Mean arterial pressure versus cardiac index for haemodynamic management and myocardial injury after hepatopancreatic surgery

BACKGROUND Myocardial injury after noncardiac surgery (MINS) frequently complicates the peri-operative period and is associated with increased mortality. OBJECTIVE(S) We hypothesised that cardiac index (CI) based haemodynamic management reduces peri-operative high-sensitive troponin-T (hsTnT) elevation and MINS incidence in patients undergoing hepatic/pancreatic surgery compared to mean arterial pressure. DESIGN A randomised controlled study SETTING A single-centre study conducted in a university-affiliated tertiary hospital between June 2022 and March 2023. PATIENTS Ninety-one patients, who were ≥ 65 years old or ≥ 45 years old with a history of at least one cardiac risk factor were randomised to either mean arterial pressure (MAP) based ( n = 45) or CI-based ( n = 46) management groups, and completed the study. INTERVENTION(S) In group-MAP, patients received fluid boluses and/or a noradrenaline infusion to maintain MAP above the predefined threshold. In group-CI, patients received fluid boluses and/or dobutamine infusion to keep CI above the predefined threshold. When a low MAP was observed despite a normal CI, a noradrenaline infusion was started. MAIN OUTCOME MEASURES The primary outcome was peri-operative hsTnT elevation. The secondary outcomes were MINS incidence and 90-day mortality. RESULTS The median absolute troponin elevation was 4.3 ng l −1 (95% CI 3.4 to 6) for the CI-based group, and 9.4 ng l −1 (95% CI 7.7 to 12.7) for the MAP-based group (median difference: 5.1 ng l −1 , 95% CI 3 to 7; P < 0.001). MINS occurred in 8 (17.4%) patients in the CI-based group and 17 (37.8%) patients in the MAP-based group (relative risk: 0.46, 95% CI: 0.22 to 0.96; P = 0.029). Two patients in group-MAP died from cardiovascular-related causes. One patient in group-CI and two in group-MAP died from sepsis-related complications (for all-cause mortality: χ 2 = 1.98, P = 0.16). MAP-AUC and CI-AUC values of the CI- and MAP-based groups were 147 vs. 179 min × mmHg ( P = 0.85) and 8.4 vs. 43.2 l m −2 min −1 × min ( P < 0.001), respectively. CONCLUSIONS CI-based haemodynamic management assures sufficient flow and consequently is associated with less peri-operative hsTnT elevation and lower incidence of MINS compared to MAP. TRIAL REGISTRATION Clinicaltrials.gov identifier: NCT05391087.