Abstract
PURPOSE: The blood urea nitrogen/creatinine ratio (BCR) is an effective marker for disease severity stratification. Its efficacy has been demonstrated under numerous conditions. This study aims to investigate the relationship between BCR and in-hospital mortality in intensive care unit (ICU) patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD). PATIENTS AND METHODS: Eligible ICU patients with AECOPD from the eICU database were included in the study. Patients were divided into high-BCR and low-BCR groups on the basis of the optimal cutoff value (22.78) of the ROC curve for predicting in-hospital mortality in AECOPD patients. Propensity score matching (PSM) was used to balance the baseline differences between the high-BCR and low-BCR groups. Multivariate logistic regression was used to analyze the relationship between BCR and in-hospital mortality in ICU patients with AECOPD. Decision curve analysis (DCA) was performed to evaluate the clinical efficacy of each model via multivariate logistic regression. RESULTS: A total of 3399 eligible ICU patients with AECOPD were included in the study, with 1559 patients in the high-BCR group and 1840 patients in the low-BCR group. After propensity score matching (PSM), 1174 pairs of patients were successfully matched. The results of the multivariate logistic regression revealed that the in-hospital mortality rate for AECOPD patients in the high-BCR subgroup was significantly greater than that in the low-BCR subgroup in both the unmatched and matched cohorts after adjusting for multiple factors. Additionally, DCA demonstrated that the models used in the multivariate logistic regression had effective clinical utility. CONCLUSION: The blood urea nitrogen/creatinine ratio (BCR) is an effective predictor of in-hospital mortality in ICU patients with AECOPD. Clinicians can use BCR to identify critically ill ICU patients with AECOPD earlier and implement interventions to improve patient outcomes.