ANOREXIA NERVOSA and EATING DISORDERS
PAUL M. COPELAND
Incidence and Epidemiology
Etiology and Pathogenesis
Clinical and Laboratory Findings
The major eating disorders of anorexia nervosa and bulimia nervosa represent a spectrum of abnormalities with overlapping clinical features. Both disorders are characterized by preoccupation with weight, and about one-half of patients with anorexia nervosa exhibit bulimic behavior. The American Psychiatric Association specifies criteria for the diagnosis of anorexia nervosa that may be summarized as a refusal to maintain weight above 85% of ideal body weight; fear of obesity; distorted body image; and, in females, amenorrhea for at least 3 months. Patients with anorexia nervosa who engage in repeated binge-eating or purging behavior are subcategorized as binge-eating/purging types, and others are classified as restricting types.
Criteria for bulimia nervosa require an average of at least two binge-eating episodes each week for at least 3 months about which the person feels a lack of control. The person has a persistent overconcern with body shape and weight and regularly engages in self-induced vomiting; use of diuretics, laxatives, or enemas; strict dieting or fasting; or vigorous exercise to prevent weight gain. “Eating disorder not otherwise specified” is a category for patients who do not meet full criteria for anorexia nervosa or bulimia nervosa. Amenorrhea, in particular, has been the subject of controversy as a criterion for anorexia nervosa, since the degree of psychopathologic behavior may be as severe in patients who menstruate. Binge-eating disorder, a newly designated diagnosis, shares many characteristics with bulimia nervosa, but patients do not regularly purge or compensate by fasting or excessive exercise. Diagnostic criteria include binge eating at least 2 days per week, on average, for a period of at least 6 months.
INCIDENCE AND EPIDEMIOLOGY
The lifetime prevalence of anorexia nervosa is about 0.5%. Most surveys show a greater preponderance among upper-socioeconomic classes, but later studies suggest that this gap is narrowing. There is a bimodal peak age at onset at 14 and 18 years. Only 4% to 10% of patients are males. Bulimia nervosa occurs in about 1% of young adults; males make up fewer than 10% of the patients. More individuals occasionally engage in bulimic behavior, but do not meet the frequency and duration criteria for bulimia nervosa. About 10% to 19% of young women binge eat and purge. Certain groups are especially vulnerable: eating disorders are more common in women with a history of obesity or diabetes mellitus, models, ballet students, or those in professions that demand high achievement and among men who are homosexual. Binge-eating disorder has a prevalence rate of 0.7% to 4%. Among obese individuals in weight-control programs, the typical prevalence is 30%; females are approximately 1.5 times more likely than males to have this eating pattern.
ETIOLOGY AND PATHOGENESIS
The factors that initiate or perpetuate anorexia nervosa are unknown. Theories include developmental, familial, sociocultural, and biochemical factors. Typically, the person with anorexia nervosa is a perfectionist and may perceive that parents are more concerned with achievements than feelings. Crisp maintained that weight loss also becomes a means of avoiding the characteristics of sexual and psychological maturity. The etiologic importance of a history of sexual abuse is debatable. Fasting has long been associated with asceticism and purity. Although medical reports of anorexia nervosa date to the seventeenth century, the modern societal thin ideal influences choices about weight control. Many patients with anorexia nervosa begin their diets after being teased about the need to lose weight.
Concordance rates of approximately 55% for monozygotic as opposed to 5% for dizygotic twins have suggested a genetic predisposition to anorexia nervosa. Some biochemical theories for the origin of anorexia nervosa have focused on neurotransmitters that affect control of appetite. Defects in central nervous system norepinephrine and serotonin (5-hydroxytryptamine, 5-HT) metabolism have been verified in acutely ill and recovered patients. One difficulty with biochemical theories related to appetite is that anorexia nervosa is not characterized by true anorexia but rather by a willful avoidance of eating, at least in the initial stages. Other genetic data may relate to the behavioral trait of perfectionism; both anorexia nervosa and the frequent comorbid condition of obsessive-compulsive disorder have been associated with polymorphism in the promoter region of the 5-HT2A-receptor gene.
The cause of bulimia nervosa is also unknown, but it typically begins during a diet. The binges often occur during times of stress or depressed mood. Patients describe an initial soothing, albeit frenetic, quality to food ingestion, followed by feelings of guilt and then resolution of the guilt by purging. One-half of those with anorexia nervosa develop bulimic behavior. The stress of chronic dieting may create a breakthrough of binge eating. Keys and co-workers’ study of conscientious objectors who volunteered to undergo 6 months of semistarvation toward the end of World War II revealed experiences of uncontrollable binge eating reminiscent of bulimia when the subjects were subsequently allowed unrestricted access to food. As in bulimia nervosa, some patients with binge-eating disorder report episodes triggered by dysphoric moods. Others report binge eating to relieve tension or describe a dissociative “numb” state during episodes.
CLINICAL AND LABORATORY FINDINGS
The behavioral features of anorexia nervosa are most notable for obsessive pursuit of thinness, body image disturbance, and denial of illness. Even those who superficially agree that a problem exists may still not be able to change their behavior. Other characteristic features are hyperactivity, social isolation, food rituals, and hoarding of food that may then be disposed of or eaten in private in small amounts or in binges. Patients with bulimic behavior tend to be more outgoing, more impulsive and more sexually active and have a higher prevalence of alcohol and other drug use than those who purely restrict food intake. Depression is common among patients with anorexia nervosa or bulimia nervosa.
The clinical manifestations of anorexia nervosa are consequences of starvation plus any of the associated behaviors, such as binge eating, vomiting, or abuse of laxatives, diuretics, or ipecac (Table 37.1). Some patients with bulimia nervosa manifest milder consequences of intermittent starvation. Food restriction is typical between binge-eating episodes. Estrogen deficiency and amenorrhea are consequences of diminished hypothalamic gonadotropin-releasing hormone (GnRH) secretion. Leptin deficiency at low weights may lead to decreased GnRH secretion. Gonadotropin secretion is also reduced in males with anorexia nervosa, and low serum testosterone levels may contribute to symptoms of impotence and loss of libido.
TABLE 37.1. IMPORTANT CLINICAL AND LABORATORY ABNORMALITIES IN PATIENTS WITH EATING DISORDERS
About 20% of patients with bulimia nervosa also have amenorrhea, and another 20% to 40% of patients have oligomenorrhea. Reduced luteinizing hormone secretion has been correlated with weight reductions of less than 85% of past high weights. Osteoporosis in anorexia nervosa may lead to vertebral compression fractures and stress fractures. Although osteoporosis is related to the duration of amenorrhea, a lack of estrogen may not be the sole cause. Increased cortisol production rates are correlated with bone loss. Nutritional depletion and resultant low levels of insulin-like growth factor I may also be responsible for bone loss.
Thyroid function in anorexia nervosa resembles that of the euthyroid sick syndrome. Peripheral conversion of thyroxine to triiodothyronine (T3) is decreased. Bradycardia, hypothermia, cold intolerance, dry skin and hair, slowed relaxation of deep tendon reflexes, and hypercarotenemia may result in part from low T3 concentrations. Diminished norepinephrine turnover may also contribute to some of these features of slowed metabolic rate.
Low serum glucose concentrations are usually well tolerated in fasting; however, hypoglycemia has been associated with coma and death in several patients with anorexia nervosa. Some of these patients had the added stress of an infection, which may have further diminished the gluconeogenic capacity of the liver. Among young women with type 1 diabetes mellitus, approximately 30% intentionally omit insulin to achieve weight control; 8% omit it frequently. Frequent omission is associated with poorer glycemic control and a higher risk of diabetic retinopathy.
Excessive caffeine or water ingestion may cause polyuria. Patients with anorexia nervosa also have diminished or erratic secretion of vasopressin in response to an osmotic challenge. Some patients manifest partial diabetes insipidus. Renal abnormalities largely reflect dehydration and a reduced glomerular filtration rate. Laxative abuse and self-induced vomiting can contribute to hypokalemia, volume depletion, and renal failure. Diuretic abuse can cause hypokalemia and, if coupled with water ingestion and vomiting, hyponatremia. During treatment of severe hyponatremia, too-rapid restoration of serum sodium has led to central pontine myelinolysis.
Patients with anorexia nervosa manifest low blood pressure and diminished cardiac output. Arrhythmia may be lethal. Retrospectively, cardiac deaths have been associated with QT prolongation. Repeated use of ipecac to induce vomiting can cause fatal cardiomyopathy. Hypomagnesemia has been associated with congestive heart failure.
Gastrointestinal changes in anorexia nervosa include decreased gastric emptying and intestinal motility, which are associated with an exaggerated sense of stomach fullness, abdominal pain, and constipation. Metoclopramide and cisapride have been used to enhance gastric emptying. Hepatic transaminases may be elevated in anorexia nervosa because of fatty infiltration of the liver. Vomiting often elevates levels of salivary amylase, and starvation may raise levels of pancreatic amylase. Vomiting during a state of decreased consciousness (with alcohol or other drug use) has resulted in aspiration pneumonia. Low neutrophil counts (less than 1,500 cells per microliter) in patients with anorexia nervosa have been associated with increased risk of infection. A mild elevation of serum creatine phosphokinase and electromyographic changes consistent with myopathy also can be found in patients with anorexia nervosa.
Outcome studies of anorexia nervosa find that 50% of patients totally recover, 30% of patients improve, and 20% remain chronically ill. Studies with up to 20 years of follow-up observation show a 15% mortality rate. Suicide and medical complications each account for one-half of disorder-related deaths. With weight restoration, there may be various degrees of psychological and physical recovery. Most physical manifestations reverse with weight gain, but menstrual disturbances are prone to continue among patients with anxiety disorders or with persistent abnormal eating attitudes. To help improve bone formation, the patient should have a calcium intake of at least 1,500 mg/day and take a vitamin supplement containing 400 IU of vitamin D. Hormone replacement therapy has improved bone density in patients with ideal body weight less than 70%.
Psychiatric treatment of anorexia nervosa consists of combinations of individual, group, and family therapy. Recovered patients have emphasized the value of a relationship with a therapist or friend. Nutritional counseling and medical follow-up is important. Hospitalization is required for severe cardiovascular or metabolic disturbances, continued or rapid weight loss, or other psychiatric crises. Regular food is given, starting with intakes of 30 to 40 kcal per kilogram per day. Intake may increase up to 70 to 100 kcal per kilogram per day during weight gain and can require as much as 40 to 60 kcal per kilogram per day during weight maintenance. Tube feeding or parenteral hyperalimentation is rarely required and usually should be avoided. Serious volume depletion can be treated with intravenous fluids. Serum phosphorus levels should be assessed.
The course of bulimia nervosa has been less well studied. About 30% to 60% of patients recover. Up to another 20% experience anorexia nervosa. In addition to cognitive-behavioral therapy (also effective in binge-eating disorder) and individual psychotherapy, pharmacotherapy with antidepressant drugs has met with some success. Tricyclic antidepressants, monoamine oxidase inhibitors, and antidepressants that more selectively inhibit serotonin reuptake have reduced binge-eating episodes in double-blind clinical trials. Fluoxetine has been approved for use in cases of bulimia nervosa. The dose of fluoxetine is generally titrated upward from 20 to 60 mg per day.
Beresin EV. Anorexia nervosa. Compr Ther 1997;23:664–671.
Cachelin FM, Maher BA. Is amenorrhea a critical criterion for anorexia nervosa? J Psychosom Res 1998;44:435–440.
Copeland PM, Sacks NR, Herzog DB. Longitudinal follow-up of amenorrhea in eating disorders. Psychosom Med 1995;57:121–126.
Crisp AH. Some psychological aspects of adolescent growth and their relevance for the fat/thin syndrome anorexia nervosa. Int J Obes 1977;1:231–238.
Enoch M-A, Kaye WH, Rotondo A, et al. 5-HT2A promoter polymorphism-1438G/A, anorexia nervosa, and obsessive-compulsive disorder. Lancet 1998;351:1785–1786.
Grinspoon S, Herzog D, Klibanski A. Mechanisms and treatment options for bone loss in anorexia nervosa. Psychopharmacol Bull 1997;33:399–404.
Keys A, Brozek J, Henschel A, et al. The biology of human starvation. Minneapolis: University of Minnesota Press, 1950.
Mayer LES, Walsh BT. The use of selective serotonin reuptake inhibitors in eating disorders. J Clin Psychiatry 1998;59(suppl 15):28–34.
Nielsen S, Moller-Madsen S, Isager T, et al. Standardized mortality in eating disorders: a quantitative summary of previously published and new evidence. J Psychosom Res 1998;44:413–434.
Rydall AC, Rodin GM, Olmsted MP, et al. Disordered eating behavior and microvascular complications in women with insulin-dependent diabetes mellitus. N Engl J Med 1997;336:1849–1854.
Stoving RK, Vinten J, Handberg A, et al. Diurnal variation of the serum leptin concentration in patients with anorexia nervosa. Clin Endocrinol (Oxf) 1998;48:761–768.