BACKGROUND: Cervical lymphadenitis (CL) cannot be easily distinguished from Kawasaki disease (KD). We therefore explored whether brain natriuretic peptide (BNP) levels are useful in this context. METHODS: We retrospectively analyzed 14 children with CL and 177 children with KD. Patients with KD were divided into three groups according to their clinical symptoms at hospitalization - 97 patients had typical KD, 35 had node-first KD (NFKD), and 45 had KD without lymphadenopathy. We reviewed data on clinical and laboratory parameters, including serum BNP levels, at hospitalization together with factors that might distinguish KD from CL. RESULTS: Patients with CL were older than those with KD. Serum BNP levels were higher in all the KD groups than in the CL group. Multivariate logistic regression analyses indicated that higher BNP levels were associated with NFKD (odds ratio: 1.12, 95% confidence interval: 1.01-1.25). The receiver operating characteristic curve yielded a BNP cutoff of 18.3 pg/mL, with a sensitivity of 0.680, a specificity of 0.857, and an area under the curve of 0.806 (95% confidence interval: 0.665-0.947). CONCLUSIONS: Serum BNP levels can be used to distinguish KD from CL, especially in patients with NFKD.
BACKGROUND: Cervical lymphadenitis (CL) cannot be easily distinguished from Kawasaki disease (KD). We therefore explored whether brain natriuretic peptide (BNP) levels are useful in this context. METHODS: We retrospectively analyzed 14 children with CL and 177 children with KD. Patients with KD were divided into three groups according to their clinical symptoms at hospitalization - 97 patients had typical KD, 35 had node-first KD (NFKD), and 45 had KD without lymphadenopathy. We reviewed data on clinical and laboratory parameters, including serum BNP levels, at hospitalization together with factors that might distinguish KD from CL. RESULTS: Patients with CL were older than those with KD. Serum BNP levels were higher in all the KD groups than in the CL group. Multivariate logistic regression analyses indicated that higher BNP levels were associated with NFKD (odds ratio: 1.12, 95% confidence interval: 1.01-1.25). The receiver operating characteristic curve yielded a BNP cutoff of 18.3 pg/mL, with a sensitivity of 0.680, a specificity of 0.857, and an area under the curve of 0.806 (95% confidence interval: 0.665-0.947). CONCLUSIONS: Serum BNP levels can be used to distinguish KD from CL, especially in patients with NFKD.