BackgroundElevated blood urea nitrogen (BUN) and reduced albumin have been prominently correlated with unfavorable outcomes in patients with cardiovascular diseases. However, whether combination BUN and albumin levels could predict the adverse outcomes of cardiac surgery patients remains to be confirmed. Here, we investigated the prognostic effect of the preoperative BUN to serum albumin ratio (BAR) in cardiac surgery patients.MethodsData were obtained from the Medical Information Mart for Intensive Care (MIMIC) III and eICU databases and classified into a training cohort and validation cohort. The BAR (mg/g) was calculated by initial BUN (mg/dl)/serum albumin (g/dl). The primary outcome was in-hospital mortality. Secondary outcomes were 1-year mortality, prolonged length at intensive care unit, and duration of hospital stay. The associations of BAR with outcomes were explored by multivariate regression analysis and subgroup analyses. Then, C statistics were performed to assess the added prognostic impact of BAR beyond a baseline risk model.ResultsPatients with in-hospital death had significantly higher levels of BAR. Multivariate regression analysis identified BAR, as a categorical or continuous variable, as an independent factor for adverse outcomes of cardiac surgery (all p < 0.05). Subgroup analyses demonstrated a significant relationship between elevated BAR and in-hospital mortality in different subclasses. The addition of BAR to a baseline model provided additional prognostic information benefits for assessing primary outcome. Results were concordant in the external validation cohort.ConclusionsIncreased preoperative BAR is a potent predictor of unfavorable outcomes in patients undergoing cardiac surgery.