Interim Analysis of Acute Hepatitis of Unknown Etiology in Children Aged <10 Years

United States, October 2021-June 2022

Jordan Cates, PhD; Julia M. Baker, PhD; Olivia Almendares, MSPH; Anita K. Kambhampati, MPH; Rachel M. Burke, PhD; Neha Balachandran, MBBS; Eleanor Burnett, MPH; Caelin C. Potts, PhD; Sarah Reagan-Steiner, MD; Hannah L. Kirking, MD; David Sugerman, MD; Umesh D. Parashar, MD, MBBS; Jacqueline E. Tate, PhD


Morbidity and Mortality Weekly Report. 2022;71(26):852-858. 

In This Article

Abstract and Introduction


On April 21, 2022, CDC issued a health advisory encouraging U.S. clinicians to report all patients aged <10 years with hepatitis of unknown etiology to public health authorities, after identification of similar cases in both the United States[1] and Europe.§ A high proportion of initially reported patients had adenovirus detected in whole blood specimens, thus the health advisory encouraged clinicians to consider requesting adenovirus testing, preferentially on whole blood specimens. For patients meeting the criteria in the health advisory (patients under investigation [PUIs]), jurisdictional public health authorities abstracted medical charts and interviewed patient caregivers. As of June 15, 2022, a total of 296 PUIs with hepatitis onset on or after October 1, 2021, were reported from 42 U.S. jurisdictions. The median age of PUIs was 2 years, 2 months. Most PUIs were hospitalized (89.9%); 18 (6.1%) required a liver transplant, and 11 (3.7%) died. Adenovirus was detected in a respiratory, blood, or stool specimen of 100 (44.6%) of 224 patients. Current or past infection with SARS-CoV-2 (the virus that causes COVID-19) was reported in 10 of 98 (10.2%) and 32 of 123 (26.0%) patients, respectively. No common exposures (e.g., travel, food, or toxicants) were identified. This nationwide investigation is ongoing. Further clinical data are needed to understand the cause of hepatitis in these patients and to assess the potential association with adenovirus.

Clinicians and health departments began retrospectively and prospectively identifying PUIs on April 21, 2022. A PUI was defined as a person aged <10 years with elevated (>500 U/L) aspartate aminotransferase (AST) or alanine aminotransferase (ALT), an unknown etiology for the hepatitis, and onset on or after October 1, 2021. Comprehensive investigations of PUIs included rapid reporting of preliminary information, medical chart abstractions, caregiver interviews, laboratory testing, and tissue specimen examination. Upon identification of a PUI, jurisdictional health departments sent preliminary information (basic demographic information, date of hepatitis diagnosis, adenovirus testing results, and patient outcome) to CDC. Medical chart abstraction used standardized forms to collect information on demographic characteristics, signs and symptoms of illness, underlying health conditions, laboratory results (pathogen testing, biomarkers, and toxicology), radiologic findings, tissue pathology findings, vaccination history, and diagnoses and treatment received. Patient caregiver interviews collected information on demographic characteristics, household structure, symptoms, health care use, medical and medication history, and potential exposures (e.g., close contacts, diet, and toxicants). Adenovirus nucleic acid amplification testing (e.g., polymerase chain reaction [PCR]) of blood, respiratory, or stool specimens or rectal swabs was requested at the discretion of the treating clinician and conducted at a diagnostic or reference laboratory.** Available specimens that yielded positive results for adenovirus were further characterized using Sanger sequencing of the six hypervariable regions of the hexon gene to determine adenovirus type.[2] Formalin-fixed, paraffin-embedded (FFPE) liver biopsy, explant, or autopsy tissue specimens underwent routine evaluation at the clinical institutions, and residual FFPE tissue specimens were submitted to CDC for additional pathologic evaluation and diagnostic testing.[3,4] This investigation was reviewed by CDC and conducted consistent with applicable federal law and CDC policy.†† Data were managed using REDCap electronic data tools hosted at CDC,§§ and SAS (version 9.4; SAS Institute) was used to conduct all analyses.

As of June 15, 2022, a total of 296 PUIs were reported from 42 U.S. jurisdictions. There was no apparent temporal clustering of hepatitis diagnoses among these children, although a peak in diagnoses coincided with the release of the health advisory (Figure 1). The median age at time of illness was 2 years, 2 months (range = 1 month–9 years, 8 months), and the largest percentage of PUIs (37.8%, 112) were Hispanic or Latino children, followed by White, non-Hispanic children (32.4%, 96) (Table). Among all reported PUIs, 266 (89.9%) required hospitalization, 18 (6.1%) required a liver transplant, and 11 (3.7%) died. Preliminary reports indicated that among 224 PUIs receiving adenovirus testing, 100 (44.6%) had a positive result in any specimen type, including 31 of 71 (43.7%) who underwent testing of whole blood (Figure 1).

Figure 1.

Patients under investigation for pediatric hepatitis of unknown etiology* reported to CDC (N = 296), by week of hepatitis presentation and stratified by results of preliminary adenovirus testing using whole blood — United States, October 2021–June 2022

Data from 123 PUIs with medical chart abstraction and interviews were available for detailed analyses; completion of data collection is pending for the remaining 173 PUIs. Compared with all reported PUIs, those with completed medical chart abstraction and interview data were similar demographically, by date of hepatitis diagnosis and by percentage of positive adenovirus test results. Systemic and gastrointestinal signs and symptoms (86.2% and 87.8%, respectively) were common and included vomiting (61.8%), fatigue (55.3%), and jaundice (57.7%) (Table). The median interval between symptom onset and clinical evaluation was 4 days (IQR = 2–9 days). The median peak AST and ALT levels were 2,254.5 U/L (IQR = 802.5–4,266.5) and 1,744.5 U/L (IQR = 710.5–3,358.5), respectively. Thirty-seven (30.1%) patients received a diagnosis of acute liver failure¶¶ and 15 (12.2%) had hepatic encephalopathy. The clinical assessments of four PUIs were consistent with potential autoimmune hepatitis based on liver biopsy results or other laboratory testing.

Medical records of the 123 PUIs with available chart abstractions and interviews indicated testing for and identification of a range of pathogens; adenovirus was detected most frequently (Figure 2). Among PUIs with adenovirus test results from any specimen type, 49.5% (48 of 97) received a positive test result (Table). Adenovirus was detected in 37.8% (14 of 37) of whole blood specimens, 34.4% (11 of 32) of plasma specimens, 30.0% (12 of 40) of stool specimens, and 30.1% (22 of 73) of respiratory specimens. Typing was attempted for 13 specimens, six of which were species F (type 41), one was species C (type C1); six could not be typed.*** Overall, 98 (79.7%) PUIs with available chart data received testing for current SARS-CoV-2 infection, 10 (10.2%) which received a positive test result. History of SARS-CoV-2 infection, based on documentation in the medical chart, antibody testing, or parental report, was reported for 32 (26.0%) patients. The median interval from prior SARS-CoV-2 infection to hepatitis diagnosis was 133 days (IQR = 77–283; nine with unknown date of prior infection). Five (4.1%) patients had received at least 1 dose of a COVID-19 vaccine. Other commonly detected pathogens included rhinovirus/enterovirus (24.5%, 24 of 98 tested), acute Epstein-Barr virus††† (11.4%, nine of 79), and rotavirus (14.0%, six of 43). Adenovirus and SARS-CoV-2 were co-detected in 3.5% (three of 86) of patients receiving testing.

Figure 2.

Pathogens*,†,§,¶ detected during illness among a subset of patients under investigation for pediatric hepatitis of unknown etiology with completed medical chart abstraction and parental interviews (N = 123) — United States, October 2021–June 2022
Abbreviations: CMV = cytomegalovirus; EA = early antigen; EBV = Epstein-Barr virus; HHV = human herpesvirus; IgG = immunoglobulin G; IgM = immunoglobulin M; PCR = polymerase chain reaction; RSV = respiratory syncytial virus; RT-PCR = reverse transcription-polymerase chain reaction; VCA = viral capsid antigen.
*Adenovirus test results: positive = adenovirus detected in any specimen type; negative = all tested specimens negative.
Current SARS-CoV-2 detection: positive SARS-CoV-2 RT-PCR or antigen test result during current illness.
§Acute EBV: positive EBV VCA IgM or EA IgG test result, or diagnosis of primary EBV in the medical chart.
Acute CMV: based on clinical diagnosis, verified by PCR/IgM test result.

Among 36 PUIs for whom information on pathologic evaluation of liver biopsy, explant, or autopsy tissues was available, 25 (65.8%) had evidence of active or acute hepatitis, and none had viral inclusions. As previously reported, liver biopsies from six patients with adenovirus infection had no immunohistochemical evidence of adenovirus and no viral particles identified by electron microscopy.[1]

Approximately one third of patients were the only child in the household aged <10 years. Fewer than one half (42.5%) of children attended a child care facility or school in the month before becoming ill, and 56.1% had never attended a child care facility or school. At present, no exposures (e.g., travel, food, or toxicants) were common among the PUIs, and no epidemiologic links were identified among PUIs in preliminary analyses.

*These authors contributed equally to this report.
Adenovirus-positive results on respiratory, blood, or stool specimen types but excluded PUIs with pending or unknown test results (test results might not have been available at the time of data collection).
††45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.
¶¶Acute liver failure (ALF) was based on the number of patients reported to have ALF in their medical chart. Twenty-four (64.9%) of the 37 patients were confirmed to meet the clinical definition for ALF based on laboratory markers and hepatic encephalopathy diagnosis (AST >500 U/L or ALT >500 U/L and either international normalized ratio (INR) >1.5 with hepatic encephalopathy or INR >2 without hepatic encephalopathy.
***Adenovirus type C1 was identified in a nasopharyngeal swab. Specimens reported as "could not be typed" were those in which sequencing was attempted, and insufficient sequencing information was obtained to identify the adenovirus type.
†††Acute Epstein-Barr virus (EBV) infection was defined as a positive EBV viral capsid antigen immunoglobulin (Ig) M or early antigen IgG test result, or diagnosis of primary EBV infection in the medical chart.