Applying neurobiology to the treatment of adults with anorexia nervosa

There have been significant advances in our understanding of etiological influences associated with anorexia nervosa. Genetic studies indicate that heritability accounts for approximately 50–80 % of the risk of developing an eating disorder (ED) [36]. Recent imaging studies reveal that individuals with anorexia nervosa tend to have common temperament and personality traits related to neural circuit function, which are heavily implicated in the development and maintenance of the disorder [31, 34, 37–42]. The temperaments are characterized by increased trait anxiety [43] and state anxiety related to food and eating, [44] high incidence of co-morbid anxiety disorders, [43] high punishment sensitivity and low reward reactivity, [45] elevated intolerance of uncertainty [46] and exaggerated harm avoidance (HA). HA is a multifaceted temperament trait [47] that contains elements of anxiety, inhibition and inflexibility [28–30, 48, 49]. There are multiple ways in which these traits may influence the etiology, maintenance and treatment of the disorder. For example, elevated anxiety and HA not only maintain ED symptoms, but may predict poor treatment outcome; [28, 50–54] and higher levels of pre-meal anxiety predict lower food intake [44]. This is offset through the anxiolytic effect from acute dietary restraint and caloric restriction [44, 55]. Food consumption stimulates a dysphoric mood, [56] suggesting that anorexia nervosa individuals may regulate eating behavior to manage anxiety.

A new anorexia nervosa model is proposed that reflects temperament and alterations in brain circuitry in order to inform and help guide treatment interventions (Fig. 1). This anorexia nervosa model shows heritable traits, such as harm avoidance, perfectionism, anxiety and inhibition, establishing a temperamental basis from which the illness may develop [28–30, 48, 49]. These traits are initially expressed behaviorally in childhood. Common temperamental expressions may include a natural compliance to rules within the family/school and social settings. Likewise, perfectionism may be expressed through high grades and excellent performance, and it may be perceived as never good enough and experienced with little reward sensation. Alterations in reward and punishment reactivity also account for the tendency for anorexia nervosa individuals to see their own errors over their successes. Individuals with anorexia nervosa often demonstrate a natural preference towards highly structured and predictable environments due to difficulties with tolerating uncertainty and set-shifting. As such, anxiety appears to reduce when children, adolescents and adults with anorexia nervosa are provided with structure verses open-ended and more ambiguous tasks. These same traits also influence response to food for individuals with anorexia nervosa. Individuals with anorexia nervosa tend to report that they do not experience a sense of reward in response to food intake, but instead experience anxiety. As such, restriction of food intake appears to provide an anxiolytic effect [57].

Recent research, including our own, has implicated neural substrates in the ventral (lower or bottom area) limbic circuitry, dorsal (top) cognitive circuitry and insula, underlying altered reward processing [37, 58, 59], cognitive or self-regulatory control [32, 60–62] and interoception [41, 63–65] in the pathophysiology of anorexia nervosa (Fig. 2). The ventral area of the limbic neural circuit includes the nucleus accumbens, putamen and caudate, as well as the orbitofrontal cortex and amygdala. These regions code for the rewarding and motivating value of eating and contribute to approach or avoidance behaviors.

The dorsal cognitive network makes and executes decisions, such as to control food consumption, based on considerations of both short- and long-term outcomes (e.g., perceived weight gain). It includes the dorsal caudate and dorsal anterior cingulate, lateral prefrontal cortex and parietal cortex. Specifically, the insula is a hub for the evaluation of interoceptive cues, such as internal pain, tastes or feelings of fullness, and plays a pivotal role in anticipation and processing of interoceptive states by conveying information about the internal milieu of the person or organism [66, 67] and perceived importance or salience of a food stimulus. These systems interactively weigh the reward value of food and consequences of consuming it, and integrate this information with homeostatic and motivational drives to guide eating behavior.

The model describes brain response changes for those with anorexia nervosa from mounting evidence that indicates that an altered balance of reward and inhibition may contribute to disordered eating [1]. Figure 1 demonstrates that in patients with anorexia nervosa, severely restricted food intake appears to be related to overactive inhibitory control (up inhibition arrow) in combination with underactive reward circuitry (down interoception and pleasure arrows). Imaging data further suggest that anticipatory anxiety contributes to restricted eating. For example, individuals with anorexia nervosa report exaggerated anxious and avoidance responses to food cues [41, 44]. This is reflected in ventral striatal limbic responses (up emotion arrow) [41] and implicates elevated HA and anxiety in anorexia nervosa. This may be expressed and experienced as cognitions becoming a cacophony of “noise” impacting the inability to decide or execute decisions, thus increasing doubt in the dorsal caudate (up arrow). This also entails altered dopamine (DA) and serotonin (5-HT) function. This disconnect between anticipating and experiencing food stimuli likely contributes to restricted eating in AN. Lowered dopamine and increased serotonin response may also contribute to the ability of persons with anorexia nervosa to delay rewards since reward may be experienced as less pleasurable [41].

Evidence also suggests parietal disturbance [68], which codes for perception of body image and shape disturbance increases (up body disturbance arrow); and the motor cortex area implements the overactive anxiety and body shape disturbance through excessive exercise (up movement arrow). Together, these findings have implications for motivation to eat and ability to evaluate reward and make decisions. It is plausible that elevated harm avoidance, perfectionism, anxiety and inhibition, present during childhood, may increase the inclination of compliance with rules for adults with anorexia nervosa.

The model shows that nurture, or environmental influences, have more influence on the child’s temperament pre-puberty compared to post-puberty [58]. The hypothalamic-pituitary gonadal axis initiates biological changes during puberty, contributing to a shift from nurture to nature taking the dominate role in trait expression through mid-twenties [69]. This may trigger changes in neurochemical circuity and contribute to how thought patterns, emotions and motor expression alter [70, 71]. In our view, to acknowledge that anorexia nervosa tends to develop around puberty and may rise in incidence again around 18, when faced with a multitude of new life and daily decisions, is not enough to understand the illness. In addition, the arrow on the right, social and traumatic experiences, influence and may even alter genetic expression during adolescent and adult life, contributing to the development of anorexia nervosa (shown as the arrow on the right in the Figure).

Neurobiological findings indicate and direct us to possible truths. However, research holds little value if clinicians and clients do not understand it or know how to interpret it. Neurobiologically based research is new to the treatment field and needs to be presented to anorexia nervosa adults and their supports in a way that they can understand the findings, so the client can identify and determine how the findings relate to their own experience. Interpreting research accurately and creatively in a manner that enhances understanding can lead to increased motivation, instead of resistance, to change.

To date, anorexia nervosa clinical treatment approaches have not been developed and/or updated to include both current neurobiological research findings and family-based approaches for adults. The most widely used behavioral treatment models, such as cognitive behavioral therapy (CBT or CBT-Enhanced), focus on symptomology. Biological underpinnings of the cognitions and behaviors are given less to no focus.

Hence, what does a neurobiologically based adult anorexia nervosa treatment look like?