Risk Factors of Intravenous Immunoglobulin Resistance in Children With Kawasaki Disease: A Meta-Analysis of Case-Control Studies

Biomarkers and Related Factors for the Diagnosis, Risk of Coronary Artery Lesions, and Resistance to Intravenous Immunoglobulin in Kawasaki Disease: An Umbrella Review of Meta-Analyses

Kawasaki disease (KD) is a self-limited febrile disease predominantly affecting infants and children under 5 years old. Coronary artery lesions (CAL) are a prevalent complication, highlighting the necessity for swift diagnosis and treatment. A comprehensive review of biomarkers applicable for the diagnosis and treatment of Kawasaki disease (KD) in clinical settings is imperative. To provide a comprehensive review and analysis of biomarkers for diagnosis of KD, incidence of CAL, and intravenous immunoglobulin (IVIG) resistance. The data included in our study were sourced from searches conducted in PubMed/MEDLINE, Embase, EBSCO, and Google Scholar until March 15, 2024. Studies investigating the association with KD or evaluating diagnostic value were included in our study. Eligibility was independently assessed by two authors, with conflicts resolved through discussion. Data extraction was performed by 2 independent authors, following Meta-analyses Of Observational Studies in Epidemiology (MOOSE) guideline. Data were pooled using a random-effects model. We assess biomarkers relevant to KD, categorizing them into three groups: diagnostic, associated with CAL incidence, and linked to IVIG resistance. For studies focusing solely on association, we present standardized mean differences (SMD). For those reporting sensitivity and specificity as diagnostic measures, we calculate the diagnostic odds ratio (DOR) to compare their efficacy. We identified 14 meta-analyses on biomarkers related to KD. 11 biomarkers exhibited diagnostic value for KD, while 21 were associated with its progression. Four biomarkers, including non-coding RNAs (DOR, 19.35 [95% CI, 13.58-27.56]), Serum ferritin (DOR, 24.90 [11.67-53.12]), N terminal proBNP (DOR, 21.03 [9.03-49.00]), and micro RNAs (DOR, 45.28 [6.30-325.52]), have significant diagnostic value for the diagnosis of KD. Seven biomarkers showed significant association with the incidence of CAL. Twenty biomarkers were for the prediction of IVIG resistance, including prognostic nutritional index (DOR, 7.72 [95% CI, 2.37-25.09]), non-coding RNAs (DOR, 14.63 [3.24-66.14]), neutrophil to lymphocyte ratio (DOR, 6.62 [4.05-10.81]), platelet to lymphocyte ratio (DOR, 3.30 [2.10-5.19]), and C reactive protein (DOR, 6.58 [3.69-11.74]). Based on the evidence, we have proposed various biomarkers associated with KD. Our aim is for these biomarkers to have wide applicability in both diagnostic and therapeutic settings.

U-shaped relationship found between fibrinogen-to-albumin ratio and systemic inflammation response index in osteoporotic fracture patients

The relationship between the Systemic Inflammatory Response Index (SIRI) and the Fibrinogen-to-albumin ratio (FAR) has not been extensively investigated. The objective of this study was to determine the independent relationship between FAR and SIRI in people with osteoporotic fractures (OPF). A cross-sectional study was conducted using retrospective data from 3431 hospitalized OPF patients. The exposure variable in this study was the baseline FAR, while the outcome variable was the SIRI. Covariates, including age, gender, BMI, and other clinical and laboratory factors, were adjusted. Cross-correlation analysis and linear regression models were applied. The generalized additive model (GAM) investigated non-linear relationships. Adjusted analysis revealed an independent negative association between FAR and SIRI in OPF patients (β = - 0.114, p = 0.00064, 95% CI - 0.180, - 0.049). A substantial U-shaped association between FAR and SIRI was shown using GAM analysis (p < 0.001). FAR and SIRI indicated a negative association for FAR below 6.344% and a positive correlation for FAR over 6.344%. The results of our study revealed a U-shaped relationship between SIRI and FAR. The lowest conceivable FAR for a bone-loose inflammatory disease might be 6.344%, suggesting that this has particular significance for the medical diagnosis and therapy of persons with OPF. Consequently, the term "inflammatory trough" is proposed. These results offer fresh perspectives on controlling inflammation in individuals with OPF and preventing inflammatory osteoporosis.

The safety and effectiveness of clopidogrel versus aspirin in Kawasaki disease with mild-to-moderate liver injury

Kawasaki disease can be combined with liver injury. As a mainstay treatment for Kawasaki disease, aspirin may cause liver injury. This study aimed to compare the safety and effectiveness of clopidogrel versus aspirin in Kawasaki disease with mild-to-moderate liver injury. This study retrospectively analysed 166 children with Kawasaki disease combined with mild-to-moderate liver injury. The children treated with clopidogrel were less likely to have aggravated liver injury than those treated with aspirin (n = 2/100 vs. n = 13/66, P < 0.001). The initial alanine aminotransferase value of the clopidogrel group was higher (131.5 [98.5, 167.5] vs. 96 [72, 133], P < 0.001), while the time of alanine aminotransferase recovery to normal was similar (5 [4, 7] vs. 4 [3, 7], P = 0.179). No significant fever differences observed between groups: 7.5 [6, 9] for aspirin vs. 7 [6, 8] for clopidogrel group, P = 0.064. The probability of nonresponse to intravenous immunoglobulin (n = 29/100 vs. n = 30/66, P = 0.030) and the days of hospitalization (n = 6 [4, 9] vs. n = 7 [5, 10], P = 0.007) in the clopidogrel group were less than those in the aspirin group. In conclusion, the application of clopidogrel is potentially superior to aspirin in Kawasaki disease combined with mild-to-moderate liver injury.

Bibliometric and visual analysis of Kawasaki disease in children from 2012 to 2022

Background

In recent years, the incidence of Kawasaki disease among the pediatric population has experienced a significant increase. With complications mainly affecting the cardiovascular system, Kawasaki disease has received widespread attention from scholars worldwide. Numerous articles on Kawasaki disease in children have been published far. However, there is a lack of studies that use visualization methods to perform a bibliometric analysis of the relevant literature. This study aims to obtain overall information on the output characteristics of publications on childhood Kawasaki disease between 2012 and 2022 through bibliometric analysis, identify research hotspots and frontiers, and provide new ideas and references for future clinical and scientific research.

Methods

Literature meeting the inclusion criteria was screened from the Web of Science Core Collection, PubMed, and Scopus databases. Visual analysis of the literature by country, institution, journal, author, keywords, and references was performed using Citespace (6.1.R6), VOSviewer (1.6.18), and the online bibliometric website (https://bibliometric.com/).

Results

A total of 4,867 eligible publications were included. The number of annual publications is generally rising, rapidly increasing since 2019. Among countries and institutions, China and KAOHSIUNG CHANG GUNG MEMORIAL HOSPITAL have the highest output of articles. With 104 publications, Ho-Chang Kuo has a high impact in the field of KD. The most cited author is Jane W. Newburger. The most prolific journal is FRONTIERS IN PEDIATRICS. CIRCULATION is the most frequently co-cited journal. The most popular keyword in frequency and centrality is "immunoglobulin". The reference with the highest burst intensity was Verdoni L, LANCET, 2020.

Conclusion

Kawasaki disease in children remains a hot topic among pediatricians worldwide and is receiving increasing attention. We innovated the "national-institutional-journal" model, which promotes further international cooperation in this field. The hot topics in the field of pediatric KD are "KD pathogenesis", "immunoglobulin resistance and complementary therapy", and "cardiovascular complications". Frontiers include disease-related ("multisystem inflammatory syndrome", "coronavirus disease 2019", "hypotension"), treatment-related ("procalcitonin", " anakinra"), and pathogenesis ("polymerase chain reaction").

The state of play in tools for predicting immunoglobulin resistance in Kawasaki disease

INTRODUCTION

Intravenous immunoglobulin (IVIG) resistance is an independent risk factor for the development of coronary artery lesions (CAL) in patients with Kawasaki disease (KD). Accurate identification of IVIG-resistant patients is one of the biggest clinical challenges in the treatment of KD.

AREAS COVERED

In this review article, we will go over current IVIG resistance scoring systems and other biological markers of IVIG resistance, with a particular focus on advances in machine-based learning techniques and high-throughput omics data.

EXPERT OPINION

Traditional scoring models, which were developed using logistic regression, including the Kobayashi score and Egami score, are inadequate at identifying IVIG resistance in non-Japanese populations. Newer machine-learning methods and high-throughput technologies including transcriptomic and epigenetic arrays have identified several potential targets for IVIG resistance including gene expression of the Fc receptor, and components of the interleukin (IL)-1β and pyroptosis pathways. As we enter an age where access to big data has become more commonplace, interpretation of large data sets that are able take into account complexities in patient populations will hopefully usher in a new era of precision medicine, which will enable us to identify and treat KD patients with IVIG resistance with increased accuracy.

Intravenous immunoglobulin-resistant Kawasaki disease: Risk factors in children in a middle-income country

Background

Identifying risk factors in patients with intravenous immunoglobulin (IVIG) resistance Kawasaki disease (KD) is vital in managing and preventing coronary artery aneurysms (CAAs). We aimed to study the risk factors for IVIG resistance KD in Malaysian children.

Methods

This retrospective observational study of children with KD was conducted at two tertiary hospitals in Malaysia from January 2014 to December 2019. Multivariable binary logistic regression was used to analyze the risk factors associated with IVIG resistance.

Results

A total of 174 patients, 118 males (67.8%) with a median age of 1.4-year-old (interquartile range: 0.1-12.1-year-old), were analyzed. Early (<5 days) and late (>10 days) IVIG treatments were observed in 14 (8.1%) and 19 (11.0%), respectively. Thirty-two patients (18.4%) had IVIG resistance. The independent factors associated with IVIG resistance were high white cell count, hypoalbuminemia, and extremities changes with an odd ratio of 4.7, 3.0, and 4.0, respectively. In addition, CAA was significantly higher in IVIG resistance compared to IVIG responder patients (57.5% [19/33] vs. 23.4% [33/141], P < 0.001). The sensitivity was high in Harada (93.8%) but low in Kobayashi and Egami (46.9% and 34.4%, respectively). The specificity was high with Egami (79.6%) but low in Harada and Kobayashi (22.5% and 64.1%, respectively).

Conclusion

Leukocytosis, hypoalbuminemia, and extremities changes were independent risk factors for IVIG resistance. The variation in sensitivity and specificity of the Japanese scoring makes it unsuitable for predicting IVIG resistance in Malaysian children.

Bridging a diagnostic Kawasaki disease classifier from a microarray platform to a qRT-PCR assay

BACKGROUND

Kawasaki disease (KD) is a systemic vasculitis that mainly affects children under 5 years of age. Up to 30% of patients develop coronary artery abnormalities, which are reduced with early treatment. Timely diagnosis of KD is challenging but may become more straightforward with the recent discovery of a whole-blood host response classifier that discriminates KD patients from patients with other febrile conditions. Here, we bridged this microarray-based classifier to a clinically applicable quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay: the Kawasaki Disease Gene Expression Profiling (KiDs-GEP) classifier.

METHODS

We designed and optimized a qRT-PCR assay and applied it to a subset of samples previously used for the classifier discovery to reweight the original classifier.

RESULTS

The performance of the KiDs-GEP classifier was comparable to the original classifier with a cross-validated area under the ROC curve of 0.964 [95% CI: 0.924-1.00] vs 0.992 [95% CI: 0.978-1.00], respectively. Both classifiers demonstrated similar trends over various disease conditions, with the clearest distinction between individuals diagnosed with KD vs viral infections.

CONCLUSION

We successfully bridged the microarray-based classifier into the KiDs-GEP classifier, a more rapid and more cost-efficient qRT-PCR assay, bringing a diagnostic test for KD closer to the hospital clinical laboratory.

IMPACT

A diagnostic test is needed for Kawasaki disease and is currently not available. We describe the development of a One-Step multiplex qRT-PCR assay and the subsequent modification (i.e., bridging) of the microarray-based host response classifier previously described by Wright et al. The bridged KiDs-GEP classifier performs well in discriminating Kawasaki disease patients from febrile controls. This host response clinical test for Kawasaki disease can be adapted to the hospital clinical laboratory.

PROGRESSION OF VASCULAR FUNCTION AND BLOOD PRESSURE IN A MOUSE MODEL OF KAWASAKI DISEASE

ABSTRACT

Kawasaki disease (KD) is a systemic vasculitis of childhood characterized by vascular damage in the acute stage, which can persist into the late stage. The vascular mechanisms in the cardiovascular risk of KD are not fully studied. We investigated the vascular function and blood pressure in a murine model of KD. We used the Candida albicans water-soluble (CAWS) fraction model. Mice were injected with 4 mg CAWS for 5 consecutive days and separated into three groups. Control, CAWS 7 days (C7), and CAWS 28 days (C28). Hearts and arteries were harvested for vascular characterization. Rat aortic smooth muscle cells were used to studies in vitro. C7 presented elevated inflammatory markers in the coronary area and abdominal aortas, whereas C28 showed severe vasculitis. No difference was found in blood pressure parameters. Vascular dysfunction characterized by higher contractility to norepinephrine in C7 and C28 in aortic rings was abolished by blocking nitric oxide (NO), reactive oxygen species, and cyclooxygenase (COX)-derived products. The CAWS complex increased COX2 expression in rat aortic smooth muscle cells, which was prevented by Toll-like receptor 4 antagonist. Our data indicate that the murine model of KD is associated with vascular dysfunction likely dependent on COX-derived products, oxidant properties, and NO bioavailability. Furthermore, vascular smooth muscle cell may present an important role in the genesis of vascular dysfunction and vasculitis via the Toll-like receptor 4 pathway. Finally, the CAWS model seems not to be appropriate to study KD-associated shock. More studies are necessary to understand whether vascular dysfunction and COXs are triggers for vasculitis.

Under-Recognized Macrophage Activation Syndrome in Refractory Kawasaki Disease: A Wolf in Sheep's Clothing

Recognition of macrophage activation syndrome (MAS) in patients with refractory Kawasaki disease (KD) can be challenging. This study aimed to investigate the incidence of MAS in patients with refractory KD and to compare the characteristics of refractory KD and MAS. Medical records of 468 patients diagnosed with KD from January 2010 to December 2019 were retrospectively reviewed. Of the 468 KD patients, 63 were enrolled in the study as a refractory KD group (n = 59) and an MAS group (n = 4). The incidence of MAS was 0.8% (4/468) in patients with KD and 6.3% (4/63) in patients with refractory KD. Compared to the refractory KD group, the MAS group had higher frequencies of incomplete KD, hepatosplenomegaly, third-line treatment, and MAS screening, and showed lower levels of albumin. No significant differences were found in other clinical and laboratory findings. In addition to four patients with MAS, five patients with refractory KD who received third-line treatment showed severe systemic inflammation and organ dysfunction, but only one in five patients underwent MAS screening, including ferritin levels. In conclusion, given the relatively high incidence of MAS in children with refractory KD and the similar phenotype between refractory KD and MAS, we propose that MAS screening should be included in routine laboratory tests for refractory KD.

[Association of liver damage with coronary artery lesion and no response to intravenous immunoglobulin in the acute stage of Kawasaki disease]

OBJECTIVES

To summarize the clinical features of liver damage in children in the acute stage of Kawasaki disease (KD), and to investigate the clinical value of liver damage in predicting coronary artery lesion and no response to intravenous immunoglobulin (IVIG) in children with KD.

METHODS

The medical data were collected from 925 children who were diagnosed with KD for the first time in Beijing Children's Hospital from January 1, 2016 to December 31, 2017. According to the presence or absence of abnormal alanine aminotransferase (ALT) level on admission, the children were divided into a liver damage group (n=284) and a non-liver damage group (n=641). A logistic regression analysis was used to investigate the clinical value of the indicators including liver damage in predicting coronary artery lesion and no response to IVIG in children with KD.

RESULTS

Compared with the non-liver damage group, the liver damage group had a significantly earlier admission time and significantly higher serum levels of inflammatory indicators (P<0.05). The liver damage group had a significantly higher incidence rate of coronary artery lesion on admission than the non-liver damage group (P=0.034). After initial IVIG therapy, the liver damage group had a significantly higher proportion of children with no response to IVIG than the non-liver damage group (P<0.001). In children with KD, coronary artery lesion was associated with the reduction in the hemoglobin level and the increases in platelet count, C-reactive protein, and ALT (P<0.05), and no response to IVIG was associated with limb changes, the reduction in the hemoglobin level, the increases in platelet count, C-reactive protein, and ALT, and coronary artery lesion (P<0.05).

CONCLUSIONS

Compared with those without liver damage, the children in the early stage of KD with liver damage tend to develop clinical symptoms early and have higher levels of inflammatory indicators, and they are more likely to have coronary artery lesion and show no response to IVIG treatment.

Risk Factors and Predictive Models for Intravenous Immunoglobulin Resistance in Children with Recurrent Kawasaki Disease

Purpose

To explore the risk factors and develop predictive models for intravenous immunoglobulin (IVIG) resistance in children with recurrent Kawasaki disease (KD).

Patients and Methods

Patients with recurrent KD were retrospectively reviewed. Clinical and laboratory data at recurrence were collected and compared between patients with and without IVIG resistance. The patients were randomly divided into training and validation cohorts for model development and validation. All variables were subjected to standard Lasso and its variant group Lasso analyses, respectively, to construct predictive models. Model performance was evaluated by receiver operating characteristics (ROC) curves, calibration curves, and Hosmer-Lemeshow tests.

Results

A total of 90 children with recurrent KD were included. A total of 16 cases were IVIG resistant. The patients with IVIG resistance had higher age and IVIG resistance probability at the first episode, increased CRP levels, neutrophil count, neutrophil percentage, direct bilirubin level, prothrombin time, and international normalized ratio, and decreased lymphocyte count, lymphocyte percentage, and serum sodium levels. Five variables including age and IVIG resistance at the first episode, lymphocytes count, serum sodium levels, and CRP levels were finally selected by standard Lasso (lLasso model) and four variables including age and IVIG resistance at the first episode, neutrophil percentage, and CRP levels were selected by group Lasso (gLasso). ROC curves suggested lLasso and gLasso models had similar excellent discrimination in both the training cohort (0.895 vs 0.906) and the validation cohort (0.855 vs 0.909). Hosmer-Lemeshow tests suggested the two models exerted a good calibration. Two nomograms were also constructed to facilitate the potential application of the two models.

Conclusion

Age and IVIG resistance at the first episode and some laboratory variables may be risk factors for IVIG resistance in recurrent KD. Two predictive models for IVIG resistance with excellent performance were established in recurrent KD. External validation should be performed before clinical use.

High-Concentration Intravenous Immunoglobulin May Influence the Course of Fever and Rate of Reported Treatment Resistance in Children With Kawasaki Disease: A Single-Center Retrospective Analysis

BACKGROUND

Intravenous immunoglobulin (IVIG) resistance in patients with Kawasaki disease (KD) is defined as persistent or recrudescent fever ≥36 hours after IVIG infusion. We have experienced an increase in IVIG resistance in patients with KD since the substitution of 10% IVIG for 5% IVIG. This study aimed to determine the independent association between increased IVIG resistance and 10% IVIG therapy.

METHODS

Medical records of pediatric patients with KD were retrospectively reviewed. Clinical and laboratory characteristics were compared between patients receiving 5% IVIG therapy and those receiving 10% IVIG therapy. Between IVIG-responsive and IVIG-resistant patients, a multivariate analysis was performed to determine the independent factors for IVIG resistance.

RESULTS

A total of 119 patients were included in this study: 81 (68.1%) and 38 (31.9%) patients received 5% and 10% IVIG therapy, respectively. IVIG resistance was identified in 34 (28.6%) patients: 44.7% of patients receiving 10% IVIG therapy and 21.0% of patients receiving 5% IVIG therapy (p = 0.008). The clinical manifestations and outcomes were comparable between patients who received 5% IVIG therapy and those who received 10% IVIG therapy. IVIG resistance was significantly associated with fewer fever days at IVIG administration (p = 0.032), a higher percentage of neutrophils (p = 0.013), and 10% IVIG treatment (p = 0.004) in the multivariate analysis.

CONCLUSION

10% IVIG therapy was significantly associated with increased reporting of IVIG resistance. However, the increase in patients with fever patterns consistent with IVIG resistance seemed to represent adverse febrile reactions resulting from using high-concentration IVIG rather than increased severity of KD.

Can AI Help Pediatricians? Diagnosing Kawasaki Disease Using DRSA

The DRSA method (dominance-based rough set approach) was used to create decision-making rules based on the results of physical examination and additional laboratory tests in the differential diagnosis of Kawasaki disease (KD), infectious mononucleosis and S. pyogenes pharyngitis in children. The study was conducted retrospectively. The search was based on the ICD-10 (International Classification of Diseases) codes of final diagnosis. Demographic and laboratory data from one Polish hospital (Poznan) were collected. Traditional statistical methods and the DRSA method were applied in data analysis. The algorithm formed 45 decision rules recognizing KD. The rules with the highest sensitivity (number of false negatives equals zero) were based on the presence of conjunctivitis and CRP (C-reactive Protein) ≥ 40.1 mg/L, thrombocytosis and ESR (Erythrocyte Sedimentation Rate) ≥ 77 mm/h; fair general condition and fever ≥ 5 days and rash; fair general condition and fever ≥ 5 days and conjunctivitis; fever ≥ 5 days and rash and CRP ≥ 7.05 mg/L. The DRSA analysis may be helpful in diagnosing KD at an early stage of the disease. It can be used even with a small amount of clinical or laboratory data.

Shaanxi Provincial Diagnosis and Treatment Center of Kawasaki Disease

Kawasaki disease is an acute, self-limited vascuitis that mainly occurs in children under 5 years of age. Intravenous immunoglobulin (IVIG) has become an effective treatment regimen, which can effectively reduce the incidence of cardiovascular complications. However, there has been no consensus or clinical guidelines for the application of IVIG in children with Kawasaki disease till now. This consensus is developed based on the current research progress on the application of IVIG in children with Kawasaki disease in China and overseas, with reference to the diagnosis and treatment guidelines for Kawasaki disease in China and overseas, and the opinions of experts. This consensus provides recommendations on the clinical application strategy of IVIG in children with Kawasaki disease and the prevention and treatment of its adverse reactions. Citation.

Is there an association between intravenous immunoglobulin resistance and coronary artery lesion in Kawasaki disease?-Current evidence based on a meta-analysis

Background

Coronary artery lesion (CAL) caused by Kawasaki disease (KD) is a leading cause of acquired heart disease in children. Initial treatment of intravenous immunoglobulin (IVIG) can reduce the incidence of CAL. Although most of the current studies have shown a certain correlation between CAL and IVIG resistance, the conclusions are not completely consistent. Thus, we performed this meta-analysis to evaluate the association between IVIG resistance and CAL in KD.

Methods

PubMed, EMBASE, the Cochrane Central Register of Controlled Trials, and China National Knowledge Infrastructure through April 21, 2020 were searched to detect relevant studies. Data analysis was performed with STATA 15.1.

Results

A total of 53 relevant studies were eligible to this analysis, including 30312 KD patients, of which 4750 were IVIG resistance and 25562 were responders. There was a significant difference found between IVIG resistance and IVIG response groups in the incidence of CAL (P < 0.001, odds ratio (OR), 3.89; 95% confidence interval (CI) (3.18, 4.75)). The heterogeneity test results showed that the I² value was 74.8%. The meta-regression analysis showed that the study regions might be the sources of heterogeneity. The subgroup analysis suggested that the incidence of CAL in the IVIG resistance group was still higher than that in the IVIG response group under different regions, IVIG resistance diagnostic criteria, CAL diagnostic criteria, and study types. Meanwhile, the sensitivity analysis did not find any significant impact from every single study.

Conclusions

This is the first meta-analysis to reveal the incidence of CAL was associated with IVIG resistance in KD patients. Further well-designed studies with uniform criteria are needed to evaluate the incidence of CAL in IVIG resistant patients.

A Machine Learning Model to Predict Intravenous Immunoglobulin-Resistant Kawasaki Disease Patients: A Retrospective Study Based on the Chongqing Population

Objective: We explored the risk factors for intravenous immunoglobulin (IVIG) resistance in children with Kawasaki disease (KD) and constructed a prediction model based on machine learning algorithms. Methods: A retrospective study including 1,398 KD patients hospitalized in 7 affiliated hospitals of Chongqing Medical University from January 2015 to August 2020 was conducted. All patients were divided into IVIG-responsive and IVIG-resistant groups, which were randomly divided into training and validation sets. The independent risk factors were determined using logistic regression analysis. Logistic regression nomograms, support vector machine (SVM), XGBoost and LightGBM prediction models were constructed and compared with the previous models. Results: In total, 1,240 out of 1,398 patients were IVIG responders, while 158 were resistant to IVIG. According to the results of logistic regression analysis of the training set, four independent risk factors were identified, including total bilirubin (TBIL) (OR = 1.115, 95% CI 1.067-1.165), procalcitonin (PCT) (OR = 1.511, 95% CI 1.270-1.798), alanine aminotransferase (ALT) (OR = 1.013, 95% CI 1.008-1.018) and platelet count (PLT) (OR = 0.998, 95% CI 0.996-1). Logistic regression nomogram, SVM, XGBoost, and LightGBM prediction models were constructed based on the above independent risk factors. The sensitivity was 0.617, 0.681, 0.638, and 0.702, the specificity was 0.712, 0.841, 0.967, and 0.903, and the area under curve (AUC) was 0.731, 0.814, 0.804, and 0.874, respectively. Among the prediction models, the LightGBM model displayed the best ability for comprehensive prediction, with an AUC of 0.874, which surpassed the previous classic models of Egami (AUC = 0.581), Kobayashi (AUC = 0.524), Sano (AUC = 0.519), Fu (AUC = 0.578), and Formosa (AUC = 0.575). Conclusion: The machine learning LightGBM prediction model for IVIG-resistant KD patients was superior to previous models. Our findings may help to accomplish early identification of the risk of IVIG resistance and improve their outcomes.