Background & Aims

Endometriosis is a chronic inflammatory condition affecting 1 in 10 women of reproductive age, with one of the main symptoms being chronic pelvic pain. Previously, we have shown that 75% of women with endometriosis-associated pain (EAP) have a possible neuropathic component if screened with the painDETECT (PD) questionnaire [1]. To explore the neural basis of this component, we used functional magnetic resonance imaging to identify brain regions in which resting state (RS) functional connectivity with other regions scaled with the individuals’ painDETECT score as a proxy for neuropathic-like mechanisms [2,3,4]. In addition, we compared these RS findings to regions of interest (ROIs) from task-based fMRI during stimulation using punctate probes of the abdomen, again correlated with the painDETECT score. We recruited women with confirmed endometriosis prior to repeat surgery for endometriosis-associated pain for this study.

Methods

Participants underwent RS and task-based fMRI in a 3T MR scanner (Siemens) and completed questionnaires including PD. Imaging analyses were carried out using FSL and Matlab. Preprocessing of RS data included physiological noise modelling and single-subject independent component analysis (ICA)(MELODIC) with manual labelling. Punctate fMRI analysis was carried out as per standard pipelines to identify where activation in response to punctate correlated with PD.
Multiple RS group level analyses comprised:
A) Seed-based whole brain analysis using dual regression with key ROI identified in the task-based scan.
B) FSLNets to determine connectivity of a group of ROIs found to be correlated with painDETECT in the task-based scan.
C) Voxel-based approach utilising group ICA followed by dual regression.
D) Static functional connectivity (sFC) between key regions of the salience network and default mode network[2]
All resulting connectivity measures were correlated with demeaned PD scores.

Results

Of the 28 women recruited, N= 26 RS datasets and N= 15 task-based datasets passed quality control checks. PD scores ranged between 1-28 (mean: 15). painDETECT scores showed no significant correlation with any psychological measure.
Analyses of punctate-induced activation revealed that PD scores were correlated negatively with activity in the pontine reticular formation (pRF; brainstem masked; Z>3.1, p<0.05). Significant negative correlations were also found with left thalamus, left insula & midcingulate cortex on whole brain analysis, at lower threshold (Z>2.3, p<0.05).
In contrast, our RS analyses using the pRF ROI showed no significant connectivity with any other brain regions at either group mean or correlation with PD. Mirroring this, Neurosynth.org, a healthy participant database, also showed no resting connectivity of pRF. Likewise, a voxel-based ICA analysis and sFC of the salience network showed no meaningful significant correlation with PD.

Conclusions

Whilst the PD score may be useful clinically to stratify patients with endometriosis-associated pain, our data suggests that the variation in these scores is not reflected in resting state connectivity. This is contrary to what has been seen in other conditions with a neuropathic component.
In response to evoked pain PD is negatively correlated with activity in regions implicated in descending modulation of pain, potentially representing a central network maintaining peripheral neuropathic pain. However, despite our thorough analysis of the resting state data, using a variety of indices, we found no evidence for changes in RS activity related to PD. This argues against a role for altered RS functional connectivity in neuropathic-like pain in EAP.
Women with endometriosis-associated pain describe a wide variety of pain symptoms and these findings suggest there may be important differences in the mechanisms generating and maintaining provoked and unprovoked pain.

References

[1]Coxon L, Wiech K and Vincent K (2021) Is There a Neuropathic-Like Component to Endometriosis-Associated Pain? Results From a Large Cohort Questionnaire Study. Front. Pain Res. 2:743812. doi: 10.3389/fpain.2021.743812
[2]Bosma RL, Kim JA, Cheng JC, et al. Dynamic pain connectome functional connectivity and oscillations reflect multiple sclerosis pain. Pain. 2018. doi:10.1097/j.pain.0000000000001332
[3]Soni A, Wanigasekera V, Mezue M, et al. Central Sensitization in Knee
Osteoarthritis: Relating Presurgical Brainstem Neuroimaging and PainDETECT-Based
Patient Stratification to Arthroplasty Outcome. Arthritis Rheumatol. 2019.
doi:10.1002/art.40749
[4]Segerdahl AR, Themistocleous AC, Fido D, Bennett DL, Tracey I. A brain-based
pain facilitation mechanism contributes to painful diabetic polyneuropathy.
Brain. 2018. doi:10.1093/brain/awx337

Presenting Author

Lydia Coxon

Poster Authors

Lydia Coxon

DPhil BA

University of Oxford

Lead Author

Katja Wiech

University of Oxford

Lead Author

Miriam Szabo

University of Oxford

Lead Author

Christian Becker

University of Oxford

Lead Author

Krina Zondervan

University of Oxford

Lead Author

Katy Vincent DPhil BSc MBBS MRCOG

University of Oxford

Lead Author

Topics

  • Specific Pain Conditions/Pain in Specific Populations: Abdominal and Pelvic Pain