Association of phase angle at different perioperative timepoints with outcomes in cardiac surgery: a post hoc analysis of two prospective cohorts

In cardiac surgery, both pre- and postoperative phase angle (PhA) measured by bioelectrical impedance analysis have been associated with postoperative outcomes. Whether preoperative or postoperative PhA is the optimal measurement timepoint remains unclear. The aim of the present study was to compare the association of pre- and postoperative PhA with postoperative outcomes. This post hoc analysis pooled individual patient data from 2 prospective European cohorts conducted between October 2016 and October 2020. A total of 521 adults undergoing planned cardiac surgery with cardiopulmonary bypass were included. Three sets of logistic regression models were constructed to assess the relationship with 30-day mortality, prolonged mechanical ventilation, prolonged ICU length of stay, and cardiac surgery-associated acute kidney injury (CSA-AKI): models including preoperative PhA alone, postoperative day-1 PhA alone, and preoperative PhA combined with the change from preoperative to postoperative day-1 PhA (ΔPhA). Perioperative PhA trajectories during intensive care unit (ICU) stay were described using a missing-data pattern based on ICU length of stay. Lower PhA values at baseline and on day 1 were consistently associated with all four outcomes in both univariable and multivariable analysis. In univariable analysis, each one-degree increase in preoperative PhA was associated with reduced odds of 30-day mortality (OR 0.35 [95% CI 0.22–0.52]), prolonged mechanical ventilation (OR 0.47 [95% CI 0.36–0.59]), prolonged ICU length of stay (OR 0.59 [95% CI 0.49–0.69]), and CSA-AKI (OR 0.61 [95% CI 0.52–0.71]). Postoperative day-1 PhA showed associations of the same direction. In multivariable models, ΔPhA was independently associated with CSA-AKI (OR 1.72 [95% CI 1.24–2.42]) and prolonged ICU length of stay (OR 1.43 [95% CI 1.01–2.04]), without significant improvement in discrimination over preoperative PhA. The trajectory analysis showed a marked postoperative decrease in PhA without recovery during ICU stay. Both preoperative and postoperative day-1 PhA were associated with postoperative outcomes after cardiac surgery. After accounting for preoperative PhA, the postoperative decrease in PhA was independently associated with CSA-AKI and prolonged ICU stay. These hypothesis-generating findings warrant confirmation in prospective studies.