Surgery for Anorexia Nervosa

This chapter includes an accompanying lecture presentation that has been prepared by the authors: .

Key Concepts

Anorexia nervosa (AN) is a severe refractory psychiatric disorder with high rates of morbidity and mortality.
Multiple factors contribute to the development and maintenance of AN, including genetic and epigenetic, biologic, cognitive, and sociopsychological mechanisms and their interactions.
Current psychotherapeutic interventions and pharmacologic therapies for AN are far from being universally effective, thus leading to the consideration of experimental neurosurgical approaches as viable options for the treatment of refractory AN.
Both deep brain stimulation (DBS) targeting the nucleus accumbens (NAcc) or subcallosal cingulate (SCC) and stereotactic ablation, including capsulotomy and cingulotomy, are potential approaches to be investigated in the treatment of refractory AN.
Novel targets for AN treatment, such as the subthalamic nucleus (STN) and zona incerta (ZI), warrant further investigations in preclinical studies.
The choice between DBS and ablation procedure may depend on the type and severity of AN as well as the patient’s beliefs and current physical condition.

Introduction

Anorexia nervosa (AN) is a refractory psychiatric disorder characterized by a voluntary and persistent restriction of energy intake, weight loss, an intense fear of gaining weight, and a distorted self-perception of body image and weight. It has also been associated with weight-loss behaviors in addition to dietary restriction, such as purging or excessive physical activity. The average lifetime prevalence rate of AN is 0.3% to 1% in young females and less than 0.5% in males. ^,

There are two subtypes of AN—the restricting (AN-R) and binge-purge (AN-BP) subtypes—with several overlapping characteristics. However, emerging evidence has demonstrated the neurobiologic differences between these two subtypes, such as sensitivity of the reward circuit, as well as neurocognitive traits and individual temperament. ^,

AN is associated with the highest rate of mortality of all mental disorders, with a weighted mortality rate (i.e., deaths per 1000 person-years) of 5.1 and a standard mortality ratio of 5.86. Long-lasting malnutrition can lead to various severe physical complications, including osteoporosis, gastrointestinal and cardiac complications, electrolyte disturbances, and eventually multiple organ failure. Moreover, patients with AN-R manifest a higher frequency of underweight-mediated medical complications (e.g., bone marrow dysfunction, hepatic dysfunction, and hypoglycemia). AN psychiatric comorbidities include major depressive disorder (MDD; 50%–70% of AN patients), anxiety disorder (>60%), and obsessive-compulsive disorder (OCD; >40%). ^, Personality disorders and substance abuse are also common (12%–27%) among those with AN-BP, in whom the rate of impulsive behavior is even higher than in those with AN-R. The majority of individuals with eating disorders report suicidal thoughts, and about 22% attempt to commit suicide.

Neurocognitive impairments in AN patients have increasingly attracted the attention of psychologists and researchers. Impaired executive functioning (e.g., set-shifting difficulty and weak decision making), weak central coherence, and lack of theory of mind are considered candidate endophenotypes for AN. For example, using neuropsychological tasks such as the Wisconsin Card Sorting Test, several studies have reported poor set-shifting abilities (i.e., the ability to move back and forth between tasks) in individuals with AN and their relatives. ^, It was also shown that these changes may persist even after weight restoration in patients. The identification of endophenotypes for AN could facilitate the development of animal models, diagnosis, patient stratification, and outcome prediction. To this date, there are no reports on the efficacy of surgical interventions for AN with regard to cognitive improvement in patients.

With regard to AN outcomes, in an extensive literature review, Steinhausen showed that less than half (46.9%) of surviving patients recovered from AN, one-third (33.5%) improved partially, and in 20.8% (0%–79%) the disease became chronic despite standard treatments. Patients with durations of illness longer than 10 years are not likely to recover. ^,

Severe and Enduring Anorexia Nervosa

To date, there is no consensus definition of severe and enduring anorexia nervosa (SE-AN) to inform research and clinical practice. The most common criteria proposed to this end are (1) persistent unremitting symptoms, (2) long illness duration, and (3) treatment resistance. ^, The severity could be defined, at least in part, based on body mass index (BMI) values (i.e., mild: ≥17 kg/m ² ; moderate: 16–16.99 kg/m ² ; severe: 15–15.99 kg/m ² ; extreme: <15 kg/m ² ). With respect to illness duration, thresholds vary across studies, with cutoffs ranging from 3 to 10 years. With regard to treatment resistance, owing to differences in the availability of eating disorder expertise centers and specialists across countries, the definition of treatment resistance should depend on the accessibility of pharmacologic and psychotherapeutic interventions.

Etiopathology and Neurocircuitry of Anorexia Nervosa

AN is a disorder of complicated etiology, the susceptibility to which seems to be shaped by genetic and epigenetic, biologic, cognitive, and sociopsychological mechanisms and their interactions.

Genetics of Anorexia Nervosa

Concerning the genetic component in AN, previous family-based epidemiologic studies have shown that the relative risk for AN susceptibility is 11.3 in female relatives of anorexics and probands, pointing to the familial aggregation of AN. Subsequently performed twin studies on heterozygotes and homozygotes demonstrated that replicated heritability estimates for AN ranged from 0.48 to 0.74. The candidate gene–driven approach and corresponding meta-analysis indicated that serotonergic genes (e.g., HTR2A and 5-HTT ) could be involved in the genetic etiology of AN. More recently, one international multicenter genome-wide association study (GWAS) meta-analysis has further suggested that AN may be better conceptualized as a disorder that arises from both psychiatric and metabolic origins.

Structural Alterations in Anorexia Nervosa

Several studies have focused on factors underlying AN pathologic mood, anxiety, reward, body perception, interoception, inhibition, alexithymia, and appetite. Neuroimaging has been widely used to demonstrate both structural and functional differences among individuals with AN, those who have recovered from AN, and healthy controls. Both globally reduced gray matter and white matter were observed in acutely ill AN patients as compared with healthy control subjects, and this abnormality was able to be reversed after weight restoration, suggesting that malnourishment might partly contribute to brain morphologic changes in acutely underweight patients. However, some regions, including the cingulate gyrus and cerebellum, appeared to be altered more frequently than others. Interesting to note, higher cortical thickness was reported in the medial orbitofrontal cortex (OFC) and insula of AN patients. These results suggest that regional mixed cortical thickness patterns could constitute biomarkers of AN.

Functional Alterations in Anorexia Nervosa

Endophenotypes and Related Functional Brain Areas

Individuals with AN typically exhibit distinctive cognitive features, temperament, and personality traits, which often first occur in childhood before the onset of AN, and may even contribute to the development of AN, often persisting following recovery. These features, which are independent of nutritional status, may serve as more stable phenotypes (i.e., endophenotype ; for a more extensive review, see Zhou et al. ) with a clear genetic connection, and are likely to be encoded in neural circuits. For example, characteristic harm avoidance is positively associated with dopamine (DA) D ₂ /D ₃ receptor binding in the dorsal caudate and dorsal putamen, and perfectionism and obsessive personality traits are associated with exaggerated cognitive control by the dorsolateral prefrontal cortex (DLPFC). The DLPFC may become excessively involved in inhibitory processes that dampen information processing through reward pathways. The anxious temperament of patients with AN leads to various dysfunctional behaviors thought to serve as a means of coping with adverse mood. Functional brain imaging shows that decreased activation of frontostriatal circuits, including the ventral striatum, anterior cingulate cortex (ACC), and supplementary motor area, may underlie the impaired cognitive flexibility in patients with AN, resulting in stereotypical and ritualistic behaviors inherent to controlling eating and weight.

Serotonergic System and Emotion in Anorexia Nervosa

Considerable data indicate that individuals with AN exhibit disturbances in serotonin 5-HT and DA systems. 5-HT has a well-established role in the modulation of mood and cognition (e.g., altered satiety and impulse control). Neuroimaging studies using positron emission tomography (PET) and single-photon emission computed tomography (SPECT) with 5-HT–specific radioligands consistently show alterations in 5-HT _1A and 5-HT _2A receptor and 5-HT transporter activity in cortical and limbic structures, which may be related to anxiety, behavioral inhibition, and body image distortions. ^, Depletion of 5-HT levels reduces anxiety in acutely ill patients and those who have recovered from AN.

Dopaminergic System and Reward in Anorexia Nervosa

DA is implicated in motivation, executive functioning (e.g., inhibitory control, salience attribution, and decision making), and the aberrant reward effects of food. The processing of food-related reward is complex and modulated by cognitive, emotional, and biologic factors involving learned behaviors and genetic predispositions. The motivation to approach food and eat is a critical part of the reward pathway and is particularly disturbed in AN. Recent imaging studies provide evidence of disturbed gustatory processing in patients with AN, involving the anterior insula and striatal regions. Genetic and epigenetic, pharmacologic, and physiologic data show that individuals with AN and those who have recovered from AN exhibit altered striatal DA function. DA disturbances could contribute to altered modulation of appetitive behaviors as well as symptoms of anhedonia, dysphoric mood, asceticism, and increased motor activity. These results raise the possibility that individuals with AN have altered appetitive mechanisms that may involve either sensory, interoceptive, or reward processes. These disturbances in the modulation of reward and emotion may increase vulnerability to dysregulated appetitive behaviors. Individuals with AN may be able to inhibit appetite and have extraordinary self-control owing to exaggerated dorsal cognitive circuit function.

Other Neurocircuitries Implicated in Anorexia Nervosa

In addition to the reward pathway, several other neural circuits may also contribute to AN. For example, the insular cortex serves to integrate processing from many of the structures relevant to AN (e.g., the limbic system, the nucleus accumbens [NAcc], and the striatum). It has therefore been hypothesized that a rate-limiting dysfunction of neural circuitry integrated by the insula may account for the clinical phenomena of AN. More recently, the zona incerta (ZI) has been shown to be involved in the regulation of feeding and may therefore play a distinct role in the pathogenesis of AN, possibly serving as a potential treatment target. The GABAergic neurons in the ZI demonstrate an unexpected robust orexigenic effect by projecting to paraventricular thalamus excitatory neurons. Binge-like eating behaviors are observed when these GABAergic neurons in the ZI are stimulated in mice.

One of the core features of AN is the intense fear of food and weight gain. Key regions involved in fear perception include the amygdala, hippocampus, insular cortex, ACC, striatum, and prefrontal cortex (PFC). In response to visual food cues, patients with AN display stronger activation of the bilateral amygdala, medial PFC, ACC, and striatum than healthy controls. Thus, food cues trigger arousal, fear, and avoidance in patients with AN. Moreover, fear circuitry is involved in the extinction of fearful memories, and inhibitory projections from the medial PFC/OFC to the amygdala are indispensable for this extinction. In this regard, individuals with eating disorders exhibit faster fear-learning processes and a heightened resistance to the extinction of conditioned responses. Neuroimaging, neurophysiologic, and lesion studies implicate ventral (limbic) and dorsal (cognitive) neural circuits, which may be of particular relevance to understanding AN behaviors. In a fluorine 18 ( ¹⁸ F) fluorodeoxyglucose (FDG)–PET study of patients with AN in 2013, we observed that hypermetabolism in the frontal lobe, hippocampus, and lentiform nucleus decreases after deep brain stimulation (DBS) of the NAcc. Lipsman et al. also reported decreased hypermetabolism in the anterior cingulate, insular, and parietal lobes after subcallosal cingulate–DBS (SCC-DBS), along with improvements in BMI and mood. ^, These results highlight the importance of these brain regions for AN.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here