameters included age (maximum difference tolerance: three years), sex (male/female), the number of bypass grafts (exact number), and the presence of myocardial infarction at baseline (yes/no).

Primary Endpoints

To assess the cardioprotective capability of the propensity-matched group of patients, biomarker release and EF were considered as primary measurements since they are routinely performed for all patients within the same time frame. High-sensitivity troponin T, with a reference range of 0-14 pg/mL, and creatine kinase-MB isoenzyme, with a reference range of 0-7 ng/mL, were measured 12 and 36 hours after surgery.

Transthoracic echocardiography and Simpson’s method were used to assess left ventricular systolic function preoperatively and on the second postoperative day.

Other Measured Parameters

Baseline parameters, including planned surgery/acute coronary syndrome, age, sex, comorbidities, and perioperative risk in European System for Cardiac Operative Risk Evaluation (or EuroSCORE) II, were recorded for each patient. Major adverse cardiac and cerebrovascular events (MACCE) (defined as death, myocardial infarction, or stroke) was evaluated throughout the hospitalization. Myocardial infarction was defined following the fourth universal definition of myocardial infarction, which includes an elevation of troponin values > 10-times the 99^th percentile upper reference limit in patients with normal baseline troponin values or a rise > 20% in patients with elevated but stable troponin values, plus one of the following: the development of new pathological Q waves, angiographically documented new graft occlusion or new native coronary artery occlusion, imaging evidence of new loss of viable myocardium, or new regional wall motion abnormality consistent with an ischemic etiology^[7].

Our institution routinely uses inotropic support for patients with significantly impaired contractility when weaning from extracorporeal circulation. Therefore, mechanical circulatory support rates (mainly intra-aortic balloon pump [IABP]) were compared to assess the issue of low cardiac output despite the use of inotropes.

Daily creatinine values were measured, and the highest values were compared with preoperative creatinine to determine renal dysfunction following the Acute Kidney Injury Network criteria: stage 1 (creatinine increase > 0.3 mg/dL or 150-200%), stage 2 (creatinine increase 200-300%), and stage 3 (creatinine increase > 300%).

Power Calculations

An independent samples t-test assuming unequal variances was used to estimate continuous data. A difference in means of 14 pg/mL troponin values and a standard deviation of 14 pg/mL would require 22 pairs at a 95% confidence level and 90% power to achieve statistical significance. A difference in creatine kinase means of 7 ng/mL and a standard deviation of 7 ng/mL would also require 22 pairs at a 95% confidence level and 90% power to achieve significance. For EF, a difference in means of 5% and a standard deviation of 5% would require 13 pairs at a 95% confidence level and 90% power for the results to become significant.

Statistical Analysis

Continuous data are presented as median (interquartile range), while categorical data are presented as number (percentage). The Shapiro-Wilk test was used to assess the distribution of continuous data (normal vs. non-normal). Normal distribution was rejected for nearly all continuous parameters except cross-clamping and extracorporeal circulation time in Group 1. Therefore, the Mann-Whitney U test was used for comparative calculations. Fisher’s exact test was used to compare categorical data due to the small sample size. No missing values were reported for the analyzed endpoints. The alpha level was set at 0.05, and calculations were performed using MedCalc v.18.5 (MedCalc Software, Ostend, Belgium).

Limitations

The study is a retrospective dataset analysis; in such cases, all issues associated with this methodology must be considered. Additionally, the endpoints only refer to the hospitalization period. Furthermore, the small sample size is one of the main limitations, indicating that further investigation is mandatory.