Heritability is high among patients with AN. The concordance rate for AN in dizygotic twins is 5% compared with 56% for monozygotic twins. The frequency of the disorder in female relatives of patients with AN is 8 to 20 times higher than in the general population (Gorwood, et al., 1998). The exact mode of inheritance and the specific genes that influence risk are not yet known. Some evidence has supported a disordered hypothalamic-pituitary-adrenal axis as a cause. Hypercortisolemia, nonsuppressive dexamethasone suppressive tests, and increased central nervous system fluid levels of corticotropin-releasing hormone have all been demonstrated in AN (Kaye, 1998). Recent studies show increased levels of cerebral serotonin in patients with AN, leading theorists to believe that high levels of serotonin may lead to reduced food intake (Walsh & Devlin, 1998). Research also has suggested that a high level of cholecystokinin, a gastric peptide, is associated with delayed gastric emptying, causing a constant sense of satiety in patients with AN (Fujimoto et al., 1997).

The psychosocial profile of children with AN demonstrates a premorbid history of high achievement and perfectionism. These children often come from families with rigid homeostatic systems. Parents of children with AN may be overinvolved or overprotective, avoid conflict, or have high and unrealistic expectations. As a result, these children struggle for autonomy, identity, and self-respect, gaining a sense of control by restricting severely the amount of food they eat. Thus, the ability to do without food becomes a way to cope with stress.

Psychological trauma, especially sexual abuse, may contribute to AN. Between 34% and 83% of children with AN report some form of sexual trauma (Schmidt et al., 1997). Children and adolescents with AN often have other associated psychological disorders, including major depression (50%–75%), bipolar disorders, obsessive-compulsive disorder (10%–13%), substance abuse, and borderline personality disorder (Halmi et al., 1993). Refer to Chapter 23, Chapter 25, and Chapter 28 for more information on these comorbidities.

The relationship between social influences and AN is worth exploring. For example, the use of tall and emaciated teen fashion models in the mass media conveys an unrealistic, unhealthy, and dangerous standard of beauty for very receptive children and adolescents. More information is needed to understand how such images may contribute to AN and other eating disorders.

Prevalence of AN in the United States is 0.5% to 1% among adolescents, and the incidence is rising (Pawluck & Gorey, 1998). Incidence almost certainly is lower in young children, but the exact numbers are not known. Pattern of onset is bimodal, with peaks at ages 13 to 14 years and later at ages 17 to 18 years. In adolescents, AN is observed overwhelmingly among females (90%–95% of cases). In children with AN, 19% to 30% of cases are observed in boys (Bryant-Waugh & Lask, 1995).

Patients with full-blown AN often wear several layers of clothing to keep warm and to hide their thinness. During examination, patients must disrobe completely for providers to determine their weight accurately.

Untreated patients with AN will show obvious signs of emaciation. Several physical signs of AN include the following:

The criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM IV), for diagnosis of AN include the following:

Timely diagnosis is crucial but can be difficult. Patients often are secretive about their problem; many of them deny they have AN. Early “red flags” in a child or adolescent include the following:

Diagnosis of AN usually is obvious and based on the patient’s history, attitude toward food, associated symptoms, and physical examination. No specific biologic test is used to diagnose this disorder. Early in the diagnosis, the only abnormality providers may observe is an elevated blood urea nitrogen that results from dehydration. Later, providers may note iron deficiency anemia with microcytosis, hypocalcemia, hypoglycemia, hypomagnesemia, and hypophosphatemia, all due to starvation. Patients with AN have severe but reversible hypercholesterolemia (Stone, 1994). Later, serum aminotransferase may be elevated.

Providers must perform an electrocardiogram and echocardiography on patients with AN to evaluate for cardiac muscle damage. Patients may develop decreased left ventricular wall thickness, reduced heart size, bradycardia, and nonspecific ST-T wave changes. They may have a low-voltage ECG. Echocardiography may reveal mitral valve prolapse (Cheng, 1997). These alterations make patients with AN prone to sudden death (Neumarker et al., 1997).

Anorexia nervosa severely alters the endocrine system; therefore, measurements of leuteinizing hormone (LH), follicle-stimulating hormone (FSH), and estradiol are necessary, as are thyroid function tests. The starvation that accompanies AN disrupts the hypothalmic-pituitary-adrenal axis, resulting in low levels of LH and FSH. For this reason, adolescents with AN who previously have experienced menarche become amenorrheic with anovulation. Because they are estrogen deficient, these patients do not experience withdrawal bleeding after progestin challenge. Between 30% and 50% of patients with AN become amenorrheic even before they experience significant weight loss. Resumption of menses often is delayed until some time after they regain weight, perhaps because the body produces high cortisol levels as a response to the stress of starvation (Kaye, 1998). In males with AN, hypogonadism with low levels of testosterone, LH, and FSH may be seen. In either sex, thyroid function tests are abnormal with elevated reverse T₃, reduced total T₃, low normal T₄, and normal TSH. Reduced metabolic rate accounts for the hypothyroid-like presentations of bradycardia, cold intolerance, constipation, and dry skin.

One of the most important complications of AN is osteoporosis resulting from estrogen deficiency. Photon absorptiometry will assist in evaluating fracture risk. When amenorrhea is prolonged, sustained bone loss may not be reversible even once menses resume. Therefore, these patients are at an increased risk for fractures throughout their lives (Hotta et al., 1998).

Several complications are associated with refeeding patients with AN:

Providers must conduct refeeding carefully to prevent complications and progression to bulimia. The goal for the female patient is to gain 2 lb per week until she reaches the weight at which she previously menstruated and is within 15% of her ideal weight. The goal for the male patient is to
achieve a weight within that same ideal range. Providers prescribe a diet that provides 2000 to 4000 calories each day (Walsh & Devlin, 1998). Open trial studies have shown zinc supplementation (100 mg/d) to double the rate of increase in the body mass index (BMI) of patients with AN (Birmingham et al., 1994). Patients require calcium supplements of at least 1 g/d and need multivitamins as well. Because refeeding may cause severe hypophosphatemia, recommendations are for patients to be started on oral phosphate when feeding is initiated. Providers need to monitor serum phosphate every 1 to 2 days for at least the first week of refeeding.

More research is needed regarding the use of force-feeding, which may be necessary in life-threatening cases that are refractory to treatment (Mehler & Weiner, 1995). Force-feeding may be performed by nasogastric tube or through total parenteral nutrition. Generally, force-feeding is used only with extreme cases of AN, including patients who may lose weight again, necessitating repeated treatments. When force-feeding is conducted, the patient should be detained in a hospital that also provides psychotherapy. Other medical therapies will not be possible until the patient has steadily gained some weight. Some experts disagree with force-feeding, suggesting that repeated episodes of it can adversely affect ultimate success (Draper, 1998). To resolve this controversy in the United Kingdom, the Mental Health Commission issued guidelines about when patients with AN can be detained, treated, and fed without consent (Mental Health Commission, 1997).

Providers must decide on the need for hospitalization versus the appropriateness of treating individuals as outpatients. Indications for the hospitalization of patients with AN are as follows:

While the patient is in the hospital, providers must maintain a caring, highly structured, nonpunitive environment. Often, health care professionals will present a written contract to the patient and family on admission (Robin et al., 1998). The contract must describe the goals of hospitalization, criteria for discharge, a nutritional rehabilitation plan, and rules regarding exercise, visitation, and system of rewards. Display 22-1 presents a sample contract. Generally, achievement of these goals takes about 4 to 6 weeks; however, in today’s era of managed care, most hospitals discharge patients within 2 weeks. Criteria for discharge from the hospital include the following:

An alternative to full hospitalization for refeeding is to discharge patients as they approach their target weight and admit them into a parital pediatric day treatment program from 8:00 AM to 10:00 PM. Robin et al. (1998) report success using this method, with 84% of patients reaching and retaining their ideal weight after 9 months of follow-up treatment.

Psychotherapy is fundamental to the treatment of AN. Ego-oriented individual therapy (EOIT) and cognitive-behavioral treatment (CBT) have both proven useful. EOIT combines weekly individual sessions for the patient with collateral sessions twice a month for the parents. Individual sessions focus primarily on the youngster’s ego strength, coping skills, individuation from the nuclear family, confusion about identity, and other interpersonal issues related to physical, social, and emotional growth. In these sessions, therapists examine the connection between such issues and eating, weight expectations, and body image. Conversely, CBT uses cognitive restructuring to modify distorted beliefs and attitudes about the “meaning” of weight gain. Compared with CBT, restoration of weight is slower with EOIT, but changes in eating, attitudes, depression, ego-functioning, and family interactions are comparable (Robin et al., 1998). Providers should refer patients whose families are interested in pursuing either EOIT or CBT to mental health professionals skilled in their application with adolescents.