Data-driven clustering identifies features distinguishing multisystem inflammatory syndrome from acute COVID-19 in children and adolescents

Abstract

Background Multisystem inflammatory syndrome in children (MIS-C) consensus criteria were designed for maximal sensitivity and therefore capture patients with acute COVID-19 pneumonia. Methods We performed unsupervised clustering on data from 1,526 patients (684 labeled MIS-C by clinicians) <21 years old hospitalized with COVID-19-related illness admitted between 15 March 2020 and 31 December 2020. We compared prevalence of assigned MIS-C labels and clinical features among clusters, followed by recursive feature elimination to identify characteristics of potentially misclassified MIS-C-labeled patients. Findings Of 94 clinical features tested, 46 were retained for clustering. Cluster 1 patients ( N = 498; 92% labeled MIS-C) were mostly previously healthy (71%), with mean age 7·2 ± 0·4 years, predominant cardiovascular (77%) and/or mucocutaneous (82%) involvement, high inflammatory biomarkers, and mostly SARS-CoV-2 PCR negative (60%). Cluster 2 patients ( N = 445; 27% labeled MIS-C) frequently had pre-existing conditions (79%, with 39% respiratory), were similarly 7·4 ± 2·1 years old, and commonly had chest radiograph infiltrates (79%) and positive PCR testing (90%). Cluster 3 patients ( N = 583; 19% labeled MIS-C) were younger (2·8 ± 2·0 y), PCR positive (86%), with less inflammation. Radiographic findings of pulmonary infiltrates and positive SARS-CoV-2 PCR accurately distinguished cluster 2 MIS-C labeled patients from cluster 1 patients. Interpretation Using a data driven, unsupervised approach, we identified features that cluster patients into a group with high likelihood of having MIS-C. Other features identified a cluster of patients more likely to have acute severe COVID-19 pulmonary disease, and patients in this cluster labeled by clinicians as MIS-C may be misclassified. These data driven phenotypes may help refine the diagnosis of MIS-C.

Association between Coagulation Profile and Clinical Outcome in Children with SARS-CoV-2 Infection or MIS-C: A Multicenter Cross-Sectional Study

“…As expected, the diagnosis of MIS-C was certainly the most relevant risk factor (OR 27.85 with CI 9.91–78.27). The decision to involve patients with two different clinical manifestations of infection (acute SARS-CoV-2 infection and MIS-C) in the analysis stems from the fact that often, especially in the early stages of the disease, the diagnosis of MIS-C may not be immediate and, more importantly, an overlap between severe COVID-19 and MIS-C has been highlighted and their distinction is not always easy, particularly in those children with a positive nasopharyngeal PCR [ 24 ]. For this reason, evaluating all patients with a documented SARS-CoV-2 infection, regardless of a confirmed diagnosis of MIS-C, allows us to better define the risk factors and alert practitioners managing children from the moment the infection is detected.…”

Frequency, Characteristics and Complications of COVID-19 in Hospitalized Infants

“…Patients whose data were included in other reports are listed in the Table, Supplemental Digital Content 5, http://links.lww.com/INF/E610 . 4 , 6 , 9 …”

The Multifaceted Presentation of the Multisystem Inflammatory Syndrome in Children: Data from a Cluster Analysis

“…The spectrum of the disease has become a subject of increasing interest, as patients with MIS-C express a phenotype similar to many diseases, such as Kawasaki disease (KD), macrophage-activating syndrome (MAS), and septic shock. Most recently, Geva et al [ 5 ] performed a data-driven cluster analysis (CA) to identify the subphenotypes of SARS-CoV-2 during childhood and showed that three clusters of disease were as follows: cluster 1 (previously healthy individuals with a mean age of 7.2 years, presenting with mainly cardiovascular and/or mucocutaneous features and negative for nasopharynx SARS-CoV-2 PCR); cluster 2 (individuals with frequent respiratory findings and positive PCR test); cluster 3 (individuals with younger age (mean 2.8), positive PCR test and less inflammation). According to their data, patients with pulmonary findings and positive SARS-CoV-2 PCR test were labeled as cluster 2, while MIS-C patients were distinguished as a separate subgroup (cluster 1) [ 5 ].…”