[Prediction of the risk of coronary arterial lesions in Kawasaki disease by N-terminal pro-brain natriuretic peptide].

OBJECTIVE: To detect plasma N-terminal pro-brain natriuretic peptide (NT-proBNP) in acute Kawasaki disease (KD) and analyze the relationship between NT-proBNP and other bio-markers in order to evaluate if NT-proBNP could be as a useful diagnostic marker to predict the risk of coronary arterial lesions in acute KD.

METHOD: Totally 106 patients with KD were recruited from January 2012 to April 2014 at Department of Pediatrics of Tongji Hospital Affiliated to Tongji Medical College, Huazhong University of Science and Technology,64 were boys and 42 were girls, their age ranged from 2 months to 8 years and 4 months. Of the 106 cases, 48 had typical KD(TKD) and 58 incomplete KD(IKD). They were divided into two groups according to echocardiography results: coronary arterial lesions (KD-CAL, n = 33) and non coronary arterial lesions (KD-nCAL, n = 73). Forty children whose age and gender matched with respiratory tract infection were selected as control group, 22 were boys and 18 were girls, age range from 7 months to 7 years and 11 months. Plasma NT-proBNP levels were measured by using the enzyme-linked fluorescence analysis (ELFA) at the day of admission, meanwhile blood routine tests, liver function tests, determination of C-reactive protein (CEP), erythrocyte sedimentation rate (ESR), electrolytes were performed in these patients. Pearson's correlation analysis was used to evaluate the association. The ROC curve analysis was done to identify the threshold of coronary 'arterial lesions.

RESULT: The levels of NT-proBNP were (1 037 271) ng/L in TKD group and (1,325 ± 264) ng/L in IKD group. The levels of NT-proBNP in control group was (125 ± 22) ng/L. Both the levels of NT-proBNP in TKD and IKD group were significantly higher than that of control group (t = 3.360, 3.590; P < 0.05). The level of NT-proBNP in KD-CAL group was (2,775 ± 842) ng/L and that of KD-nCAL group was (830 ± 145) ng/L, NT-proBNP levels of KD-nCAL group was significantly higher than that of control group (t = 3.660, P = 0.007) ; moreover the level of NT-proBNP of KD-CAL group was also significantly higher than that of KD-nCAL group ( t = 3.860, P = 0.005). The levels of total white blood cell count, neutrophil percentage, platelet count, CRP and ESR of KD-CAL group were significantly higher than those of the control group, however there was no significant difference between KD-CAL group and KD-nCAL group. The levels of albumin and Na of KD-nCAL group were significantly lower than those of the control group. Plasma NT-proBNP level in KD group was positively correlated with white blood cell count, neutrophil percentage, and CRP (r = 0.239, P = 0.025; r = 0.359, P = 0.001; r = 0.474, P = 0.001), there was a negative correlation between albumin and Na (r = -0.303, P = 0.015; r = -0.338, P = 0.002). When the level of NT-proBNP was higher than 950 ng/L, the sensitivity for diagnosis of coronary arterial lesions in the KD was 88.1% and the specificity was 89.0%.

CONCLUSION: The plasma NT-proBNP can be used as a useful parameter in early diagnosis of KD. Plasma NT-proBNP could be used to predict the risk of coronary arterial lesions in KD.