Positive association between severity of COVID-19 infection and liver damage: a systematic review and meta-analysis

Abstract

Aim: The current study aimed to report a pooled analysis of the association of the circulating levels of liver enzymes and total bilirubin with severe and non-severe COVID-19.

Background: The ongoing coronavirus outbreak is an important threat to health worldwide. Epidemiological data representing greater risk of liver failure in patients infected with Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2).

Methods: Electronic databases were comprehensively searched using Medline, ISI Web of Science, EMBASE, and the Cochrane Library up to July 2020. Outcomes from each relevant study were pooled using a random-effects model. Heterogeneity was analyzed by Q test and I² statistics. Sensitivity analysis was also evaluated.

Results: A total of 24 studies were included (4,246 patients) in this study. We found a significant association of COVID-19 severity with increased levels of ALT [SMD: 1.40 U/L; 95% CI (0.93, 1.88); P < 0.05, I² = 96.5%, P _{Heterogenity =} 0.000 ], AST [SMD: 2.11 U/L; 95% CI (1.40, 2.83); P < 0.05, I² = 97.9%, P _{Heterogenity =} 0.000], LDH [SMD: 3.88 U/L; 95% CI (2.70, 5); P < 0.05, I² = 98.7%, P _{Heterogenity =} 0.000] and TBil [SMD: 1.08 mol/L; 95% CI (0.44, 1.72); P = 0.001, I² = 97.7, P _{Heterogenity =} 0.000], whereas, ALP values [SMD: 0.31; 95% CI (-1.57, 2.20); P = 0.74] was not significant between severe and non-severe COVID-19 patients. Moreover, elevated liver enzymes were found more in males [OR: 1.52, (95% CI 1.26, 1.83), P < 0.05] with severe COVID-19 infection than in females.

Conclusion: The alterations of liver function indexes caused by SARS-CoV-2 infection suggested a potential prognosis biomarker for screening of severe patients at early stages of the disease.

Keywords: COVID-19; Coronavirus; Liver damage; Liver enzymes; Meta-analysis.

Figure 1

Flow chart of the number of studies identified and proper in the meta-analysis

Figure 2

Forest plots assessing standardized mean difference (SMD) and 95% confidence intervals for the association between circulating levels of a) ALT, b) AST, c) LDH, and d) ALP with severity of COVID-19 in admitted and hospitalized patients. Meta-analysis was performed using a random-effects model with inverse variance weighting

Figure 3

a) Forest plot assessing standardized mean difference (SMD) and 95% confidence intervals for the association between TBil concentration with severity of COVID-19 in admitted and hospitalized patients, and b) Random-effects funnel plot detailing publication bias after trimming and filing in the studies investigating the association between TBil concentration with severity of n-COVID-19 in admitted and hospitalized patients. Open circles represent observed published studies; closed circles represent imputed unpublished studies. Meta-analysis was performed using a random-effects model with inverse variance weighting

Figure 4

Random-effects funnel plots detailing publication bias after trimming and filing in the studies investigating the association between circulating levels of a) ALT, b) AST, c) LDH, and d) ALP with severity of COVID-19 in admitted and hospitalized patients. Open circles represent observed published studies; closed circles represent imputed unpublished studies

Figure 5

Forest plots detailing Odds Ratio (OR) and 95% confidence intervals between variables of sex, a) male and b) female, and drug treatment, c) antiviral and d) antibiotic drugs, and risk of elevated liver enzymes in admitted and hospitalized patients with COVID-19 infection. Meta-analysis was performed using a random-effects model with inverse variance weighting