Article Figures & Data
Figures
- Figure 1.
Midbrain PAG seed region, pain block, and sliding-window connectivity design. A, With coordinates derived from a previous investigation utilizing this same dataset, a 1 mm sphere surrounding the PAG was generated and used as a seed region for subsequent connectivity analyses. B, Visual representation of the bock analysis used to assess corticobrainstem functional connectivity during periods of spontaneously high versus spontaneously low pain. In this analysis, connectivity between the PAG seed and each cortical voxel was assessed within a 50 s period overlying participants lowest reported pain (blue shaded bar) against a 50 s period overlying participants’ highest reported pain (orange shaded bar). Pain intensity was assessed using a VAS from 0, no pain, to 10, worst pain imaginable. C, In addition to the block design, a sliding-window analysis was performed where PAG connectivity was assessed during a 50 s period (Window 1 light gray box) which was then shifted forward by one volume (Window 2 dark gray box). This resulted in 336 contrast images representing 50 s mean PAG connectivity values across the entire 12 min scan.
- Figure 2.
Ongoing pain intensity ratings and pain intensity variability in fluctuating (n = 13) and stable (n = 11) pain groups. A, Trace of VAS responses recorded throughout the entire 12 min fMRI scan series. All participants fit the diagnostic criteria for trigeminal neuropathic pain; however, note that multiple spontaneous shifts in perceived pain are observed in only our fluctuating pain group (left plot). B, Mean pain intensity and pain rating variability in fluctuating (orange bars) and stable (blue bars) pain groups across the 12 min scan. Fluctuating pain participants exhibited significantly higher mean intensity and variability of pain than stable participants. C, Linear regression analysis comparing participant mean pain intensity and variability throughout the 12 min scan period. No significant interaction between these variables was observed despite variable participants demonstrating both greater intensity and variability than stable pain participants.
- Figure 3.
Significant differences in PAG connectivity strength during periods of high versus low pain in fluctuating pain participants (n = 13). Across the 12 min scanning window, the 25 volumes where participants rated their pain as the most intense were entered into a second-level, paired design with the 25 volumes where participants rated their pain the least intense. The top panel shows significant cortical clusters with greater midbrain PAG connectivity during high versus low pain (red color scale) and vice versa (blue color scale). The slice locations in MNI space are shown at the top right of each slice. The middle panel shows plots of mean (±SEM) PAG connectivity extracted from the above clusters and compared between fluctuating (high- and low-blocks) and stable (evenly dispersed blocks) participants. Note that the magnitude of PAG connectivity change in these sites only differs in those participants where pain fluctuates. Within the bottom panel is a post hoc analysis of the dose-dependent response within these clusters, as across the scan course our fluctuating pain group exhibited periods of perceived pain between the lowest and highest reported blocks. Note that in the hippocampus, dlPFC, and PCC, PAG connectivity appears to particularly track with the magnitude of pain only in those participants displaying fluctuating pain. Please see Extended Data Figure 3-1.
- Figure 4.
Cortical region demonstrating ongoing PAG connectivity change relating to fluctuations in pain intensity perceived across the 12 min scan period. The left overlay shows firstly the midbrain PAG seed (top box) sectioned within a render also showing a significant cluster in the contralateral (to side of pain) dlPFC where PAG connectivity was anticorrelated with perceived pain throughout the entire scanning period in fluctuating pain patients. The right panel provides visual examples of two fluctuating pain PTN participants where PAG–dlPFC connectivity (orange line) was continuously anticorrelated with perceived pain (blue line) over the 12 min scanning window. FC, functional connectivity; VAS, visual analog scale; dlPFC, dorsolateral prefrontal cortex.
Tables
Age (years) Sex Side of pain Diary pain intensity (VAS) Diary pain variability (SD) Pain duration (months) Current medication use 1 29 F Left 5.3 1.5 49 - 2 45 M Left 3.1 0.4 21 - 3 35 M Right 1.5 0.7 96 PEA 4 78 F Left 1.9 0.7 314 Paracetamol, Ostelin, Pristiq, Somac 5 47 F Right 2.1 0.4 13 PEA 6 34 M Right 3.7 1.7 10 PEA, Mersyndol 7 66 M Left 3.9 0.6 48 - 8 44 F Right - - 5 Valium, Panadeine Forte 9 47 F Right 4.1 1.6 7 - 10 44 F Left 8.1 0.4 75 PEA 11 46 F Left 7.5 2.5 27 Endep 12 28 F Right 5.8 0.3 52 - 13 25 M Right 4.5 1.7 8 Panadeine Forte 14 69 F Right 3.9 0.4 259 Endep, Lyrica, Oroxine, Crestor, Effexor, folic acid, prednisone 15 45 F Bilateral 2.8 2.8 24 Duloxetine 16 66 M Right 1.8 0.7 81 Endep 17 36 F Left 1.9 0.6 37 PEA 18 40 F Left 5.8 0.3 33 - 19 58 F Left 4.0 0.4 11 Namipril, Zoloft, Panadol, PEA 20 35 M Left 4.2 0.7 60 - 21 67 M Bilateral 2.0 0.9 96 Endep 22 51 F Left 1.5 0.3 13 Mobic, PEA 23 35 F Right 3.4 1.0 15 Amitriptyline 24 64 F Bilateral - - 144 - Participants 1–13 were placed in the fluctuating pain group, and Participants 14–23 were placed in the stable pain group. PEA, palmitoylethanolamide; VAS, visual analog scale; SD, standard deviation.
- Table 2.
MNI coordinates, cluster sizes, and significance values for clusters showing midbrain PAG connectivity differences during spontaneous low- versus high-pain intensity blocks in fluctuating pain participants (n = 13)
Region MNI coordinate Cluster size (kE) t value z value X Y Z Highest > lowest pain Ipsilateral posterior insula −36 −32 8 21 6.08 4.03 PCC −6 −36 30 18 5.16 3.68 Ipsilateral hippocampus −52 0 2 11 3.86 3.05 Ipsilateral dlPFC −20 34 28 11 3.77 3.00 Highest < lowest pain sgACC 6 −28 −6 12 3.86 3.05 - Table 3.
MNI coordinates, cluster sizes, and significance values for clusters showing correlations pain intensity and PAG connectivity across the entire 12 min scan length in fluctuating pain participants (n = 13)
Region MNI coordinate Cluster size (kE) t value z value X Y Z Negative correlation Contralateral dlPFC 32 40 16 21 4.55 3.27
Figure 3-1
Mean connectivity and two-factor ANOVA interaction significance within cortical regions identified as altering in connectivity with the midbrain periaqueductal gray between low and high pain periods in fluctuating pain PTN participants. Note that only fluctuating PTN pain participants that displayed pain intensity ratings in all four pain blocks were included in this post-hoc analysis. Ipsi = ipsilateral, PI = posterior insula, PCC = posterior cingulate cortex, dlPFC = dorsolateral prefrontal cortex, sgACC = subgenual anterior cingulate cortex. Download Figure 3-1, DOCX file.
In this issue
