TY - JOUR
T1 - Data-driven clustering identifies features distinguishing multisystem inflammatory syndrome from acute COVID-19 in children and adolescents
AU - Overcoming COVID-19 Investigators
AU - Geva, Alon
AU - Patel, Manish M.
AU - Newhams, Margaret M.
AU - Young, Cameron C.
AU - Son, Mary Beth F.
AU - Kong, Michele
AU - Maddux, Aline B.
AU - Hall, Mark W.
AU - Riggs, Becky J.
AU - Singh, Aalok R.
AU - Giuliano, John S.
AU - Hobbs, Charlotte V.
AU - Loftis, Laura L.
AU - McLaughlin, Gwenn E.
AU - Schwartz, Stephanie P.
AU - Schuster, Jennifer E.
AU - Babbitt, Christopher J.
AU - Halasa, Natasha B.
AU - Gertz, Shira J.
AU - Doymaz, Sule
AU - Hume, Janet R.
AU - Bradford, Tamara T.
AU - Irby, Katherine
AU - Carroll, Christopher L.
AU - McGuire, John K.
AU - Tarquinio, Keiko M.
AU - Rowan, Courtney M.
AU - Mack, Elizabeth H.
AU - Cvijanovich, Natalie Z.
AU - Fitzgerald, Julie C.
AU - Spinella, Philip C.
AU - Staat, Mary A.
AU - Clouser, Katharine N.
AU - Soma, Vijaya L.
AU - Dapul, Heda
AU - Maamari, Mia
AU - Bowens, Cindy
AU - Havlin, Kevin M.
AU - Mourani, Peter M.
AU - Heidemann, Sabrina M.
AU - Horwitz, Steven M.
AU - Feldstein, Leora R.
AU - Tenforde, Mark W.
AU - Newburger, Jane W.
AU - Mandl, Kenneth D.
AU - Randolph, Adrienne G.
N1 - Publisher Copyright:
© 2021 The Authors
PY - 2021/10
Y1 - 2021/10
N2 - 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.
AB - 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.
KW - COVID-19
KW - Clustering
KW - Critical care medicine
KW - Multisystem inflammatory syndrome
KW - Pediatrics
UR - http://www.scopus.com/inward/record.url?scp=85115439485&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85115439485&partnerID=8YFLogxK
U2 - 10.1016/j.eclinm.2021.101112
DO - 10.1016/j.eclinm.2021.101112
M3 - Article
C2 - 34485878
AN - SCOPUS:85115439485
SN - 2589-0042
VL - 40
JO - iScience
JF - iScience
M1 - 101112
ER -