679: Brain-Muscles Network Decomposition in a Dynamic Upper-Limb Task With Acute Shoulder Pain

Topic: Experimental (Non-clinical) Human Studies
Subtopic: Pain Category (ICD-11), Acute musculoskeletal pain, Electrophysiology/Clinical neurophysiology (e.g. microneurography, EEG, MEG, OPMs, TMS), Pain mechanisms (e.g. descending inhibitory control, placebo, inflammation), Pain models (e.g. saline injection, capsaicin, UVB, HFS, incision, exercise, itch)

Background and Aims: Shoulder movement requires coordinated neural drive to multiple muscles to achieve stable, accurate upper limb actions. Acute nociceptive input is known to modify motor behaviour [3,4], but how it reorganises functional coupling between the cortex and shoulder musculature during dynamic movement remains unclear [1]. Cortico-muscular and intermuscular coherence provide insight into shared neural control across the movement cycle. By characterising low dimensional brain–muscle networks [2] during a repetitive pointing task, this study examines how experimentally induced shoulder pain alters the organisation and timing of these networks when a larger set of ten upper limb muscles is considered.

Methods: Eighteen healthy adults performed a paced repetitive pointing task before and during acute shoulder pain induced by injection of 6% hypertonic saline into the anterior deltoid. EEG was recorded from C3/C4, and EMG was collected from ten right side muscles: upper, middle, and lower trapezius; anterior, middle, and posterior deltoid; biceps; triceps; serratus anterior; and infraspinatus. Signals were filtered, cleaned, and segmented by repetitions. Time frequency cortico-muscular and intermuscular coherence were estimated, and non negative matrix factorisation was applied to extract low dimensional brain–muscle networks across baseline and pain conditions.

Results: NNMF identified four networks explaining 65% of coherence variance, indicating a low-dimensional structure underlying CMC and IMC. Acute shoulder pain modulated all networks, with phase-specific changes across movement onset, arm lifting, and target reaching. Networks differed in timing, with Networks 1 and 2 showing overlapping modulation around initiation and reaching. Networks 3 and 4 exhibited similar but less pronounced changes. Muscle weightings varied across components, with some networks dominated by shoulder muscles and others by more distal groups. Beta–gamma differences were smaller than network level distinctions, and several muscles contributed to multiple components whereas others loaded strongly onto only one.

Conclusions: Acute shoulder pain altered brain–muscle interactions, reflected in a small set of recurrent low dimensional networks. Four BMNs accounted for most coherence structure, and pain influenced their activation in a phase dependent way, with some networks changing most around movement onset and others during arm elevation or reaching. Muscle contributions also differed across networks, with certain components dominated by shoulder muscles and others involving more distal groups; beta–gamma differences were present but secondary. Overall, nociception shifted how existing control networks were recruited while task performance remained stable, supporting BMNs as a useful framework for studying pain related motor adaptations.

References: 1. Hodges PW, Tucker K., Moving differently in pain: a new theory to explain the adaptation to pain. Pain. 2011;152(3 Suppl):S90–S98.
2. Roeder L, et al. Dynamics of brain-muscle networks reveal effects of age and somatosensory function on gait. iScience. 2024;27(3):109162.
3. Kinsella R, et al. Electromyographic activity of the shoulder muscles during rehabilitation exercises in subjects with and without subacromial pain syndrome: a systematic review. Shoulder Elbow. 2016;9(2):112-126.

Presenting Author: Giacomo Nardese
Submitting Author: Giacomo Nardese

Senior Author: Graham Kerr

IASP 2026 Global Year theme of Neuropathic Pain: No

• Giacomo Nardese: MRes - School of Exercise and Nutrition Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia.
• Tjeerd W Boonstra: PhD - Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, Netherlands.
• Paul W Hodges: PhD - CIPHeR, School of Health and Rehabilitation Sciences, University of Queensland, Brisbane, Queensland, Australia.
• Wolbert van den Hoorn: PhD - School of Health and Rehabilitation Sciences, Faculty of Health, Medicine and Behavioural Sciences, The University of Queensland, Brisbane, Queensland, Australia.
• Arthur Fabre: MSc - School of Exercise and Nutrition Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia
• Simon J Summers: PhD - School of Biomedical Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia.
• Johan van der Meer: PhD - Centre for Biomedical Technologies, Queensland University of Technology, Brisbane, Queensland, Australia
• Eric Su: PhD - School of Exercise and Nutrition Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia
• Graham Kerr: PhD - School of Exercise and Nutrition Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia.