A Classification Tool for Differentiation of Kawasaki Disease from Other Febrile Illnesses

Abstract

Objective To develop and validate a novel decision tree-based clinical algorithm to differentiate Kawasaki disease (KD) from other pediatric febrile illnesses that share common clinical characteristics. Study design Using clinical and laboratory data from 801 subjects with acute KD (533 for development, and 268 for validation) and 479 febrile control subjects (318 for development, and 161 for validation), we developed a step-wise KD diagnostic algorithm combining our previously developed linear-discriminant-analysis (LDA)-based model with a newly developed, tree-based algorithm. Results The primary model (LDA) stratified the 1,280 subjects into FC (276), indeterminate (247), and KD (757) subgroups. The subsequent model (decision trees) further classified the indeterminate group into FC (103) and KD (58) subgroups, leaving only 29 of 801 (3.6%) KD and 57 of 479 (11.9%) FC subjects indeterminate. The 2-step algorithm had a sensitivity of 96.0% and a specificity of 78.5% and correctly classified all KD subjects who later developed coronary artery aneurysms. Conclusion The addition of a decision tree step increased sensitivity and specificity in KD/FC classification over our previously described LDA model. A multicenter trial is needed to prospectively determine its utility as a point of care diagnostic test for KD.

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center ( 17 ) and multi-center ( 23 ) validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects.…”

“…In this Taiwan single center study (Figure 2), the algorithm's testing performance is in line with our 2016 US single center (17) and US 2020 multi-center (23) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/…”

“…In this Taiwan single center study ( Figure 2 ), the algorithm’s testing performance is in line with our 2016 US single center ( 17 ) and US 2020 multi-center ( 23 ) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center ( 17 ) and multi-center ( 23 ) validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects.…”

“…In this Taiwan single center study (Figure 2), the algorithm's testing performance is in line with our 2016 US single center (17) and US 2020 multi-center (23) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/…”

“…In this Taiwan single center study ( Figure 2 ), the algorithm’s testing performance is in line with our 2016 US single center ( 17 ) and US 2020 multi-center ( 23 ) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Asia

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center [21] and multi-center [31] validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects (Table 6).…”

“…In this Taiwan single center study (Figure 3), the algorithm’s testing performance is in line with our 2016 US single center [21] and US 2020 multi-center [31] analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center ( 17 ) and multi-center ( 23 ) validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects.…”

“…In this Taiwan single center study (Figure 2), the algorithm's testing performance is in line with our 2016 US single center (17) and US 2020 multi-center (23) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/…”

“…In this Taiwan single center study ( Figure 2 ), the algorithm’s testing performance is in line with our 2016 US single center ( 17 ) and US 2020 multi-center ( 23 ) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Asia

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center [21] and multi-center [31] validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects (Table 6).…”

“…In this Taiwan single center study (Figure 3), the algorithm’s testing performance is in line with our 2016 US single center [21] and US 2020 multi-center [31] analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center ( 17 ) and multi-center ( 23 ) validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects.…”

“…In this Taiwan single center study (Figure 2), the algorithm's testing performance is in line with our 2016 US single center (17) and US 2020 multi-center (23) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/…”

“…In this Taiwan single center study ( Figure 2 ), the algorithm’s testing performance is in line with our 2016 US single center ( 17 ) and US 2020 multi-center ( 23 ) analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Asia

“…The frequencies of the five KD principal criteria and illness days in this Taiwan single-center study were in line with those in the US single-center [21] and multi-center [31] validation dataset of the algorithm except for cervical lymph node (25.5%/19.5% versus 49.4%), conjunctival injection (44.1%/58.9% versus 19.7%), oropharyngeal changes (48.4%/46.8% versus 86.1%), and days of illness (5.0/5.0 versus 2.0) in FC subjects (Table 6).…”

“…In this Taiwan single center study (Figure 3), the algorithm’s testing performance is in line with our 2016 US single center [21] and US 2020 multi-center [31] analyses: sensitivity of 90.7% (379/418), specificity of 86.1% (223/259), PPV of 92.7% (379/409), and NPV of 90.3% (223/247); 3.5% (15/418) of the patients with KD and 2.3% (6/259) FCs were classified as indeterminate. Overall, 5.7% (24/418) of the patients with KD and 11.5% (30/259) of FCs were misclassified by the algorithm.…”