Short summary

In severe anorexia nervosa (SAN) it remains to be clarified, which psychiatric symptoms are the direct consequence of malnutrition and adaptation to starvation, and which are not. The main objective of this study is in detail to elucidate the short-term changes in psychopathology, depression, anxiety, and cognitive functions in relation to intensive nutritional rehabilitation with weight gain of 10-40 % in a specialized medical stabilization unit. Secondarily, it is examined whether cortisol levels are associated with depression, anxiety and cognitive impairments. The hypothesis is that improvement in nutritional status leads to clinically significant improvements in psychopathology and cognitive functions which are sustained over four to six weeks after discharge.

---

Rationale

Anorexia nervosa (AN) is a syndrome characterized by distorted body image and anxiety of gaining weight. Restrictive food intake and compensating behaviour (e.g. obsessive exercise and self induced purging) result in malnutrition. Severe AN (SAN) has the largest mortality rate among all psychiatric disorders. The majority of the deaths are related to malnutrition and its consequences. No effective evidence-based treatment is available. The aetiology is unknown, and it is undetermined which psychological symptoms are a consequence of malnutrition and which are premorbid or comorbid symptoms. Depression and anxiety are coexisting and well-known traits in patients with SAN. However, antidepressant treatment has not shown a therapeutic effect in AN. Varying degrees of cognitive impairment are likewise well-known traits in AN, but are only scanty elucidated in the literature.

There is a rather clinically consensus on depression, anxiety, and cognitive impairment in SAN as a sequalae to malnutrition. Knowledge about the relation between psychopathology and malnutrition in SAN is currently limited to a small amount of studies with contradictory conclusions. Furthermore, it seems, patients with AN have impairment in psychomotor speed, attention, and higher cognitive functions as flexibility, abstraction, and problem-solving. However, there are no published longitudinal re-nutrition studies of the relations between weight restoration and cognitive functions in SAN. A consequence of starvation stress is that the endocrine axes are altered, leading to protein- and energy preserving adaptation. Thus, in particular increased levels of cortisol are well described in AN. Depression, anxiety, and cognitive impairment is a well-described side effect of pharmacological treatment with corticosteroids and is well-known symptoms in excess endogenous cortisol production. There is preliminary experimental evidence that pharmacological cortisol synthesis inhibition may have therapeutic efficacy in depression. Cortisol has been found to correlate positively with anxiety and depression. However, this is not investigated in AN.

In international and national guidelines, it is recommended that treatment efforts in AN are multidisciplinary and two-form psychiatric/somatic. However, the effect of of nutritional rehabilitation on depression, anxiety, and cognitive impairment remains to be clarified. We have conducted a pilot project on five patients before and after weight gain. The pilot project yielded preliminary evidence that patients with extremely low BMI were not cognitively impaired, except for the executive function mental flexibility. However, after re-nutrition the patients performed better on psychomotor functioning, attention, and memory.

The primary aim of the project is in detail to clarify short term changes in depression, anxiety and cognitive functions before and after intensive nutritional rehabilitation treatment with weight gain on 10-40 % in a medical specialized medical stabilization unit for SAN. As a control the retest effect will be assessed on standard weight outpatients with an eating disorder. Secondarily, it will be examined whether the above mentioned effect is related to subtypes (restrictive/binge eating - purging), psychiatric or somatic co-morbidity, psychopharmacological treatment, illness duration and BMI during hospitalization, and if cortisol levels are associated with depression and anxiety symptomatology and cognitive impairment.

---

Description of the cohort

Center of Eating Disorders at Odense University Hospital is one of three highly specialized centres in Denmark. Within the centre there is a strongly formalized cooperation between psychiatric and somatic units. Patients with life-threatening weight loss are hospitalized in the Nutrition unit for somatic stabilization and weight gain on typical 10-40 %, after which they are either transferred to a psychiatric specialist unit or are discharged to a tw- form psychiatric/somatic outpatient treatment course. The unit receives patients from all over the country. The median hospitalization BMI is about 13.5 (span 7.8 to 25.8). Internationally, there are only few somatic units with corresponding specialization and patient volume, and therefore the unit makes a suitable fundament for studying the effect of intensive re-nutrition. The centre's organization and patient population are described in various observational and intervention studies.

N = 30 patients with SAN, hospitalized in the Nutrition unit, will undergo three sessions of a psychometric assessment programme, at admission, at discharge (or drop-out), and 2-4 months after. N = 15 weight-stable outpatients with eating disorders will undergo two sessions of the test battery to test the retest effect, with a 4-6 weeks' time interval.

Inclusion criteria:

Inpatients:

• Patients with fulfilling Diagnostic and Statistical Manual of Mental Disorders -fifth edition (DSM-5)-criteria for AN.

Outpatients:

• Patients who are weight stable (<5 % weight change) and have fulfilling DSM-5-criteria for AN, bulimia nervosa (BN), or other eating or feeding disorders (OSFED).

Exclusion criteria:

• Patients who cannot complete a Danish-language questionnaire
• Patients who primarily are hospitalized for a few days for liquid-electrolyte correction
• Patients with active drug or alcohol abuse
• Patients with comorbid schizophrenia
• Patients where the psychopharmacological treatment is qualitatively changed during the re-nutrition
• Patients under the age of 16

Data and biological material

Clinical/somatic data:

• Morning weight
• Height
• Blood pressure
• Illness duration
• Comorbid psychiatric or somatic diagnoses
• Psychopharmacological drugs

Biological material:

• Cortisol excretion in urine (nmol Cortisol pr. day)

Questionnaires:

• Eating Disorder Inventory - third version (EDI-3)
• Toronto Alexithymia Scale (TAS-20)
• Beck Depression Inventory - second version (BDI-2)
• Hospital Anxiety and Depression Scale (HADS)
• The Perceived Stress Scale (PSS)

Measuring eating disorder, anxiety, and depression symptomatology.

Cognitive test battery:

• Wechsler Memory Scale - third version (WMS-III)
• Coding and Symbol Search from Wechsler Adult Intelligence Scale - fourth version (WAIS-IV)
• D 2 test of attention
• Trail making A & B
• Design Fluency test
• Test of word mobilization (ordmobilisering)
• Wisconsin Card Sorting Test - revised (WCST-R)

Measuring: Psychomotor speed, attention, memory, executive functions (e.g. mental flexibility, abstraction, and problem-solving). 

Publications associated with the project

Department of Endocrinology, Odense University Hospital

• Associate professor René Klinkby Støving, MD, PhD
• Assistant professor Mia Beck Lichtenstein, Cand. Psych., PhD
• Research assistant Simone Nejrup Hemmingsen, Cand. Psych.
• Project nurse Kirsten Gitte Hansen

The Nutritional Unit, Department of Endocrinology, Centre for Eating Disorders, Odense University Hospital

The Child- and Adolescent Psychiatry, Odense University Hospital, The Psychiatry of Southern Denmark

University of Southern Denmark.