BACKGROUND: The aim of this study was to evaluate the prognostic value of plasma N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) in unselected critically ill patients with acute respiratory failure (ARF). METHODS: Prospective, observational cohort study in 25 intensive care units in Finland. This study included a total of 602 patients with laboratory samples from 958 consecutive patients with ARF treated either with invasive or with non-invasive ventilatory support (the FINNALI study). Plasma NT-pro-BNP samples were drawn after the onset of ventilatory support (day 0) and on the morning of the second day. RESULTS: The median [interquartile ranges (IQR)] NT-pro-BNP-values were significantly higher at baseline in 90-day non-survivors than the survivors, 4378 pg/ml (1400-13,943 pg/ml) vs. 1052 pg/ml (232-4076 pg/ml), respectively. The median (IQR) NT-pro-BNP values were significantly higher in patients with chronic cardiac disease or cardiac surgery than in non-cardiac patients, 1947 pg/ml (801-4687 pg/ml) vs. 417 pg/ml (153-1735 pg/ml), respectively, if renal function was normal. With deteriorating renal function, the NT-pro-BNP values showed a significant increase. The area under curve for baseline NT-pro-BNP predicting 90-day mortality was moderate: 0.718 (95% confidence interval 0.674-0.761). Baseline NT-pro-BNP over 1765 pg/ml was independently associated with 90-day mortality by logistic regression analysis (P<0.001). CONCLUSIONS: NT-pro-BNP on admission is commonly elevated and independently associated with 90-day mortality in critically ill ARF patients. However, the routine use of NT-pro-BNP for prognostic purpose does not seem to add value to clinical data in ARF patients.
BACKGROUND: The aim of this study was to evaluate the prognostic value of plasma N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) in unselected critically illpatients with acute respiratory failure (ARF). METHODS: Prospective, observational cohort study in 25 intensive care units in Finland. This study included a total of 602 patients with laboratory samples from 958 consecutive patients with ARF treated either with invasive or with non-invasive ventilatory support (the FINNALI study). Plasma NT-pro-BNP samples were drawn after the onset of ventilatory support (day 0) and on the morning of the second day. RESULTS: The median [interquartile ranges (IQR)] NT-pro-BNP-values were significantly higher at baseline in 90-day non-survivors than the survivors, 4378 pg/ml (1400-13,943 pg/ml) vs. 1052 pg/ml (232-4076 pg/ml), respectively. The median (IQR) NT-pro-BNP values were significantly higher in patients with chronic cardiac disease or cardiac surgery than in non-cardiac patients, 1947 pg/ml (801-4687 pg/ml) vs. 417 pg/ml (153-1735 pg/ml), respectively, if renal function was normal. With deteriorating renal function, the NT-pro-BNP values showed a significant increase. The area under curve for baseline NT-pro-BNP predicting 90-day mortality was moderate: 0.718 (95% confidence interval 0.674-0.761). Baseline NT-pro-BNP over 1765 pg/ml was independently associated with 90-day mortality by logistic regression analysis (P<0.001). CONCLUSIONS: NT-pro-BNP on admission is commonly elevated and independently associated with 90-day mortality in critically ill ARFpatients. However, the routine use of NT-pro-BNP for prognostic purpose does not seem to add value to clinical data in ARFpatients.