Can plasma B-type natriuretic peptide levels predict need for mechanical ventilation after injury?

B-type natriuretic peptide (BNP) is a neurohormone released from cardiomyocytes in response to volume expansion and increased ventricular wall distension. Increased plasma BNP levels are associated with mortality in critically ill patients cared for in medical intensive care units (ICUs). Additionally, plasma BNP levels may serve as a biomarker for excessive fluid resuscitation after injury. The utility of plasma BNP levels as a prognosticator of outcomes after injury has not been previously described. The purpose of this study was to describe the change in plasma BNP levels over the first 48 hours after injury and determine if there was a correlation between plasma BNP levels and clinical outcomes.

In this prospective observational cohort trial, plasma BNP levels were followed in injured patients admitted to a surgical ICU. Levels were obtained at admission to the emergency room (baseline) and at 12, 24, and 48 hours. Change in plasma BNP levels from baseline were calculated for each time point (ΔBNP). Demographic information was collected, including age, gender, injury severity score, ventilator days, ICU length of stay (LOS), hospital LOS, net fluid balance at 24 hours, and in-hospital mortality. Spearman's rank-order correlation coefficients were determined for plasma ΔBNP levels and outcome measures (days ventilated, ICU LOS, hospital LOS, and mortality).

Forty-four ICU patients were prospectively enrolled. Thirty-six patients (82%) were male, and 30 patients (68%) required mechanical ventilation. The mean age was 40.3 years. The median injury severity score was 19.5 (range, 9-29), and overall mortality was 14%. The mean baseline BNP level was 48 ± 66 pg/mL (range, 10-274 pg/mL). Mean ΔBNP at 24 hours was 74 ± 147 pg/mL. Net fluid balance at 48 hours after admission ranged from -1.6 to 15.6 L. Plasma ΔBNP levels at 24 and 48 hours did not correlate with net resuscitation volume at their respective time points of 24 and 48 hours. ΔBNP at 24 hours correlated with the number of days on mechanical ventilation (Spearman's ρ = .428, P = .007). ΔBNP at 24 hours also correlated with Acute Physiology and Chronic Health Evaluation II scores (Spearman's ρ = .430, P = .046). Patients with increases in plasma BNP of ≥ 75 pg/mL at 24 hours were nearly 3 times more likely to be ventilated for > 3 days (relative risk, 2.9; 95% confidence interval, 1.1-7.7).

Changes in plasma BNP levels over the first 24 hours after ICU admission may have prognostic value in determining the need for mechanical ventilation in patients admitted to ICUs after injury. Additionally, the correlation between plasma BNP levels at 24 hours and mechanical ventilation requirements is not due solely to resuscitation volume. Further studies examining the prognostic value of plasma BNP levels after injury are warranted.