Article Text

Download PDFPDF

Cerebellum is more concerned about visceral than somatic pain
  1. Jens Claassen1,
  2. Laura Ricarda Koenen2,
  3. Thomas M Ernst1,
  4. Franziska Labrenz2,
  5. Nina Theysohn3,
  6. Michael Forsting3,
  7. Ulrike Bingel1,
  8. Dagmar Timmann1,
  9. Sigrid Elsenbruch2
  1. 1Department of Neurology, Essen University Hospital, University of Duisburg-Essen, Essen, Germany
  2. 2Institute of Medical Psychology and Behavioral Immunobiology, Essen University Hospital, University of Duisburg-Essen, Essen, Germany
  3. 3Institute for Diagnostic and Interventional Radiology and Neuroradiology, Essen University Hospital, University of Duisburg-Essen, Essen, Germany
  1. Correspondence to Professor Dagmar Timmann, Department of Neurology, Essen University Hospital, University of Duisburg-Essen, 45147 Essen, Germany; dagmar.timmann-braun{at}

Statistics from

Chronic pain disorders are extremely common, including chronic back pain and headaches, but also chronic visceral pain disorders, such as irritable bowel syndrome. Treatment is notoriously difficult. A detailed understanding of the neural pain circuitry is a prerequisite for the development of new treatment options. The cerebellum has become an interesting target for non-invasive and invasive brain stimulations in a wide range of brain disorders and may be a future option in treating chronic pain. The possible contribution of the cerebellum to the pathophysiology of chronic pain has become of interest only recently.1 Although the cerebellum has frequently been shown to respond to painful stimuli, knowledge about the specific contributions of the cerebellum to pain processing remains elusive. Electrophysiological studies in rodents provide evidence that the cerebellum receives afferent input coming from cutaneous and visceral nociceptors.2 Our group and others have found that neural processing of somatic and visceral pain is partly overlapping but reveals also significant differences.3 As yet, however, it is unknown to what extent cerebellar responses differ between visceral and somatic pain. To address this question, we compared pain-related activations of the cerebellum between carefully matched rectal distensions and cutaneous heat stimuli.

Functional MRI (fMRI) data were acquired in 22 healthy female participants as part of a previous study.3 fMRI data were reanalysed using a normalising method optimised for the cerebellum. Participants were on average 24.4±0.6 years old with a mean body mass index of 21.9±0.5. Questionnaires confirmed lack of gastrointestinal, anxiety or depressive symptoms. For the application of visceral pain, pressure-controlled rectal distensions were carried out. For somatic pain, cutaneous thermal heat stimuli were applied to the left ventral forearm. Stimulus intensities were individually calibrated and matched to pain intensity. Likewise, heating and cooling times of the thermode were individually matched to the barostat inflation and deflation times. Participants received 10 visceral and 10 somatic pain stimuli in a pseudorandomised order (stimulus duration=30 s). Pain stimuli were cued with modality-specific visual symbols that were presented for 9–11 s (jittered) prior to painful stimulation. Trial-by-trial pain intensity ratings were obtained using a digitised visual analogue scale (VAS). While stimuli were well matched at the beginning, a mild habituation was observed for somatic stimuli, resulting in slightly different mean pain intensity ratings (visceral: 73.7±2.5 mm; somatic: 61.9±4.2 mm on 0–100 mm VAS scales; see Koenen et al3 for details). fMRI images were acquired on a 3T MR scanner using a multiecho echo-planar imaging sequence (repetition time 2.5 s). They were analysed using SPM12 (Wellcome Trust Centre for Neuroimaging, UCL, London, UK) implemented in MATLAB R2015a (MathWorks, Sherborn, MA, USA). Normalisation of the cerebellum was performed using the spatially unbiased template (SUIT) toolbox ( Images were smoothed with an isotropic Gaussian kernel (6 mm full width at half maximum). A high-pass filter with a cut-off period of 180 s was used.

First-level analysis of preprocessed data was performed using a general linear model. Three different time windows were considered: (1) visual cue presentation (anticipation phase, 9–11 s); (2) thermode and balloon inflation, respectively (ascending phase, 10 s); and (3) constant balloon pressure and cutaneous heat stimulation, respectively (plateau phase, 10 s). To address differences across modalities, one-sample t-tests were computed based on first-level differential contrasts for visceral compared with somatic pain stimulation. Individual differences in mean pain intensity ratings were considered as covariate of no-interest. Clusters are reported at p<0.05 (family-wise error (FWE) corrected), based on the threshold-free cluster-enhancement (TFCE) statistic image (

During cued pain anticipation, no significant differences were detected between modalities. During painful stimulation, visceral pain resulted in significantly enhanced cerebellar activation compared with somatic pain (figure 1). These differences were more pronounced during the ascending phase of pain stimulation with significantly enhanced cerebellar activation to visceral pain in most of the vermis, with a focus on vermal lobule VI (figure 1A). It was also significantly enhanced within the cerebellar hemispheres bilaterally with a focus on lobule VI (extending into lobule V and Crus I) and lobule VIII (extending into lobule VIIb). In the plateau phase, the differences between visceral and somatic pain stimulations were restricted to the vermis and left lobule VI with a small extension to Crus I (figure 1B).

Figure 1

Differences of cerebellar activation comparing visceral and somatic pain stimuli (visceral pain>somatic pain) in ascending (A) and plateau (B) phases. Clusters are shown, which are significant at p<0.05 (family-wise error (FWE) corrected), based on the TFCE statistic image. Individual differences in mean pain intensity ratings were considered as covariates of no interest. Functional MRI data are displayed on a surface-based flatmap of the cerebellum ( (C) Cluster characteristics are summarised. In the two main clusters, only the three largest maxima are listed. L, left; R, right; SUIT, spatially unbiased template; TFCE, threshold-free cluster enhancement.

Increased cerebellar activation related to acute visceral compared with somatic pain provides further evidence that the neural processing of these two pain modalities differ.3 Modality-specific differences in cerebellar fMRI signal were more pronounced during the ascending phase of stimulation, a phase that is characterised by markedly changing afferent sensory input. Our data fit to the observation that the cerebellum is involved in the detection of changes in sensory input.4 Large parts of the cerebellum that were differentially activated belong to the sensorimotor cerebellum (ie, anterior cerebellar lobe extending into lobule VI, and lobule VIII in the posterior cerebellar lobe).5 These regions include the known foot, hand and facial representations within the cerebellum, which is—at first glance—an unexpected finding. Increased cerebellar activation to visceral pain may indicate more demanding sensory discrimination because interoceptive stimuli are more difficult to localise than exteroceptive stimuli and result in a more diffuse perceptual experience. Increased differential activation in the vermis may also reflect increased autonomic responses to visceral pain, in line with evidence that they are perceived as more unpleasant and fear provoking than somatic pain stimuli.3 The latter may also explain increased visceral pain-related activation in cerebellar parts known to contribute to cognitive and emotional functions (ie, parts of lobule VI and Crus I, but also lobules VIIb and IX/X).5 Overall, the cerebellum has to be added to the neural structures, which are more active in the processing of visceral compared with somatic pain, likely because of the unique biological salience of interoceptive, visceral pain. Adding to the still growing spectrum of cerebellar functions, the cerebellum may be a new component of the brain–gut axis that remains to be explored in future studies.


View Abstract


  • DT and SE contributed equally.

  • Contributors JC, LRK, TME, FL, UB, DT and SE conceived and designed the research; LRK and NT performed the experiments; JC, LRK, TME, FL, MF, UB, DT and SE analysed the data; JC, LRK, TME, FL, NT, MF, UB, DT and SE interpreted the results of the experiments; JC, LRK, TME, FL and DT prepared the figures; JC, LRK, TME, FL, DT and SE drafted the manuscript; JC, LRK, TME, FL, NT, MF, UB, DT and SE edited and revised the manuscript; JC, LRK, TME, FL, NT, MF, UB, DT and SE approved the final version of the manuscript.

  • Funding This study was funded by the Deutsche Forschungsgemeinschaft (German Research Foundation), project number 316803389, SFB 1280, subprojects A05, A10 and A11. The funding agency had no role in the conception, analysis or interpretation of the data.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.