Abstract
BACKGROUND: Post-acute COVID-19 syndrome (PACS; long COVID) is characterized by persistent or delayed symptoms at least 4 weeks from acute COVID-19 infection. Given the well-documented incidence of liver injury in acute COVID-19, this systematic review aims to assess the odds of liver injury in earlier experiencers of PACS. METHODS: Observational studies published prior to March 2022 were screened for data describing liver injury (defined per primary study) in patients with PACS. RESULTS: A total of 2,117 abstracts and 35 full texts were screened, of which 26 met the inclusion criteria. The mean time since acute COVID infection across all studies was 195.5 days. Seven studies included COVID-negative control groups. Twenty-three studies measured lab findings, and nine studies measured imaging or elastography. Five studies were eligible for meta-analysis of odds ratios, which did not demonstrate a statistically significant difference in odds for liver injury in patients with PACS compared with COVID-negative patients (OR 2.22 [95% CI 0.51-9.61; p = 0.28]). Newcastle-Ottawa Scale assessments for all studies found 24 of 26 studies with high to very high risk of bias. ROBINS-E assessments for studies included in the meta-analysis found five of five studies with high to very high risk of bias. CONCLUSIONS: Overall, our findings demonstrate no statistical difference in odds ratios of liver injury in patients with PACS compared with COVID-negative controls. As such, routine assessment and monitoring of liver injury in patients with PACS may not be required; however, higher quality data with lower risk of bias are required to make recommendations of higher certainty.