Cardiopulmonary exercise testing to evaluate post-acute sequelae of COVID-19 ("Long COVID"): a systematic review and meta-analysis

Abstract

Importance: Reduced exercise capacity is commonly reported among individuals with Long COVID (LC). Cardiopulmonary exercise testing (CPET) is the gold-standard to measure exercise capacity to identify causes of exertional intolerance.

Objectives: To estimate the effect of SARS-CoV-2 infection on exercise capacity including those with and without LC symptoms and to characterize physiologic patterns of limitations to elucidate possible mechanisms of LC.

Data sources: We searched PubMed, EMBASE, and Web of Science, preprint severs, conference abstracts, and cited references in December 2021 and again in May 2022.

Study selection: We included studies of adults with SARS-CoV-2 infection at least three months prior that included CPET measured peak VO ₂ . 3,523 studies were screened independently by two blinded reviewers; 72 (2.2%) were selected for full-text review and 36 (1.2%) met the inclusion criteria; we identified 3 additional studies from preprint servers.

Data extraction and synthesis: Data extraction was done by two independent reviewers according to PRISMA guidelines. Data were pooled with random-effects models.

Main outcomes and measures: A priori primary outcomes were differences in peak VO ₂ (in ml/kg/min) among those with and without SARS-CoV-2 infection and LC.

Results: We identified 39 studies that performed CPET on 2,209 individuals 3-18 months after SARS-CoV-2 infection, including 944 individuals with LC symptoms and 246 SARS-CoV-2 uninfected controls. Most were case-series of individuals with LC or post-hospitalization cohorts. By meta-analysis of 9 studies including 404 infected individuals, peak VO ₂ was 7.4 ml/kg/min (95%CI 3.7 to 11.0) lower among infected versus uninfected individuals. A high degree of heterogeneity was attributable to patient and control selection, and these studies mostly included previously hospitalized, persistently symptomatic individuals. Based on meta-analysis of 9 studies with 464 individuals with LC, peak VO ₂ was 4.9 ml/kg/min (95%CI 3.4 to 6.4) lower compared to those without symptoms. Deconditioning was common, but dysfunctional breathing, chronotropic incompetence, and abnormal oxygen extraction were also described.

Conclusions and relevance: These studies suggest that exercise capacity is reduced after SARS-CoV-2 infection especially among those hospitalized for acute COVID-19 and individuals with LC. Mechanisms for exertional intolerance besides deconditioning may be multifactorial or related to underlying autonomic dysfunction.