Abstract
Most people experience acute pain as a temporary condition, while a small subset develops chronic pain. The role of pain-related circuits driving this transition remains unclear. Using UK Biobank data and an independent dataset (OpenPain), we analyzed MRI scans of participants with acute musculoskeletal pain (n=160), categorizing them based on those who recovered (AMPR) and those who developed chronic pain (CMPO) later. A machine learning model was applied to predict follow-up outcomes in two independent validation cohorts. AMPR participants showed increased functional connectivity (FC) between the ventral posterolateral thalamus (VPL-Thal) and left dorsolateral prefrontal cortex (DLPFC) compared to CMPO. Increased right NAc-mPFC FC was found in CMPO participants. These FC changes predicted pain chronification with AUCs of [0.74-0.83] across validation cohorts. Our results suggest that multiple circuits, particularly a newly observed VPL -left DLPFC pathway, alongside a previously established right NAc-mPFC pathway are involved in CMP development. These findings may inform the development of more innovative prevention strategies. PERSPECTIVE: This study identifies distinct brain connectivity patterns that differentiate acute pain outcomes (recovery vs. chronic pain development). The VPL-Thal-DLPFC and NAc-mPFC circuits underlie pain chronification, which enables early prediction and may guide targeted interventions to prevent transition from acute to chronic musculoskeletal pain.</p>