BackgroundThe factors predicting high-risk Kawasaki disease (KD) remain unclear. Therefore, we aimed to determine the risk factors for resistance to intravenous immunoglobulin (IVIG) treatment and coronary artery aneurysm (CAA) development in a Chinese pediatric population with high-risk KD.MethodsWe compared the performances of 11 scoring systems that have been reported to predict IVIG resistance among patients with KD hospitalized from January 2013 through August 2021. Patients were risk-stratified based on the optimal scoring system. The association of baseline characteristics with IVIG treatment resistance and CAA development was investigated within the high-risk group of KD.ResultsIn total, 346 pediatric patients with KD were included, of whom 63 (18.2%) presented with IVIG resistance. The Kobayashi score and five Chinese scoring system scores (Tang et al., Yang et al., Lan et al., Liping et al., and Wu et al.) were significantly higher in the IVIG non-responsive KD group than in the IVIG responsive KD group, and the results of the receiver operating characteristic (ROC) curves analysis were observed to be highest in the Xie Liping scoring system for IVIG resistance (area under the curve, 0.650). Especially, 87 (25.1%) patients comprised the high-risk KD group based on this optimal scoring system (≥5 points). IVIG resistance was significantly associated with the total bilirubin-to-albumin ratio (B/A ratio) [odds ratio, 7.427; 95% confidence interval (CI): 1.022–53.951]. The area under the ROC was 0.703 (95% CI: 0.586–0.821), and the cutoff point was 0.383, which indicated a sensitivity and specificity for predicting treatment resistance of 58% and 80%, respectively. The serum albumin level (odds ratio, 1.401; 95% CI: 1.049–1.869) and Z score of the left main coronary artery (odds ratio, 9.023; 95% CI: 1.070–76.112) were independent predictors of CAA development.ConclusionsIn the Chinese pediatric population with KD, the Xie Liping scoring system is the most appropriate method for identifying high-risk patients, and IVIG resistance could be predicted based on the B/A ratio. Serum albumin level and Z score of the left main coronary artery at baseline were warning indicators for CAA development. More intensified or adjunctive therapies and close follow-up should be considered for high-risk patients with these risk factors.

Objectives We aimed to evaluate the clinical and laboratory characteristics of patients with Kawasaki disease (KD) before and after therapy. Methods Patients with KD were divided into different groups according to their responsiveness to initial intravenous immunoglobulin (IVIG) treatment and coronary status. The clinical and laboratory parameters before and after therapy were compared. Multivariate analysis was performed to identify the independent risk factors, and the receiver operating characteristic (ROC) curve was applied to assess and compare the prediction ability of risk factors and their fluctuations. Results Of the 153 patients included in the study, 41 (26.8%) had IVIG resistance and 37 (24.2%) had developed CAA. After stratifying by therapy response, the two groups differed in the levels of total bilirubin (TSB), albumin, and sodium, neutrophil-to-lymphocyte count ratio (NLR), platelet-to-lymphocyte count ratio (PLR), TSB-to-albumin (B/A) ratio, and prognostic nutritional index (PNI) before IVIG, and in the white blood cell count (WBC), neutrophil count, levels of hemoglobin, C-reactive protein (CRP), alanine aminotransferase (ALT), and albumin, NLR, PNI, capillary leakage index (CLI), and systemic immune-inflammation index (SII) after IVIG. Multivariate analysis revealed that the B/A ratio before IVIG and CLI and SII after IVIG were significantly and positively associated with IVIG resistance and that there was a larger decline in the B/A ratio and smaller decline in CLI and SII pre- and post-treatment in the IVIG-resistant group than in the IVIG-responsive group. However, no statistical differences in the fluctuations of the B/A ratio, CLI, and SII as well as all parameters before and after therapy were observed in patients with and without CAA. ROC curve analyses found a greater AUC value of post-treatment parameters (0.751 and 0.706 for CLI and SII, respectively) compared with pre-treatment parameters (0.654 for B/A ratio) in predicting IVIG resistance; however, the predictive ability of the fluctuations in risk factors before and after therapy was not superior to that of baseline values. Conclusions The B/A ratio before IVIG and CLI and SII after IVIG were risk factors for IVIG resistance in patients with KD, independent of CAA development. Key Points• A high total bilirubin-to-albumin ratio before IVIG and high capillary leakage and systemic immune-inflammation indices after IVIG may indicate an increased risk of intravenous immunoglobulin resistance in patients with Kawasaki disease.• Post-treatment parameters were superior to pre-treatment parameters in terms of prediction; therefore, rapid and repeated assessment of risk factors before and after treatment must be considered in children in whom the vital signs and symptoms do not improve after treatment.

Purpose: Multisystem inflammatory syndrome in children, which has overlapping clinical features with Kawasaki disease (KD), has generated considerable interest in the relationship between KD and infectious diseases during the ongoing global outbreak of coronavirus disease 2019. However, few studies have focused on the relationship between KD and concomitant infection, and reports on the relationship between infections and recovery from coronary artery aneurysms (CAA) are even rarer.Methods: Patients were classified into case and control groups according to the results of their pathogen examinations, and the baseline characteristics of the two groups were compared. The Kaplan–Meier survival analysis was used to compare the medium-term recovery time of CAA between patients with and without infections, and multivariable analyses were performed to evaluate potential risk factors associated with CAA without recovery between 1 and 2 years of follow-up.Results: A total of 353 pediatric patients with KD were included, of whom 83 (23.5%) had confirmed co-infection. There were no significant differences in patients’ response to treatment and coronary artery outcome when compared between patients with and without infections. Among the 90 patients diagnosed with CAA, 20 (22.2%) had confirmed co-infection, and no significant differences were observed in coronary artery changes from baseline at 2 weeks, 4 weeks, and 3 months in patients with CAA with and without infections. The estimated median time (6 months, 95%CI:1.920–10.080) was higher in the CAA co-infected group than in the CAA non-infected group (3 months, 95%CI:2.366–3.634), with no significant difference. Multivariate analysis revealed that a high Z-score of the coronary artery internal diameter at 1 month after onset was significantly associated with CAA without recovery. The Z score of the left main coronary artery was ≥3.215 with an 89% sensitivity and 77% specificity in predicting CAA without recovery within 1 year of onset. The Z score of the right coronary artery was ≥3.845 with a 64% sensitivity and 98% specificity in predicting CAA without recovery within 1 year of onset and with an 83% sensitivity and 88% specificity within 2 years of onset.Conclusions: Concomitant infection with KD diagnosis did not affect the patients’ response to IVIG treatment and coronary artery outcome when compared with patients without infections; however, the time to coronary artery normalization was not notably prolonged in CAA patients with infections, but a larger maximum Z score at 1 month after onset was a risk factor significantly associated with coronary artery dilatation without recovery within 2 years of onset.