The severity of early fluid overload assessed by bioelectrical vector impedance as an independent risk factor for longer patient care after cardiac surgery.

ABSTRACT

BACKGROUND AND AIMS: Fluid overload is a common postoperative complication in patients undergoing cardiac surgery. Although this condition is notably associated with relevant adverse outcomes, assessment of hydration status in clinical practice is challenging. Bioelectrical impedance vector analysis (BIVA) has emerged as a potentially effective method to monitor hydration changes, but the available evidence in critically ill patients undergoing cardiac surgery is limited and sometimes conflicting. The aim of this study was to explore by mean of BIVA the evolution over time of hydration status and its impact on relevant outcomes. METHODS: Prospective observational study enrolling 130 patients undergoing cardiac surgery. Height normalized impedance was calculated both before surgery (baseline) and in the first five postoperative days. Relevant clinical and laboratory data were collected daily close to BIVA measurements. Length of mechanical ventilation (MV), intensive care unit (ICU) and hospital stay exceeding the 75th percentile of the study population were considered as study endpoints. RESULTS: Compared to baseline, a significant reduction in impedance was found at first postoperative day, demonstrating a relevant fluid overload. An adjusted impedance at first postoperative day shorter than the best respective threshold was associated to longer MV (7.4 times), ICU stay (4.7 times) and hospital stay (5.6 times). A significant change in impedance and phase angle was documented throughout the observation days (p < 0.001), without returning to the baseline value. The co-existence of low impedance and high plasma osmolarity increased significantly the risk of incurring the study outcomes. CONCLUSIONS: In patients with cardiac surgery-induced fluid overload, recovery to baseline conditions occurs slowly. A relevant early fluid overload should be considered predictive for longer time of MV, ICU and total hospital stay.