This report is a pre-specified exploratory substudy of the EXER-COVID randomized 2 × 2 crossover trial that evaluated the immunological consequences of supervised exercise in adults with long COVID (post-acute sequelae of SARS-CoV-2 infection). Participants completed a 6-week supervised exercise program performed twice weekly or received usual care, then crossed over to the alternate condition after a 3–5 day washout period. The exercise regimen is described as moderate intensity in the abstract. The trial is registered at ClinicalTrials.gov (NCT04797871).
Details such as sample size, participant age and sex distribution, baseline clinical characteristics, and adherence are not reported in the abstract and therefore are not included here.
Plasma cytokine concentrations were measured using a multiplex immunoassay. The cytokines assayed were IL‑1β, IL‑6, IL‑10, TNF‑α, MCP‑1/CCL2, MIP‑1α/CCL3, MIP‑1β/CCL4, and IP‑10/CXCL10.
Immunophenotyping focused on adaptive and innate compartments. The adaptive panel included 20 CD4+ and CD8+ T‑cell phenotypic subsets. Innate immune assessment included five populations: neutrophils, low‑density neutrophils (LDN), monocytes, M2‑like monocytes, and natural killer (NK) cells.
Treatment effects were estimated using within-period change scores for the crossover design. The analytic plan accounted for multiple comparisons using the Benjamini–Hochberg procedure, controlling the false discovery rate at less than 5%. Period and carryover effects were assessed; p values for these models were obtained from linear mixed-effects models as noted in figure captions.
When comparing within-participant paired treatment contrasts, five variables reached nominal statistical significance before adjustment for multiple testing. These included two CD8+ T‑cell subsets and three circulating cytokines: IL‑1β, IL‑10, and MIP‑1α. However, after controlling for multiple comparisons with the Benjamini–Hochberg method, none of these variables remained statistically significant.
No detailed effect sizes, confidence intervals, or exact p values for the corrected analyses are reported in the abstract; the figures in the full text illustrate mean percentage changes and variability for multiple T‑cell subsets and cytokines, and the treatment‑effect p values correspond to the pre‑specified one‑sample z test of within‑participant paired treatment contrasts.
No changes were observed in the measured innate immune populations, which included neutrophils, low‑density neutrophils, monocytes, M2‑like monocytes, and NK cells. The analyses did not detect evidence of carryover effects across the crossover periods.
In this secondary exploratory analysis of a randomized 2 × 2 crossover trial, supervised, moderate‑intensity exercise delivered twice weekly for 6 weeks was not associated with statistically significant immunological alterations after correction for multiple comparisons in adults with long COVID who were PESE‑negative. The authors interpret these results as supporting the short‑term immunological safety of the supervised exercise program in this specific population.
Clinicians and researchers should note that nominal signals were observed for a small number of adaptive and cytokine measures prior to adjustment, but these did not survive false discovery rate correction. The absence of detected changes in innate populations also argues against major short‑term perturbations of the innate immune compartment under the conditions studied.
The abstract does not report key methodological details needed to fully appraise the findings, including participant numbers, baseline characteristics, magnitude of changes, statistical estimates (effect sizes and confidence intervals), and adherence or adverse event data. The exercise program intensity descriptors are limited to "moderate intensity" without further operational details in the abstract. The full text (Physiol Rep; DOI: 10.14814/phy2.71031; PMCID PMC13525194) should be consulted for complete methods and numeric results.
Overall, based on the information presented in the abstract, supervised moderate‑intensity exercise did not induce statistically robust immunological changes in this cohort of PESE‑negative adults with long COVID when controlling for multiple comparisons.