RESUMEN
BACKGROUND: The mechanisms driving osteoarthritic pain remain poorly understood, but there is increasing evidence for a role of the central nervous system in the chronification of pain. We used functional magnetic resonance imaging to investigate the influence of a model of unilateral knee osteoarthritis on nociceptive processing. RESULTS: Four to five weeks post intra-articular injection of monosodium iodoacetate (MIA, 1 mg) into the left knee, Sprague Dawley rats were anesthetized for functional magnetic resonance imaging studies to characterize the neural response to a noxious stimulus (intra-articular capsaicin injection). In a two-arm cross-over design, 5 µM/50 µl capsaicin was injected into either the left knee (n = 8, CAPS-MIA) or right control knee (n = 8, CAPS-CON), preceded by contralateral vehicle (SAL) injection. To assess neural correlates of mechanical hyperalgesia, hindpaws were stimulated with von Frey hairs (8 g: MIA; 15 g: control knee, based on behavioral withdrawal responses). The CAPS-MIA group exhibited significant activation of the periaqueductal gray, unilateral thalamus and bilateral mensencephalon, superior-colliculus, and hippocampus, with no significant activation in the other groups/conditions. Capsaicin injection increased functional connectivity in the mid-brain network and mediodorsal thalamic nucleus, hippocampus, and globus pallidus, which was significantly stronger in CAPS-MIA compared to CAPS-CON groups. Mechanical stimulation of the hyperalgesic (ipsilateral to MIA knee) and normalgesic (contralateral) hindpaws evoked qualitatively different brain activation with more widespread brainstem and anterior cingulate (ACC) activation when stimulating the hyperalgesic paw, and clearer frontal sensory activation from the normalgesic paw. CONCLUSIONS: We provide evidence for modulation of nociceptive processing in a chronic knee osteoarthritis pain model with stronger brain activation and alteration of brain networks induced by the pro-nociceptive stimulus. We also report a shift to a medial pain activation pattern following stimulation of the hyperalgesic hindpaw. Taken together, our data support altered neural pain processing as a result of peripheral and central pain sensitization in this model.