Use of Cardiopulmonary Exercise Testing to Evaluate Long COVID-19 Symptoms in Adults: A Systematic Review and Meta-analysis.

Durstenfeld MS; Sun K; Tahir P; Peluso MJ; Deeks SG; Aras MA; Grandis DJ; Long CS; Beatty A; Hsue PY

JAMA network open [JAMA Netw Open] 2022 Oct 03; Vol. 5 (10), pp. e2236057.
Date of Electronic Publication: 2022 Oct 03.

Importance: Reduced exercise capacity is commonly reported among individuals with COVID-19 symptoms more than 3 months after SARS-CoV-2 infection (long COVID-19 [LC]). Cardiopulmonary exercise testing (CPET) is the criterion standard to measure exercise capacity and identify patterns of exertional intolerance.
Objectives: To estimate the difference in exercise capacity among individuals with and without LC symptoms and characterize physiological patterns of limitations to elucidate possible mechanisms of LC.
Data Sources: A search of PubMed, EMBASE, Web of Science, preprint servers, conference abstracts, and cited references was performed on December 20, 2021, and again on May 24, 2022. A preprint search of,, and was performed on June 9, 2022.
Study Selection: Studies of adults with SARS-CoV-2 infection more than 3 months earlier that included CPET-measured peak oxygen consumption (V̇o2) were screened independently by 2 blinded reviewers; 72 (2%) were selected for full-text review, and 35 (1%) met the inclusion criteria. An additional 3 studies were identified from preprint servers.
Data Extraction and Synthesis: Data extraction was performed by 2 independent reviewers according to the PRISMA reporting guideline. Data were pooled using random-effects models.
Main Outcomes and Measures: Difference in peak V̇o2 (in mL/kg/min) among individuals with and without persistent COVID-19 symptoms more than 3 months after SARS-CoV-2 infection.
Results: A total of 38 studies were identified that performed CPET on 2160 individuals 3 to 18 months after SARS-CoV-2 infection, including 1228 with symptoms consistent with LC. Most studies were case series of individuals with LC or cross-sectional assessments within posthospitalization cohorts. Based on a meta-analysis of 9 studies including 464 individuals with LC symptoms and 359 without symptoms, the mean peak V̇o2 was -4.9 (95% CI, -6.4 to -3.4) mL/kg/min among those with symptoms with a low degree of certainty. Deconditioning and peripheral limitations (abnormal oxygen extraction) were common, but dysfunctional breathing and chronotropic incompetence were also described. The existing literature was limited by small sample sizes, selection bias, confounding, and varying symptom definitions and CPET interpretations, resulting in high risk of bias and heterogeneity.
Conclusions and Relevance: The findings of this systematic review and meta-analysis study suggest that exercise capacity was reduced more than 3 months after SARS-CoV-2 infection among individuals with symptoms consistent with LC compared with individuals without LC symptoms, with low confidence. Potential mechanisms for exertional intolerance other than deconditioning include altered autonomic function (eg, chronotropic incompetence, dysfunctional breathing), endothelial dysfunction, and muscular or mitochondrial pathology.