I recently spoke to Dr. David Herzog about new research in eating disorders over the last few years. Dr. Herzog, who is director of the eating disorders program at Mass. General Hospital, suggested three hot topics to discuss in this issue: upcoming diagnostic changes in DSM-V, evidence on the efficacy of family therapy in adolescents with eating disorders, and binge eating disorder, a diagnosis that has been receiving more airplay lately.

I’ve covered each of these topics below, and a few more as well. As you’ll see, eating disorders are still among the most challenging treatment issues in psychiatry. “We still have no magic bullets in terms of drugs,” commented Dr. Herzog. “These are very complex disorders and require treatment from different disciplines.”

There used to be only two eating disorders: anorexia nervosa and bulimia nervosa. Anorexia is defined as disordered body image along with refusal to maintain weight above 85% of ideal body weight, and, in women, amenorrhea. Bulimia is defined as frequent episodes of binge eating, along with extreme efforts to lose weight, such as purging or excessive exercise.

While these are real diagnostic entities, it turns out the majority of people who are referred to eating disorders specialists end up being diagnosed with neither AN nor BN, but rather with something called EDNOS – “eating disorder, not otherwise specified.” In fact, up to 60% of people with eating disorders are diagnosed with EDNOS, a heterogeneous entity that includes people with subthreshold eating disorders, binge eating disorder (not yet an official DSM diagnosis), and a smattering of other diagnostic entities. The problem is that the EDNOS diagnosis is too broad to guide our treatment decisions.

Well, a recent article in the International Journal of Eating Disorders lays out a blueprint for DSM-V in order to remedy this problem (Wilfley DE et al., 2007, early view, published online Aug 8 2007). According to the authors, all of whom are prominent researchers in the field, the following changes are likely to be incorporated in DSM-V.

1. The amenorrhea criteria will be removed from AN, since there are no prognostic differences between anorexics with and without amenorrhea.

2. The minimum required frequency of binging will be lowered to once weekly (from the current twice weekly), because there is no evidence that patients who binge weekly are less sick than those who binge more frequently.

3. Binge eating disorder (BED) will become an officially recognized diagnosis. BED, currently listed as a “disorder worthy of study,” is now defined as recurrent binge eating (at least twice weekly for 6 months) without efforts to lose weight (i.e., bulimia without the purging). As with bulimia, the criteria for BED will be loosened a bit, to once weekly binges for 3 months.

Of course, anytime the DSM adds or broadens diagnoses, there is a concern among some that a hidden agenda is to bring more potential customers into the psychiatric fold. In this case, however, it’s likely that the total number of patients diagnosed with eating disorders won’t change drastically. Instead, the “EDNOS” category will shrink at the expense of AN, BN, and BED, all of which will expand. Hopefully, this will allow us to make our eating disorder diagnoses significantly more precise.

How is all this helpful for front-line clinicians right now? Be less insistent about applying rigid binge frequency criteria in your bulimia diagnoses, and throw out the amenorrhea criterion for AN. Focus less on checklists of symptoms and more on the core psychopathology of most eating disorders, which is, in its essence, the excess concern about body shape and weight.

As I have adopted this different mind-set in my own practice, I’ve been picking up on important body image and weight concerns in many more patients, even in those who would not qualify for a formal eating disorder. This leads to productive discussions and in some cases, informs treatment.

Anorexia is a difficult condition to treat, and the prognosis is not particularly good. In one observational study, 75 adolescent patients with AN were followed for 2-7 years. Overall, only 45% had a good outcome, meaning maintenance of weight above 85% of ideal weight and good social functioning. Interestingly, the 21 patients who had received inpatient care did significantly worse than those who did not, which may imply that inpatient care is counterproductive, or, more likely, that those who are sick enough to be admitted to a hospital have a poorer prognosis regardless of treatment (Gowers et al., Brit J Psychiatry 2000;176 (2): 138-141).

While we often think of anorexia as one of the more lethal illnesses in psychiatry, in fact death is unusual when weight loss is due purely to self-starvation. According to one study, about 5% of patients died over an 11 year follow-up, and death was most likely in anorexics who binge and purge, who are alcoholic, or who have affective disorders (Herzog DB, et al., Int J Eating Disord 2000;28:20-6).

Antidepressants do not help much for anorexia. Early trials found SSRIs to be ineffective for anorexia in the underweight stage. Because of concerns that the starved state itself might account for lack of medication response, researchers recently conducted a randomized controlled trial of fluoxetine in anorexia after weight restoration. In that study, 93 patients with AN who regained weight to a minimum body mass index of 19 were randomly assigned to placebo or fluoxetine (mean dose 63.5 mg/day). There was no significant difference in the percentage of patients who were able to maintain their weight for 52 weeks – 26.5% for the fluoxetine group versus 31.5% for the placebo group (Walsh BT et al., JAMA 2006;295(22):2605-2612).

The hot treatment for anorexia in adolescents these days is a type of family therapy originally developed at Maudsley Hospital in London, called Family-Based Therapy (FBT). Several studies have shown that this technique outperforms individual therapy (for a good review, see Le Grange D, World Psychiatry 2005 October; 4(3): 142-146). What is FBT, exactly? I spoke briefly with Dr. Le Grange, who is an Associate Professor of Psychiatry at University of Chicago, and who is one of the leading researchers on the technique.

FBT is a three-phase therapy in which parents are encouraged to temporarily take full control of food decisions. This means that the parents prepare all the food and watch the adolescents while they eat. While it sounds like it is modeled after eating disorders inpatient units, Dr. Le Grange prefers to emphasize that the technique encourages parents to become re-involved in their child’s care. “FBT is a reinvigoration of the kinds of things any parent would want to do with a sick child. In the past, parents were encouraged to remain on the sidelines of anorexia treatment, and were sometimes blamed for their children’s condition. In my experience, parents often see this technique as giving them permission to be good parents again.”

In FBT, once the adolescent achieves normal weight, phase 2 involves giving control over food back to the child, and phase 3 examines the effects the eating disorder has had on the child’s school and social life. FBT has been shown to be effective primarily in the age range of 12-18; studies with older patients have yielded unimpressive results. This is likely due to the fact that it is harder to locate family members who have sufficient leverage over older patients.

While most of the studies have focused on FBT for anorexia, Le Grange and colleagues published a new study of FBT for adolescents with bulimia in the most recent issue of Archives of General Psychiatry (Le Grange et al., 2007; 64(9):1049-1056). The study randomized 41 patients to FBT and 39 patients to supportive treatment (ST). After an average of 18 weekly to biweekly sessions, the FBT group had a higher rate of binge-purge abstinence than the ST group (FBT, 39%, ST, 18%). At 6 month follow-up, about half of abstinent patients relapsed, though FBT was still statistically superior to ST at that later time point.

Those interested in learning more about the FBT method can find treatment manuals and a book for parents, Help Your Teenager Beat Your Eating Disorder, all via Guilford Press (www.guilford.com).

An excellent review of the research on bulimia was recently published (Shapiro JR et. al, Int J Eat Disord 2007; 40:321-336) covering medication and psychotherapy studies. In terms of medications, that old standard, high-dose Prozac, is still considered the first line. 60 mg per day has been convincingly shown to be superior to 20 mg per day. Oddly enough, even though clinical lore has it that all SSRIs work about the same for bulimia, at least if dosed high enough, the only other SSRI to have been adequately evaluated is fluvoxamine, which was superior to placebo at an average dose of 182 mg/day. The biggest surprise? Trazodone effectively treated bulimia when dosed at 400 mg/day. (Maybe the patients slept through their meals!) Other meds shown effective in single studies have included desipramine (200-300 mg/day), topiramate (100 mg/day), and odansetron. As a reminder, it's best to avoid Wellbutrin in patients with bulimia or anorexia, because one early study indicated a higher seizure risk in such patients.

In terms of therapy, the review endorsed cognitive behavioral therapy (CBT) as being the most effective psychotherapy, although the new data on family-based therapy for bulimia (see above) had not been released in time for the review.

Think of binge eating disorder – BED – as bulimia without the purging. As you might imagine, many patients with BED also suffer from obesity – at least 50%, according to studies. This is about twice the rate seen in bulimia. On the other hand, it would not be correct, politically or scientifically, to think of obesity as an eating disorder, since only about 2% of obese people in the community have BED, although the prevalence goes up to 25% among severely obese patients seeking treatment (Yanovski SV, Int J Eat Disord 2003; 34:S117-S120.) BED also differs from other eating disorders in that men are almost as likely to receive the diagnosis as are women.

SSRIs appear to work as well for BED as they do for BN. Studies have endorsed fluoxetine (avg. dose 70 mg/day), fluvoxamine (50-300 mg/day), sertraline (187 mg/day), and citalopram (40-60 mg/day). Other controlled studies have found the following medications significantly more effective than placebo: imipramine, 25 mg TID; topiramate, 212 mg QD; and sibutramine (Meridia), 15 mg/day. Among psychotherapies, CBT has the best evidence base for effectiveness (Brownley KA et al., Int J Eat Disord 2007; 40:337-348).

Recently, Night Eating Disorder (NED) reached the big time when the American Journal of Psychiatry published a placebo-controlled trial of sertraline for its treatment (O’Reardon JP et al., 2006; May,163(5):893-8.) The study randomly assigned 34 outpatients to either sertraline, 50-200 mg/day, or placebo. The response rate on sertraline was 71%, significantly greater than the 18% response rate on placebo.

But what is NED, exactly, and is it a valid diagnosis? The authors of the AJP study defined the disorder as: “morning anorexia” (i.e., skipping breakfast); “evening hyperphagia” (binging at night); and "insomnia with awakenings followed by nocturnal ingestions" (getting up in the middle of the night to eat even more food). Unfortunately, many researchers are not convinced that NED qualifies as legitimate disorder. In a recent review, eating disorders specialists wrote that NED does not meet the five criteria usually required to judge an entity ready for inclusion as a valid diagnosis in DSM (Striegel-Moore RH et al.,  Int J Eat Disord 2006;39:544-549). Specifically, they argue that:

1. Researchers have not been able to agree on a definition of NED;

2. There are no generally accepted diagnostic instruments for it, and

3. It is not clearly distinguishable from other diagnoses, such as binge eating disorder.

To make matters even more confusing, another “disorder” – sleep-related eating disorder (SRED) – has been making the rounds of the journals. This one is related to sleep-walking, and is defined by waking up at night and sleep-eating, often with little awareness or memory of the activity. According to a recent review in Current Psychiatry, SRED can be triggered by zolpidem (Ambien) and other meds, and has been successfully treated with topiramate, 25 mg to 150 mg at bedtime (Howell MJ et al., 2007;6[7], accessed online).

What’s the bottom line? Regardless of whether they get a DSM diagnosis, many patients have difficulty controlling urges to overeat at night. Asking specifically about this is a good idea in any patient with eating disorders or with weight problems.