Cardiorespiratory fitness and plasma metabolites in adults: Sex‐dependent insights from a combined regression and machine learning approach
Physiological Reports · 2026-11 · 14 (21)
DOI: 10.14814/phy2.71208 ↗At a glance
Journal article in Physiological Reports, published online on 2026-11-13. Written by Malin Enarsson and 4 co-authors. Institution: Division of Family Medicine and Primary Care, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm, Sweden. Major MEDLINE topics: Cardiorespiratory Fitness, Machine Learning, Metabolome. 37 references. The full text is freely available (open access, CC BY 4.0). Funded by: Center for Clinical Research Dalarna, Centrum fÖr Klinisk Forskning Dalarna.
Abstract (original language)
Cardiorespiratory fitness (CRF) is associated with reduced cardiovascular disease risk, but the underlying molecular mechanisms remain poorly understood. Metabolomics offers a promising strategy to elucidate these mechanisms. Our aim was to investigate sex‐dependent associations between CRF and plasma metabolites. This cross‐sectional study included 450 participants (225 females, 225 males) from the Swedish population‐based POEM cohort. CRF (VO 2 peak) was assessed by maximal bicycle ergometer testing. Plasma levels of 790 metabolites were quantified using the Metabolon platform. Associations between VO 2 peak and metabolites were examined using sex‐stratified multivariable linear regression and random forest models, adjusted for cardiovascular risk factors. VO 2 peak was associated with 27 metabolites in females and 24 in males ( p < 0.000063), of which 4 overlapped between sexes and 10 showed sex‐specific associations. Most associations remained significant after adjustment for cardiovascular risk factors. Significant metabolites substantially improved model fit for VO 2 peak (adjusted R 2 : females 26%–53%; males 27%–37%; p < 0.001). Random forest variable importance rankings showed substantial overlap with regression findings. In conclusion, CRF was associated with distinct plasma metabolite profiles in females and males, with most metabolites showing sex‐dependent associations and a smaller subset being shared between the sexes. These findings provide new insight into the biological pathways connecting CRF to cardiometabolic health.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Division of Family Medicine and Primary Care, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm, Sweden.
- Center for Clinical Research Dalarna, Falun, Sweden.
- Medical Epidemiology, Department of Surgical Sciences, Uppsala University, Uppsala, Sweden.
- Department of Medical Sciences, Uppsala University, Uppsala, Sweden.
- School of Health and Social Studies, Dalarna University, Falun, Sweden.
ORCID profiles: Malin Enarsson
Funding
- Center for Clinical Research Dalarna
- Centrum fÖr Klinisk Forskning Dalarna
Cite
APA
Enarsson, M., Schillemans, T., Brooke, H. L., Lind, L., & Ärnlöv, J. (2026). Cardiorespiratory fitness and plasma metabolites in adults: Sex‐dependent insights from a combined regression and machine learning approach. Physiological Reports, 14(21). https://doi.org/10.14814/phy2.71208
Vancouver
Enarsson M, Schillemans T, Brooke HL, Lind L, Ärnlöv J. Cardiorespiratory fitness and plasma metabolites in adults: Sex‐dependent insights from a combined regression and machine learning approach. Physiological Reports. 2026;14(21). doi:10.14814/phy2.71208
BibTeX
@article{Enarsson2026,
author = {Enarsson, Malin and Schillemans, Tessa and Brooke, Hannah L and Lind, Lars and Ärnlöv, Johan},
title = {Cardiorespiratory fitness and plasma metabolites in adults: Sex‐dependent insights from a combined regression and machine learning approach},
journal = {Physiological Reports},
year = {2026},
volume = {14},
number = {21},
doi = {10.14814/phy2.71208}
}
EUVIMED