Objective: To evaluate the value of the change of B-type natriuretic peptide (BNP) in predicting weaning outcome of patients with mechanical ventilation in ICU.
Methods: We carried out a prospective cohort study on patients in ICU at Fuxing Hospital from July 2013 to January 2014. Patients expected to use mechanical ventilation for more than 24 hours were enrolled in our study. The plasma concentration of BNP was measured at the beginning of spontaneous breathing trial (SBT) and 2 hours after. According to the outcome of SBT, patients were divided into SBT failure group (group A), weaning failure group (group B) and weaning success group (group C). The levels of BNP before and after the SBT were compared among three groups. The receiver operating characteristic (ROC) curve analysis was used to evaluate the changes of BNP (ΔBNP) and the rate of change (ΔBNP%) in predicting weaning outcome.
Results: A total of 56 patients were enrolled in our study, 12 in group A, 11 in group B and 33 in group C. In group C, both ΔBNP and ΔBNP% [7.0, (-1.2, 45.5) ng/L; 6.1 (-2.1, 11.5)%] were significantly lower than those in group A [88.0 (58.0, 140.0) ng/L, P < 0.05; 20.5 (15.3, 40.3)%, P < 0.001] and group B [46.0 (17.5, 91.2) ng/L, P < 0.001; 24.3 (13.0, 32.5) %, P < 0.001]. The area under ROC curve (AUC) of ΔBNP% to predict weaning success was 0.88 [95% CI 0.79-0.97; P < 0.001] and the cut-off value is 13.4% with relatively high sensitivity (84.8%) and specificity (82.6%). The AUC of ΔBNP to predict weaning success was 0.80 [95%CI 0.69-0.92; P < 0.001] and the cut-off value is 80ng/L with high sensitivity 93.4% and mediocre specificity (47.9%).
Conclusion: Either ΔBNP% or ΔBNP in the SBT can predict weaning outcome, in which ΔBNP% seems better than ΔBNP.