Rofo 2009; 181 - A8
DOI: 10.1055/s-0028-1124039

Modulation of pain processing with physical exercise in athletes

L Scheef 1, J Jankowski 1, M Daamen 1, G Weyer 1, M Klingenberg 1, J Renner 1, S Mueckter 1, M Wagner 2, H Boecker 1
  • 1Functional Neuroimaging Group, Dep. of Radiology, University of Bonn, Bonn, Germany
  • 2Department of Psychiatry, University of Bonn, Bonn, Germany

Purpose: Physical exercise is known to promote various psycho-physiological effects, among these mood elevation, stress reduction, anxiolysis, and modulation of pain perception with reduced sensitivity to pain stimuli. How the modulation of pain perception in athletes is mediated via the distributed human pain matrix is not yet understood. Assuming underlying central opioidergic mechanisms, we expected physical exercise (two hours outdoor RUN) to affect the affective dimensions of pain processing on the behavioral and the systemic (brain activation) level, as compared to a control condition (two hours outdoor WALK).

Materials and Methods: 18 right-handed athletes (mean age 39.2 years, range 23–59 years, mean training distance 58.7km/week, range 30–130km/week) were included in this study. All participants underwent four fMRI-examinations on two separate occasions (order randomly assigned: group I: walk-run, group II: run-walk). Eight volunteers were assigned to group I, ten to group II. Before and after the two hours aerobic exercise, a thermal heat-pain fMRI-experiment was performed using a blocked design. The paradigm consisted of 2×5 blocks of 20sec stimulation phases (5x undulating heat pain, 46.5°C/5x non-painful heat, 43.0°C). Each stimulation phase was followed by a 30s baseline phase (33.0°C). Before and after each pain stimulation the pain intensity was recorded using a visual-analog scale (VAS), and the affective pain component was measured using the McGill Affective Pain Questionnaire Ratings. All examinations were performed on a 3T-Philips Achieva whole body scanner using a 8-channel SENSE head coil. The sequence parameter were as follows: T2*-weighted gradient echo single-shot EPI, TR/TE/Flip=2600ms/35ms/90°, acquired spatial resolution 3.6×3.6×3.6mm3. The fMRI-data sets were preprocessed and statistically analyzed (random-effect analysis, two sample t-test) using SPM5. The preprocessing steps included slice-time correction, realignment, normalization and isotropic smoothing (8×8x8mm3). The statistical analysis of the behavioral data was performed using SPSS 16.0 (repeated measures ANOVA, Post hoc comparisons using a one-sided t-test).

Results: The covered distance during the walk condition was in average 10.3±1.8km at an average heart rate of 81.5±4.6/min. During the aerobic exercise 23.0±2.7km were covered at an average heart rate of 146.4±10.5/min. No pain experience was reported during exercise. The VAS pain intensity ratings after the fMRI sessions did not show any significant effect regarding exercise type (RUN vs. WALK condition) but a trend for increased pain intensity in the post exercise type (p=0.083). The McGill Affective Pain Questionnaire did not show a significant dependency on exercise type (RUN vs. WALK condition) or time (pre. vs. post scan). However, the affective pain ratings showed a significant time x exercise interaction: The post hoc comparison revealed a significant decrease in the pre vs. post RUN condition (p<0.05, one-sided t-Test). In all four scans we were able to delineate the affective-motivational and sensory-discriminative part of the pain network. The contrast PreWALK – PostWALK (Pain > Heat) did not show any significant effect, whereas the contrast PreRUN – PostRUN (Pain > Heat) was significant in the perigenual anterior cingulate cortex (ACC, p<0.001, uncorrected; p<0.05 FWE-corrected for ACC-ROI). The interaction term: (PRE-POST)RUN > (PRE-POST)WALK was significant for the antinociceptive pathway, i.e. the perigenual ACC and the periaqueductal gray (p<0.001 and SVC (10 voxel sphere).

Conclusion: Endurance exercise in athletes is capable of specifically modulating the affective dimensions of pain processing. This effect is mediated via the descending antinociceptive pathway. The specific effect on the affective dimension of pain and loco-regional convergence with previous ligand PET studies suggests an underlying opioidergic effect promoted by exercise. This neurobiological evidence may provide a basis for more concerted use of physical exercise as a therapeutic option in patients with chronic pain states.