Anorexia nervosa (AN) is a serious mental illness with one of the highest mortality rates of all psychiatric disorders. It is characterized by reduced caloric intake, pre-meal anxiety and avoidance of food; behaviors that often persist after weight restoration. How the experience of eating provokes such anxiety in anorexia nervosa is unknown. Altered anxiety expression has been suggested as one explanation, on the basis that anxiety disorders are well known antecedents to AN, are frequently comorbid, and due to the increased aggregation of anxiety disorders among first degree family members of affected individuals. We propose that the altered processing of interoceptive signals is an important mechanism contributing to the expression of meal associated anxiety and dysfunctional eating behaviors in AN, and that determination of the processes contributing to this dysregulation will yield novel insights into the illness pathophysiology. To evaluate the neurobiological underpinnings of meal associated anxiety and interoception, the current proposal will investigate how individuals with AN experience cardiorespiratory sensations during meal anticipation relative to two control groups: age and weight-matched healthy comparison women, and an anxious comparison group of women with generalized anxiety disorder (GAD). Cardiorespiratory interoception and meal anxiety will be assessed using a validated protocol of intravenous infusions of isoproterenol and saline, during the pre-meal anticipatory time period. Isoproterenol, a rapid peripherally acting sympathomimetic agent, is a reliable method to measure changes in cardiorespiratory sensation and the double-blinded bolus infusion approach was developed by the PI. To understand the neural processes underlying this interoceptive phenotype the PI has successfully adapted the isoproterenol infusion paradigm to the functional MRI environment, and proposes using Arterial Spin Labeling (ASL) to identify cerebral blood flow changes associated with interoceptive stimulation. This pharmacological-fMRI (phMRI) approach is optimal for identifying changes in brain activity induced by peripherally induced sensation.
Aims 1 and 2 of the research will identify which interoceptive biomarkers are similar and different between patients with AN and GAD. Once identified, these neural interoceptive biomarkers will be used to predict illness outcomes at one year, for each patient group (Aim 3). At the conclusion of these studies, we will know whether interoceptive-based biomarkers can be used as predictors of poor mental health outcomes in individuals with AN or GAD. Collectively, these findings will lay the groundwork for determining whether this approach can be translated into a novel treatment intervention for anxiety in AN, for example, by augmenting cognitive behavioral therapy with pre-meal interoceptive exposure training, to enhance inhibitory fear learning and reduce the fear of food in AN.
AND PUBLIC HEALTH SIGNIFICANCE Anorexia nervosa (AN) shares substantial psychopathology with anxiety disorders such as generalized anxiety disorder (GAD), in that both disorders show exaggerated anticipatory responses to aversive stimuli. The current proposal will investigate how similarly or differently individuals with AN or GAD experience meal- associated cardiorespiratory sensations relative to unaffected healthy individuals, will identify the brain circuitry underlying these shared and separate experiences, and will evaluate how interoceptive phenotypes predict illness outcomes in AN and GAD. By comparing and contrasting AN and GAD, the study will clarify the degree to which dysfunction is restricted to a dysregulated eating domain, or a factor of anxious psychopathology inherent to both disorders.