RR:C19 Evidence Scale rating by reviewer:
Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.
***************************************
Review:
This is a retrospective study from a large database looking at elevated liver enzymes and bilirubin in children with COVID 19 infection at one, three and six months with certain exclusion criteria.
1. Aims and objectives of the study are relevant and a large global database is an appropriate source for the information.
2. For a diagnosis of hepatitis, including children with elevated ALT or AST is incorrect. It should have been both or only ALT, since AST has a non-hepatic source also.
3. Reference No 3 quoted in introduction is a study from India on 47 children with post COVID hepatitis (PCR negative, Antibody positive), of whom 37 were transient with spontaneous resolution. This is what we have been seeing and the acute hepatitis resolves without any intervention in a few weeks. That study compared the features and outcome between the 37 with transient hepatitis and the 10 with hepatitis as part of post COVID Multisystem disease. This study is reporting persisting hepatitis at 3 and 6 months and therefore that study is not a very relevant reference.
4. In the methodology section there is no clear explanation about how the propensity matching of the subjects done on so many parameters and also about how the statistical analysis was performed. In biomedical science we usually go with chi-square and t-test for comparison between parameters, with a p value. Instead the authors have chosen MSD. The rationale for use of this is not mentioned.
5. While using hazard ratio at different intervals, was it for the same patients
6. In table 1, the authors report propensity matching for BMI Pediatric, giving numbers. What do they mean? Is it a cut off BMI?
7. This study is reporting elevated ALT/ AST and Bilirubin at 6 months, which means chronic liver disease. As far as we know now, COVID-19 only causes acute transient hepatitis in the majority of children that resolves in few weeks. The implication of COVID-19 being a cause of asymptomatic chronic liver disease is huge from a global perspective. This is a major finding and requires a thorough review of the data before publication.
8. The discussion is very brief and unclear. The recent spike in hepatitis of unknown etiology in many countries is acute severe hepatitis with high propensity of progression to liver failure. This study is reporting that COVID-19 can lead to chronic liver disease. The two are different. Could it be that COVID-19 is resulting in increased incidence of fatty liver disease?
9. The reviewers are clinicians. We think this paper should be thoroughly vetted by a medical statistical person.