The periaqueductal grey in chronic low back pain: dysregulated metabolites and function

Mechanisms underlying chronic pain are insufficiently understood. Preclinical evidence suggests a potential contribution of excitatory glutamatergic and inhibitory GABAergic imbalances in pain-relevant brain areas, such as a lower excitatory/inhibitory tone in the brainstem periaqueductal grey (PAG). This cross-sectional magnetic resonance spectroscopy (MRS) study investigated whether a lower excitatory/inhibitory tone is also observed in the PAG of patients with non-specific chronic low back pain (CLBP) and whether this would relate to altered psychophysical measures of descending pain modulation and experimental pressure pain sensitivity. Specifically, the ratio between pooled glutamate and glutamine and GABA levels (Glx/GABA), Glx and GABA in the PAG were compared between CLBP patients and pain-free controls. Further, associations of Glx/GABA with conditioned pain modulation (CPM) effects and pressure pain thresholds (PPTs) were assessed. MRS was acquired on a 3T Philipps MR system using a point-resolved spectroscopy sequence optimized with very selective saturation pulses (OVERPRESS) and voxel-based flip angle calibration in a 1.1 mL volume of interest. Data from 41 CLBP patients (median [interquartile range]: 54 years [41 - 65], 22 females) and 29 age- and sex-matched controls (47 years [34 - 67], 17 females) fulfilled MRS quality criteria. CPM and PPTs were assessed at the lower back as most painful area and the non-dominant hand as pain-free control area. The CPM paradigm consisted of PPTs applied before, during (parallel CPM effect) and after a cold water bath and an ambient temperature water bath as control paradigm to identify true CPM effects. In the PAG of CLBP patients, a lower Glx/GABA ratio, i.e. a lower excitatory/inhibitory tone, was observed (P = 0.002, partial {eta}2 = 0.14) driven by decreased Glx (P = 0.012, partial {eta}2 = 0.11) and increased GABA (P = 0.038, d = 0.46). CLBP patients showed disrupted associations between Glx/GABA and PPTs compared to controls in both areas (lower back: P = 0.004, partial {eta}2 = 0.12; hand: P = 0.002, partial {eta}2 = 0.16). In controls, lower Glx/GABA was associated with lower PPTs (lower back: r = 0.48, P = 0.009, hand: r = 0.53, P = 0.003), but this link was missing in CLBP patients (rs > -0.23, Ps > 0.150). Additionally, CLBP patients with more severe clinical pain showed smaller CPM effects at the hand (rho = 0.54, P = 0.003). These findings suggest a dysfunction of the PAG in patients with CLBP and might indicate altered descending inhibition of deep tissue afferents.