# Integration of patient-reported outcomes and myokine profiling for the detection of physical inactivity in COPD: an exploratory multicentre study

## TL;DR

This study explores combining patient-reported outcomes and myokine levels to better detect physical inactivity in COPD patients.

## Contribution

The study introduces a multidimensional approach combining PROs and myokine profiling for improved inactivity detection in COPD.

## Key findings

- PROs like mMRC and PROMs-D correlated negatively with moderate activity and step count.
- Myokines GDF-15, FABP3, and BDNF showed moderate associations with different physical activity metrics.
- Combined PRO–myokine models outperformed single markers in detecting inactivity.

## Abstract

Physical inactivity strongly predicts poor prognosis in chronic obstructive pulmonary disease (COPD) but is often underrecognized. We investigated whether combining patient-reported outcomes (PROs) with myokine profiling enhances detection of inactivity in COPD. In this multicentre cross-sectional study, 73 patients with stable COPD underwent PRO assessment (modified Medical Research Council dyspnea scale (mMRC), dyspnea-specific PROs (PROMs-D), COPD Assessment Test (CAT), Shortness of Breath Daily Activities Questionnaire (SOBDA-Q), and Kihon Checklist (KCL)), serum myokine measurement, and accelerometer-based physical activity evaluation, stratified into 1.0–1.5 METs (low-intensity/sedentary), ≥ 3.0 METs (moderate), total activity (METs·h), and step count. Correlation and logistic regression analyses were performed. mMRC and PROMs-D correlated negatively with moderate activity and step count. Among myokines, growth differentiation factor-15 (GDF-15), fatty acid binding protein 3 (FABP3), and brain-derived neurotrophic factor (BDNF) showed moderate associations with physical activity: GDF-15 and BDNF with low-intensity, GDF-15 with moderate, and FABP3 and BDNF with step count. Combined PRO–myokine models outperformed single markers, with areas under the curve of 0.77 for low-intensity activity, 0.82 for moderate activity, and 0.86 for step count. In conclusion, integrating PROs and myokines improves the specificity and accuracy of inactivity detection in COPD. This multidimensional strategy may facilitate early, personalized interventions.

The online version contains supplementary material available at 10.1038/s41598-025-34436-y.