Neutrophil-to-Lymphocyte Ratio Predicts Intravenous Immunoglobulin-Resistance in Infants Under 12-Months Old With Kawasaki Disease

A machine learning-based model to predict intravenous immunoglobulin resistance in Kawasaki disease

Accurate prediction of intravenous immunoglobulin (IVIG) resistance is crucial for the effective treatment of Kawasaki disease(KD). This study aimed to develop a predictive model for IVIG resistance in patients with Kawasaki disease and to identify the key predictors. The training set underwent cross-validation, and models were constructed using six machine learning algorithms. Model performance was validated through cross-validation, test set evaluation, and two external validation sets evaluation. The model constructed using the random forest algorithm demonstrated the best overall performance among six models. The areas under the receiver operating characteristic curve (AUCs) for 5-fold cross-validation, internal validation, and external validations from Shaoxing and Quzhou were 0.711, 0.751, 0.827, and 0.735, respectively. According to the Shapley additive explanation (SHAP) method, C-reactive protein-to-albumin ratio, prognostic nutritional index, and sex were identified as the most important predictors. Our model demonstrates strong predictive capability for assessing IVIG resistance in Kawasaki disease patients.

Establishment and validation of risk prediction model to predict intravenous immunoglobulin-resistance in Kawasaki disease based on meta-analysis of 15 cohorts

BACKGROUND

Pediatric Kawasaki disease (KD) patients showing resistance to intravenous immunoglobulin (IVIG) are at risk of coronary artery lesions; thus, early prediction of IVIG resistance is particularly important. Herein, we aimed to develop and verify a novel predictive risk model for IVIG resistance in KD based on meta-analyses.

METHODS

PubMed, Embase, and Web of Science databases were searched for cohort studies on the risk factors for IVIG resistance from January 2006 to December 2022. Data were extracted from the screened literature, followed by quality assessment using the Newcastle-Ottawa scale. meta-analyses used Stata 17.0 software to extract the risk factors with significant combined effect sizes and combined risk values, followed by logistic regression prediction model construction. The model was prospective validated using data from 1007 pediatric KD cases attending the Children's Hospital of Soochow University. The model's predictive ability was assessed using the Hosmer-Lemeshow test and area under the receiver operating characteristic curve (AUC) and the clinical utility was assessed using decision curve analysis(DCA).

RESULTS

Fifteen cohort studies reporting 4273 patients with IVIG resistance were included. The incidence of IVIG resistance was 16.2%. Six risk factors were reported ≥ 3 times with significant results for the combined effect size: male sex, rash, cervical lymphadenopathy, % neutrophils ≥ 80%, Age ≤ 12 months and platelet count ≤ 300 × 109/L. The logistic scoring model had 83.8% specificity, 70.4% sensitivity, and an optimal cut-off value of 23.500.

CONCLUSION

The risk prediction model for IVIG resistance in KD showed a good predictive performance, and pediatricians should pay high attention to these high-risk patients and develop an appropriate individual regimens to prevent coronary complications.

A machine learning model for distinguishing Kawasaki disease from sepsis

KD is an acute systemic vasculitis that most commonly affects children under 5 years old. Sepsis is a systemic inflammatory response syndrome caused by infection. The main clinical manifestations of both are fever, and laboratory tests include elevated WBC count, C-reactive protein, and procalcitonin. However, the two treatments are very different. Therefore, it is necessary to establish a dynamic nomogram based on clinical data to help clinicians make timely diagnoses and decision-making. In this study, we analyzed 299 KD patients and 309 sepsis patients. We collected patients' age, sex, height, weight, BMI, and 33 biological parameters of a routine blood test. After dividing the patients into a training set and validation set, the least absolute shrinkage and selection operator method, support vector machine and receiver operating characteristic curve were used to select significant factors and construct the nomogram. The performance of the nomogram was evaluated by discrimination and calibration. The decision curve analysis was used to assess the clinical usefulness of the nomogram. This nomogram shows that height, WBC, monocyte, eosinophil, lymphocyte to monocyte count ratio (LMR), PA, GGT and platelet are independent predictors of the KD diagnostic model. The c-index of the nomogram in the training set and validation is 0.926 and 0.878, which describes good discrimination. The nomogram is well calibrated. The decision curve analysis showed that the nomogram has better clinical application value and decision-making assistance ability. The nomogram has good performance of distinguishing KD from sepsis and is helpful for clinical pediatricians to make early clinical decisions.

Identification of key signaling pathways and hub genes related to immune infiltration in Kawasaki disease with resistance to intravenous immunoglobulin based on weighted gene co-expression network analysis

Background: Kawasaki disease (KD) is an acute vasculitis, that is, the leading cause of acquired heart disease in children, with approximately 10%-20% of patients with KD suffering intravenous immunoglobulin (IVIG) resistance. Although the underlying mechanism of this phenomenon remains unclear, recent studies have revealed that immune cell infiltration may associate with its occurrence. Methods: In this study, we downloaded the expression profiles from the GSE48498 and GSE16797 datasets in the Gene Expression Omnibus database, analyzed differentially expressed genes (DEGs), and intersected the DEGs with the immune-related genes downloaded from the ImmPort database to obtain differentially expressed immune-related genes (DEIGs). Then CIBERSORT algorithm was used to calculate the immune cell compositions, followed by the WGCNA analysis to identify the module genes associated with immune cell infiltration. Next, we took the intersection of the selected module genes and DEIGs, then performed GO and KEGG enrichment analysis. Moreover, ROC curve validation, Spearman analysis with immune cells, TF, and miRNA regulation network, and potential drug prediction were implemented for the finally obtained hub genes. Results: The CIBERSORT algorithm showed that neutrophil expression was significantly higher in IVIG-resistant patients compared to IVIG-responsive patients. Next, we got differentially expressed neutrophil-related genes by intersecting DEIGs with neutrophil-related module genes obtained by WGCNA, for further analysis. Enrichment analysis revealed that these genes were associated with immune pathways, such as cytokine-cytokine receptor interaction and neutrophil extracellular trap formation. Then we combined the PPI network in the STRING database with the MCODE plugin in Cytoscape and identified 6 hub genes (TLR8, AQP9, CXCR1, FPR2, HCK, and IL1R2), which had good diagnostic performance in IVIG resistance according to ROC analysis. Furthermore, Spearman's correlation analysis confirmed that these genes were closely related to neutrophils. Finally, TFs, miRNAs, and potential drugs targeting the hub genes were predicted, and TF-, miRNA-, and drug-gene networks were constructed. Conclusion: This study found that the 6 hub genes (TLR8, AQP9, CXCR1, FPR2, HCK, and IL1R2) were significantly associated with neutrophil cell infiltration, which played an important role in IVIG resistance. In a word, this work rendered potential diagnostic biomarkers and prospective therapeutic targets for IVIG-resistant patients.

Lung Ultrasound and Neutrophil Lymphocyte Ratio in Early Diagnosis and Differentiation between Viral and Bacterial Pneumonia in Young Children

Lung ultrasound (LUS) is a crucial diagnostic tool for identifying pneumonia in the pediatric age group. However, it plays a limited role in the early distinction between viral and bacterial pneumonia in children. The objectives of our study were to determine if LUS and the neutrophil-lymphocyte ratio (NLR) were useful in identifying and distinguishing between viral and bacterial pneumonia in Egyptian children under the age of two. Within the first 12 h of being admitted to our department, 52 children with clinical symptoms and signs suggestive of community-acquired pneumonia (CAP) underwent LUS and the NLR. LUS and the NLR strongly differentiated children with viral from those with bacterial pneumonia. For the early diagnosis and differentiation between viral and bacterial pneumonia in young Egyptian children, LUS was proven to be a noninvasive and reliable method. Combining the NLR with LUS increased the diagnostic accuracy when evaluating children suspected of having pneumonia.

Value of blood inflammatory markers for predicting intravenous immunoglobulin resistance in Kawasaki disease: A systematic review and meta-analysis

BACKGROUND

Previous studies have assessed the diagnostic accuracy of blood inflammatory markers like neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and C-reactive protein (CRP), and CRP to albumin ratio (CAR) to predict the resistant Kawasaki disease (KD). The aim of the current meta-analysis and systematic review is to compare the prognostic ability of these inflammatory markers to predict the resistance to IVIG in patients with Kawasaki disease.

METHODS

A systematic search of online academic databases and search engines such as EMBASE, PubMed Central, MEDLINE, Cochrane library, Google Scholar, and ScienceDirect was conducted for papers that report the diagnostic accuracy of inflammatory markers for resistant KD. Meta-analysis was performed using STATA software.

RESULTS

Twenty-two studies met the inclusion criteria. Pooled sensitivity and specificity of NLR as a predictor of resistant Kawasaki disease was 72% (95% CI: 62%, 80%) and 71% (95% CI: 63%, 78%), with AUC of 0.77 for PLR was 60% (95% CI: 50%, 69%) and 68% (95% CI: 61%, 75%), with area under the curve (AUC) of 0.69. Pooled sensitivity and specificity of CRP was 75% (95% CI: 68%, 81%) and 66% (95% CI: 55%, 76%), respectively, with an AUC value of 0.78. Pooled sensitivity and specificity of combined NLR and PLR was 58% (95% CI: 46%, 69%) and 73% (95% CI: 65%, 79%), respectively, with an AUC value of 0.72.

CONCLUSION

Our study found that NLR, CRP, PLR, and combined NLR/PLR have a good prognostic value in patients with resistant Kawasaki disease, with moderate to high sensitivity and specificity. More research on the accuracy of these indexes in multiple combinations is needed.

SYSTEMATIC REVIEW REGISTRATION

[https://www.crd.york.ac.uk/prospero/], identifier [CRD42022322165].

Prognostic value of pretreatment prognostic nutritional index in intravenous immunoglobulin-resistant Kawasaki disease

The aim of the present study was to investigate the potential predictive significance of pretreatment prognostic nutritional index (PNI) in intravenous immunoglobulin (IVIG) resistant Kawasaki disease (KD). The PNI, neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) were analyzed in 1257 eligible patients with KD. Receiver operating curve analysis was used to explore the prediction accuracy for IVIG-resistant KD. The optimal cut-off values were identified as 49.5 for PNI, 3.58 for NLR and 164.00 for PLR, respectively. Lower pretreatment PNI (< 49.5) was demonstrated to be associated with lower age, serum sodium levels and platelet counts, and with a higher incidence of IVIG resistance and higher C-reactive protein levels. There was a significantly negative association between the PNI and NLR, and PLR. Univariate and multivariate analyses revealed that PNI, NLR and PLR were independent predictive factors for IVIG resistance. The discriminatory ability of PNI was not inferior to NLR, PLR and their combination (NLR > 3.58 and PLR > 164) for predicting IVIG resistance, respectively. Pretreatment PNI may serve as a novel surrogate independent predictor for IVIG-resistant KD.

Difference in serum miRNA expression between immunoglobulin-sensitive and -insensitive incomplete Kawasaki disease patients

The present study aimed to investigate the expression of microRNAs (miRNAs/miRs) and inflammatory factors in patients with immunoglobulin-sensitive and IVIG-insensitive incomplete Kawasaki disease (KD). One hundred and eighty-five patients with incomplete KD were included as the study group (KD group), and 182 patients with respiratory infection as the control group. Neutrophil to lymphocyte ratio (NLR), C-reactive protein (CRP) levels, alanine aminotransferase (ALT), aspartate aminotransferase (AST), white blood cell count (WBC), hemoglobin level (Hb), platelet count (PLT) and T cell subsets (CD3⁺, CD3⁺ CD4⁺) were compared. Patients in the KD group received aspirin (30 mg/kg orally daily) and gamma globulin (IVIG, 1 g/kg intravenously daily). According to the sensitivity to IVIG, patients were divided into IVIG-sensitive group and IVIG-insensitive KD group. The relative expression levels of miRNA-21, miRNA-145, miRNA-155 and miRNA-199b-5p in the serum were detected by RT-qPCR. Serum TNF-α, IL-6 and IL-1β levels were assessed using ELISA. Before treatment, the neutrophil to lymphocyte ratio (NLR), levels C-reactive protein, and leukocytes in the KD group were significantly higher compared with the control group (P<0.05). After medical intervention, the relative expression of miRNA-21, miRNA-145 and miRNA-155 in the serum of patients in IVIG-sensitive and IVIG-insensitive KD groups were increased when compared with these levels in the control group (P<0.05). Meanwhile, the relative expression of miRNA-199b-5p was decreased (P<0.05). Compared with the IVIG-sensitive KD group, the relative expression levels of miRNA-145 and miRNA-155 were increased in the serum of patients in the IVIG-insensitive KD group (P<0.05). Compared with the control group, the levels of TNF-α, IL-6 and IL-1β were increased in the serum of patients in the IVIG-sensitive and IVIG-insensitive KD groups (P<0.05). Compared with the IVIG-sensitive KD group, the serum levels of TNF-α and IL-6 were increased in patients of the IVIG-insensitive KD group (P<0.05). Except for NLR and CRP, there were differences in the expression of peripheral blood miRNA-145, miRNA-155 and serum TNF-α and IL-6 in patients with immunoglobulin-sensitive and -insensitive incomplete KD.

Prediction of intravenous immunoglobulin resistance in Kawasaki disease in children

BACKGROUND

We aimed to explore predictive measures for intravenous immunoglobulin (IVIG) resistance in children with Kawasaki disease (KD).

METHODS

Patients diagnosed with KD were enrolled in this study. Univariate analysis and multiple logistic regression were utilized to analyze the clinical features and laboratory results prior to IVIG-treatment of the two groups. Independent predictors of IVIG resistance were analyzed, and a predictive model for KD children with IVIG resistance was constructed.

RESULTS

A total of 277 children with KD, 180 boys and 97 girls, aged 2-128 (median 23) months, were enrolled in the study. Compared with the IVIG-responsive group, the IVIG-resistant group had higher levels of the peripheral neutrophil count, mean platelet volume, mean platelet volume-to-lymphocyte ratio and C-reactive protein, and total serum bilirubin, but lower levels of peripheral lymphocyte count, serum albumin and serum prealbumin. Age (in months), peripheral neutrophil count, lymphocyte count and mean platelet volume and serum albumin were independent indicators for IVIG resistance by multivariate logistic regression analysis. A logistic regression model and a scoring system were set up, where cut-off values of - 0.46 and 6.5 points yielded sensitivities of 83.9% and 77.4%, and specificities of 74.8% and 61.0%, respectively. The areas under the curve (AUC) were 0.808 in the logistic regression model, and 0.750 in the scoring system.

CONCLUSION

Our model for predicting IVIG-resistant children with KD, involving age (months), peripheral neutrophil count, lymphocyte count and mean platelet volume and serum albumin prior to IVIG-treatment, is helpful for clinical prediction of children with IVIG-resistant KD.

The Risk Prediction of Coronary Artery Lesions through the Novel Hematological Z-Values in 4 Chronological Age Subgroups of Kawasaki Disease

Background and Objectives: Most cases of Kawasaki disease (KD) occur between the ages of 6 months and 5 years. Differences in immunological reaction and CAL (coronary artery lesion) by the age subgroups classified according to the prevalence of KD and those particularly in the earlier life of KD should be investigated. Materials and Methods: The laboratory data of 223 infantile and 681 non-infantile KD cases from 2003 to 2018 at Korea University Hospital were retrospectively analyzed. Patients with KD were divided into infants and non-infants and further subdivided into four subgroups by age. The age-adjusted Z-values were compared among the subgroups. Febrile controls were identified as patients with fever for >5 days and who showed some of the KD symptoms. Results: IVIG (intravenous immunoglobulin) resistance at the age of 6 months or less was significantly lower than that at the ages of 7–12 months and 13–60 months (respectively, p < 0.05). The significant risk factors for CAL in total KD patients were age, incomplete KD, post-IVIG fever, IVIG resistance, convalescent Z-eosinophil, and subacute platelet (p < 0.05). The significant risk factors for CAL at the age of 6 months or less were IVIG resistance, acute Z-neutrophil, subacute Z-neutrophil, subacute NLR (neutrophil to lymphocyte ratio), and subacute platelet (respectively, p < 0.05). Conclusion: Younger age and incomplete presentation in KD can be independent risk factors for CAL. The immune reactions of KD at a younger age are more tolerated compared with those at older ages during the acute phase. The immune response at the age of 6 months or less showed immune tolerance in terms of incomplete presentation and IVIG responsiveness. The risk factors such as IVIG resistance, subacute platelet, subacute NLR, and acute or subacute Z-neutrophil at the age of 6 months or less can be very useful parameters to predict CAL in young, incomplete KD.

Neutrophil-to-lymphocyte ratio and scoring system for predicting coronary artery lesions of Kawasaki disease

BACKGROUND

Kawasaki disease (KD) causes coronary artery lesions (CAL) and is the leading cause of acquired heart disease in children. The aim of this study is to evaluate the risk factors and set-up a scoring system for predicting CAL of KD.

METHODS

We retrospectively reviewed a total of 478 patients diagnosed with KD. We compared age, gender, laboratory data, and treatment response in two groups and developed a scoring system for predicting CAL.

RESULTS

During the study period, 365 of these patients had complete medical records of coronary surveys by echocardiography. Anemia, hypoalbuminemia, C reactive protein (CRP), alanine aminotransferase, neutrophil count, and neutrophil/lymphocyte ratio (NLR) showed significant differences with CAL formation. We determined the cut-off value using a receiver-operating-characteristic (ROC) curve, and following multivariate logistic regression analysis, four independent risk factors demonstrated a significant difference with CAL formation, including CRP > 103 mg/L, NLR > 3.5, male gender, and intravenous immunoglobulin (IVIG) resistance. We established a score system based on the above evaluation, for which a ROC curve was performed, and a total score of ≥ 2 points showed a sensitivity of 60.8% and a specificity of 70.6%, with an area under the ROC curve of 0.696.

CONCLUSIONS

Identifying children at risk is important in order to prevent CAL from developing. Four independent risk factors that can predict CAL formation were CRP > 103 mg/L, NLR > 3.5, male gender, and IVIG resistance. This first report incorporated NLR into score systems to predict CAL reinforces previously well-known risk factors for the CAL formation among KD patients.

Neutrophil-to-lymphocyte ratio as a biomarker for predicting the intravenous immunoglobulin-resistant Kawasaki disease

BACKGROUND

In recent years, many studies focused on the association between the neutrophil-to-lymphocyte ratio (NLR) and the risk of intravenous immunoglobulin (IVIG)-resistant Kawasaki disease (rKD), with inconsistent results. Therefore, we aimed to investigate the role of NLR as a biomarker in detecting rKD.

METHODS

We searched PubMed, EMBASE, the Cochrane Central Register of Controlled Trials, and China National Knowledge Infrastructure through May 18th, 2019. Meta-disc 1.4 and STATA 15.1 were used to perform this metaanalysis in a fixed/random-effect model.

RESULTS

A total of 7 relevant studies were eligible to analyze pooled accuracy. The overall performance of NLR detection was: pooled sensitivity, 0.66 (95% confidence interval [CI], 0.63 - 0.70); pooled specificity, 0.71 (95%CI, 0.69 - 0.73); and area under the summary receiver operating characteristic curves value (SROC), 0.7956. The meta-regression analysis showed that the type of samples was the sources of heterogeneity. The subgroup analysis suggested that NLR detection after the initial treatment of IVIG had the largest area under curve of SROC in all the subgroups: pooled sensitivity, 0.58 (95%CI, 0.53 - 0.63); pooled specificity, 0.77 (95%CI, 0.75 - 0.79); and SROC, 0.8440.

CONCLUSIONS

This is the first meta-analysis demonstrated that NLR might be a biomarker for detecting rKD, especially NLR value after the initial treatment of IVIG. More well-designed researches need to be done to launch the application of NLR for predicting rKD in the clinic.

Elevated Levels of Pentraxin 3 Correlate With Neutrophilia and Coronary Artery Dilation During Acute Kawasaki Disease

Kawasaki disease (KD) is the leading cause of acquired pediatric heart disease in the developed world as 25-30% of untreated patients and at least 5% of treated patients will develop irreversible coronary artery lesions (CAL). Pentraxin-3 (PTX-3) has been well-studied in inflammatory diseases, particularly in cardiovascular diseases associated with vascular endothelial dysfunction. We hypothesized that PTX-3 plays an important role in the development of KD-associated CAL and investigated the circulating levels of PTX-3 in the serum of KD patients. Children with acute KD were followed from diagnosis through normalization of the clinical parameters of inflammation (convalescent phase). Serum samples were obtained and echocardiograms were conducted at several phases of the illness: acute [prior to intravenous immunoglobulin (IVIG) treatment], sub-acute (5-10 days after IVIG treatment), and convalescent (1-4 months after KD diagnosis). Seventy children were included in the final cohort of the study, of whom 26 (37%) presented with CAL and 18 (26%) developed IVIG resistance. The patients included in this study came from diverse ethnic backgrounds, mostly with mixed ancestry/ ethnicity. Significantly increased PTX-3 levels were observed during the acute phase of KD compared to the sub-acute and the convalescent phases. The PTX-3 levels during acute KD were significantly higher among KD patients with CAL compared to patients with normal coronary arteries (NCA). Also, the PTX-3 levels were significantly higher in patients with IVIG resistance. Furthermore, the PTX-3 levels were significantly higher in IVIG-resistant KD patients with CAL as compared to the NCA group. Moreover, the PTX-3 levels were significantly correlated to coronary artery z-score during acute KD and to neutrophil counts throughout KD progression regardless of coronary artery z-score. Elevated PTX-3 levels correlated to elevated neutrophil counts, a known source of PTX-3 in acute inflammation and an important player in the development of KD vasculitis. We, therefore, suggest PTX-3 as a novel factor in the development of KD-associated CAL and propose neutrophil-derived PTX-3 as contributing to KD vascular dysfunction.