Hospitalizations for Multisystem Inflammatory Syndrome in Children (MIS-C) in US Children’s Hospitals in 2023 vs 2021

Abstract

This cross-sectional study evaluates hospitalizations for multisystem inflammatory syndrome in children in US pediatric hospitals in 2023 vs 2021.

Multisystem inflammatory syndrome in children (MIS-C) is a rare, serious condition characterized by fever and multiorgan involvement that occurs after SARS-CoV-2 infection. US Centers for Disease Control and Prevention (CDC) surveillance data demonstrate that cases peaked in January 2021, with more than 200 cases weekly. Only 117 cases were reported to the CDC with illness onset during 2023, reflecting decreased incidence and likely incomplete reporting. As the International Classification of Diseases, Version 10 (ICD-10), diagnosis code for MIS-C was validated, administrative databases may complement traditional surveillance efforts. We describe MIS-C hospitalizations identified in the Pediatric Health Information System (PHIS; Children’s Hospital Association) with admission in 2023 and compare to cases during peak incidence in 2021.

Methods

PHIS is an administrative database containing encounter-level data from 47 US tertiary care children’s hospitals. We identified hospitalizations in PHIS among patients aged 30 days to 21 years in 2021 and 2023 with an MIS-C diagnosis code (M35.81), excluding those with oncologic or transplant diagnoses, using validated methods. We further excluded hospitalizations without evidence of treatment for MIS-C (intravenous immunoglobulin, steroids, or a biologic agent [anakinra, infliximab, or tocilizumab]) and excluded repeat hospitalizations. We compared characteristics of patients hospitalized with MIS-C between years using χ² tests. Clinical characteristics included presence of a Kawasaki disease diagnosis code (KD codiagnosis), intensive care unit (ICU) admission, shock, extracorporeal membrane oxygenation, and death, identified using previously described methods. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study was deemed exempt from review based on policies of the Cincinnati Children's Hospital Medical Center institutional review board.

Results

We identified 221 patients with MIS-C hospitalizations in 2023 (127 [57.5%] male and 94 [42.5%] female). Monthly hospitalizations are displayed in the Figure. The median (IQR) age was 6 (3-11) years (Table). A KD codiagnosis was identified in 67. Almost half received ICU care (98), 82 had treatment for shock, 6 required extracorporeal membrane oxygenation, and 5 died.

Figure. Number of Multisystem Inflammatory Syndrome in Children (MIS-C) Hospitalizations in the Pediatric Health Information System (PHIS) by Month of Admission.

Includes hospitalizations in the PHIS database among pediatric patients with a diagnosis of MIS-C and no oncologic or transplant diagnoses with evidence of treatment for MIS-C, including intravenous immunoglobulin, steroids, or a biologic agent (anakinra, infliximab, or tocilizumab). Hospitalizations were reported from 45 to 47 hospitals per month and were categorized by date of admission. Six repeat hospitalizations were identified and are not displayed.

Table. Sociodemographic and Clinical Characteristics of Pediatric Patients Hospitalized With Multisystem Inflammatory Syndrome in Children (MIS-C) in the Pediatric Health Information System (PHIS), 2021 and 2023.

Characteristic	No. (%)		P value
	2021	2023
Patients hospitalized with MIS-C, No.a	3578	221	NA
Age, median (IQR), y	9 (5-12)	6 (3-11)	<.001
Age group
<1 y	68 (1.9)	11 (5)	<.001
1-4 y	707 (19.8)	69 (31.2)
5-11 y	1749 (48.9)	91 (41.2)
12-15 y	727 (20.3)	25 (11.3)
16-20 y	327 (9.1)	25 (11.3)
Sex
Male	2187 (61.1)	127 (57.5)	.28
Female	1390 (38.9)	94 (42.5)
Race and ethnicityb
Asian	88 (2.5)	17 (7.7)
Non-Hispanic Black	1117 (31.2)	52 (23.5)
Hispanic	818 (22.9)	59 (26.7)
Non-Hispanic White	1273 (35.6)	77 (34.8)	<.001
Otherc	282 (7.9)	16 (7.2)
Payor
Government	1932 (54)	111 (50.2)	.21
Private	1423 (39.8)	100 (45.2)
Other	223 (6.2)	10 (4.5)
Geographic region
Midwest	775 (21.7)	47 (21.3)	.004
Northeast	292 (8.2)	23 (10.4)
South	1798 (50.3)	88 (39.8)
West	713 (19.9)	63 (28.5)
Complex chronic conditions
0	1586 (44.3)	61 (27.6)	<.001
1-2	1837 (51.3)	133 (60.2)
≥3	155 (4.3)	27 (12.2)
Kawasaki disease codiagnosis	304 (8.5)	67 (30.3)	<.001
Intensive care unit admission	1939 (54.2)	98 (44.3)	.004
Treatment for shock	1435 (40.1)	82 (37.1)	.38
Receipt of ECMO	43 (1.2)	6 (2.7)	.05
Death	22 (0.6)	5 (2.3)	.005

Abbreviations: ECMO, extracorporeal membrane oxygenation; NA, not applicable.

^a Includes pediatric patients with hospitalizations in the PHIS database with a diagnosis of MIS-C and no oncologic or transplant diagnoses who received treatment for MIS-C, including intravenous immunoglobulin, steroids, or a biologic agent (anakinra, infliximab, or tocilizumab).

^b Race and ethnicity data in PHIS are submitted by each participating hospital for each encounter and are collected using hospital-specific practices. Race and ethnicity data were analyzed as social constructs and included because of known racial and ethnic disparities in infections and complications from COVID-19.

^c Other race and ethnicity groups included all categories submitted by participating hospitals that were not Asian, Non-Hispanic Black, Hispanic, or Non-Hispanic White, and were consolidated because of small numbers.

In comparison, we identified 3578 patients with MIS-C hospitalizations in 2021. Patients were younger in 2023 (median [IQR] age, 6 [3-11] vs 9 [5-12] years; P < .001). More patients had a codiagnosis of KD in 2023 (67 of 221 [30.3%] vs 304 of 3578 [8.5%]; P < .001). Similar proportions were treated for shock. A higher proportion died in 2023 (5 of 221 [2.3%]) compared with 2021 (22 of 3578 [0.6%]) (P = .005).

Discussion

In this cross-sectional study, we identified 221 patients hospitalized with MIS-C in PHIS in 2023, with some differences compared to those hospitalized in 2021. Compared to hospitalizations in 2021, those in 2023 were younger, which was also described by the CDC. We found a higher proportion of deaths among recent hospitalizations, which warrants further investigation. We also note a higher proportion of KD codiagnoses over time. Increasing KD codiagnoses may reflect increasing diagnostic uncertainty or clinical overlap in a population of younger patients with evidence of SARS-CoV-2 antibodies from prior infection or immunization.

The characteristics of patients we identified in PHIS appear similar to the 117 individuals reported to the CDC from 2023: 58 (49.6%) with ICU care and 40 (34.2%) with shock, and 3 (2.6%) died. Similarities in demographic and clinical characteristics suggest that similar cohorts were identified. The higher number in PHIS could reflect identification of cases that were not reported to the CDC or differences between clinical diagnoses and surveillance case definitions, especially since the 2023 case definition considers KD an alternative diagnosis.

As with all administrative data analyses that lack the ability to adjudicate diagnoses, the potential for misclassification exists. However, this risk is mitigated by using a previously validated ICD-10 code and limiting the cohort to those with treatment. Additionally, these data are from 47 children’s hospitals that may not be nationally representative. We identified more patients with MIS-C hospitalizations in PHIS than what was reported to the CDC in 2023, but we were unable to compare patient-level data between the 2 cohorts.

In conclusion, MIS-C continues to occur, although rarely, with severe outcomes. Administrative databases, like PHIS, can complement ongoing surveillance for MIS-C and potentially other diseases that are treated in children’s hospitals.

Supplement.

Data sharing statement