Does the type of cardioplegia solution affect intraoperative glucose levels? A propensity-matched analysis

Myocardial protection during cardiac surgery is a multi-faceted process that is structured to limit injury and preserve function. Evolving techniques use solutions with varying constituents that enter the systemic circulation and alter intrinsic systemic concentrations. This study compared two distinct cardioplegia solutions on affecting intraoperative glucose levels. Data were abstracted from a multi-institutional perfusion registry, including a total of 1,188 propensity-matched cases performed from January through October 2016, at 17 cardiac surgical centers across the United States in which both del Nido and 4: 1 cardioplegia were used during the study period. Covariate data included insulin administration, crystalloid cardioplegia volume, diabetes history, glucose at operating room entry, and nine additional variables. Primary and secondary endpoints were the highest intraoperative glucose level and maximum glucose in excess of 180 mg/dL. Mixed-effects multivariable linear and logistic regression models were used to assess the primary and secondary endpoints, respectively, allowing for statistical control of center and surgeon effects. Greater median crystalloid cardioplegia volume was given in the del Nido group (n = 594) 1,040 mL [interquartile range (IQR) = {800, 1,339}] compared with the 4: 1 group (n = 594) 466 mL [IQR = {360, 660}] in the 4: 1 group (p < .001) despite these groups being statistically indistinguishable in terms of bypass and cross-clamp times as well as seven other patient covariates. More patients required intraoperative insulin drip in the 4: 1 group compared with del Nido (65.7% vs. 56.2%, p < .001). Multivariable linear mixed-effects analysis yielded an estimated maximum intraoperative glucose for the del Nido group of 177.8 mg/dL compared with that of the 4: 1 group, 183.5 mg/dL - a statistically significant reduction of 5.7 mg/dL (p = .03). Multivariable logistic mixed-effects analysis showed a statistically nonsignificant reduction in the likelihood of crossing the 180 mg/dL threshold for del Nido compared with 4: 1 (odds ratio [OR] = .79, p = .214). After controlling for known confounding variables, intraoperative maximum glucose levels for the del Nido group were 5.7 mg/dL lower than that of the 4: 1 group; there was limited evidence suggesting a difference between methods in the likelihood of exceeding the threshold of 180 mg/dL intraoperatively. Further research is warranted to examine the differential effects of cardioplegia solution on intraoperative glucose levels.