A neurodevelopmental model for anorexia nervosa

https://doi.org/10.1016/S0031-9384(03)00101-XGet rights and content

Abstract

This paper integrates genetic and biological data on aetiological risk for anorexia nervosa (AN) with cognitive and psychosocial explanatory models. We have reviewed clinical and basic science data from each of these domains and then used a developmental perspective to formulate a multifactorial threshold model. By positioning interpersonal stress as a central component of this model, psychological, social and biological conceptualisations of AN can be used to generate a data driven, neurodevelopmental hypothesis for the aetiology of this complex disorder.

Introduction

Anorexia nervosa (AN) is a disorder of complex aetiology, in which genetic, biological, psychological and sociocultural factors appear to contribute significantly to susceptibility. However, few of these risk factors are specific to AN and no single factor has been shown to be either necessary or sufficient to express the disorder. A multifactorial threshold model is therefore an appropriate explanatory model.

In the proposed model, genetic factors and early life experience interact to generate susceptibility to a chronic submissive type stress response and to hypothalamic–pituitary–adrenal (HPA) axis dysregulation. Psychosocial and biological changes associated with puberty exacerbate vulnerability, such that when stress is encountered, the coping response is maladaptive and an aberrant HPA axis response is elicited. Specifically, the HPA axis fails to adapt to the chronicity of a stressor, in that there is persistently elevated corticotrophin releasing hormone (CRH) activity rather than a switch to an alternative adrenocorticotrophic hormone (ACTH) secretagogue, such as arginine vasopressin (AVP). Prolonged elevation of CRH release leads to a persistent loss of nutritional homeostasis. Data from both basic sciences and clinical research will be presented to support this neurodevelopmental model for the aetiology of AN.

Before describing any model of aetiology, it is necessary to define the clinical condition. Although this may seem to be self-evident, in that the condition is easily recognised by the characteristic severe underweight, the criteria used to define the illness in modern diagnostic systems have been criticised for overemphasis upon pathoplastic features of the disorder such as fear of fat, which may be absent in a significant proportion of cases [1]. Furthermore, there is disagreement about whether there is disturbance of appetite. Reduced hunger and desire to eat, coupled with increased fullness and satiety after a test meal, have been reported in AN [2], [3], although the capacity to respond to physiological hunger and satiety cues may not be entirely absent [4], [5]. Some authors argue that these findings reflect tight cognitive control of normal appetite [6]. However, relative to healthy comparison women, those with AN show reduced salivation [7], a heightened autonomic response to food [8] and fear and disgust in response to images of food [9]. Following recovery, there is a diminished response to the appetite suppressing effects of fenfluramine [10]. These objective data suggest that appetite regulation may indeed be impaired in AN [11].

The neurodevelopmental model of AN that is proposed does not address the fact that there are established restricting and binge-purge subtypes. Arguably, it is applicable to both subtypes because they both escape from normal weight and satiety-related mechanisms and differ more in terms of personality and behaviours [12].

Section snippets

Genetic factors

Family and twin studies indicate an increased risk of both AN and bulimia nervosa (BN) in relatives of AN and BN probands [13], [14], [15]. In addition, subthreshold forms of eating disorders (ED) appear to lie on a continuum of liability with full ED [16]. Genetic factors are estimated to contribute 58–88% of the risk for developing AN [17], but it is extremely unlikely that a single gene effect accounts for this heritability. Multiple genes probably contribute to the genetic liability via a

Attachment

Prior to the birth of a child who goes on to develop AN, 25% of parents may have experienced severe obstetric difficulty and loss compared with only 7.5% of matched comparison parents [23]. In addition, the incidence of prematurity and birth trauma is elevated by two- to threefold in the birth histories of those with AN [24]. It is perhaps understandable in this context that mothers report heightened anxiety during pregnancy and the perinatal period and are possibly overcontrolling and

Adolescence: transition and maturation

Adolescence is a time of profound biological, psychological and sociocultural change and demands a considerable degree of flexibility to successfully manage the transition into adulthood. Change may challenge the rigidity of those vulnerable to AN and open a window of vulnerability to dysregulation in relevant biopsychosocial systems. This may contribute to the timing of onset. In addition, the changes associated with adolescence differ in males and females and may therefore contribute to the

The HPA axis in AN

The HPA axis shows several abnormalities in underweight patients with AN. Plasma cortisol is raised in the context of normal plasma levels of ACTH [81], [82], [83], [84] and raised levels of CRH in cerebrospinal fluid (CSF) [85]. Peripheral metabolism of cortisol is reduced, and the adrenal cortisol response to ACTH is increased, suggesting adrenal hyperplasia and hypertrophy [86]. Dexamethasone fails to suppress cortisol release in over 90% of patients with AN [86] despite intact feedback

The serotonergic system in AN

Altered serotonergic system function is well recognised in AN. During the acute disorder, indices of serotonin turnover are reduced [93], and some serotonergic challenge tests are suggestive of altered 5HT1aR and 5HT2R activity [94], [95], [96]. Again, these changes tend to normalise with weight gain (for example, Ref. [93]), although there is some evidence of persistent 5HT2R dysfunction and elevated serotonin turnover after long-term weight restoration [94], [97], [98]. More recently,

Submissive responses and chronic stress

In social ranking theory, individuals low in the social hierarchy have little prospect of winning conflicts and must therefore resolve social conflict either by submission or by escape [101]. If escape is barred, an individual becomes trapped in a submissive stance. It has been hypothesised that it is this perception of involuntary submission to dominant others that gives rise to depression in humans [102].

A severe life event or difficulty, generally of an interpersonal nature, was identified

The HPA axis response to chronic stress in AN

Despite high levels of depressive symptomatology in AN, the mechanism of HPA axis hyperactivity appears to differ from that of depressive disorder. The latter is associated with evidence of heightened AVP release and sensitivity [124], [125], in keeping with findings in animal models of chronic stress. In contrast, the lack of an enhanced response to a combined DXM/CRH challenge test suggests that AVP activity is not increased in AN [126] despite elevated levels of AVP in the CSF [127]. Thus,

Appetite regulation and the catabolic spiral caused by dysfunctional HPA axis response

Persistent, dysregulated release of CRH can significantly affect nutritional homeostasis [129]. CRH is a key effector in the catabolic network of the hypothalamus and produces anorexia when injected into the ventricles of rats [130]. This is at least in part because CRH is a powerful inhibitor of the synthesis of neuropeptide Y (NPY) [131], which is one of the key hypothalamic anabolic effectors, increasing food intake and fat storage [132]. Accordingly, stress-induced CRH release results in

The broader role of CRH

CRH-containing neurones and CRH receptors are widely distributed in the brain and are particularly dense in the hypothalamus, pituitary, limbic system, prefrontal and cingulate cortices and autonomic structures. CRH projections are therefore well placed to mediate emotional, behavioural and physiological responses to stress (see Ref. [136] for review). In addition to activation of the HPA axis, CRH release stimulates autonomic system activity and endorphin release, modulates noradrenergic,

A neurodevelopmental model of AN

Persistently elevated and dysregulated CRH release is hypothesised to be the central common pathway generating sustained suppression of appetite and thus severe weight loss in AN. Clearly, stress-dependent changes in HPA axis function are not in themselves the cause of the disorder: were it the case, subjects with PTSD would develop AN. In a similar way, it is unclear why some individuals who experience chronic severe childhood adversity are prone to depression rather than AN. However,

Implications

In taking a developmental, biopsychosocial perspective on the aetiology of AN, this neurodevelopmental model may be easily and meaningfully applied to the individual experience of patients. Whilst many of the risk factors for AN may not be reversible, the tendency to maintain or trigger the disorder may be attenuated through psychotherapy. For example, the hypothesised vulnerability associated with anxious, insecure attachment relationships and impaired stress management may be ameliorated with

Acknowledgements

PPP Foundation “Modelling the risk factors for anorexia nervosa: a genetic and environmental study of sister pairs” 1206/87.

Frances Connan was supported by a Clinical Research Training Fellowship, Medical Research Council, London, UK.

References (143)

  • M. Durand et al.

    Differential effects of neonatal handling on anxiety, corticosterone response to stress, and hippocampal glucocorticoid and serotonin (5-HT)2A receptors in Lewis rats

    Psychoneuroendocrinology

    (1998)
  • J.W. Smythe et al.

    Median eminence corticotrophin-releasing hormone content following prenatal stress and neonatal handling

    Brain Res. Bull.

    (1996)
  • S.J. Lupien et al.

    The acute effects of corticosteroids on cognition: integration of animal and human model studies

    Brain Res. Brain Res. Rev.

    (1997)
  • C.S. Monk et al.

    The effects of hydrocortisone on cognitive and neural function: a behavioral and event-related potential investigation

    Neuropsychopharmacology

    (2002)
  • J. McIntosh et al.

    Brain Res. Dev. Brain Res.

    (1999)
  • M. Winick et al.

    Cellular response in rats during malnutrition at various ages

    J. Nutr.

    (1966)
  • T. Dalgleish et al.

    Perceived control over events in the world in patients with eating disorders: a preliminary study

    Pers. Individ. Differ.

    (2001)
  • I.C. Gillberg et al.

    The cognitive profile of anorexia nervosa: a comparative study including a community-based sample

    Compr. Psychiatry

    (1996)
  • K. Tchanturia et al.

    Perceptual illusions in eating disorders: rigid and fluctuating styles

    J. Behav. Ther. Exp. Psychiatry

    (2001)
  • D.R. Rubinow et al.

    Estrogen–serotonin interactions: implications for affective regulation

    Biol. Psychiatry

    (1998)
  • R.F. McGivern et al.

    Cognitive efficiency on a match to sample task decreases at the onset of puberty in children

    Brain Cogn.

    (2002)
  • J. Licinio et al.

    The hypothalamic–pituitary–adrenal axis in anorexia nervosa

    Psychiatry Res.

    (1996)
  • W.H. Kaye et al.

    CSF 5-HIAA concentrations in anorexia nervosa: reduced values in underweight subjects normalize after weight gain

    Biol. Psychiatry

    (1988)
  • T.D. Brewerton et al.

    Studies of serotonin function in anorexia nervosa

    Psychiatry Res.

    (1996)
  • G.K. Frank et al.

    Reduced 5-HT2A receptor binding after recovery from anorexia nervosa

    Biol. Psychiatry

    (2002)
  • S. Lee et al.

    Rationales for food refusal in Chinese patients with anorexia nervosa

    Int. J. Eat. Disord.

    (2001)
  • P.H. Robinson

    Perceptivity and paraceptivity during measurement of gastric emptying in anorexia and bulimia nervosa

    Br. J. Psychiatry

    (1989)
  • P. Cugini et al.

    Hunger sensation: a chronobiometric approach to its within-day and intra-day recursivity in anorexia nervosa restricting type

    Eat. Weight Disord.

    (1998)
  • R.L. Palmer

    Management of eating disorders

    (2000)
  • T. Leonard et al.

    Assessment of test meal induced autonomic arousal in anorexic, bulimic and control females

    Eur. Eat. Disord. Rev.

    (1998)
  • A. Ward et al.

    Neuroendocrine, appetitive and behavioural responses to d-fenfluramine in women recovered from anorexia nervosa

    Br. J. Psychiatry

    (1998)
  • J.P.J. Pinel et al.

    Hunger, eating, and ill health

    Am. Psychol.

    (2000)
  • A. Ward et al.

    Anorexia nervosa subtypes: differences in recovery

    J. Nerv. Ment. Dis.

    (2003)
  • E.E. Walters et al.

    Anorexia nervosa and anorexic-like syndromes in a population-based female twin sample

    Am. J. Psychiatry

    (1995)
  • L.R. Lilenfeld et al.

    A controlled family study of anorexia nervosa and bulimia nervosa: psychiatric disorders in first-degree relatives and effects of proband comorbidity

    Arch. Gen. Psychiatry

    (1998)
  • M. Strober et al.

    Controlled family study of anorexia nervosa and bulimia nervosa: evidence of shared liability and transmission of partial syndromes

    Am. J. Psychiatry

    (2000)
  • K.S. Kendler et al.

    The genetic epidemiology of bulimia nervosa

    Am. J. Psychiatry

    (1991)
  • C.M. Bulik et al.

    Twin studies of eating disorders: a review

    Int. J. Eat. Disord.

    (2000)
  • J. Hebebr et al.

    Anorexia nervosa viewed as an extreme weight condition: genetic implications

    Hum. Genet.

    (1995)
  • B. Devlin et al.

    Linkage analysis of anorexia nervosa incorporating behavioral covariates

    Hum. Mol. Genet.

    (2002)
  • P. Shoebridge et al.

    Parental high concern and adolescent-onset anorexia nervosa. A case-control study to investigate direction of causality

    Br. J. Psychiatry

    (2000)
  • S. Cnattingius et al.

    Very preterm birth, birth trauma, and the risk of anorexia nervosa among girls

    Arch. Gen. Psychiatry

    (1999)
  • A. Ward et al.

    Attachment research in eating disorders

    Br. J. Med. Psychol.

    (2000)
  • A. Ward et al.

    Attachment in anorexia nervosa: a transgenerational perspective

    Br. J. Med. Psychol.

    (2001)
  • L.A. Sroufe

    Psychopathology as an outcome of development

    Dev. Psychopathol.

    (1997)
  • M.J. Meaney et al.

    Molecular basis for the development of individual differences in the hypothalamic–pituitary–adrenal stress response

    Cell. Mol. Neurobiol.

    (1993)
  • L.A.M. Welberg et al.

    Prenatal stress, glucocorticoids and the programming of the brain

    J. Neuroendocrinol.

    (2001)
  • S. Maccari et al.

    Adoption reverses the long-term impairment in glucocorticoid feedback induced by prenatal stress

    J. Neurosci.

    (1995)
  • R.M. Sapolsky et al.

    The neuroendocrinology of stress and aging: the glucocorticoid cascade hypothesis

    Endocr. Rev.

    (1986)
  • Y.I. Sheline et al.

    Depression duration but not age predicts hippocampal volume loss in medically healthy women with recurrent major depression

    J. Neurosci.

    (1999)
  • Cited by (0)

    View full text