Does Stress play a role in Central Sensitisation in Migraine? A Cross Sectional Analysis using Quantitative Sensory Testing and Questionnaires.

Tom Mulhearn, Dr. Aimie Peek, Prof. Trudy Rebbeck, Prof. Luke Henderson

Background

Migraine is a primary headache disorder with a significant disease burden. Clinically, it is broken into two groups based on migraine frequency, chronic migraine (>15 days/month) and episodic migraine (<15 days/month). A proportion of people with migraine develop unhelpful pain processing such as central sensitisation (CS). Stressful life events are postulated to be involved in the genesis of CS and migraine chronification. 

Aims

 Firstly to describe any differences in self-reported questionnaires and CS in chronic migraineurs, episodic migraineurs and healthy controls. Secondly to determine if there is a relationship between self-reported stress questionnaires and clinical evidence of CS. 

Methods

This was a cross-sectional study on 75 participants. Participants were recruited from neurology clinics and from within the community. Migraine groups were categorised as per the ICHD-3 classification. Participants completed stress-related questionnaires being the Childhood Abuse and Trauma Scale (CATS), Perceived Stress Scale (PSS), and the Depression Anxiety and Stress Scale (DASS-42) along with central sensitisation questionnaires, the Central Sensitisation inventory (CSI) and the Pain-Catastrophising-Scale (PCS). Quantitative sensory testing (QST), a validated clinical assessment tool for CS, was used as our criteria of CS by being outside normative ranges for 1 dynamic and 1 static test and a CSI score of >25.5. ANOVA or Mann-Whitney-U testing was conducted to answer aim 1 and logistic regression analysis to answer aim 2 in SPSS. 

Results

Both chronic and episodic migraineurs were significantly more centrally sensitised and reported greater levels of adulthood stress compared to healthy controls, but not between the migraine groups. There was no significant difference in childhood stress between migraineurs and controls. Univariate analysis showed stress questionnaires could predict CS, but when controlled for by the CSI and PCS on multivariate logistic regression analysis, they did not. 

Conclusion

Our data supports two main arguments within the literature, firstly that chronic and episodic migraine should be thought of as a continuum of migraine rather than two separate pathologies. Secondly that the relationship between stress and CS is likely complex given our result and the heterogeneity of effects reported in literature.