The Relationship of Overlapping Affective and Somatization Symptoms With Homeostatic Afferent Processing Network Activations in Irritable Bowel Syndrome (IBS)

Abstract

Background: Brain activation differences in response to painful visceral stimuli previously have been reported in IBS compared to healthy controls, and often are observed within the homeostatic afferent processing network (HAPN). While HAPN regions are involved in emotional and cognitive responses to pain (Mayer E., et al. Gastroenterol 2006), many of these same brain regions also are also implicated in affective disorder pathogenesis. IBS brain activation patterns with noxious stimulation are inconsistent across studies, possibly in part due to the psychiatric heterogeneity in IBS. We hypothesized that affective and somatization symptoms may influence observed abnormal HAPN pain responses in IBS subjects. Methods: 16 right-handed female IBS subjects (45.8 ±10.3 yr old) and 16 healthy controls (40.7 ±10.7 yr old) completed the PHQ-15 somatization index and recent affective symptom measures (Beck Depression (BDI) and Anxiety (BDI) Inventories). All subjects underwent functional MRI (fMRI) imaging in a Seimens Trio 3.0T scanner, during randomized visceral stimulation (rectal balloon distension) at low(LP, 25-mmHg) and higher-pressure (HP, 50-mmHg) levels. Regions where IBS response to rectal distention was significantly different from controls were identified in ANOVA analyses. Linear regression models subsequently were developed to assess affective and somatization symptoms as predictors of HAPN activations, and mediation analysis explored affective symptoms as potential intervening variables in the relationship between somatization and brain activations. Results: BDI, BAI, and PHQ-15 scores all were higher in IBS vs. control subjects (p 0.70, p<0.001). Significant IBS-control activation differences were identified in several HAPN regions, including the L amygdala, pregenual/rostral ACC (BA 32), and BA 10, as well as B anterior/posterior insula, middle frontal gyrus (BA 8/9, 47) and thalamus (p <0.01 for each). In separate regression models, BAI scores predicted HAPN activation in the L amygdala, pregenual ACC, mid/posterior insula, BA 8/9, BA10, and R BA 47, posterior insula, and BA 9. PHQ-15 scores predicted HAPN activations in these same regions except L BA 10 (BDI scores predicted activation only in the L amygdala and BA32.) Mediation analyses failed to identify either anxiety or depression as intermediates in the observed somatization-HAPN activation relationship. Discussion: IBS-control HAPN activation differences were predicted by anxiety and somatization, and to a limited extent by depression symptoms. Despite their high correlation, anxiety and somatization appear to independently influence HAPN activations. Routine measurement of recent somatization and affective symptoms, particularly anxiety, may be important in the interpretation of future IBS neuroimaging studies.