The neuroinflammatory component of negative affect in patients with chronic pain

Abstract

Negative affect (NA) is a significant cause of disability for chronic pain patients. While little is known about the mechanism underlying pain-comorbid NA, previous studies have implicated neuroinflammation in the pathophysiology of both depression and chronic pain. Here, we tested the hypothesis that NA in pain patients is linked to elevations in the brain levels of the glial marker 18 kDa translocator protein (TSPO), and changes in functional connectivity. 25 cLBP patients (42.4 ± 13 years old; 13F, 12M) with chronic low back pain (cLBP) and 27 healthy control subjects (48.9 ± 13 years old; 14F, 13M) received an integrated (i.e., simultaneous) positron emission tomography (PET)/magnetic resonance imaging (MRI) brain scan with the second-generation TSPO ligand [¹¹C]PBR28. The relationship between [¹¹C]PBR28 signal and NA was assessed first with regression analyses against Beck Depression Inventory (BDI) scores in patients, and then by comparing cLBP patients with little-to-no, or mild-to-moderate depression against healthy controls. Further, the relationship between PET signal, BDI and frontolimbic functional connectivity was evaluated in patients with mediation models. PET signal was positively associated with BDI scores in patients, and significantly elevated in patients with mild-to-moderate (but not low) depression compared with controls, in anterior middle and pregenual anterior cingulate cortices (aMCC, pgACC). In the pgACC, PET signal was also associated with this region's functional connectivity to the dorsolateral PFC (pgACC-dlPFC), and mediated of the association between pgACC-dlPFC connectivity and BDI. These observations support a role for glial activation in pain-comorbid NA, identifying in neuroinflammation a potential therapeutic target for this condition.

Figure 1.. ROI [¹¹C]PBR28 signal is associated with depressive symptoms.

Scatterplots display the relationship between [¹¹C]PBR28 SUVR in each ROI (displayed in green) and BDI in cLBP patients (n=25). All data have been adjusted for TSPO polymorphism. BDI – Beck Depression Inventory; ACC – anterior cingulate cortex; PFC – prefrontal cortex

Figure 2.. Voxelwise [¹¹C]PBR28 signal is associated with depressive symptoms and elevated in patients with mild-to-moderate depression.

A. Results from the voxelwise analysis showing clusters where [¹¹C]PBR28 SUVR is significantly positively associated with BDI. B. For visualization purposes, average SUVR from the aMCC and pgACC clusters in panel A are plotted against BDI, both adjusted for TSPO polymorphism. C. Results from the ANCOVA analysis comparing average aMCC and pgACC SUVR between cLBP patients with little-to-no depression, mild-to-moderate depression, and controls. P-values represent results from post-hoc Dunnett’s tests comparing both patient subgroups against to controls. All values have been adjusted for age, injected dose, and TSPO polymorphism.

Figure 3.. Frontocingulate connectivity is associated with pgACC [¹¹C]PBR28 signal, and with BDI

A. Results from the voxelwise nonparametric permutation analysis showing a significant positive association between pgACC [¹¹C]PBR28 SUVR and functional connectivity between pgACC and right dlPFC. B. For visualization purposes, the average z-statistic was extracted from the dlPFC cluster in panel A and plotted against pgACC SUVR. All data has been adjusted for TSPO polymorphism. C. A scatterplot shows the regression between average z-statistic connectivity extracted from dlPFC in Panel A and BDI.

Figure 4.. pgACC [¹¹C]PBR28 signal mediates the relationship between pgACC-dlPFC connectivity and BDI

A bootstrapped mediation analysis revealed that pgACC significantly mediated the relationship between pgACC-dlPFC connectivity and BDI. Values within parentheses represent bootstrap standard errors for each path. ^#p = 0.054, *p < 0.05, **p < 0.01.