Introduction Chronic pain is a characteristic feature of functional gastrointestinal disorders (FGID), which are more prevalent in neurotic individuals. A recent meta-analysis has demonstrated that abnormalities in the parasympathetic nervous system (PNS) and hypothalamic pituitary adrenal (HPA) axis may be implicated in the pathogenesis of FGID.1 Moreover, accumulating evidence suggests that the PNS plays an important role in anti-nociception.

Methods 120 healthy subjects (68 males, median age 29.2 years, range 19–55 years) had personality and anxiety variables assessed using validated questionnaires. All subjects had validated autonomic nervous system (ANS) parameters measured at baseline and continuously thereafter. Seven painful visceral stimuli (mid-oesophageal balloon distension) followed by seven painful somatic stimuli (nail bed pressure) were administered to the subject's pain tolerance, with an inter-stimulus interval of two minutes. Subjects were asked to rate stimuli for pain intensity and unpleasantness. Serum cortisol was measured at baseline, following visceral and somatic stimulation. A cohort of 30 subjects (18 males, median age 29 years, range 20–54 years) had the study repeated at 1 year to evaluate reproducibility of responses.

Results Neuroticism correlated positively with trait anxiety (p<0.0001, r=0.52). Neuroticism was negatively correlated with the mean intensity of visceral and somatic stimuli (p=0.001, r=−0.38 and p=0.003, r=−0.45, respectively) but positively correlated with baseline, post-somatic and post-visceral pain cortisol (p<0.0001, r=0.42, p<0.0001, r=0.48 and p<0.0001, r=0.53, respectively). Hierarchical cluster analysis identified two psychophysiological clusters: Cluster 1 (n=59) were more neurotic (p<0.0001), less extrovert (p<0.0001), more anxious (p<0.0001), had lower resting PNS tone (p=0.01), higher baseline cortisol (p=0.006), habituated less to visceral and somatic pain (p=0.005 and p=0.04, respectively) with an increase in PNS tone to visceral and somatic pain (p<0.0001 and p<0.0001, respectively). Cluster 2 (n=61) had the opposite profile. In a demographically representative reproducibility cohort, intra–subject personality traits, baseline cortisol and ANS parameters were not significantly different between studies. Bland-Altman plots showed no proportional, magnitude-related, or systematic error between the degree and direction of ANS responses to visceral or somatic pain between studies.

Conclusion In health, two distinct and reproducible psychophysiological profiles exist in response to visceral and somatic pain. The clinical relevance and prevalence of these pain endophenotypes as vulnerability factors for FGID warrants further investigation.