1 Department of Health, Kinesiology and Applied Physiology, Concordia University, Montreal, Canada. Electronic address: brent.rosenstein@icloud.com.
2 Department of Health, Kinesiology and Applied Physiology, Concordia University, Montreal, Canada.
3 Department of Health, Kinesiology and Applied Physiology, Concordia University, Montreal, Canada. Electronic address: alexarou@live.ca.
4 School of Health and Social Care, Edinburgh Napier University, Edinburgh, United Kingdom. Electronic address: S.Valentin@napier.ac.uk.
5 Department of Companion Animals and Horses, University of Veterinary Medicine Vienna, Vienna, Austria; Royal (Dick) School of Veterinary Studies, University of Edinburgh, Edinburgh, United Kingdom. Electronic address: Theresia.Licka@vetmeduni.ac.at.
6 Centre de Recherche du Centre Hospitalier Universitaire de Sherbrooke (CHUS), Department of Anesthesiology, Université de Sherbrooke, Sherbrooke, Canada. Electronic address: monica.sean@USherbrooke.ca.
7 Centre de Recherche du Centre Hospitalier Universitaire de Sherbrooke (CHUS), Department of Anesthesiology, Université de Sherbrooke, Sherbrooke, Canada. Electronic address: Pascal.Tetreault@USherbrooke.ca.
8 University of Sydney, School of Heath Science, Department of Medicine and Health, Sydney, Australia. Electronic address: jim.elliott@sydney.edu.au.
9 Department of Health, Kinesiology and Applied Physiology, Concordia University, Montreal, Canada. Electronic address: maryse.fortin@concordia.ca.

It remains unclear whether paraspinal muscle fatty infiltration in low back pain (LBP) is i) solely intramuscular, ii) is lying outside the epimysium between the muscle and fascial plane (epimuscular) or iii) or combination of both, as imaging studies often use different segmentation protocols that are not thoroughly described. Epimuscular fat possibly disturbs force generation of paraspinal muscles, but is seldomly explored. This project aimed to 1) compare epimuscular fat in participants with and without chronic LBP, and 2) determine whether epimuscular fat is different across lumbar spinal levels and associated with BMI, age, sex and LBP status, duration or intensity. Fat and water lumbosacral MRIs of 50 chronic LBP participants and 41 healthy controls were used. The presence and extent of epimuscular fat for the paraspinal muscle group (erector spinae and multifidus) was assessed using a qualitative score (0-5 scale; 0 = no epimuscular fat and 5 = epimuscular fat present along the entire muscle) and quantitative manual segmentation method. Chi-squared tests evaluated associations between qualitative epimuscular fat ratings and LBP status at each lumbar level. Bivariate and partial spearman's rho correlation assessed relationships between quantitative and qualitative epimuscular fat with participants' characteristics. Epimuscular fat was more frequent at the L4-L5 (X² = 13.781, p = 0.017) and L5-S1 level (X² = 27.825, p < 0.001) in participants with LBP compared to controls, which was not found for the higher lumbar levels. The total qualitative score (combined from all levels) showed a significant positive correlation with BMI, age, sex (female) and LBP status (r = 0.23-0.55; p < 0.05). Similarly, the total area of epimuscular fat (quantitative measure) was significantly correlated with BMI, age and LBP status (r = 0.26-0.57; p < 0.05). No correlations were found between epimuscular fat and LBP duration or intensity. Paraspinal muscle epimuscular fat is more common in chronic LBP patients. The functional implications of epimuscular fat should be further explored.