Книги по МРТ КТ на английском языке / Functional Neuroimaging in Child Psychiatry Ernst 1 ed 2000

may contribute to the increased vigilance seen in patients with anorexia nervosa, since stimulation of the caudate inXuences alertness and responses to external stimuli. The head of the caudate may also be important in anorexia nervosa because it receives aVerents from not only the frontal cortex but also the limbic system. However, the small patient sample size involved and the use of a male control group make it diYcult to interpret the Wndings. Indeed, most PET studies (Baxter et al., 1987; Yoshii et al., 1988; Andreason et al., 1994; Ernst et al., 1998) but not all (Miura et al.,1990; Gur et al., 1995), report higher global CMRGlu in women than in men.

Delvenne et al. (1995) observed global and regional absolute cerebral glucose hypometabolism in 20 underweight female patients with anorexia nervosa (14±33 years; mean age 20.5 years; #85% normal body weight) compared with 10 healthy female volunteers (18±30 years; mean age 24 years). All patients were under 85% of normal weight. The abnormally low CMRGlu was particularly signiWcant in the frontal and parietal cortices bilaterally. The eVect of age was not addressed. In addition, there was no correlation between rCMRGlu and depression or anxiety levels.

To assess the reversibility of this hypometabolism, Delvenne et al. (1996) measured rCMRGlu in 10 female patients with anorexia nervosa (aged 14±29 years; mean age 20.0 years) ®ve of whom had participated in the 1995 study and rescanned them after weight gain. All patients with anorexia nervosa were under 85% normal weight at the time of the Wrst scan and had an average weight gain of 20% at the time of the second scan. Compared with 10 female controls (same controls as in the 1995 study), the patients with anorexia nervosa had global and regional absolute glucose hypometabolism (i.e., not normalized to global metabolism) signiWcant in the superior frontal and parietal cortices. This Wnding was associated with relative hypermetabolism (after normalization to global metabolism) in the caudate nuclei and inferior frontal cortex. After weight gain, the absolute hypometabolism returned to normal in all regions, whereas relative rCMRGlu remained low in the parietal cortex and high in the inferior frontal cortex (orbitofrontal gyri). The authors suggested that parietal cortex hypometabolism might reXect a primary cerebral dysfunction or a particular regional sensitivity to the consequences of starvation. Cognitive studies in anorexia nervosa have also shown low performances in arithmetic (Fox, 1981; Hamsher et al., 1981), consistent with parietal dysfunction. Abnormalities in orbitofrontal cortex and caudate nucleus were reminiscent of CMRGlu patterns found in obsessive-compulsive patients, suggesting their possible involvement with obsessive thoughts about food,

weight, and body shape, which remained in weightrestored anorexic patients.

In a further study, Delvenne et al. (1997a) compared PET brain images from 10 female patients with anorexia nervosa (mean age 24.5 years; from the previous 1985 study group) and 20 female depressed patients, 10 of whom were underweight as a direct consequence of depression (normal-weight group mean age 29.3 years; underweight group mean age 23.4 years). The mean body mass index (BMI: weight over squared height ratio, used as a measure of body size) of the patients with anorexia nervosa was 14.29, while that of the depressed group who were underweight was 16.52. Ten healthy female volunteers were also scanned as controls. It is unclear whether the sample of patients and controls were the same as in the previous study. Nevertheless, the results showed that absolute hypometabolism of glucose appeared to be a consequence of low weight, as it was found in both low-weight patients with anorexia nervosa and low-weight patients with depression. In relative values, patients with anorexia nervosa showed signiWcant hypometabolism in the parietal lobe and signiWcant hypometabolism in the caudate nuclei. It was also noted that the inXuence of depressive symptoms did not inXuence absolute metabolism, as depressed patients without low weight did not diVer from controls.

Although one has to apply a degree of caution when interpreting the above results, it is notable that two diVerent research centers have reported caudate hypermetabolism in patients with anorexia nervosa. Whether other research groups replicate the Wnding of parietal hypometabolism seen in the work by Delvenne et al. (1995, 1996) remains to be seen.

PET studies of bulimia nervosa in adults

Patients with bulimia nervosa tend to maintain a relatively normal body weight, thus avoiding the confounds introduced by weight loss. Wu et al. (1990) compared PET images of CMRGlu of eight female adults with bulimia nervosa (mean age 28.6 years; six right-handed, two lefthanded) with those of eight female healthy controls (all right-handed; mean age 28.9 years). All were evaluated with the Hamilton Rating Scale for Depression, and all performed a visual continuous performance task, a measure of sustained attention. The use of an attention task was designed to control the state in which subjects were scanned. The healthy control group showed relatively higher rCMRGlu in the right hemisphere, whereas patients with bulimia nervosa lacked this asymmetry. Also, the right parietal/temporal region and the cingulate and ventral

306U. Chowdhury, I. Gordon and B. Lask

putamen appeared to show a reversal of the normal ªright greater than leftº pattern in the patients with bulimia nervosa.

In contrast to an earlier study of anorexia nervosa (Herholz et al., 1987), Wu et al. (1990) did not Wnd abnormally high rCMRGlu in the caudate or temporal lobes in patients with bulimia nervosa. The authors suggested that bulimia nervosa and anorexia nervosa were likely to be mediated by diVerent neural pathways and that the loss of right-predominant temporal lobe asymmetry in patients with bulimia raised questions about the possible role of aberrant hemispheric lateralization in the etiology of bulimia nervosa. In support of this hypothesis, the authors cited a PET study by Mayberg et al. (1988) that found that lateralized changes in brain function caused by stroke damage had diVerential eVects on mood and serotonin receptor density. Wu et al. (1990) proposed that this may have some implications for bulimia nervosa since the serotonin system has been shown to be involved in this disorder.

Andreason et al. (1992) studied 11 adult female inpatients with bulimia nervosa (mean age 25.5 years) and 18 healthy matched volunteers (mean age 25.3 years). Obsessive-compulsive symptoms were assessed using the Maudsley Obsessive Compulsive Inventory. The Hamilton Rating Scale was used to measure depression. All subjects were scanned 2±3h after a light meal and performed a continuous auditory discrimination task for 30min to control state during scanning. The results showed no diVerence between the global metabolic rates of patients with bulimia nervosa and normal control subjects. There was a left-predominant hemispheric asymmetry in the temporal lobes of patients with bulimia nervosa that was not observed in comparison subjects and a relative bilateral inferior temporal lobe hypermetabolism.

Orbitofrontal regions were analyzed to compare the results with the studies of Nordahl et al. (1989) and Baxter et al. (1987), which found that patients with obsessivecompulsive disorder had increased orbitofrontal metabolic rates. There was no diVerence in mean rCMRGlu nor was there a positive correlation between rCMRGlu and obsessive-compulsive symptoms in the orbitofrontal regions.

Metabolism in the left anterior and posterior lateral prefrontal cortical areas correlated negatively with Hamilton Depression Scale scores.When patients with scores of 15 or greater were considered separately, mean rCMRGlu values in the left anterior and posterior lateral prefrontal cortex were found to be signiWcantly lower than those of comparison subjects. Andreason et al. (1992) concluded that left anterior lateral prefrontal cortex hypometabolism varied with the depressive symptoms observed in bulimia

nervosa. In contrast, the lack of correlation between temporal CMRGlu and depressive symptoms in the patient group suggested that temporal lobe hypermetabolism and asymmetries were independent of the mood state.

Delvenne et al. (1997b) investigated CMRGlu in patients with bulimia nervosa at rest. Eleven female patients with bulimia nervosa (mean age 26.2 years) and 11 healthy matched controls underwent brain PET imaging. Compared with the control group, the patients with bulimia showed hypometabolism in most of the cerebral regions and relative hypometabolism in the parietal cortex. Since previous studies (Delvenne et al., 1995, 1996) demonstrated similar Wndings in patients with anorexia nervosa, the authors suggested that the observations might support a possible primary cerebral dysfunction or a particular regional sensitivity to the consequences of nutritional deWciencies. They cited previous works by Bowden et al. (1989) and Horne et al. (1991) showing that patients with anorexia and bulimia nervosa present with disturbed body image perception. Therefore, the Wndings of parietal involvement in bulimia would be consistent with the proposed involvement of the parietal cortex in the perception of body image.

SPECT studies of adults and older adolescents with eating disorders

In an early study using inhaled 133Xe to measure regional cerebral blood Xow (rCBF), Krieg et al. (1989) compared SPECT (single photon emission CT) images of 12 female inpatients with anorexia nervosa (aged 16±27 years) with a control group of Wve female and seven male patients (aged 19±17 years) whose rCBF had been measured to exclude cerebrovascular disease. The authors do not give the diagnosis of the control group but state that these patients showed no signs of a neuropsychiatric disorder after clinical examination. Eleven of the patients with anorexia nervosa were rescanned after discharge, an average of 102 days after admission. At this time, the mean weight gain after behavioral therapy was 11% of ideal body weight. The results showed that the mean global rCBF rate assessed at the Wrst examination did not diVer signiWcantly from the second examination or from those of the control group.

Using technetium-99m-labeled hexamethylpropyleneamine oxime (99mTC-HMPAO) to measure rCBF, Nozoe et al. (1993) measured rCBF in a single session before eating and again during eating and at an average interval of 71 days before therapy and again after therapy. The authors hypothesized that when patients with anorexia nervosa ate, the rCBF in ªcertain cortical areas would increaseº because of the abnormal sensitivity of patients to food

stimuli. Seven female adults with anorexia nervosa (mean age 19.0 years) were scanned prior to eating their breakfast. The procedure involved injecting a dose of 99mTC-HMPAO and scanning the subjects at rest and with eyes closed. After the Wrst scan, the subjects were asked to eat a piece of cake with their left hand with their eyes remaining closed. During this time, another dose of isotope was injected and 5min later the patients were rescanned. The same procedure was applied to Wve healthy female volunteers.

Only Wve symmetrical pairs of cortical regions were analyzed (superior frontal, inferior frontal, temporal, parietal, and occipital). All regional values were normalized to the cerebellum by using the ratio of the region of interest to the cerebellar activity. Unfortunately, whether the cerebellar metabolism is aVected by the disorder is unknown and, therefore, normalization using this region as a reference structure may have introduced some artifacts. The authors repeated the procedure in the patients following therapy but did not identify the type of therapy. The results showed that the patients' eating increased rCBF in the left inferior frontal cortex compared with that in the resting condition (8.9% increase).

Other methodologic issues include the potential inXuence of the motor and sensory variables involved in eating on brain activity. The involvement of the left inferior frontal lobe in the eating process may be related to the role of the dorsolateral prefrontal cortex in complex learned feeding behavior elicited in monkeys (Ono et al., 1984) or the inXuence of emotional state and arousal on this brain region (Nauta, 1971).

Nozoe et al. (1995) conducted a second SPECT brain imaging study of Wve female patients with bulimia nervosa (mean age 21.0 years), eight females with anorexia nervosa (mean age 24.1 years) (although it is not clear whether there was any overlap between this sample and that of the previous study), and nine healthy female volunteers (mean age 20.3 years). Using the same procedure as in their previous study (Nozoe et al., 1993), they measured values for rCBF in 10 cerebral cortical regions before and after eating. Patients with bulimia nervosa showed higher rCBF values before eating in inferior frontal and left temporal regions compared with the other groups. The patients with anorexia nervosa showed signiWcantly lower rCBF values in the left parietal region compared with the control group. There were no signiWcant diVerences among the three groups following food intake. The authors suggested that diVerences in cerebral function of patients with bulimia nervosa and anorexia nervosa could be characterized through SPECT imaging. Given the small numbers involved, as well as the design and methodologic issues mentioned above, interpretation of these studies is problematic.

Kuruoglu et al. (1998) reported on two female teenagers (aged 16 and 18 years) with anorexia nervosa who underwent SPECT brain imaging using 99mTC-HMPAO. Both presented with a history of intentional weight loss, for 14 and 16 months, respectively, and both were diagnosed with anorexia nervosa of the binge eating and purging type. Neither was depressed according to the Hamilton Depression Rating Scale. They were both scanned 4±6 weeks following diagnosis. Scanning was repeated after 1.5 years of treatment when the patients had remained free of symptoms for at least 3 months. No control subjects were included in the study.

The pretreatment SPECT scan showed diVuse bilateral hypoperfusion in frontal, parietal, and frontotemporal areas. It is unclear how ªabnormalityº was deWned. These abnormalities were judged to be more prominent in the left hemisphere. Post-treatment SPECT studies showed normal brain perfusion in both patients.

Overall, SPECT studies in adults and adolescents with eating disorder have failed to yield consistent and conclusive results. In the following section, we will discuss the only SPECT study published to date in children with anorexia nervosa.

SPECT in early-onset anorexia nervosa

Using SPECT, Gordon et al. (1997) measured rCBF in 15 children and adolescents (14 girls, one boy; age range 8±16 years) with a DSM-IV diagnosis of anorexia nervosa. Three of the patients had a follow-up scan after returning to normal weight. Initially, 99mTC-HMPAO was the radiotracer of choice used by the hospital, but it was later replaced by 99mTC-ECD (ethyl cysteinate dimer). Since the study was carried out during this period of change, only the Wrst few patients received an injection of 99mTC-HMPAO; the remainder received ECD. The comparison group included Wve children who underwent rCBF as part of an investigation for cerebrovascular disease because of hypertension. Their SPECT rCBF scans and cerebral angiograms were considered normal. These control SPECT scans showed either no asymmetry of rCBF or an asymmetry below 5% (Gordon, 1996).

The scans of the patients with early-onset anorexia nervosa were evaluated clinically by nuclear medicine physicians, who judged them to be abnormal if there was an asymmetry above 10% between the two sides of the brain on more than one contiguous slice. By this criterion, 13 of 15 patients showed an abnormal degree of temporal lobe asymmetry (Fig. 17.1, p. 242). Temporal lobe rCBF was decreased on the left in eight patients and on the right side in Wve patients. In the follow-up study of the three children

308U. Chowdhury, I. Gordon and B. Lask

who had regained weight, all continued to show asymmetries of rCBF in the temporal lobe that fell in the same direction as that seen on their initial scans. The magnitude of the asymmetry in the repeat scans was not indicated by the authors. Since publication of the above study, an additional 9 of 14 children studied by this group have asymmetry in the temporal lobes on SPECT images (I. Gordon, personal communication).

Further support for the above Wnding comes from a similar SPECT study carried out by the Childrens' Hospital in Sydney, Australia, which has shown unilateral reduced temporal lobe rCBF in three of four children with anorexia nervosa (Howman-Giles, personal communication). Some of the patients in the study by Gordon et al. (1997) had rCBF measured with 99mTC-HMPAO, while others had rCBF measured with 99mTC-ECD. This change in tracers makes comparison between subjects diYcult, because the tracers will have varying properties, such as the rate of uptake. Nonetheless, the frequency of the Wnding of temporal lobe asymmetry is striking.

This, together with the persistence of an abnormality in those patients re-scanned by Gordon et al. (1997) after weight restoration, suggests that the data cannot be explained as being secondary to starvation. Changes in rCBF that are purely secondary to starvation would be likely to produce global and symmetrical changes. Therefore, such asymmetry may reXect an underlying primary cerebral abnormality that is contributing to the development of anorexia nervosa in these young patients. The abnormality in the temporal lobe (although it is not clear which part of the temporal lobe is involved) might be associated with functional imbalance in the limbic system. Such imbalance in the limbic system may lead to a disturbance in the hypothalamic±pituitary axis, which is thought to be responsible for the clinical changes seen in anorexia nervosa.

Summary and future directions

Structural neuroimaging studies in patients with anorexia nervosa have revealed abnormalities such as sulcal widening and ventricular enlargement that appear to be related to the consequences of anorexia nervosa, rather than to the cause. Results of PET studies in adults and older adolescents with anorexia nervosa have not yielded consistent Wndings; however some of these, too, have been linked to consequences of anorexia nervosa rather than to its causes (Delvenne et al., 1997a).

The majority of functional imaging studies in adults were discordant with rCBF studies in children. Studies in

adults with anorexia nervosa found abnormalities in the caudate (Herholz et al., 1987; Delvenne et al., 1996) and parietal lobe (Delvenne et al. 1997a,b). Technical variations may account for the discrepancy in the Wndings, but the eVects of a maturing and developing brain may be the critical factors. In addition, there may be a dissociation between rCBF and glucose metabolism in anorexia nervosa. Gordon et al. (1997) has pointed out a number of other possible explanations for the discrepancy in Wndings: childhood-onset anorexia nervosa may represent a biologically distinct group or a subgroup of patients with more severe involvement than is typically seen in patients with later onsets or childhood-onset anorexia nervosa may be a separate illness with a diVerent etiology from that in adult populations. This seems unlikely as the clinical syndrome is similar no matter what the age of onset. And Wnally, it is possible that rCBF abnormalities result from a lesion that is inXuenced by age or chronicity.

The Wndings of asymmetrical temporal lobe hypoperfusion in a large percentage of the children (13/15) studied by Gordon et al. (1997) and by the team at the Childrens' Hospital in Sydney, Australia, who found it in three of four children studied, suggest that the temporal lobe may be involved in anorexia nervosa. This hypothesis is further strengthened by the observation that following full weight restoration in the Gordon et al. (1997) study, the focal rCBF abnormality persisted in all three children who were rescanned. The continuation of this longitudinal study presently shows three of Wve children with unilateral temporal lobe hypoperfusion. One child continues to show rCBF assymetry in the temporal lobe almost 4 years after her Wrst SPECT scan despite weight gain (I. Gordon, unpublished data). These Wndings may reXect an underlying biological factor contributing to the development and/or maintenance of anorexia nervosa.

Current understanding of the physical symptoms of anorexia nervosa suggests a hypothalamic±pituitary abnormality, but whether this abnormality is a primary or secondary event remains unclear. The limbic system has clear connections between the frontal and temporal lobes and the hypothalamic±pituitary axis. The work of MacLean (1955) and Papez (1937) suggest a close interdependent role of portions of the temporal lobe, including the amygdala, and hippocampus, in maintaining equilibrium of the limbic system and so allowing normal hypothalamic function. Dysfunction in the limbic system provides a possible mechanism that links disturbances in cortical functioning, emotional processes, and appetite control. Such an underlying cerebral vulnerability may be of genetic origin and may help to explain the heritability of anorexia nervosa (Holland et al., 1988). An important question arising from

the studies of early-onset anorexia nervosa is whether asymmetry in the temporal lobe reXects a primary abnormality or a vulnerability, i.e., an underlying limbic system imbalance that, if exposed to emotional stress, could lead to the development of anorexia nervosa.

As noted above, the SPECT study of Gordon et al. (1997) is the Wrst published functional imaging study involving children with anorexia nervosa. Additional controlled imaging studies using larger samples and examining relationships of functional imaging Wndings to clinical and neuropsychologic proWles are needed. Longitudinal studies will be important for monitoring the eVects of treatment, as well as CNS maturational eVects on the clinical expression of these disorders. Cognitive activation studies using fMRI or PET may further our understanding of the interaction of cognitive and emotional processing in relationship to eating disorders, e.g., the role of fear or disgust in anorexia nervosa. The investigation of neurotransmitter systems using PET may elucidate the role of serotonin, dopamine, noradrenaline, and the opioids in these disorders.

Acknowledgements

We would like to thank the Eating Disorders Team at Great Ormond Street Hospital who have helped with the SPECT study, especially Rachel Bryant-Waugh, Deborah Christie, Dasha Nicholls, and Kate Wigley. We would also like to thank the Gordon Carlton Memorial Fund for Wnancial support.

iReferencesi

American Psychiatric Association (1994). Diagnostic and Statistical

Manual of Mental Disorders, 4th edn.Washington, DC: American

Psychiatric Association.

Anand, B. K. and Brobeck, J. R. (1951). Localisation of a ªfeeding centreº in hypothalamus of the rat. Proc. Soc. Exp. Biol. Med., 77, 323±4.

Andreason, P. J., Altemus, M., Zametkin, A. J., King, A. C., Lucinio, J. and Cohen, R. M. (1992). Regional cerebral glucose metabolism in bulimia nervosa. Am. J. Psychiatry, 149, 1506±13.

Andreason, P. J., Zametkin, A. J., Guo, A. C., Baldwin, P. and Cohen, R. M. (1994). Gender-related diVerences in regional cerebral glucose metabolism in normal volunteers. Psychiatr. Res., 51, 175±83.

Artmann, H., Grau, H., Adelmann, M. and SchleiVer, R. (1985). Reversible and non-reversible enlargement of cerebrospinal

Xuid space in anorexia nervosa. Biol. Psychiatry, 23, 377±87. Barry, A. and Lippman, B. (1990). Anorexia in males. Postgrad.

Med., 87, 161±5.

Baxter, L. R., Phelps, M. E., Mazziotta, J. C., Guze, B. H., Schwartz, J. M. and Selin, C. E. (1987). Local cerebral glucose metabolism rates in obsessive compulsive disorder. Arch. Gen. Psychiatry, 44, 211±18.

Bentovim, D. and Morton, J. (1990). Anorexia in males. Postgrad. Med., 87, 161±5.

Bowden, P. K., Touyz, S. W., Rodriguez, P. J., Hensely, R. and Beumont, P. J. V. (1989). Distorting patient or distorting instrument? Br. J. Psychiatry, 155, 196±201.

Bruins-Slot, L., Gorwood, P., Bouvard, M. et al. (1998). Lack of association between anorexia nervosa and D3 dopamine receptor gene. Biol. Psychiatry, 43, 76±8.

Bryant-Waugh, R. and Lask, B. (1995). Annotation: Eating disorders in children. J. Child Psychol. Psychiatry, 36, 191±202.

Campbell, D. A., Sundaramurthy, D., Markham, A. F. and Pieri, L. F. (1998). Lack of assoication between 5-HT2A gene promoter polymorphism and susceptibility to anorexia nervosa. (Letter)

Lancet, 351, 499.

Casper, R. C. (1984). Hypothalamic dysfunction and symptoms of anorexia nervosa. Psychiatr. Clin. North Am., 7, 201±13.

Christie, D., Lambert, S., Wigley, K. et al. (1998). IQ, memory and attainments in childhood onset anorexia nervosa. In Proceedings of the 8th International Conference on Eating Disorders, New York.

Collier, D. A., Arranz, M. J., Li, T., Mupita, D., Brown, N. and Treasure, J. (1997). Association between 5-HT2A gene promoter polymorphism and anorexia nervosa. (Letter) Lancet, 350, 412.

Dally, P. (1969). Anorexia Nervosa. London: Heinemann Medical. De Vile, C. J., Sofrraz, R., Lask, B. D. and Stanhope, R. (1995). Occult intracranial tumours masquerading as early onset anorexia

nervosa. BMJ, 311, 1359±60.

Doraiswamy, P. M., Krishnan, K. R., Bogko, O. B. et al. (1991). Pituitary abnormalies in eating disorders: further evidence from MRI studies. Prog. Neuropsychopharmacol. Biol. Psychiatry, 15, 351±6.

Delvenne, V., Lotstra, F., Goldman, S. et al. (1995). Brain hypometabolism of glucose in anorexia nervosa: a PET-scan study.

Biol. Psychiatry, 37, 161±9.

Delvenne, V., Goldman, S., de Maertelaer, V., Simon, Y., Luxen, A. and Lotstra, F. (1996). Brain hypometabolism in anorexia nervosa; normalization after weight gain. Biol. Psychiatry, 40, 761±8.

Delvenne, V., Goldmann, S., de Maertelaer, V., Wikler, D., Damhaut, P. and Lostra, F. (1997a). Brain glucose metabolism in anorexia nervosa and aVective disorders: inXuence of weight loss or depressive symptomatology. Psychiatr. Res., 74, 83±92.

Delvenne, V., Goldman, S., Simon, Y., de Maertelaer, V. and Lostra, F. (1997b). Brain hypometabolism of glucose in bulimia nervosa.

Int. J. Eating Disord., 21, 313±20.

Erb, J. L., Gwirtsman, H. E., Fuster, J. M. and Richeimer, S. H. (1989). Bulimia associated with frontal lobe lesions. Int. J. Eating Disord., 8, 117±21.

Ernst, M., Zametkin, A. J. Phillips, R. L. and Cohen, R. M. (1998). Age-related changes in brain glucose metabolism in adults with attention-deWcit/hyperactivity disorder and control subjects. J. Neuropsychiatry Clin. Neurosci., 10, 168±77.

Fosson, A., Knibbs, J., Bryant-Waugh, R. and Lask, B. (1987). Early onset anorexia nervosa. Arch. Dis. Childhood, 621, 114±18.

Fox, C. F. (1981). Neuropsychological correlates of anorexia nervosa. Int. J. Psychiatr. Med., 11, 285±90.

GarWnkel, P. E. (1974). The perception of hunger and satiety in anorexia nervosa. Psychol. Med., 4, 309±15.

Gillberg, C. (1995). Clinical Child Neuropsychiatry. Cambridge, UK:

Cambridge University Press.

Gillberg, I. C., Gillberg, C., Rastam, M. and Johansson, M. (1996). The cognitive proWle of anorexia nervosa: a comparative study including a community-based sample. Compr. Psychiatry, 37, 23±30.

Goldrey, R. D. (1978). Craniopharyngioma simulating anorexia nervosa. J. Nerv. Mental Dis., 166, 533±6.

Gordon, I. (1996). Cerebral blood Xow imaging in paediatrics: a review. Nuc. Med. Commun., 17, 1021±9.

Gordon, I., Lask, B., Bryant-Waugh, R., Christie, D. and Timimi, S. (1997). Childhood-onset anorexia nervosa: towards identifying a biological substrate. Int. J. Eating Disord., 22, 159±65.

Gowers, S., Crisp, A., Joughin, N. and Bhat, A. (1991). Premenarcheal anorexia nervosa. J. Child Psychol. and Psychiatry, 32, 515±24.

Gur, R. C., Mozley, L. H., Mozley, P. D. et al. (1995). Sex diVerences in regional cerebral glucose metabolism during a resting state.

Science, 267, 528±31.

Hamsher, K. S., Halmi, K. A. and Benton, A. L. (1981). Prediction of outcome of anorexia nervosa from neuropsychological status.

Psychiatr. Res., 4, 79±88.

Hawkins, R. A. and Biebuyck, J. F. (1979). Ketone bodies are selectively used by individual brain regions. Science, 205, 325±7.

Hawley, R. (1985). The outcome of anorexia nervosa in younger subjects. Brit. J. Psychiatry, 146, 657±60.

Herholz, K. (1996). Neuroimaging in anorexia nervosa. Psychiatr. Res., 62, 105±10.

Herholz, K., Krieg, J. C., Emrich, H. M. et al. (1987). Regional cerebral glucose metabolism in anorexia nervosa measured by positron emission tomography. Biol. Psychiatry, 22, 43±51.

Heron, G. B. and Johnston, D. A. (1976). Hypothalamic tumour presenting as anorexia nervosa. Am. J. Psychiatry, 133, 580±2.

Herzog, D., Keller, M. B., Sacks, N. R., Yeh, C. J. and Lavori, P. W. (1992). Psychiatric comorbidity in treatment-seeking anorexics and bulimics. J. Am. Acad. Child Adolesc. Psychiatry, 31, 810±18.

Higgs, J., Goodyer, I. and Birch, J. (1989). Anorexia nervosa and food avoidance emotional disorder. Arch. Dis. Childhood, 64, 346±51.

Hinney, A., Barth, N., Ziegler, A. et al. (1997). Serotonin transporter gene-linked polymorphic region: allele distributions in relationship to body weight and in anorexia nervosa. Life Sci, 61, 295±303.

Holland, A., Sicotte, N. and Treasure, J. (1988). Anorexia nervosa: evidence for a genetic basis. J. Psychosomat. Res., 32, 549±54.

Horne, R. L., van Vactor, J. C. and Emerson, S. (1991). Disturbed body image in patients with eating disorders. Am J. Psychiatry,

148, 211±15.

Hsu, L. K. G. (1990). Eating Disorders. New York: Guilford Press.

Husain, M. M., Black, K. J., Doraiswamy, P. M. et al. (1992). Subcortical brain anatomy in anorexia and bulimia. Biol. Psychiatry, 31, 735±8.

Jacobs, B. and Isaacs, S. (1986). Pre-pubertal anorexia nervosa. A retrospective controlled study. J. Child Psychol. Psychiatry, 27, 237±50.

Johnson, R. D. (1995). Opioid involvement in feeding behavior and the pathogenesis of certain eating disorders. Med. Hypoth., 45, 491±7.

Jonas, J. M. and Gold, M. S. (1986). Naltrexone reverses bulimic symptoms. Lancet, 1, 807.

Katzmann, D. K., Zipursky, R. B., Lambe, E. K. and Mikulis, D. J. (1997). A longitudinal magnetic resonance imaging study of brain changes in adolescents with anorexia nervosa. Arch. Paediatr. Adolesc. Med., 151, 793±7.

Kaye, W. H., Pickar, D., Nabar, D. and Ebert, M. H. (1982). Cerebrospinal Xuid opioid activity in anorexia nervosa. Am. J. Psychiatry, 139, 643±5.

Kaye, W., Weltzin, T. E., Hsu, L. K. G., Bulik, C., McConaha, C. and Sobkiewicz, T. (1992). Patients with anorexia nervosa have elevated scores on Yale±Brown Obsessive Compulsive Scale. Int J. Eating Disord., 12, 57±62.

Kingston, K., Szmukler, G., Andrewes, D., Tress, B. and Desmond, P. (1996). Neuropsychological and structural brain changes in anorexia nervosa before and after refeeding. Psychol. Med., 26, 15±28.

Krieg, J. C., Backmund, H., Pirke, K. M. (1987). Cranial computed tomography Wndings in bulimia. Acta Psychiatr. Scand., 75, 144±9.

Krieg, J. C., Pirke, K. M., Lauer, C. and Backmund, H. (1988). Endocrine, metabolic, and cranial computed tomographic

Wndings in anorexia nervosa. Biol. Psychiatry, 23, 377±87. Krieg, J. C., Lauer, C., Leinsinger, G. et al. (1989). Brain morphology

and regional cerebral blood Xow in anorexia nervosa. Biol. Psychiatry, 25, 1041±8.

Krieg, J. C., HoltoV, V., Schreiber, W., Pirke, K. M. and Herholz, K. (1991). Glucose metabolism in the caudate nuclei of patients with eating disorders, measured by PET. Eur. Arch. Psychiatry Clin. Neurosci., 240, 331±3.

Kuruoglu, A. C., Kapucu, O., Atasever, T., Arikan, Z., Isik, E. and Unlu, M. (1998). Technetium-99m-HMPAO brain SPECT in anorexia nervosa. J. Nucl. Med., 39, 304±6.

Laessle, R. G., Zoettl, H. and Pirke, K. (1987). Meta-analysis of treatment studies for bulimia. Int. J. Eating Disord., 11, 97±110.

Laessle, R. G., Krieg, J. C., Fichter, M. M., Pirke, K. M. (1989). Cerebral atrophy and vigilance performance in patients with anorexia nervosa and bulimia nervosa. Neuropsychobiology, 21, 187±91.

Lambe, E. K., Katzman, D. K., Mikulis, D. J., Kennedy, S. H. and Zipursky, R. B. (1997). Cerebral grey matter volume deWcits after weight recovery from anorexia nervosa. Arch. Gen. Psychiatry,

54, 537±42.

Lask, B. and Bryant-Waugh, R. (1986). Childhood onset anorexia nervosa. In Recent Advances in Paediatrics, No. 8, ed. R. Meadow, pp. 21±31. London: Churchill Livingstone.

Lask, B. and Bryant-Waugh, R. (eds.) (1999). Anorexia Nervosa and Related Eating Disorders in Childhood and Adolescence, 2nd edn. Hove, UK: Psychology Press.

Lask, B., Britten, C., Kroll, L., Magagna, J. and Tranter, M. (1991). Pervasive refusal in children. Arch. Dis. Childhood, 66, 866±9.

Lewin, K., Mattingley, D. and Millin, R. R. (1972). Anorexia nervosa associated with hypothalamic tumour. BMJ, 2, 629±30.

Mayberg, H. S., Robinson, R. G.andWong, D. F. (1988). PET imaging of cortical S2 serotonin receptors after stroke: lateralized changes and relationship to depression. Am. J. Psychiatry, 145, 937±43.

Maxwell, J. K., Tucker, D. M. and Towes, B. D. (1984). Asymmetric cognitive function in anorexia nervosa. Int. J. Neurosci., 24, 37±44.

MacLean, P. (1955). The limbic system and emotional behaviour.

Arch. Neurol. Psychiatry, 73, 130±4.

Miura, S. A., Schapiro, M. B., Grady, C. L. et al. (1990). Effect of gender on glucose utilization rates in healthy humans: a positron emission tomography study. J. Neurosci. Res., 27, 500±4.

Nauta,W. J. H. (1971). The problem of the frontal lobe: a reinterpretation. J. Psychiatry, 147, 838±49.

Nordahl, T. E., Benkelfat, C., Semple,W. E., Gross, M., King, A. C. and Cohen, R. M. (1989). Cerebral glucose metabolic rates in obsessive compulsive disorder. Neuropsychopharmacology, 2, 23±8.

Nozoe, S., Naruo, T., Nakabeppu, Y., Soejima, Y., Nakajo, M. and Tanaka, H. (1993). Changes in regional cerebral blood Xow in patients with anorexia nervosa detected through single photon emission tomography imaging. Biol. Psychiatry, 34, 578±80.

Nozoe, S., Naruo, T., Yonekura, R. et al. (1995). Comparison of regional cerebral blood Xow in patients with eating disorders.

Brain Res. Bull., 36, 251±5.

Ono, T., Nishino, H., Fukunda, M., Sasaki, K. and Nishijo, H. (1984). Single neuron activity in dorsolateral prefrontal cortex of monkey during operant behaviour sustained by food reward.

Brain Res., 311, 332.

Palazidou, E., Robinson, P. and Lishman, W. (1990). Neuroradiological and neuropsychological assessment in anorexia nervosa. Psychol. Med., 20, 521±7.

Papez, J. (1937). A proposed mechanism of emotion. Arch. Neurol. Psychiatry, 38, 725±43.

Pendleton-Jones, B., Duncan, C., Browers, P. and Mirsky, A. F. (1991). Cognition in eating disorders. J. Clin. Exp. Neuropsychol.,

13, 711±28.

Phillipp, E., Pirke, K. M., Keller, M. and Krieg, J. (1991). Disturbed cholecystokinin secretion in patients with eating disorders. Life Sci., 48, 2443±50.

Pirke, K. M., Pahl, J., Schweiger, U. and WarnoV, M. (1985). Metabolic and endocrine indices of starvation in bulimia: a comparison with anorexia nervosa. Psychiatr. Res., 15, 33±9.

Russell, G. (1985). Anorexia and bulimia nervosa. In Child and

Adolescent Psychiatry, 2nd edn, eds. M. Rutter and Hersov, pp. 625±37. Oxford: Blackwell ScientiWc.

Schmidt, U., Hodes, M. and Treasure, J. (1992). Early onset bulimia nervosa-who is at risk? Psychol. Med., 22, 623±8.

Sorbi, S., Nacmias, B., Tedde, A., Ricca, V., Mezzani, B. and Rotella, C. M. (1998). 5-HT2A promoter polymorphism in anorexia nervosa. (Letter) Lancet, 351, 1785.

Silver, A. J. and Morley, J. E. (1991). The role of CCK regulation of food intake. Prog. Neurobiol., 36, 23±34.

Strober, M., Lampert, C., Morrell, W., Burroughs, J. and Jacobs, C. (1990). A controlled family study of anorexia nervosa. Int. J. Eating Disord., 9, 239±54.

Szmukler, G. I. and Tam, D. (1984). Anorexia nervosa: starvation dependence. Br. J. Med. Psychol., 57, 303±10.

Szmukler, G. I., Andrewes, D., Kingston, K., Chen, L., Stargatt, R. and Stanley, R. (1992). Neuropsychological impairment in anorexia nervosa before and after refeeding. J. Clin. Exp. Neuropsychol., 14, 347±52.

Tanner, J., Whitehouse, R. and Takaishi, M. (1966). Standards from birth to maturity for height, weight, height velocity and weight velocity: British children, 1965, Parts 1 and 2. Arch. Dis. Childhood, 41, 454±71, 613±35.

Teitelbaum, P. and Stellar, E. (1954). Recovery from failure to eat produced by hypothalamic lesions. Science, 120, 894±5.

Terzian, H. and Ore, G. (1955). Syndrome of Kluver and Bucy reproduced in man by bilateral removal of the temporal lobes.

Neurology, 15, 373±80.

Treasure, J. (1997). Eating disorders. In Essentials of Postgraduate Psychiatry, 3rd edn, eds. R. Murray, P. Hill and P. McGuYn, pp. 192±221. Cambridge, UK: Cambridge University Press.

Treasure, J. and Campbell, I. (1994). The case for biology in the aetiology of anorexia nervosa. Psychol. Med., 24, 3±8.

Whitaker, A., Johnson, J., ShaVer, D., Rapoport, J. and Kalikow, K. (1990). Uncommon troubles in young people: prevalence estimates of selected psychiatric disorders in a non-psychiatric population. Arch. Gen. Psychiatry, 47, 487±96.

White, L. E. and Hain, R. F. (1959). Anorexia in association with a destructive lesion of the hypothalamus. Arch. Pathol., 68, 275±81.

Wise, R. A. and Rompre, P. P. (1989). Brain dopamine and reward.

Annu. Rev. Psychol., 40, 191±255.

Witt, E. D., Ryan, C. and Hsu, E. (1985). Learning deWcits in adolescents with anorexia nervosa. J. Nerv. Mental Dis., 173, 182±4.

Wu, J. C., Hagman, J., Buchsbaum, M. S. et al. (1990). Greater left cerebral hemispheric metabolism in bulimia assessed by positron emission tomography. Am. J. Psychiatry, 147, 309±12.

Yoshii, F., Barker, W. W., Chang, J. Y. et al. (1988). Sensitivity of cerebral glucose metabolism to age, gender, brain volume, brain atrophy, and cerebrovascular risk factors. J. Cereb. Blood Flow Metab., 8, 654±61.