Perioperative myocardial injury in patients receiving cardiac output-guided haemodynamic therapy: a substudy of the OPTIMISE Trial

Abstract

Evidence suggests that cardiac output-guided haemodynamic therapy algorithms improve outcomes after high-risk surgery, but there is some concern that this could promote acute myocardial injury. We evaluated the incidence of myocardial injury in a perioperative goal-directed therapy trial. Patients undergoing major gastrointestinal surgery (n=723) were randomly assigned to cardiac output-guided haemodynamic therapy (intervention group) or usual care as part of the OPTIMISE trial. At four participating sites, 288 patients were enrolled in a biomarker substudy. Serum high-sensitivity cardiac troponin I (TnI) concentration and N-terminal pro-brain natriuretic peptide (NT-proBNP) concentration were measured before and at 24 and 72 h after surgery. Median preoperative TnI and NT-ProBNP concentrations were 4.3 ng litre(-1) and 144 pg ml(-1), respectively. After surgery, 67 (46%) patients in the intervention group and 68 (48%) patients receiving usual care had TnI concentrations above the 99th centile upper reference limit (P=0.82). Peak serum TnI concentration was similar in the intervention and usual care groups (median [interquartile range]: 10.0 [5.3-21.5] vs 7.8 [5.0-21.8] ng litre(-1); P=0.85), and no differences were observed in serum TnI concentrations over 72 h (repeated-measures anova, P=0.51). Likewise, there were no differences in peak NT-proBNP concentration between intervention and usual care groups (645 [362-1169] vs 659 [381-1028] pg ml(-1); P=0.86) or in serial NT-proBNP concentrations over 72 h (P=0.20). Myocardial injury is common among patients undergoing major gastrointestinal surgery. In this study, the frequency was not affected by cardiac output-guided fluid and low-dose inotropic therapy.