Background The clinical features of Kawasaki disease (KD) overlap with those of other paediatric febrile illnesses. A missed or delayed diagnosis increases the risk of coronary artery damage. Our computer algorithm for KD and febrile illness differentiation had a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 94.8%, 70.8%, 93.7% and 98.3%, respectively, in a single-centre validation study. We sought to determine the performance of this algorithm with febrile children from multiple institutions across the USA.
Methods We used our previously published 18-variable panel that includes illness day, the five KD clinical criteria and readily available laboratory values. We applied this two-step algorithm using a linear discriminant analysis-based clinical model followed by a random forest-based algorithm to a cohort of 1059 acute KD and 282 febrile control patients from five children’s hospitals across the USA.
Results The algorithm correctly classified 970 of 1059 patients with KD and 163 of 282 febrile controls resulting in a sensitivity of 91.6%, specificity of 57.8% and PPV and NPV of 95.4% and 93.1%, respectively. The algorithm also correctly identified 218 of the 232 KD patients (94.0%) with abnormal echocardiograms.
Interpretation The expectation is that the predictive accuracy of the algorithm will be reduced in a real-world setting in which patients with KD are rare and febrile controls are common. However, the results of the current analysis suggest that this algorithm warrants a prospective, multicentre study to evaluate its potential utility as a physician support tool.
- Kawasaki disease
- febrile controls
- multicenter study
Statistics from Altmetric.com
Correction notice This paper has been amended since it was published online. We have corrected the spelling of author Jane W Newburger.
Collaborators Pediatric Emergency Medicine Kawasaki Disease Research Group (Pediatrics, University of California San Diego, La Jolla, California, USA; Rady Children's Hospital, San Diego, California, USA): members include Amy Bryl, J Joelle Donofrio, Arit Edwin-Enyenihi, Michael Gardiner, Jim R Harley, Simon J Lucio, Margaret Nguyen, Kristy Schwartz, Seema Shah and Stacey Ulrich.
Contributors SH, XBL, AHT and JCB designed the study. DBM, JB and HJC supervised the study. JTK, EB, SRD, HH, P-NJ, MSA, PJ, AB, MBS, JWN, NA, AHT and Pediatric Emergency Medicine Kawasaki Disease Research Group managed the data collection. SH, EB, and JCW did the data analyses and result interpretation. SH and AHT wrote the paper. All the authors critically reviewed the draft of the paper before submission.
Funding This work was funded in part by a grant from the Gordon and Marilyn Macklin Foundation (awarded to JB and AHT). Data for this study were collected at the coauthors’ home institutions and match their current affiliation, except for PJ where data were collected from Nationwide Children’s Hospital in Columbus, Ohio.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval The study was approved by the institutional review board at each centre.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available on reasonable request. For any request regarding the data access, please contact Dr Adriana H Tremoulet (tel: (858) 246-0012; fax: (858) 246-0156; email: email@example.com).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.