More than 50 Long-Term Effects of COVID-19: A Systematic Review and Meta-Analysis

Abstract

Background: COVID-19, caused by SARS-CoV-2, can involve sequelae that last weeks to months after initial recovery. The objective of this systematic review and meta-analysis is to identify studies assessing the long-term effects of COVID-19 and estimate the prevalence of each symptom, sign, or laboratory parameters of patients at a post-COVID-19 stage. \n \nMethods: In this systematic review and meta-analysis, LitCOVID (PubMed and Medline) and Embase were searched by two independent researchers. Studies published before 1st of January 2021 and with a minimum of 100 patients were included. For effects reported in two or more studies, meta-analyses using a random-effects model were performed using the MetaXL software to estimate the pooled prevalence with 95% CI. Heterogeneity was assessed using the I2\xa0statistics. PRISMA guidelines were followed. \n \nFindings: A total of 18,251 publications were identified, of which 15 met the inclusion criteria. The prevalence of 55 long-term effects was estimated, 21 meta-analyses were performed, and 47,910 patients were included. The follow-up time ranged from 15 to 110 days post-viral infection. The age of the study participants ranged between 17 and 87 years. It was estimated that 80% (95% CI 65-92) of the patients that were infected with SARS-CoV-2 developed one or more symptoms. The five most common symptoms were fatigue (58%), headache (44%), attention disorder (27%), hair loss (25%), and dyspnea (24%). In order to have a better understanding, there is a need for studies to stratify by sex, age, previous comorbidities, severity of COVID-19 (including asymptomatic), and duration of each symptom. \n \nInterpretation: From the clinical perspective, multi-disciplinary teams are crucial to developing preventive measures, rehabilitation techniques, and clinical management strategies with whole-patient perspectives designed to address after-COVID-19 care. \n \nFunding: National Institute for Neurological Disorders and Stroke (NINDS), and Houston Methodist Research Institute, Houston, TX. \n \nDeclaration of Interests: SLL is an employee of Novartis Pharmaceutical Company; the statements presented in the paper do not necessarily represent the position of the company. The remaining authors have no competing interests to declare.