Obesity Research & Clinical Practice (2007) 1, 139—146
ORIGINAL ARTICLE
Metabolic dysfunction in anorexia nervosa Leonie K. Heilbronn a, Kerry-Lee Milner a, Adamandia Kriketos c, Janice Russell b, Lesley V. Campbell a,∗ a
Diabetes and Obesity Research Program, Garvan Institute for Medical Research, 384 Victoria Street, Darlinghurst, NSW 2010, Australia b Eating Disorders Program, Northside Clinic, University of Sydney, 2 Greenwich Road, Greenwich, NSW 2065, Australia c Department of Medicine, University of Melbourne, Vic. 3010, Australia Received 21 November 2006 ; received in revised form 8 March 2007; accepted 8 March 2007
KEYWORDS Glucose metabolism; Lipid metabolism; Anorexia nervosa; Body composition
∗
Summary Context: Anorexia nervosa (AN) is an eating disorder characterized by self-induced energy deficit and low body weight with major consequences for most organ systems and a tendency towards self-perpetuation. Objectives: To compare metabolic responses to glucose and exercise in women hospitalized with AN (n = 10) before and after 6-weeks weight gain program and in lean healthy weight women (BMI < 22 kg/m2 ) (n = 7). Main outcomes: Weight, body composition, indirect calorimetry, and response of serum insulin, glucose, adiponectin and leptin to oral glucose (75 g) and to 30-min of cycling at 50 rpm. Results: Patients with AN had similar lean mass to controls, but had significantly less body fat. Adiponectin was 43% higher (p < 0.01) and leptin 47% lower in AN subjects versus controls (p = 0.04). In response to moderate exercise, fasting glucose increased in AN (p < 0.05), but was unchanged in controls. After glucose ingestion, a trend towards a greater increase in diet-induced thermogenesis was also observed in patients with AN (p = 0.07). Despite a further 6 weeks as in patients, weight was not significantly changed in AN. Similarly, glucose, insulin, leptin or adiponectin were not altered. Conclusions: AN patients appear metabolically healthy under resting conditions, but their responses to physiological stressors differed from those of controls. Potential impediments to weight gain should be further investigated to define mechanisms with a view to improving the effectiveness of nutritional management. © 2007 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
Corresponding author. Fax: +61 29295 8201. E-mail address: [email protected] (L.V. Campbell).
Anorexia nervosa (AN) is an eating disorder affecting mostly adolescent and young adult females characterized by self-induced energy deficit, per-
1871-403X/$ — see front matter © 2007 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.orcp.2007.03.003
140 sistently low body weight and resistance to weight gain, associated with unique psychopathology, a self-perpetuating and not infrequently chronic course. Metabolic abnormalities described in this condition include reduced body temperature and energy expenditure, bradycardia, low thyroxine, amenorrhoea and osteoporosis [1]. Resistance to weight gain is common and usually attributed to poor compliance. However, disproportionately high diet-induced thermogenesis (DIT), increased energy requirements with exercise and increased physical activity have been reported in AN [2—5], suggesting that metabolic impediments to weight gain might operate to increase the difficulty of weight restoration. Anorexic patients have low levels of adipose tissue. In addition to being a storage depot for triglyceride, adipose tissue is now recognized to be an active endocrine organ that secretes numerous hormones and cytokines, including leptin and adiponectin. These adipocytokines affect many metabolic parameters including insulin sensitivity, energy expenditure and food intake. Leptin is secreted in proportion to adipose tissue mass with high concentrations in obese individuals [6]. Conversely, complete lack of leptin results in severe obesity in humans and mice [7,8]. AN patients have low circulating leptin levels [9], although the physiological consequences of this are not entirely clear apart from demonstrated effects on activity levels [10]. Adiponectin production is inversely proportional to fat mass, and obese subjects have lower concentrations of adiponectin compared to lean subjects [11]. Adiponectin concentrations are reported to be higher in AN [9] and are related to insulin sensitivity via upregulation of AMP-activated protein kinase (AMPK) [12]. Reports in the literature are conflicting as to whether individuals with anorexia are more or less insulin sensitive or whether there is no change; both increased or decreased insulin sensitivity having been described [13—15]. To our knowledge, few studies have reported response to moderate stressors, such as exercise in patients with anorexia nervosa. The aim of this study was to investigate the metabolic response to moderate metabolic challenges in subjects with AN before and after 6-weeks in-patient re-feeding and compared to healthy lean controls.
Materials and methods Subjects Participants with anorexia nervosa (AN) were recruited whilst inpatients of a specialized eating
L.K. Heilbronn et al. disorders program in a university affiliated psychiatric clinic. AN patients were included if they were >18 years and had a BMI < 18 kg/m2 . AN patients were first tested following a mean weight gain of 3.5 kg (range 0.9—6.0 kg) in a mean period of 41 days after admission to the clinic (range 14—85 days, mode 28—35 days) of nutritional intervention. Weight histories for the 12 months preceding admission were available in 6 of 10 anorexic subjects. Mean weight change in the past 12 months was −2.95 kg (range −8.2kg to +6.9 kg). The majority of AN subjects were taking a selective serotonin reuptake inhibitor (SSRI) anti-depressant (n = 4) and/or a small dose of a second generation antipsychotic medication (n = 4) at the time of testing. Seven AN subjects were amenorrhoeic and four were bulimic (Appendix A). Controls were recruited through advertisements at the university and hospital campuses and were weight stable for the past 6 months (
ORIGINAL ARTICLE
Metabolic dysfunction in anorexia nervosa Leonie K. Heilbronn a, Kerry-Lee Milner a, Adamandia Kriketos c, Janice Russell b, Lesley V. Campbell a,∗ a
Diabetes and Obesity Research Program, Garvan Institute for Medical Research, 384 Victoria Street, Darlinghurst, NSW 2010, Australia b Eating Disorders Program, Northside Clinic, University of Sydney, 2 Greenwich Road, Greenwich, NSW 2065, Australia c Department of Medicine, University of Melbourne, Vic. 3010, Australia Received 21 November 2006 ; received in revised form 8 March 2007; accepted 8 March 2007
KEYWORDS Glucose metabolism; Lipid metabolism; Anorexia nervosa; Body composition
∗
Summary Context: Anorexia nervosa (AN) is an eating disorder characterized by self-induced energy deficit and low body weight with major consequences for most organ systems and a tendency towards self-perpetuation. Objectives: To compare metabolic responses to glucose and exercise in women hospitalized with AN (n = 10) before and after 6-weeks weight gain program and in lean healthy weight women (BMI < 22 kg/m2 ) (n = 7). Main outcomes: Weight, body composition, indirect calorimetry, and response of serum insulin, glucose, adiponectin and leptin to oral glucose (75 g) and to 30-min of cycling at 50 rpm. Results: Patients with AN had similar lean mass to controls, but had significantly less body fat. Adiponectin was 43% higher (p < 0.01) and leptin 47% lower in AN subjects versus controls (p = 0.04). In response to moderate exercise, fasting glucose increased in AN (p < 0.05), but was unchanged in controls. After glucose ingestion, a trend towards a greater increase in diet-induced thermogenesis was also observed in patients with AN (p = 0.07). Despite a further 6 weeks as in patients, weight was not significantly changed in AN. Similarly, glucose, insulin, leptin or adiponectin were not altered. Conclusions: AN patients appear metabolically healthy under resting conditions, but their responses to physiological stressors differed from those of controls. Potential impediments to weight gain should be further investigated to define mechanisms with a view to improving the effectiveness of nutritional management. © 2007 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
Corresponding author. Fax: +61 29295 8201. E-mail address: [email protected] (L.V. Campbell).
Anorexia nervosa (AN) is an eating disorder affecting mostly adolescent and young adult females characterized by self-induced energy deficit, per-
1871-403X/$ — see front matter © 2007 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.orcp.2007.03.003
140 sistently low body weight and resistance to weight gain, associated with unique psychopathology, a self-perpetuating and not infrequently chronic course. Metabolic abnormalities described in this condition include reduced body temperature and energy expenditure, bradycardia, low thyroxine, amenorrhoea and osteoporosis [1]. Resistance to weight gain is common and usually attributed to poor compliance. However, disproportionately high diet-induced thermogenesis (DIT), increased energy requirements with exercise and increased physical activity have been reported in AN [2—5], suggesting that metabolic impediments to weight gain might operate to increase the difficulty of weight restoration. Anorexic patients have low levels of adipose tissue. In addition to being a storage depot for triglyceride, adipose tissue is now recognized to be an active endocrine organ that secretes numerous hormones and cytokines, including leptin and adiponectin. These adipocytokines affect many metabolic parameters including insulin sensitivity, energy expenditure and food intake. Leptin is secreted in proportion to adipose tissue mass with high concentrations in obese individuals [6]. Conversely, complete lack of leptin results in severe obesity in humans and mice [7,8]. AN patients have low circulating leptin levels [9], although the physiological consequences of this are not entirely clear apart from demonstrated effects on activity levels [10]. Adiponectin production is inversely proportional to fat mass, and obese subjects have lower concentrations of adiponectin compared to lean subjects [11]. Adiponectin concentrations are reported to be higher in AN [9] and are related to insulin sensitivity via upregulation of AMP-activated protein kinase (AMPK) [12]. Reports in the literature are conflicting as to whether individuals with anorexia are more or less insulin sensitive or whether there is no change; both increased or decreased insulin sensitivity having been described [13—15]. To our knowledge, few studies have reported response to moderate stressors, such as exercise in patients with anorexia nervosa. The aim of this study was to investigate the metabolic response to moderate metabolic challenges in subjects with AN before and after 6-weeks in-patient re-feeding and compared to healthy lean controls.
Materials and methods Subjects Participants with anorexia nervosa (AN) were recruited whilst inpatients of a specialized eating
L.K. Heilbronn et al. disorders program in a university affiliated psychiatric clinic. AN patients were included if they were >18 years and had a BMI < 18 kg/m2 . AN patients were first tested following a mean weight gain of 3.5 kg (range 0.9—6.0 kg) in a mean period of 41 days after admission to the clinic (range 14—85 days, mode 28—35 days) of nutritional intervention. Weight histories for the 12 months preceding admission were available in 6 of 10 anorexic subjects. Mean weight change in the past 12 months was −2.95 kg (range −8.2kg to +6.9 kg). The majority of AN subjects were taking a selective serotonin reuptake inhibitor (SSRI) anti-depressant (n = 4) and/or a small dose of a second generation antipsychotic medication (n = 4) at the time of testing. Seven AN subjects were amenorrhoeic and four were bulimic (Appendix A). Controls were recruited through advertisements at the university and hospital campuses and were weight stable for the past 6 months (
