As of June 15, 2022, the Centers for Disease Control and Prevention has reported 296 pediatric patients under investigation for hepatitis of unknown etiology in the United States; the World Health Organization has reported 650 probable cases worldwide. One of the leading hypotheses for this cluster of cases is adenovirus, a virus that commonly causes respiratory or gastrointestinal symptoms in healthy children but rarely causes severe hepatitis or acute liver failure in immunocompetent children. The other leading hypothesis is that prior infection with SARS-CoV-2 may predispose children to developing liver injury from a normally innocuous agent. We describe a case of a previously healthy child presenting with acute liver failure who had detectable adenovirus DNA in his stool, whole blood, and in liver explant tissue, suggesting adenovirus as the likely etiology for the liver failure. He had no evidence of prior or current SARS-CoV-2 infection, nor had he received COVID vaccination, suggesting that SARS-CoV-2 did not play a role. Additionally, we report on the ability to provide rapid evaluation of a living donor within 72 hours and successfully perform a lifesaving, left-lobe, living donor liver transplant.
On April 8, 2022, the World Health Organization (WHO) reported 74 cases of severe acute hepatitis (defined by aspartate transaminase or alanine transaminase level higher than 500 U/L) of unknown etiology in previously healthy children younger than aged 10 years.1 Over the ensuing 2 months, WHO reported the count had increased to 650 probable cases of acute hepatitis (including retrospective reports of cases dating to October 2021) in children from 33 countries, with an additional 99 cases pending classification as of May 27, 2022.2 Among the 650 children, adenoviral infection was investigated in 181 cases, 110 of whom had positive results by at least 1 specimen type (61%), making adenovirus, particularly type 41F, a leading hypothesis for the etiology of pediatric hepatitis cases.2,3 Adenovirus 41 is a common virus that causes illnesses characterized primarily by gastrointestinal or respiratory signs and symptoms, but before the WHO case compilation had not been reported to have caused severe hepatitis or acute liver failure (ALF) in healthy children. Interestingly, despite the severe liver injury, none of the children with ALF who had liver tissue sampled had positive immunohistochemical evidence of adenovirus. In the minority of children who also had adenovirus DNA testing performed by polymerase chain reaction (PCR) in the liver tissue, only a few cases were positive.4–7 It is unclear what triggered this new wave of ALF in children, but its timing suggested to some investigators that prior SARS-CoV-2 infections may have predisposed the children to liver pathology caused by a normally innocuous agent.8
In this novel case report, we describe a 2-year-old previously healthy boy who developed ALF and required liver transplantation (LT). In addition to detectable adenovirus DNA in stool and whole blood pretransplant, his explanted liver tissue was positive by PCR for adenovirus DNA. He had no evidence of prior SARS-CoV-2 infection, nor did he receive any COVID-19 vaccines. These findings provide insight into etiology and pathophysiology of ALF.
A previously healthy 2-year-old male presented with 2 weeks of diarrhea followed by acute onset of jaundice in May 2022. He developed fatigue, jaundice, and icterus 2 days before presentation and was noted to have severe cholestatic hepatitis with aspartate transaminase 6233 U/L, alanine transaminase 6286 U/L, total bilirubin 14.0 mg/dL, and conjugated bilirubin 9.7 mg/dL. At presentation, he had hypoglycemia (blood glucose, 55 mg/dL), mild coagulopathy (international normalized ratio [INR] 1.5), and normal mental status. Over the next 5 days, he developed worsening coagulopathy (INR >5), rising total and direct bilirubin (peak, 19.7 mg/dL and 11.9 mg/dL), increasing glucose infusion requirements, and grade 2 encephalopathy, consistent with progressive ALF.
He underwent an extensive evaluation for causes of ALF. His infectious evaluation was notable for stool positive for adenovirus 40/41 DNA by a multiplex PCR panel, whole blood positive for adenovirus by PCR on hospital days 1 and 7 (but negative on a simultaneous plasma specimen on both of those days), and an otherwise negative infectious evaluation (including blood testing for hepatitis A/B/C, Epstein-Barr virus, cytomegalovirus, and herpes simplex virus). The whole blood specimen was qualitatively positive for adenovirus, but the viral load was not quantifiable. He had no evidence of current or previous SARS-CoV-2 infection, including a negative nasopharyngeal swab for SARS-CoV-2 by PCR on presentation, no previous COVID-19–like illness, negative nucleocapsid and spike antibodies, and negative cell-mediated immune responses to SARS-CoV-2 spike and non-spike overlapping peptide pools as previously described using overlapping spike and non-spike peptides obtained from the La Jolla Institute (courtesy of Dr. D. Weiskopf; all SARS-CoV-2 testing was performed on samples collected at presentation).9 Acetaminophen level was negative, and he had no history of recent medication use. Serum automarkers (antinuclear, anti-smooth muscle, and anti–liver-kidney-microsome antibodies) were negative. Screening tests for causes of metabolic liver disease were negative and rapid whole-genome sequencing was unrevealing. Abdominal imaging demonstrated normal hepatic vasculature and biliary anatomy. Percutaneous liver biopsy performed on hospital day 2 demonstrated severe liver injury with diffuse hepatocellular ballooning, mixed portal and lobular inflammation, cholestasis, and piecemeal necrosis. There were viral cytopathic changes present on histology, but immunohistochemical staining for adenovirus was negative, as were herpes simplex virus and cytomegalovirus immunostains (Fig 1). In summary, an extensive evaluation for causes of ALF was negative, including previous SARS-CoV-2 infection, except for positive stool and whole-blood adenovirus PCR.
Because of a lack of evidence to support antiviral therapy in cases of presumed adenovirus hepatitis in immunocompetent hosts, cidofovir therapy was not used. Similarly, with histology supporting an infectious process as opposed to primary immune-mediated disease, immunomodulatory therapies were not used in the pretransplant period. With progressively worsening liver function by hospital day 5, the patient underwent urgent evaluation for LT and was listed with status 1A, the designation for ALF that places patients highest on the wait list. His 23-year-old brother began his rapid evaluation as a potential living donor the following day. After 5 days on the United Network for Organ Sharing wait list without an appropriate deceased organ offer and after approval of his brother for a living donation, he was successfully transplanted on hospital day 10 using the left lobe from his brother. His perioperative course was uncomplicated and he was discharged with excellent graft function on postoperative day 8. His explanted liver showed increased necrosis, bile ductular reaction, and parenchymal hemorrhage, relative to the findings of the percutaneous biopsy. The explanted liver was positive for adenovirus by PCR and had occasional viral cytopathic effect, but no immunohistochemical evidence of adenovirus (Fig 1). A multiplex PCR target specific for adenovirus 40/41 was positive in stool, but further typing was not performed for stool, whole-blood, or tissue adenovirus specimens.
We report a case of ALF requiring LT in a child with adenovirus detected by PCR in the stool, whole blood, and explanted liver tissue with no evidence of previous or current SARS-CoV-2 infection despite extensive testing. Two leading hypotheses have emerged to explain the recent pediatric severe hepatitis cases worldwide. The first hypothesis is that isolation associated with the COVID pandemic reduced infectious exposures in children, resulting in increased susceptibility to certain infections (such as adenovirus).8 With adenovirus detected in 61% of the worldwide severe hepatitis cases in which it was investigated, adenoviral infection is a leading hypothesis for the severe hepatitis of unknown etiology.2 In a single-center series of 9 children with hepatitis of unknown cause, 8 (89%) tested positive for adenovirus by whole-blood PCR, with 3 children demonstrating positive adenovirus PCR testing of liver tissue among the 6 who had this testing performed, implicating adenovirus infection in a US cluster of hepatitis cases.6 In England, 52 of the severe acute hepatitis cases in which adenovirus was detected in blood were successfully subtyped, with 48 (92.3%) identified as type 41F.10 The alternative leading hypothesis is that infection with SARS-CoV-2 (past or present) results in altered host response to previously innocuous infections (such as adenovirus), resulting in severe inflammation that damages the liver.11,12 In a report from the United Kingdom, 18% of cases were positive for SARS-CoV-2.11
Our case sheds light on these hypotheses by demonstrating a role for adenovirus infection and no role for SARS-CoV-2 in the development of this child’s ALF. The diagnosis of adenovirus liver infection was made by positive adenovirus PCR and viral cytopathic effect on histopathologic examination of the liver biopsy and explanted tissue, although adenovirus antigen staining was negative in the liver biopsy. We cannot rule out that the positive PCR in the liver tissue resulted from the presence of adenovirus in the blood circulating in the liver. Collectively, however, the positive adenovirus PCR result in both blood and liver and the viral cytopathic effect after having ruled out other infectious etiologies with extensive testing support the diagnosis of adenovirus hepatitis.
Although the majority of children with hepatitis recover with supportive care, some require LT. Children with severe ALF who are listed as status 1A on the United Network for Organ Sharing wait list are not expected to survive more than a few days without LT.13 This patient demonstrated progressive liver failure characterized by worsening coagulopathy and encephalopathy despite supportive management and would not likely have survived without an urgent transplant. Unfortunately, the supply of deceased livers does not meet the number of people in need. Living donation (LD) expands the number of available organs. LDLTs have equivalent if not better graft and patient survival compared with deceased donor LTs.14 However, LDLT is underused, accounting for only 13% of pediatric LTs in 2020.14 LDLT is infrequently considered an option for children in ALF because the length of time needed to evaluate a potential living donor often exceeds the time that the candidate can survive without LT. This case illustrates the utility of simultaneously listing a child in ALF for deceased grafts and rapidly evaluating potential living donors. Ultimately, our patient was saved through living donation when no suitable deceased grafts were offered. He continues to thrive 4 months posttransplantation with excellent graft function.
In conclusion, adenovirus infection may cause hepatic injury resulting in severe hepatitis and/or ALF in immunocompetent children, although the precise pathophysiology is still unclear. Pediatric providers should consider adenovirus PCR testing as part of the evaluation of ALF of unknown etiology in young children and ensure that testing is sent on whole blood.6 Previous COVID history and/or COVID vaccination is not required for development of liver injury. Further studies are needed to understand the mechanism of this injury and to understand whether there may be a role for antiviral agents or immune modulatory therapy in children with adenovirus induced ALF. LDLT can be a lifesaving option for children in ALF when a deceased graft is not available.
The authors thank Michael J. Johnson for performing the SARS CoV-2 cell-mediated immunity testing; and Dr Michael Wachs, Dr Elizabeth Pomfret, and the University of Colorado Health Transplant Center for their dedication to saving the lives of children with liver disease through their living donor program.
Drs Boster, Dominguez, Messacar, Adams, Weinberg, Black, and Feldman contributed to the interpretation of clinical data. Drs Boster and Feldman participated in drafting the article and all authors participated in revising it critically for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: Dr Feldman is supported by AHRQ K08 HS26510.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest to disclose. The sponsor did not provide any products.
acute liver failure
international normalized ratio
living donor liver transplant
polymerase chain reaction
World Health Organization