Association of postoperative blood pressure and bleeding after cardiac surgery.

Abstract

We sought to characterize the relationship between postoperative blood pressure on the day of surgery and metrics of bleeding. In a preplanned secondary analysis of prospectively collected data from the Limiting IV Chloride to Reduce AKI trial (NCT02020538), univariate and multivariable regression analyses explored the association between peak systolic blood pressure, peak mean arterial pressure, and peak central venous pressure recorded postoperatively on the day of surgery and multiple metrics of bleeding. Patients at increased bleeding risk due to specific criteria were excluded from analysis. The primary outcome was chest tube drainage (milliliters per hour) on the day of surgery. Secondary outcomes included red blood cell transfusion, surgical re-exploration for bleeding, and hospital mortality. The study cohort comprised 793 patients. Mean ± standard deviation peak systolic blood pressure, mean arterial pressure, and central venous pressure were 125 ± 15 mm Hg, 83 ± 9 mm Hg, and 12 ± 4 mm Hg, respectively. Median (interquartile range) chest tube drainage on the day of surgery was 33 mL/hour (interquartile range, 23 mL/hour-51 mL/hour). Adjusted for prespecified variables, there was no positive association between peak systolic blood pressure and bleeding outcomes, including chest tube drainage (-2.2 mL/10 mm Hg; 95% confidence interval, -3.9 to -0.5 mL/h/10 mm Hg; P = .01) or volume of transfusion (-15 mL/10 mm Hg; 95% confidence interval, -29 to -1 mL/h/10 mm Hg; P = .04). Results remained broadly consistent across multiple secondary outcomes and regardless of whether systolic blood pressure or mean arterial pressure was the explanatory variable. The lack of positive association between peak systolic blood pressure or peak mean arterial pressure with metrics of bleeding after cardiac surgery promotes equipoise for testing the influence of higher blood pressure targets during the early postoperative period.