Medical Management of Eating Disorders: An Update

Ulrich Voderholzer; Verena Haas; Christoph U. Correll; Thorsten Körner


Curr Opin Psychiatry. 2020;33(6):542-553. 

In This Article

Recent Findings Regarding the Medical Management of Medical Abnormalities in Patients With Eating Disorders

Nutritional Management of Anorexia Nervosa Patients

Accelerated Instead of Slow Refeeding. It was previously assumed that starting the refeeding process with low initial caloric intake (1200 kcal/day) and advancing slowly by 200 kcal every other day would prevent the potentially life-threatening refeeding syndrome. However, a more recent systematic review supported higher calorie approaches in mildly and moderately malnourished patients under close medical monitoring.[22] In addition in 103 extremely malnourished adults targeting higher caloric intake (i.e., 2000 kcal/day at the initiation of treatment) was not associated with refeeding complications when closely monitoring and supplementing phosphate and thiamine, but with normalization of weight and starvation-related laboratory abnormalities.[23] Accelerated refeeding led to rapid normalization of abnormal laboratory values (Figure 2a–c, 4-week course of creatine kinase, aspartate transferase and phosphate). Accelerated refeeding leads to quicker reversal of underweight, which is a positive predictor of mid-term and long-term outcomes.[24] Although there is still no international consensus on refeeding practices in patients with anorexia nervosa, recent data suggest that the 'start slow' approach may be outdated.[25]

Figure 2.

(a) Course of creatine kinase over 4 weeks in extremely underweight patients with anorexia nervosa during rapid refeeding according to Koerner et al. [23]. Creatine kinase: outliers above 303 U/l not shown in the graph: day 0, n = 6 (359–9941 U/l); day 7, n = 2 (464–1537 U/l); day 14, n = 1 (303 U/l); day 21, n = 0; day 28, n = 1 (374 U/l). (b) Course of aspartate transferase over 4 weeks in extremely underweight patients with anorexia nervosa during rapid refeeding according to Koerner et al. [23]. Outliers above 120 U/l not shown in the graph: day 0, n = 12 (123–1078 U/l); day 7, n = 3 (185–432 U/l); day 14, n = 0; day 21, n = 0; day 28, n = 0. (c) Course of phosphate over 4 weeks in extremely underweight patients with anorexia nervosa during rapid refeeding and phosphate supplementation according to Koerner et al. [23].

Food Composition and Route of Intake. So far, there is very little research on the ideal macronutrient content and route of delivery for patients with anorexia nervosa. While Kohn et al.[26] suggested to reduce carbohydrate content to decrease the risk of refeeding syndrome, to our knowledge, no randomized controlled trial comparing refeeding protocols with different macronutrient composition has been performed. Refeeding practices also seem to be a result of different patient characteristics (i.e. with respect to the degree of malnutrition), settings (i.e. medical or psychiatric wards), healthcare systems (i.e. long vs. short inpatient care) as well as traditional and regional developments over time. Some institutions focus on quick normalization of eating behavior by introducing oral and well balanced nutrition on the first day of treatment which was feasible even in extremely malnourished adults.[23] Other authors describe a range of benefits achieved by administering a nasogastric tube, such as less electrolyte imbalances and cardiac disturbances,[27] decreased frequency and severity of binge/purge episodes, and, in particular in combination with higher caloric refeeding, potential reduction of anxiety, gastric distention, abdominal pain and early satiety.[28] A systematic review examining the role of nasogastric feeding concluded that contrary to common concerns, nasogastric tube feeding was not associated with an increased risk for adverse outcomes, well tolerated by the patients, and effectively increased caloric intake and weight gain, both in adolescents and adults.[28,29] However, to date, there are still insufficient, well designed, controlled studies to conclude on the superiority of nasogastric tube vs. oral refeeding in patients with anorexia nervosa.

Micronutrient Deficiencies

Some clinical symptoms that are observed in anorexia nervosa could be partially explained by micronutrient deficiencies.[30] For example, Suzuki et al.[31] described an association between zinc deficiency and restrictive eating behavior; cases of sensory neuropathy were found in anorexia nervosa patients with vitamin B12 deficiency;[32] fasting can cause neurological complications due to severe vitamin B1 deficiency.[16] Hanachi et al.[30] determined micronutirent status in 374 severely malnourished anorexia nervosa patients and found that zinc deficiency was most prevalent (64.3%), followed by vitamin D (54.2%), copper (37.1%), selenium (20.5%), vitamin B1 (15%), vitamin B12 (4.7%) and vitamin B9 (8.9%). The authors concluded that micronutrients status of anorexia nervosa patients should be monitored and supplemented to prevent deficiency-related complications and improve nutritional status. Prospective studies are needed to explore the symptoms and consequences of each deficiency, which can aggravate the prognosis during recovery.

Prevention and Treatment of Bone Loss

Osteopenia and osteoporosis are common medical findings in anorexia nervosa and their degree depends on the extent and duration of weight loss and starvation.[33] As standard measures, supplementation with vitamin D and calcium are recommended. Dual energy x-ray absorbtiometry (DXA) is the measure of choice to diagnose bone mass density. Major causes of osteopenia and osteoporosis are the endocrine dysregulation associated with underweight, that is hypercortisolism and estrogene deficiency and vitamin D and calcium deficiency.

Supplementing estrogene (transdermally) has therefore been recommended in anorexia nervosa to reduce the risk of bone loss. Many patients with anorexia nervosa are using oral contraceptives, which has been criticized for several reasons. First, earlier studies did not prove a protective effect of oral contraceptives on bone mineral density in anorexia nervosa.[34] Second, combined hormonal contraceptives induce cyclic bleeding and may therefore disguise that weight loss causes amenorrhea. This masking may negatively impact on the motivation for further weight gain. In a recent cross-sectional analysis of 301 adult females with anorexia nervosa (99 on oral contraceptives) and 121 age matched healthy controls, however, whole body, lumbar spine, femoral neck, hip and radius bone mineral density values (DEXA and bone turnover markers) were systematically higher in anorexia nervosa patients taking vs. not taking oral contraceptives.[35] These differences even persisted after multiple adjustments. The data further showed that bone mineral density preservation improved with longer durations of oral contraceptives use and shorter delays between disease onset and the start of oral contraceptives. The authors concluded from their data that oral contraceptives might be prescribed for young women with anorexia nervosa to limit bone loss. Although prospective data are sorely needed, the so far rather critical restraint toward oral contraceptives in anorexia nervosa must be challenged.

The Gut Microbiome and Eating Disorders

There is a growing body of literature that implicates a potential role of the gut microbiota in the cause and progression of eating disorders.[36,37] Gut bacteria may act on the gut–brain axis altering appetite control and brain function as part of the genesis of eating disorders.[36] As the illnesses progress, extreme feeding patterns and psychological stress potentially feed back to the gut ecosystem, which can further compromise physiological, cognitive and social functioning. However, the microbiome is a quite complex entity and initial studies analyzing altered microbiomes in patients with anorexia nervosa have reported heterogeneous results.[37] As alterations of the microbiome in anorexia nervosa patients persist after short-term weight restoration,[38] microbiome-directed interventions could become an adjunctive treatment, for example, by special nutritional supplies, use of probiotics or prebiotics or drugs influencing the microbiome.[39]

New Findings on the Pharmacotherapy of Anorexia Nervosa

One aspect of medical management of eating disorders is the use of psychotropic medications[40,41] that are a key treatment approach for many psychiatric illnesses. However, the role of psychotropic medications in the treatment of eating disorders differs somewhat from that in other psychiatric disorders. For all eating disorders, psychopharmacological agents should not be a primary treatment approach, but rather a potential adjunctive treatment embedded in nutritional and psychotherapeutic treatment approaches. Current evidence for the efficacy of psychotropic medications and regulatory approval differs across eating disorders. Numerous studies have shown some efficacy of antidepressants for bulimia nervosa. High-dose fluoxetine (i.e. 60 mg/day) is the standard of care medication approach for bulimia nervosa. Fluoxetine is approved as pharmacotherapy for bulimia nervosa in many countries. There are also many studies showing moderate effects of antidepressants on symptoms of BED, but with apparently little effect on body weight change. There is also evidence for the efficacy of stimulants for BED and lisdexamfetamine is approved for treatment of BED in the USA.

Regarding anorexia nervosa, for a long-time studies with psychopharmacological agents had been negative.[42] Recently a placebo-controlled randomized controlled trial (RCT) was conducted in 152 adult outpatients with anorexia nervosa with a mean BMI of 16.7 kg/m2 who received either olanzapine 5–10 mg/day (mean = 6.2 mg/day) or placebo for 16 weeks.[43] The study documented a modest therapeutic effect of olanzapine on BMI, but not on psychological symptoms. Importantly, different from nonanorexia nervosa populations,[44,45] olanzapine was not associated with cardiometabolic adverse effects on the lipid and glucose metabolism. This is an important new finding, because for the first time a significant therapeutic effect of a pharmacological agent has been shown in patients with anorexia nervosa in a large RCT. Olanzapine may therefore be a new therapeutic option for anorexia nervosa patients who do not respond to standard nutritional and psychotherapeutic treatment and who accept and tolerate this treatment.

Currently, there is no evidence for the use of antidepressants in anorexia nervosa. Notably, however, absence of evidence does not necessarily mean absence of an effect. Almost all studies with antidepressants and antipsychotics in anorexia nervosa apart from the olanzapine study by Attia et al.[43] had been underpowered. To date, there is no study with an antidepressant in anorexia nervosa with a sufficiently large sample size to detect significant differences. Such a study is urgently needed in the view of the frequent comorbidity of depression with anorexia nervosa and the fact that mean self-rated depression scores of patients with eating disorders are comparable with patients with depressive disorders.[46]