• Hailing Li, Hongping Wang, Yunpeng Lou, Wenli Miao, and Ning Sha.
• Department of Intensive Care Unit, 401st Hospital of PLA, Qingdao 266071, Shandong, China. Corresponding author: Li Hailing, Email: lihailing608@163.com.
• Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2014 Jul 1;26(7):508-12.

ObjectiveTo investigate N-terminal pro-B-type natriuretic peptide (NT-proBNP) cutoff value for the mortality in different age groups in critically ill patients.MethodsA retrospective study was conducted. 295 patients admitted to the intensive care unit (ICU) of 401st Hospital of PLA from January 2011 to October 2012 were divided into two groups according to age [group with age<65 years old (n=105) and group with age≥ 65 years old (n=190)]. The concentrations of serum NT-proBNP, hematocrit (HCT), procalcitonin (PCT), C-reactive protein(CRP), serum creatinine(SCr), estimated glomerular filtration rate (eGFR), acute physiology and chronic health evaluationII (APACHEII) score and probability of survival (PS) were recorded within 24 hours. The primary outcome was ICU mortality. Receiver operator characteristic curve (ROC curve) was used to evaluate the value of NT-proBNP for predicting the mortality.Results(1) There were no significant differences in the length of stay in ICU, mechanical ventilation rate, the mortality, the incidence of cardiovascular disease, digestive disease, neurologic disease, and the number of patients having received operation, HCT, PCT and CRP between the two groups (all P>0.05). The percentage of the male, the APACHEII score, the percentage of respiratory disease, and NT-proBNP in group with age≥65 years old were higher than those of the group with age<65 years old [the percentage of the male: 51.6% vs. 33.0%, χ (2)=9.093, P=0.003; APACHEII score: 22.94±8.10 vs. 19.44±8.51, Z=-3.259, P=0.001; the percentage of respiratory disease: 29.47% vs. 17.14%, χ(2)=5.472, P=0.024; NT-proBNP: 5 859.00 (2 050.75, 23 802.75) ng/L vs. 2 882.00 (275.15, 6 236.00) ng/L, Z=-5.514, P=0.000]; PS, the percentage of patients having multiple injuries and other diseases and eGFR in group with age≥65 years old were lower than those of the group with age <65 years old [PS: 59.0 (31.5,79.0)% vs. 70.0 (40.0, 84.0)%, Z=-3.431, P=0.001; the percentage of multiple injuries: 0.53% vs. 17.14%, χ(2)=30.987, P=0.000; the percentage of other disease: 5.79% vs. 13.33%, χ(2)=4.962, P=0.030; eGFR (ml×min(-1)×1.73 m(-2)): 81.07 (45.77, 131.80) vs. 95.54 (33.64, 165.55), Z=-2.214, P=0.027]. (2) The area under the ROC curve (AUC) [95% confidence interval (95%CI)] of NT-proBNP in patients with age<65 years old was significantly higher than that of group with age≥65 years old and the entire group [0.825 (0.738-0.892) vs. 0.664 (0.592-0.731) and 0.725 (0.670-0.775), Z1=-2.835, P1=0.005; Z2=-1.995, P2=0.046]. (3) The sensitivity (76.10% vs. 64.10%), specificity (82.35% vs. 67.12%), positive predictive value (90.0% vs. 75.8%), and negative predictive value (62.2% vs. 53.8%) with cutoff value of NT-proBNP (2 882 ng/L) in group with age<65 years old were significantly higher than those with NT-proBNP cutoff value (6 062 ng/L) in group with age≥65 years old.ConclusionsNT-proBNP cutoff value in different age groups for the prediction of mortality in the critically ill patients maybe more objective and accurate.

18 keywords...

hide…