ABC | Volume 115, Nº2, August 2020

Original Article Gatti et al. Custodiol®-HTK solution & prolonged cross-clamp time Arq Bras Cardiol. 2020; 115(2):241-250 Methods Between July 2009 and October 2016, Custodiol-HTK cardioplegia was used in 106 consecutive patients who underwent isolated coronary artery bypass grafting at the Division of Cardiac Surgery of the University Hospital of Trieste, Italy; for 38 (35.8%) of them, XCT was of 120 minutes or longer. The early postoperative outcomes of these 38 patients (HTK cohort) were compared with those of 110 consecutive coronary surgery patients who were operated on in Trieste during the same period, had XCT ≥120 minutes, and received MCB cardioplegia (MCB cohort). Because the two cohorts differed significantly in risk profile and operative data, a propensity-matched analysis was performed as well. Unless otherwise stated, the definitions and cut-off values of the preoperative variables were those used for the European System for Cardiac Operative Risk Evaluation II (EuroSCORE II). 13 The risk profile of each patient was established preoperatively according to EuroSCORE II. The definitions of postoperative complications were in accordance with the internationally agreed definitions of complications after cardiac surgery. 14 Permanent neurological dysfunction (stroke with focal lesion by computed tomography examination), prolonged (>48 h) invasive ventilation, myocardial infarction, low cardiac output (requiring high doses of inotropic agents, intra-aortic balloon pumping or use of extracorporeal membrane oxygenator), acute kidney injury (with or without renal replacement therapy), multiorgan failure, multiple blood transfusion (three red blood cells units or more), mediastinal re-exploration for bleeding or tamponade, deep sternal wound infection (deep incisional infection or mediastinitis) and sepsis were defined as major complications. In-hospital death and major complications were included in a combined endpoint. Cardioplegia delivering protocols Either HTK or MCB cardioplegia was always delivered in both antegrade and retrograde mode at a temperature of ~4°C. Each patient received 20-25 ml of HTK solution per kg of body weight. The pressure of perfusion in aortic root was strictly maintained at 100 mmHg until cardiac arrest, and then at 40–50 mmHg. During the retrograde perfusion, the coronary sinus pressure was 20-25 mmHg. The cumulative perfusion time was 6-8 minutes. When hyponatremia occurred, it was corrected with sodium chloride (3 mEq/ ml) solution; systemic hypotension was treated using norepinephrine infusion. Hemodilution was mitigated by removing part of the HTK solution from the aortic root during the retrograde mode. 5,7,8 Conventional (Buckberg) MCB cardioplegia was delivered every 20 minutes, according to standardized protocols. 15 Blood tests Blood levels of haemoglobin, creatinine, sodium, potassium and calcium were measured early before surgery and immediately following patient admission in intensive care unit; platelet count and blood coagulation profile before and after surgery were explored as well. Serum levels of creatine kinase, creatine kinase-MB, cardiac troponin I and aspartate aminotransferase were measured during patients’ stay in ICU. The values were compared between the HTK and MCB cohorts of propensity-matched patients. Patients were informed about the study but were not required to provide individual consent, in accordance with the Italian legislation. Although it involves human subjects, this retrospective study was not registered in a publicly accessible database. Statistical methods Continuous variables with normal distribution were expressed as mean ± standard deviation and those without normal distribution, as median and the range between the first and the third quartile. Categorical variables were expressed as frequencies and percentages. Statistical comparison of baseline patient characteristics, operative data and postoperative complications was performed using the Chi-square or the Fisher’s exact test for categorical variables, and the unpaired Student’s t -test or the Mann- Whitney U -test, for continuous variables. Since the HTK and MCB cohorts significantly differed in risk profile, number of coronary anastomoses, and length of XCT, a multivariable analysis was performed using the backward stepwise logistic regression. The area under the receiver- operating characteristic curve, with a 95% confidence interval (95%CI), was used to represent the regression probabilities. To estimate the probability of a patient being assigned to either one or other group, a propensity score (PS) was calculated in a non-parsimonious way, including the following preoperative patient characteristics: age, sex, hypertension, body mass index, haemoglobin, diabetes on insulin, chronic lung disease, glomerular filtration rate as estimated by the Cockcroft-Gault formula, chronic dialysis, extracardiac arteriopathy, New York Heart Association functional class IV, Canadian Cardiovascular Society class 4 of angina, recent myocardial infarction, left main coronary artery disease, number of diseased coronary vessels, left ventricular ejection fraction, intra-aortic balloon pump use, surgical priority, expected operative risk by EuroSCORE II, number of coronary anastomoses, and length of XCT. One-to-one PS-matching was performed employing the nearest neighbour method and a caliper of 0.2 of the standard deviation of the logit of the PS. To evaluate the balance between the matched groups, the McNemar test for dichotomous variables, the Student’s t -test for paired samples or the Wilcoxon test for continuous variables, and the analysis of the standardized differences after matching were used. Standardized difference <10% was considered an acceptable imbalance between the treatment groups. The same tests were adopted to test differences in operative data and postoperative complications of matched groups. Two- way analysis of variance was used to observe the interaction between the type of cardioplegia and any major complication for relevant baseline characteristics among PS-matched pairs. All tests were two-sided with the p-value set at 0.05 for statistical significance. Acquisition of the data entries was performed using Microsoft Office Excel, version 2007. Data analyses were performed using the SPSS software package for Windows, version 13.0 (SPSS, Inc., Chicago, IL, USA). 242