Brain Aging and Chronic Pain: A Replication Study in Knee Osteoarthritis

We have previously reported that older adults with chronic musculoskeletal pain had significantly older-appearing brains compared to older controls in a small, mainly Non-Hispanic white sample of individuals. We employed a machine-learning derived brain aging biomarker that compares voxel-wise gray and white matter volume images to a statistical model that accurately predicts chronological age from neuroimaging data in healthy people. The present study aimed to extend our previous findings to a larger, younger, and more racially diverse sample of individuals with knee osteoarthritis (OA) pain. Participants (mean age=58 years) with severe (n=100) and mild knee pain (n=51) as well as age-matched controls (n=54) completed demographic, psychological and self-reported and experimental pain assessments along with a T1-weighted MRI scan. We estimated a brain-predicted age difference (brain-PAD) calculated as brain-predicted age minus chronological age. Analyses of covariances and correlations were used to determine associations of brain-PAD with pain and psychological variables. Age-matched controls had significantly "younger" brains for their age compared to individuals with the most severe knee pain (Bonferroni-p=0.023). Lower self-reported pain intensity and disability were significantly associated with a "younger" brain (p's<0.05). A "younger" brain was also significantly associated with lower mechanical and thermal pain sensitivity, greater endogenous pain inhibition, lower negative affect, lower in-vivo passive coping, and lower pain catastrophizing (p's<0.05). Our findings replicate and extend our previous findings in a larger diverse cohort of individuals with and without knee pain. Even in this younger sample, our findings suggest that chronic pain is associated with added "age-like" brain atrophy, and future studies are needed to determine whether a brain aging biomarker may help identify people with chronic pain who are at a greater risk of functional decline and poorer health outcomes. R01AG059809, R01AG067757. We have previously reported that older adults with chronic musculoskeletal pain had significantly older-appearing brains compared to older controls in a small, mainly Non-Hispanic white sample of individuals. We employed a machine-learning derived brain aging biomarker that compares voxel-wise gray and white matter volume images to a statistical model that accurately predicts chronological age from neuroimaging data in healthy people. The present study aimed to extend our previous findings to a larger, younger, and more racially diverse sample of individuals with knee osteoarthritis (OA) pain. Participants (mean age=58 years) with severe (n=100) and mild knee pain (n=51) as well as age-matched controls (n=54) completed demographic, psychological and self-reported and experimental pain assessments along with a T1-weighted MRI scan. We estimated a brain-predicted age difference (brain-PAD) calculated as brain-predicted age minus chronological age. Analyses of covariances and correlations were used to determine associations of brain-PAD with pain and psychological variables. Age-matched controls had significantly "younger" brains for their age compared to individuals with the most severe knee pain (Bonferroni-p=0.023). Lower self-reported pain intensity and disability were significantly associated with a "younger" brain (p's<0.05). A "younger" brain was also significantly associated with lower mechanical and thermal pain sensitivity, greater endogenous pain inhibition, lower negative affect, lower in-vivo passive coping, and lower pain catastrophizing (p's<0.05). Our findings replicate and extend our previous findings in a larger diverse cohort of individuals with and without knee pain. Even in this younger sample, our findings suggest that chronic pain is associated with added "age-like" brain atrophy, and future studies are needed to determine whether a brain aging biomarker may help identify people with chronic pain who are at a greater risk of functional decline and poorer health outcomes. R01AG059809, R01AG067757.