Liver Involvement in Children With SARS-COV-2 Infection

Two Distinct Clinical Phenotypes Caused by the Same Virus

Adriana Perez MD; Amanda Cantor MD; Bryan Rudolph MD, MPH; Jonathan Miller MD; Debora Kogan-Liberman MD; Qi Gao PhD; Bernardo Da Silva; Kara G. Margolis MD; Nadia Ovchinsky MD, MBA; Mercedes Martinez MD


Liver International. 2021;41(9):2068-2075. 

In This Article


In this retrospective study, we included patients ≤21 years old evaluated between March 14, 2020 and June 30, 2020, in the inpatient or outpatient clinical setting of two large children's hospitals in New York City (Morgan Stanley Children's Hospital of New York-Presbyterian and Children's Hospital at Montefiore). The Institutional Review Board approved this study at both centers. All research was conducted following the Declaration of Helsinki guidelines of good practice.

Patients were identified by the international classification of diseases 10th revision (ICD-10) code for a positive COVID-19 test (U07.1) and/or by an institutional COVID-19 or MIS-C database. Only participants with confirmed SARS-CoV-2 infection with detection of the virus via nasal swab-derived real-time reverse polymerase chain reaction were included in the COVID-19 cohort. MIS-C was defined by the modified criteria of the Centers for Disease Control and Prevention (CDC) (see supplement).[19] We excluded children without liver tests.

Clinical, demographic, laboratory, and anthropometric data were collected. We have defined E-ALT as a peak elevation of ALT >40 U/L, as these values were reported to fall above the 97th percentile for all ages and both sexes in a large cohort of healthy children.[20] Measures of hepatic synthetic dysfunction, such as international normalized ratio (INR) and bilirubin, were not included in this definition, given the multifactorial reasons for abnormal values in this clinical setting.[13] Each patient's medical records were manually reviewed for the presence of comorbidities.

E-ALT was further categorized as mild to moderate (ALT > 40 ≤ 200 U/L) and severe (ALT > 200 U/L). Due to the small number of patients with severely E-ALT and unchanged statistical significance when combined with mild to moderate elevations in ALT, all analyses were subsequently dichotomized as E-ALT present (ALT > 40 U/L) or absent (ALT ≤ 40 U/L) within each cohort. We also evaluated AST, bilirubin, albumin, and INR values. Obesity was defined as body mass index (BMI) above 95th percentile and/or BMI ≥30 kg/m2 as applicable. An immunocompromised state was defined by the presence of malignancy requiring chemotherapy or radiation, recipients of bone marrow or solid organ transplant, or among patients receiving other immunosuppressive therapy or biologics including those with inflammatory bowel disease.

Student's t-test or Mann–Whitney U-test was performed for continuous variables. Chi-square or Fisher's exact test was used for categorical variables to identify differences between the presence or absence of E-ALT within each cohort (COVID-19 and MIS-C) and to compare the differences between them. Variables were presented as frequency (percentage), mean with standard deviation (SD), or median with interquartile range (IQR), as appropriate.

Multivariable logistic regression was used to examine the potential risk factors for E-ALT by cohort (COVID-19 and MIS-C) and for the combined data set. Clinically significant variables and/or those with P < .25 in bivariate analysis were selected. Due to the small sample size in each cohort, the association between the potential risk factors and E-ALT was tested sequentially, while controlling for age, gender, and race in each model, as well as for other predictors. Backward variable selection was used to compare E-ALT in MIS-C to E-ALT in COVID-19 until all variables were significant, while controlling for age, gender, and race. P-values <0.05 were considered statistically significant. All analyses were performed using SAS 9.4.