Impact of pulmonary artery catheter hemodynamic monitoring on post-operative morbidity and mortality in elective bentall procedures

Abstract

Introduction: Because of debated effectiveness and safety, the perioperative use of pulmonary artery catheter (PAC) has decreased in cardiac surgery (1). However, PAC keeps some specifics indications in high risk cardiac surgery (2). This study evaluates the impact of perioperative PAC monitoring on postoperative morbidity and mortality after Bentall procedures. Methods: We designed a single center retrospective observational study comparing patients undergoing Bentall procedure with or without perioperative CAP monitoring. Patients included had no major risk of hemodynamic complications (Left ventricular ejection fractions (LVEF) > 40%, no pulmonary arterial hypertension, no history of cardiac surgery). Primary endpoint was a composite criteria of postoperative morbidities and mortality defined by the occurrence of any the following: death in hospital, cardiac dysfunction (myocardial infarction or congestive heart failure), cerebral dysfunction (stroke or encephalopathy), renal dysfunction (acute kidney injury or dialysis), or pulmonary dysfunction (acute respiratory distress syndrome). Epidemiology of CAP use, others postoperative complications, biological markers (lactates, troponin), hemodynamic management and length of stay in intensive care unit (ICU) and hospital were also assessed. Results: Among the 134 patients enrolled between 2012 and 2019, the primary endpoint occurred in 29 of 59 PAC patients vs. 17 of 75 patients without PAC (49.2%, 95% confidence interval (CI), 37-62, vs 22.7%, 95% CI, 15-33; P<0.01). The PAC group had an increased risk of cardiac (22% vs 2.7%, P<0.01) and cerebral dysfunction (22% vs 6.7%, P=0.02). No difference was found between patients with and without PAC for hospital mortality (1.7% in PAC group vs 0%, P=0.44), renal (33.9% in PAC group vs 18.7%, P=0.07) and pulmonary dysfunction (1.7% in PAC group vs 0%, P=0.44). PAC patients were significantly older (65.5 vs 59.7 years old, P<0.01), had higher EuroSCORE 2 (2.7 vs 1.8, P=0.02) and a lower LVEF (57.7% vs 65.1%, P<0.01). In PAC group, we also founded more frequent postoperative use of Dobutamine (35.6% vs 4%, P<0.01), Norepinephrine (61% vs 33.3%, P<0.01) and first 24 hours volume expansion (69.5% vs 46.7%, P=0.01). In PAC patients, the length of stay in ICU (4.9 days vs 3.2 without PAC, P=0.01) and in hospital (15.9 days vs 11.9 without PAC, P<0.01) were significantly higher. In multivariate logistic regression analysis, with independent factors (CAP use, length of cardiopulmonary bypass, EuroSCORE 2), CAP use was not associated with the occurrence of primary endpoint (adjusted odds ratio 2.07, 95% CI [0.9-4.7], P=0.08). Discussion: In our study, a PAC was used in a highest risk group. The use of PAC was associated with increased postoperative mortality and morbidity in univariate but not in multivariate analysis. A randomized clinical trial is needed to clarify the link between CAP use and postoperative complications.