Japanese scoring systems to predict resistance to intravenous immunoglobulin in Kawasaki disease were unreliable for Caucasian Israeli children

Arane, Karen; Mendelsohn, Kerry; Mimouni, Michael; Mimouni, Francis; Koren, Yael; Simon, Dafna Brik; Bahat, Hilla; Helou, Mona Hanna; Mendelson, Amir; Hezkelo, Nofar; Glatstein, Miguel; Berkun, Yackov; Eisenstein, Eli M.; Butbul, Yonatan; Brik, Riva; Hashkes, Philip J.; Uziel, Yosef; Harel, Liora; Amarilyo, Gil

doi:10.1111/apa.14418

Cited by 27 publications

(22 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Assuming that both research methods were appropriately and strictly designed, we may presume that an unknown ongoing process involved nutrition status, in addition to vascular inflammation. However, early validation research on Japan scoring models yield inconsistent result between different races [2,[22][23][24]. It showed multiple ethnicityexclusive models are required.…”

Section: Pni Predicts Kd With Caa and Ivig Non-respondermentioning

confidence: 99%

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

Tai

Guo

et al. 2020

BMC Pediatr

View full text Add to dashboard Cite

Background: Kawasaki Disease (KD) is considered a major acquired heart disease in children under the age of 5. Coronary artery aneurysm (CAA) can occur in serious cases despite extreme therapy efforts. Previous studies have reported low serum albumin level was associated with disease outcome, but no further investigation was addressed yet. Method: This retrospective (case-control) study randomly included children with KD who were admitted and underwent laboratory tests before undergoing IVIG treatment in this institution, the largest tertiary medical center in southern Taiwan from 2012 to 2016. Prognostic nutrition index (PNI), an albumin-based formula product, was evaluated as a predictor of CAA the first time. The progression of CAA was monitored using serial echocardiography for six months. We performed multivariable logistic regression analysis on the laboratory test and PNI with the disease outcome of the KD patients. Result: Of the 275 children, 149 had CAA, including transient dilatation, while the other 126 did not develop CAA during the 6-month follow-up period. A multivariate logistic regression model revealed that PNI, gender, IVIG nonresponder, and platelet count are significant predictors of CAA with a 95% confidence interval estimator of 1.999, 3.058, 3.864 and 1.004, respectively. Using PNI to predict CAA presence gave an area under the receiver-operatingcharacteristics (ROC) curve of 0.596. For a cutoff of 0.5 in the logistic regression model and the PNI cutoff point is taken as 55 together with IVIG non-responder, boy gender, and platelet count take into account, sensitivity and specificity were 65.7 and 70.4%. Conclusion: PNI could be a candidate of adjunctive predictor of coronary artery aneurysm in addition to IVIG nonresponder. Together with low PNI, IVIG non-responder, male gender and platelet count will give high odds to predict coronary artery aneurysm within 6 months of illness.

show abstract

Section: Pni Predicts Kd With Caa and Ivig Non-respondermentioning

confidence: 99%

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

Tai

Guo

et al. 2020

BMC Pediatr

View full text Add to dashboard Cite

show abstract

“…Assuming that both research methods were appropriately and strictly designed, we may presume that an unknown ongoing process involved nutrition status, in addition to vascular inflammation. However, early validation research on Japan scoring models yield inconsistent result between different races (2,(22)(23)(24). It showed multiple ethnicity-exclusive models are required.…”

Section: Pni Predicts Kd With Caa and Ivig Non-respondermentioning

confidence: 99%

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

Tai

Guo

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: Kawasaki Disease (KD) is considered a major acquired heart disease in children under the age of 5. Coronary artery aneurysm (CAA) can occur in serious cases despite extreme therapy efforts. Previous studies have reported low serum albumin level was associated with disease outcome, but no further investigation was addressed yet. Method: This retrospective (case-control) study randomly included children with KD who were admitted and underwent laboratory tests before undergoing IVIG treatment in this institution, the largest tertiary medical center in southern Taiwan from 2012-2016. Prognostic nutrition index (PNI), an albumin-based formula product, was evaluated as a predictor of CAA the first time. The progression of CAA was monitored using serial echocardiography for six months. We performed multivariable logistic regression analysis on the laboratory test and PNI with the disease outcome of the KD patients. Result: Of the 275 children, 149 had CAA, including transient dilatation, while the other 126 did not develop CAA during the 6-month follow-up period. A multivariate logistic regression model revealed that PNI, gender, IVIG non-responder, and platelet count are significant predictors of CAA with a 95% confidence interval estimator of 1.999, 3.058, 3.864 and1.004, respectively. Using PNI to predict CAA presence gave an area under the receiver-operating-characteristics (ROC) curve of 0.596. For a cutoff of 0.5 in the logistic regression model and the PNI cut-off point is taken as 55 together with IVIG non-responder, boy gender, and platelet count take into account, sensitivity and specificity were 65.7% and 70.4%. Conclusion: PNI could be a candidate of adjunctive predictor of coronary artery aneurysm in addition to IVIG non-responder. Together with low PNI, IVIG non-responder, male gender and platelet count will give high odds to predict coronary artery aneurysm within 6 months of illness.

show abstract

“…Many study groups have evaluated these two well-known scoring models in different regions outside of Japan. Both scoring models have an acceptable validation outcome in East China [ 14 ] but showed a relatively unsatisfactory performance in North America [ 15 ], North China [ 16 , 17 ], Israel [ 18 ] and Italy [ 19 ]. Thus, studies on IVIG resistance prediction have accumulated in recent years in different regions.…”

Section: Introductionmentioning

confidence: 99%

A machine learning approach to predict intravenous immunoglobulin resistance in Kawasaki disease patients: A study based on a Southeast China population

Wang¹,

Liu²,

Lin

2020

PLoS ONE

View full text Add to dashboard Cite

Kawasaki disease is the leading cause of pediatric acquired heart disease. Coronary artery abnormalities are the main complication of Kawasaki disease. Kawasaki disease patients with intravenous immunoglobulin resistance are at a greater risk of developing coronary artery abnormalities. Several scoring models have been established to predict resistance to intravenous immunoglobulin, but clinicians usually do not apply those models in patients because of their poor performance. To find a better model, we retrospectively collected data including 753 observations and 82 variables. A total of 644 observations were included in the analysis, and 124 of the patients observed were intravenous immunoglobulin resistant (19.25%). We considered 7 different linear and nonlinear machine learning algorithms, including logistic regression (L1 and L1 regularized), decision tree, random forest, Ada-Boost, gradient boosting machine (GBM), and lightGBM, to predict the class of intravenous immunoglobulin resistance (binary classification). Data from patients who were discharged before Sep 2018 were included in the training set (n = 497), while all the data collected after 9/1/2018 were included in the test set (n = 147). We used the area under the ROC curve, accuracy, sensitivity, and specificity to evaluate the performances of each model. The gradient GBM had the best performance (area under the ROC curve 0.7423, accuracy 0.8844, sensitivity 0.3043, specificity 0.9919). Additionally, the feature importance was evaluated with SHapley Additive exPlanation (SHAP) values, and the clinical utility was assessed with decision curve analysis. We also compared our model with the Kobayashi score, Egami score, Formosa score and Kawamura score. Our machine learning model outperformed all of the aforementioned four scoring models. Our study demonstrates a novel and robust machine learning method to predict intravenous immunoglobulin resistance in Kawasaki disease patients. We believe this approach could be implemented in an electronic health record system as a form of clinical decision support in the near future.

show abstract

Japanese scoring systems to predict resistance to intravenous immunoglobulin in Kawasaki disease were unreliable for Caucasian Israeli children

Abstract: The established risk scoring methods created for Japanese populations with Kawasaki disease were not suitable for predicting IVIG resistance in Caucasian Israeli children, and we were unable to create a specific scoring method that was able to do this.

Cited by 27 publications

References 22 publications

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

Prognostic nutrition index as a predictor of coronary artery aneurysm in Kawasaki Disease

A machine learning approach to predict intravenous immunoglobulin resistance in Kawasaki disease patients: A study based on a Southeast China population

Contact Info

Product

Resources

About