Aetna considers the following services and procedures medically necessary for the management of members with anorexia or bulimia.

1. Assessment:

   • Blood count and serum chemistry (e.g., CBC, electrolytes, BUN/creatinine)
   • Bone density measurement (for evaluation of anorexic members) (see CPB 0134 - Bone Mass Measurements)
   • Electrocardiography
   • Liver function tests
   • Medical evaluation (complete medical history and physical examination)
   • Psychiatric/psychological consultation and testing
   • Urinalysis.
      

2. Treatment:

   • Nutritional counseling (see CPB 0049 - Nutritional Counseling)
   • Pharmacotherapy for the treatment of anorexia (e.g., selective serotonin reuptake inhibitors and anti-psychotics)**
   • Pharmacotherapy for the treatment of bulimia (e.g., selective serotonin reuptake inhibitors such as fluoxetine, tricyclic anti-depressants, trazodone, topiramate)**
   • Psychotherapy (e.g., cognitive behavioral therapy, family psychotherapy, interpersonal psychotherapy, and psychodynamic psychotherapy).

** Note: Coverage of particular drugs within each class may be subjected to formulary restrictions, where applicable.

Aetna considers the following services/procedures experimental and investigational for the diagnosis and treatment of anorexia and bulimia because of insufficient evidence in the peer-reviewed literature.

1. Assessment:

   • Brain imaging
   • Individual Optimal Nutrition (ION) analysis/profile
   • Measurement of serum concentration of brain derived neurotrophic factor (for anorexia)
      

2. Treatment:

   • Acamprosate calcium (Campral) for the treatment of binge eating disorder
   • Bisphosphonates and other anti-resorptive agents in the management of osteopenia in anorexic members
   • Bupropion (Zyban), lithium, and naltrexone for the treatment of bulimia
   • Cholinesterase inhibitors (e.g., donepezil) for the treatment of anorexia
   • Deep brain stimulation (treatment-refractory anorexia)
   • Oral contraceptives in preventing bone loss in amenorrheic persons with an eating disorder
   • Repetitive transcranial magnetic stimulation
   • The Mandometer treatment.

Eating disorders are characterized by marked disturbances in eating behavior.  There are 2 severe forms of eating disorders -- (i) anorexia nervosa and (ii) bulimia nervosa.  Anorexia usually commences in the years between adolescence and young adulthood, with 90 % of the patients being female.  In the female gender, anorexia has a prevalence of approximately 1 % with a lifetime mortality rate of 15 to 20 %.  There are 3 classical symptoms associated with this eating disorder: (i) refusal to maintain a minimally normal body weight (e.g., weight loss leading to maintenance of body weight less than 85 % of that expected; or failure to attain expected weight gain during period of growth, leading to body weight less than 85 % of that expected), (ii) disturbance of body image and intense fear of being fat, and (iii) in post-menarcheal females, amenorrhea (i.e., absence of 3 consecutive menstrual cycles).  A diagnosis of anorexia should be considered for a young woman with symptoms of an eating disorder, amenorrhea, and a body mass index of 17.5 kg/m² or lower.  Similar considerations apply to a male patient with unexplained weight loss.

Bulimia is more common than anorexia.  In females, bulimia has a prevalence of 2 to 5 %, but a lesser mortality rate.  It is characterized by 4 key symptoms: (i) over-concern with weight and body shape, (ii) recurrent episodes of binge eating, (iii) recurring subsequent purging, restriction, or excessive exercise, and (iv) binge eating and subsequent inappropriate compensatory behaviors, occurring a minimum average of twice a week for at least 3 months.  In contrast to patients with anorexia, individuals with bulimia are generally in normal weight range, although recurrent weight changes are frequently observed.

The majority of patients with eating disorders can be treated in the outpatient settings.  Hospitalization is usually reserved for severely symptomatic patients such as individuals with extremely low body weight (75 % or less of expected body weight) whose condition must be hemodynamically stabilized, or those with medical problems requiring intensive monitoring such as patients with electrolyte imbalances, cardiac arrhythmias, profound hypoglycemia, self-mutilation, impaired capacity for self-care, or active suicidal ideation.  Furthermore, failure of outpatient treatment may also constitute grounds for inpatient treatment.  It should be noted that patients with bulimia rarely need hospitalization unless binge-purge cycle has led to anorexia resulting in severe metabolic deficiencies such as severe electrolyte imbalances, or suicidal depression is present.

A complete blood count may reflect anemia due to nutritional deficiency.  Serum electrolyte imbalances may occur in patients with bulimia.  Other laboratory tests include blood urea nitrogen/creatinine levels, serum measurements of calcium, magnesium, phosphorus, urinalysis, and liver function tests.  An electrocardiogram may aid to identify cardiac abnormalities such as sinus bradycardia, as well as signs of hypokalemia or ipecac-induced myopathy.  In general, brain imaging and bone mineral density studies are not necessary.  A psychiatric assessment of patients with an eating disorder is appropriate for identification of any concurrent psychiatric illness, evaluation of the risk of suicide, and exploration of the psychosocial context of the symptoms.

Treatments for patients with eating disorders include nutritional counseling; psychotherapy such as cognitive behavioral therapy, family psychotherapy, inter-personal psychotherapy, and psychodynamic psychotherapy; as well as pharmacotherapy.  Nutritional counseling, with a reasonable, graduated eating plan tied to specific weight goals, as well as psychotherapy are essential.  Medication plays an important, but limited role in the management of eating disorders.  In general, drug therapy is not effective in treating anorexia -- zinc, cyproheptadine, anti-depressants, and neuroleptic agents -- have not been shown to improve symptoms.

On the contrary, pharmacotherapy is moderately effective in treating bulimia.  High-dose (60-mg) fluoxetine (Prozac) and other selective serotonin reuptake inhibitors (SSRIs) such as trazodone (Desyrel) have been shown to be helpful in treating bulimia.  Tricyclic anti-depressants such as imipramine (Tofranil) and desipramine (Norpramin) have been demonstrated to reduce binge eating and vomiting in bulimic patients.  Monoamine oxidase inhibitors such as phenelzine (Nardil) should not be used as initial pharmacotherapy for bulimia because of their considerable side effects.  Bupropion (Zyban) is contraindicated in the treatment of bulimia because of increased risk of seizures.  Neither naltrexone nor lithium has been shown to be effective in treating bulimia. The role of bisphosphonates (e.g., alendronate and risedronate) and other anti-resorptive agents in the management of osteopenia in anorexic patients has not been established.  In a randomized controlled trial (RCT), Golden et al (2005) concluded that in adolescents with anorexia nervosa, weight restoration is the most important determinant of bone mineral density, but treatment with alendronate did increase the bone mineral densities of the lumbar spine and femoral neck within the group receiving alendronate, but not compared with placebo in the primary analysis.  Until additional studies have demonstrated efficacy and long-term safety, the use of alendronate in this population should be confined to controlled clinical trials.  Dietary supplements are usually not recommended for anorexia.  In a RCT, Barbarich et al (2004) concluded that supplement strategies are not a substitute for adequate nutrition and are ineffective in increasing the efficacy of fluoxetine in underweight anorexia nervosa subjects.  Tube or intravenous feeding is rarely needed or recommended unless the patient's condition is life threatening.

An evidence review on the management of eating disorders prepared for the Agency for Healthcare Research and Quality (AHRQ) (Berkman et al, 2006) stated that no medications are available that effectively treat patients suffering from anorexia nervosa, but a few behavioral therapies may help prevent a relapse and offer other limited benefits.  A Cochrane review on anti-depressants for anorexia nervosa (Claudino et al, 2006) also concluded that a lack of quality information precludes definite conclusions or recommendations being rendered on the use of anti-depressants in acute anorexia nervosa.  Future studies testing safer and more tolerable anti-depressants in larger, well-designed studies are needed to provide guidance for clinical practice.

The review by AHRQ also noted that both medications (e.g., fluoxetine, tricyclic anti-depressants) and behavioral therapies were found helpful in treating bulimia nervosa; however, there was no clear information about how to combine medications with behavioral treatments (Berkman et al, 2006).

In a clinical trial, Miljic et al (2006) evaluated the effects of ghrelin, a gastric hormone, on appetite, sleepiness, and neuroendocrine responses in patients with anorexia nervosa.  A total of 25 young women, including 9 patients diagnosed with anorexia nervosa with very low body weight, 6 patients who partially recovered their body weight but were still amenorrheic, and 10 constitutionally thin female subjects, without history of eating disorder, weight loss, with regular menstrual cycles, were included in the study.  Each patient received 300-min intravenous infusion of ghrelin 5 pmol/kg/min and was asked to complete visual analog scale (VAS) questionnaires hourly.  Main outcome measures were VAS scores for appetite and sleepiness, growth hormone (GH), prolactin, and cortisol responses were measured.  At baseline, patients with anorexia nervosa had significantly higher ghrelin, GH, and cortisol levels and significantly lower leptin than constitutionally thin subjects.  Responses of GH to ghrelin infusion were blunted in patients with anorexia nervosa. Ghrelin administration did not significantly affect appetite but tended to increase sleepiness in patients with anorexia nervosa.  These investigators concluded that ghrelin is unlikely to be effective as a single appetite stimulatory treatment for patients with anorexia nervosa.  These results suggested that patients with anorexia nervosa are less sensitive to ghrelin in terms of GH response and appetite than healthy controls.  Ghrelin effects on sleep need further studies.

The Mandometer treatment is a controversial program for patients with eating disorders.  It is a residential program that averages approximately 12 months in duration.  While management of patients with eating disorder has often included psychiatric treatment, advocates of the Mandometer treatment assert that standard psychiatric treatment is largely ineffective for these patients.  They believe anorexia and bulimia to be essentially the same disorder.  The Mandometer treatment for both anorexics and bulimics consists of re-teaching eating habits with a computerized, hand-held Mandometer (it gives continuous biofeedback about food intake over the course of meals), re-learning sensations of satiety, external heating by resting in warm rooms and using warm jackets, restriction of physical activity, and social re-construction to restore normal social interactions without the use of psychoactive drugs.

Evidence for the effectiveness of the Mandometer treatment came primarily from a Swedish group (Bergh et al, 1996; Bergh et al, 2002; Court et al, 2005).  In a RCT, Bergh and colleagues (2002) assessed the effectiveness of the Mandometer treatment. A total of 16 patients, randomly selected out of a group composed of 19 patients with anorexia nervosa and 13 with bulimia nervosa, were trained to eat and recognize satiety by using computer support.  They rested in a warm room after eating, and their physical activity was restricted. The patients in the control group (n = 16) received no treatment.  Remission was defined by normal body weight (anorexia), cessation of binge eating and purging (bulimia), a normal psychiatric profile, normal laboratory test values, normal eating behavior, and resumption of social activities.  Fourteen patients went into remission after a median of 14.4 months (range of 4.9 to 26.5 months) of treatment, but only 1 patient went into remission while waiting for treatment (p = 0.0057).  Relapse is considered a major problem in patients who have been treated to remission.  Thus, these researchers reported results on a total of 168 patients who have entered their treatment program.  The estimated rate of remission was 75 %, and estimated time to remission was 14.7 months (quartile range of 9.6 greater than or equal to 32).  Six patients (7 %) of 83 who were treated to remission relapsed, but the others (93 %) have remained in remission for 12 months (quartile range of 6 to 36 months).  Because the risk of relapse is maximal in the first year after remission, the authors suggested that most patients treated with this method recover.  Furthermore, these investigators noted that although these results are promising, they realized the necessity to further develop their method.  For example, it is necessary to examine if one of their interventions is more important than another, and if their procedures should be modified.  More importantly, however, is that a RCT comparing this method with the standard of care for eating disorders is needed.  Court et al (2005) presented the case of a girl with severe anorexia nervosa who had previously been resistant to treatment, and who was subsequently treated successfully by the Mandometer program.

The clinical value of the Mandometer treatment for the management of patients with eating disorders has not been established.  Its effectiveness need to be validated by well-designed studies.

In a single-center, randomized, double-blind, sham-controlled study, Walpoth et al (2008) examined the effectiveness of repetitive transcranial magnetic stimulation in the treatment of bulimia nervosa.  A total of 14 women meeting DSM-IV criteria for bulimia (BN) were included in this trial.  In order to exclude patients highly responsive to placebo, all patients were first submitted to a 1-week sham treatment.  Randomization was followed by 3 weeks of active treatment or sham stimulation.  The main outcome criterion was the change in binges and purges.  Secondary outcome variables were the decrease of the Hamilton Depression Rating Scale (HDRS), the Beck Depression Inventory (BDI) and the Yale-Brown Obsessive Compulsive Scale (YBOCS) over time.  The average number of binges per day declined significantly between baseline and the end of treatment in the 2 groups.  There was no significant difference between sham and active stimulation in terms of purge behavior, BDI, HDRS and YBOCS over time.  The authors concluded that these findings indicated that repetitive transcranial magnetic stimulation in the treatment of BN does not exert additional benefit over placebo.

McElroy and colleagues (2009) discussed the role of anti-epileptic drugs (AEDs) in the management of patients with eating disorders.  Of the available AEDs, topiramate appears to have the broadest spectrum of action as an anti-binge eating, anti-purging and weight loss agent, as demonstrated in 2 placebo-controlled studies in BN and 3 placebo-controlled studies in binge-eating disorder (BED) with obesity.  Topiramate may also have beneficial effects in night-eating syndrome and sleep-related eating disorder, but controlled trials in these conditions are needed.  The results of 1 small controlled study suggest that zonisamide may have efficacy in BED with obesity.  However, both topiramate and zonisamide are associated with adverse effect profiles that may limit their use in patients with eating disorders.  Phenytoin may be effective in some patients with compulsive binge eating, especially if co-morbid EEG abnormalities are present, but available data are too varied to allow definitive conclusions to be made.  Carbamazepine and valproate may be effective in treating patients with BN or anorexia nervosa when they are used to treat an associated psychiatric (e.g., mood) or neurological (e.g., seizure) disorder; otherwise, both agents, particularly valproate, are associated with weight gain.  The authors concluded that AEDs have an emerging role in the management of some eating disorders.

Mehler and MacKenzie (2009) reviewed the evidence on the treatment of osteopenia and osteoporosis in patients with anorexia nervosa.  These researchers identified controlled clinical studies of interventions for low bone mass in patients with anorexia nervosa via searches of MEDLINE; the Cochrane Library; EMBASE; PsycINFO; and cumulative index to nursing and allied health literature.  Outcomes of interest were changes in bone mineral density and fracture incidence.  A total of 6 RCTs and 2 cohort trials examined 5 classes of medical therapy on bone mineral density outcomes.  One RCT of bisphosphonates showed no benefit and a second flawed RCT showed some benefit; 1 RCT showed a benefit of insulin-like growth factor-I; none of the 5 trials evaluating estrogen therapy showed benefit.  The authors concluded that although patients with anorexia nervosa are often losing bone mass when they should be optimizing bone growth, there is no good evidence to guide medicinal interventions.  Thus, early detection and weight restoration are of utmost importance whereas ongoing trials define effective therapies.

Halabe Bucay (2009) stated that anorexia nervosa is a serious, multi-factorial disease, characterized by psychiatric and neurological disturbances, which would appear to be similar to the manifestations of dementia.  Patients with anorexia nervosa present compromised affectivity, characterized by hypomanic, manic and depressive symptoms, and their cholinergic system is altered with a decrease in the release of acetylcholine.  Donepezil is a drug that been proven to be effective in the treatment of dementia, including Alzheimer's; it has been used for affective disorders and its mechanism of action is to inhibit the acetylcholinesterase enzyme to increase acetylcholine.  Thus, donepezil could be effective in treating anorexia nervosa.  However, there is a lack of evidence regarding the clinical value of cholinesterase inhibitors in treating anorexia.

In a review on the identification and treatment of eating disorders in the primary care setting, Sim and colleagues (2010) focused on the practical issues faced by primary care physicians in the management of these conditions and other issues central to the care of these complex patients with medical and psychiatric co-morbid conditions.  These investigators noted that there is little evidence to support the use of oral contraceptives in preventing bone loss in amenorrheic patients with an eating disorder.  They also stated that SSRIs have little benefits in treating eating disorder symptoms or preventing relapse in patients with anorexia nervosa.

In a placebo-controlled trial, McElroy and colleagues (2011) evaluated preliminarily the effectiveness of acamprosate in binge eating disorder (BED).  In this 10-week, flexible dose RCT, a total of 40 outpatients with BED received acamprosate (n = 20) or placebo (n = 20).  The primary outcome measure was binge eating episode frequency.  While acamprosate was not associated with a significantly greater rate of reduction in binge eating episode frequency or any other measure in the primary longitudinal analysis, in the endpoint analysis it was associated with statistically significant improvements in binge day frequency and measures of obsessive-compulsiveness of binge eating, food craving, and quality of life.  Among completers, weight and bone mass index decreased slightly in the acamprosate group but increased in the placebo group.  The authors concluded that although acamprosate did not separate from placebo on any outcome variable in the longitudinal analysis, results of the endpoint and completer analyses suggested the drug may have some utility in BED.

Brandys et al (2011) noted that brain derived neurotrophic factor (BDNF) is involved in neuroplasticity, and in the homeostatic regulation of food intake and energy expenditure.  It also has a role in stress responsivity and reward processing.  On the basis of its involvement in these various processes, BDNF can be hypothesized to be an important factor in the development and maintenance of anorexia nervosa (AN).  These reserchers meta-analytically summarized investigations of serum BDNF concentrations in people currently ill with AN, in comparison to healthy controls.  A total of 7 studies measuring BDNF in serum of individuals with AN (n = 155) and healthy controls (n = 174) were identified and included in the meta-analysis of the mean differences between case and control groups.  This study confirmed that AN is associated with decreased serum BDNF concentrations, in comparison to healthy controls.  The combined effect size (standardized mean difference, SMD) was large (SMD = -0.96; 95 % confidence interval [CI]: -1.33 to -0.59; p < 0.001).  Significant heterogeneity of effect sizes was identified (I(2) = 58.3 %; p < 0.001), which emerged as being primarily attributable to the first published study on the investigated association.  The authors concluded that meta-analytical summary of studies measuring circulating BDNF concentrations in women with AN and healthy controls confirms that it is significantly reduced in this patient group.  Moreover, difficulties associated with the measurement of BDNF have been identified and potential confounding factors have been discussed.  They stated that current data do not allow inferences to be made about causal links between levels of circulating BDNF and AN.  However, possible explanations for the relationship between BDNF and AN have been presented.

Dodds et al (2012) noted that the dopamine D(3) receptor is thought to be a potential target for treating compulsive disorders such as drug addiction and obesity.  These researchers used functional magnetic resonance imaging (fMRI) to investigate the effects the selective dopamine D(3) receptor antagonist GSK598809 on brain activation to food images in a sample of over-weight and obese binge-eating subjects.  Consistent with previous studies, processing of food images was associated with activation of a network of reward areas including the amygdala, striatum and insula.  However, brain activation to food images was not modulated by GSK598809.  The results demonstrated that D(3) receptor manipulation does not modulate brain responses to food images in over-weight and obese subjects.

Kontis and Theochari (2012) stated that dopamine has been implicated in the pathophysiology of AN by pre-clinical and clinical evidence.  Pre-clinical studies have examined 2 main characteristics of AN: (i) reduction in food intake (diet restriction), and (ii) hyperactivity.  Diet restriction has been associated with reduced dopamine levels in the hypothalamus, hippocampus, and the dorsal striatum.  Animal hyperactivity following diet restriction has been linked to increased dopamine in the hypothalamus.  Increased dopamine in the nucleus accumbens was associated with food administration, but not food expectation.  Tyrosine and dopaminergic antagonists normalized anorexia-like behaviors in animal models of AN, but did not restore body weight.  Clinical studies on the etiology of AN have produced contradictory findings.  Cerebrospinal fluid concentrations of dopamine and its metabolites have been reported to be decreased or normal under conditions of low weight, whereas they tended to normalize when the weight was restored.  Plasma and urinary levels of dopamine and its metabolites have been found to be normal, increased, and decreased.  Neuroendocrine studies suggested that dopaminergic neurotransmission is increased in AN.  However, recent neuroimaging studies lend support to the increase in binding of dopaminergic receptors in the striatum, which favors the opposite theory that intra-synaptic dopamine is indeed decreased.  Genetic studies implicate dopamine D2 receptors, the dopamine transporter, and the enzyme COMT.  The authors concluded that there are promising results with respect to the use of atypical antipsychotics against symptoms of AN beyond weight gain, but further trials are required.

Anorexia nervosa is characterized by a chronic course that is refractory to treatment in many patients and has one of the highest mortality rates of any psychiatric disorder.  Deep brain stimulation (DBS) has been applied to circuit-based neuropsychiatric diseases, such as Parkinson's disease and major depression, with promising results.  In a pilot study, Lipsman et al (2013) evaluated the safety of DBS to modulate the activity of limbic circuits and examined how this might affect the clinical features of AN.  In this prospective trial, subcallosal cingulate DBS was administered in 6 patients with chronic, severe, and treatment-refractory AN.  Eligible patients were aged 20 to 60 years, had been diagnosed with restricting or binge-purging AN, and showed evidence of chronicity or treatment resistance.  Patients underwent medical optimization pre-operatively and had baseline body-mass index (BMI), psychometric, and neuroimaging investigations, followed by implantation of electrodes and pulse generators for continuous delivery of electrical stimulation.  Patients were followed-up for 9 months after DBS activation, and the primary outcome of adverse events associated with surgery or stimulation was monitored at every follow-up visit.  Repeat psychometric assessments, BMI measurements, and neuroimaging investigations were also done at various intervals.  Deep brain stimulation was associated with several adverse events, only one of which (seizure during programming, roughly 2 weeks after surgery) was serious.  Other related adverse events were panic attack during surgery, nausea, air embolus, and pain.  After 9 months, 3 of the 6 patients had achieved and maintained a BMI greater than their historical baselines.  Deep brain stimulation was associated with improvements in mood, anxiety, affective regulation, and AN-related obsessions and compulsions in 4 patients and with improvements in quality of life in 3 patients after 6 months of stimulation.  These clinical benefits were accompanied by changes in cerebral glucose metabolism (seen in a comparison of composite PET scans at baseline and 6 months) that were consistent with a reversal of the abnormalities seen in the anterior cingulate, insula, and parietal lobe in the disorder.  The authors concluded that subcallosal cingulate DBS seems to be generally safe in this sample of patients with chronic and treatment-refractory AN.  The effectiveness of this approach awaits results from well-designed studies.

The ION™ (Individual Optimal Nutrition) analysis/profile (Metametrix, Norcross, GA) is a comprehensive combination of nutritional analyses that measures blood and urine levels of amino acids including arginine, glycine, and tryptophan  (plasma), anti-oxidants, fatty acids including monounsaturated fatty acids, polyunsaturated omega-3 and omega-t fatty acids (plasma), minerals/elements including calcium, copper, manganese, magnesium, potassium, selenium, and zinc  (red blood cells), organic acids (urine), oxidation products, toxins, and vitamins including vitamin A, B, C, D and E (serum for fat-soluble vitamins).  It supposedly offers a complete evaluation of functions that impact patients' mental and physical well-being.  http://www.gdx.net/core/interpretive-guides/ION-IG.pdf.  However, there is a lack of evidence regarding the clinical value of the ION analysis/profile in the diagnosis/evaluation of patients with eating disorders.