COVID-19 Related Liver Injury: Call for International Consensus.

Dear Editor:

We read with interest the article by Fan et al regarding the clinical characteristics of COVID-19 patients with liver damage. They defined abnormal liver damage in their study, and found that liver function abnormality was associated with a longer hospital stay and might have been related to the use of lopinavir/ritonavir during hospitalization. This study is interesting and provides the direction for future research, however, there is a need to address the importance of a standardized definition of COVID-19–related liver injury, which currently is unavailable; it also calls for an international consensus in this regard.

Liver function abnormality was frequent, but usually mild, in COVID-19 patients. In patients with COVID-19, abnormal liver function may result from direct viral damage, immune-mediated inflammation, drug hepatotoxicity, hypoxia-reperfusion dysfunction, or reactivation of pre-existing liver diseases. It is worthwhile to explore the mechanisms and clinical outcomes of liver dysfunction for better understanding and treatment of COVID-19. However, as a new contagious disease, there is no consensus on the definition of COVID-19–associated liver injury, and its definition has varied in recent studies (Table 1 ).

Table 1

Summary of Published Studies Regarding the Definition of Liver Injury in COVID-19

Study	Article type	Definitions	Sample size	Male/female	Mean age, y	Pre-existing liver disease	ALT, U/L	AST, U/L	TBIL, μmol/L	ALP, U/L	GGT, U/L	Main conclusion
Xie et al5	Original article	Increased levels of ALT, AST, or TBIL	79	44/35	60	No	34 (18–67)	30 (23–50)	13.6 (8.8–17.6)	79.0 (59.0–100.0)	31.5 (19.0–81.3)	Liver injury is common in non-ICU hospitalized COVID-19 patients and may relate to severe pulmonary imaging lesions
Zhang et al7	Original article	Abnormal liver function: initial test >ULN, including ALT, AST, TBIL, ALP, GGT, and ALBLiver injury: according to the criteria of iSAEC	115	49/66	49.5	NA	25.71 ± 21.08	28.30 ± 15.66	11.31 ± 5.18	73.72 ± 24.37	36.14 ± 45.02	Although abnormalities of liver function indexes are common in COVID-19 patients, the impairment of liver function is not a prominent feature of COVID-19, and also may not have serious clinical consequences
Ji et al6	Original article	Hepatocellular type: ALT >30 for males and ALT > 19 for femalesDuctular type: ALP >ULN accompanied by GGT >ULNMixed type: both hepatocellular and ductular enzyme levels were >ULN	202	113/89	44.5	NAFLD	NA	NA	NA	NA	NA	Liver injury in COVID-19 patients was frequent but mild The pattern was mostly hepatocellular rather than cholestatic in COVID-19 patients with NAFLD
Cai et al9	Original article	Liver injury: ALT and/or AST >3 ULN; ALP, GGT, and/or TBIL >2 ULNHepatocyte type: ALT and/or AST >3 ULNCholangiocyte type: ALP or GGT >2 ULN Mixed type: an increased combination of both types	417	198/219	47.0	21 patients with NAFLD, ALD, or hepatitis B	21 (15–31)	26.5 (21–35)	10.9 (8.3–16.3)	61 (50.5–74.5)	33.45 (37.41)	Patients with abnormal liver tests, especially in hepatocyte type or mixed type, had significantly higher odds of developing severe pneumonia
Fan et al1	Original article	Any parameter >ULN on admission, includingALT, AST, LDH, GGT, ALP, and TBIL	148	73/75	50	9 patients with CLD	NA	NA	NA	NA	NA	More than a third of patients admitted to the hospital have abnormal liver function, and this is associated with a longer hospital stay
Qi et al4	Letter to the editor	Any parameter >ULN on admission, including ALT, AST, and TBIL	70	39/31	NA	No	NA	NA	NA	NA	NA	Dynamic monitoring of the liver function of patients with liver injury are needed, especially those in the ICU
Sun et al11	Comment	COVID-19–associated liver injury is defined as any liver damage occurring during disease progression and treatment of COVID-19 in patients with or without pre-existing liver diseases
Xu et al3	Review	Liver injury was indicated mainly by abnormal ALT/AST levels accompanied by slightly increased bilirubin levels

ALB, albumin; ALD, alcoholic liver disease; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CLD, chronic liver disease; GGT, γ-glutamyl transpeptidase; ICU, intensive care unit; iSAEC, international Serious Adverse Event Consortium; LDH, lactate dehydrogenase; NA, not applicable; NAFLD, nonalcoholic fatty liver disease; TBIL, total bilirubin; ULN, upper limit of normal.

First, the parameters for evaluating liver function were different among studies, although alanine aminotransferase, aspartate aminotransferase, and total bilirubin were mostly included.3, 4, 5 Several studies also included alkaline phosphatase and γ-glutamyl transpeptidase for the assessment of liver function, , probably because the angiotensin converting enzyme 2, the severe acute respiratory syndrome coronavirus 2 receptor, was reported to be highly expressed in cholangiocytes. Interestingly, the current study included lactate dehydrogenase as a parameter for liver damage in COVID-19 patients, which normally was regarded as an indicator for heart, kidney, and liver dysfunction, and thus may be less specific in this regard. The difference on diagnostic parameters may affect the grouping of liver injury and non–liver injury patients, and thus compromise the results.

More importantly, because of the different criteria for increased liver enzyme levels, researchers and clinicians may overestimate the value of abnormal liver function both in a scientific aspect and in clinical practice. Several studies have reported the potential clinical outcome of COVID-19 patients with liver injury, including longer hospitalization, severe pulmonary lesions, and higher odds of developing severe pneumonia. In contrast, Zhang et al reported that although impairment of liver function did exist in COVID-19, it was not a prominent feature. Notably, in these studies, some researchers defined liver injury as any liver function test result that was higher then the upper limit of normal (ULN), , however, other investigators have defined it as liver enzyme levels higher than 2 or 3 times the ULN, or according to the guideline of drug-induced liver injury, which may be the reason for the mixed results. Together with the comment published by Bangash et al, we think it may be inappropriate to consider mild increases and fluctuations of liver enzyme levels (ie, just slightly >ULN) as clinically significant liver injury, which may be predominantly a clinical distraction.

Furthermore, the diagnostic time point (ie, on admission or during disease progression) of liver injury, , and the standardized definition of liver injury patterns , (ie, hepatocellular type, cholangiocytes type, and mixed type) in COVID-19, also varies and remains open questions.

Overall, from a scientific point of view, the lack of standardization in the definition is undoubtedly an important issue because it may jeopardize the generalizability of the conclusions from these studies. Although liver function abnormalities and clinically significant liver injury in COVID-19 should be investigated further, we suggest researchers pay extreme attention to the terminology and its definition to avoid ambiguity in future analysis and overtreatment in clinical practice.