A Data Science Approach to Identify and Manage Multisystem Inflammatory Syndrome in Children (MIS-C) Associated with SARS-CoV-2 Infection and Kawasaki Disease in Pediatric Patients

Background: The primary objective of this study was to design and validate a predictive decision support system for the identification, treatment, and management of SARS-CoV-2 associated with multisystem inflammatory syndrome in children (MIS-C), specifically, Kawasaki Disease (KD).

Materials/Methods: Clinical questionnaires/surveys, data gathered from electronic medical records, and interviews of family history were conducted as part of a prospective, multicenter registry of pediatric patients with KD, MIS-C, and/or confirmed/suspected COVID-19 infection. The study adapted and retrained machine learning algorithms that had previously been trained in patients with KD, a pediatric inflammatory vasculopathy with clinical overlap with MIS-C but different etiology. This study was performed in collaboration with the International Kawasaki Disease Registry (IKDR) consortium.

Outcome/Impact: Initial observations suggested shared clinical features with KD and potential cardiac complications in pediatric patients. Severe systemic multi-system inflammatory syndrome with some overlap of clinical features with KD may be prevalent and potentially a delayed pathologic immunologic response that manifests after the infection itself has resolved. Additional research is needed to detect and characterize the cardiac complications related to manifestations of COVID-19, determine associated factors amenable to intervention, and determine long-term outcomes.