Utility of B-type natriuretic peptide for the evaluation of intensive care unit shock

Objectives: Among patients with congestive heart failure, B-type natriuretic peptide measurement is useful to estimate filling pressures and to prognosticate adverse outcome. However, among critically ill intensive care unit patients with shock, the utility of B-type natriuretic peptide to assess cardiac hemodynamics or prognosis has not been explored. Design: Clinical investigation. Setting: Hospital. Patients. Forty-nine patients with shock and indication for pulmonary artery catheterization. Interventions. Analysis for B-type natriuretic peptide was performed on blood obtained at the time of catheter placement. Measurements and Main Results: Correlations between B-type natriuretic peptide and pulmonary artery occlusion pressure as well as cardiac index were calculated using Spearman analysis. Mortality at the time of study completion was correlated with B-type natriuretic peptide values and Acute Physiology and Chronic Health Evaluation II scores, and logistic regression identified independent predictors of mortality. A wide range of B-type natriuretic peptide concentrations was seen in intensive care unit patients (<5 to >5000 pg/mL); only eight patients (16%) had normal B-type natriuretic peptide concentrations. Log-transformed B-type natriuretic peptide concentrations did not correlate with interpatient cardiac index or pulmonary artery occlusion pressure (all p = not significant); however, a B-type natriuretic peptide <350 pg/mL had a negative predictive value of 95% for the diagnosis of cardiogenic shock. Median B-type natriuretic peptide concentrations were higher in those who died than those who survived (943 pg/mL vs. 378 pg/mL, p < .001). In multivariable analysis, a B-type natriuretic peptide concentration in the highest log-quartile was the strongest predictor of mortality (odds ratio = 4.50, 95% confidence interval = 1.87-99.0, p < .001). Conclusion: B-type natriuretic peptide concentrations are frequently elevated among critically ill patients in the intensive care unit and cannot be used as a surrogate for pulmonary artery catheterization. B-type natriuretic peptide concentrations in intensive care unit shock may provide powerful information for use in mortality prediction.