[1] 254w Background: Sex differences in eating behavior are well documented, but it is not known whether these differences have neuroanatomical correlates. Recent neuroimaging studies have provided functional maps of the human cerebral areas activated in response to hunger and satiation. Objective: The objective of this study was to assess whether the brain's response to a meal is sex-specific. Design: Using positron emission tomography, we measured regional cerebral blood flow, a marker of neuronal activity, to investigate the functional neuroanatomy of hunger (36-h fast) and satiation (in response to a liquid meal) in 22 women and 22 men. Results: We observed extensive similarities, as well as some differences, between the sexes. In response to hunger, the men tended to have greater activation in the frontotemporal and paralimbic areas than did the women (P < 0.005). In response to satiation, the women tended to have greater activation in the occipital and parietal sensory association areas and in the dorsolateral prefrontal cortex than did the men (P < 0.005); in contrast, the men tended to have greater activation in the ventromedial prefrontal cortex than did the women (P < 0.005). Conclusions: Despite extensive similarities in the brain responses to hunger and satiation between the men and women, our study showed sex-specific brain responses to a meal that indicate possible differences between men and women in the cognitive and emotional processing of hunger and satiation. This study provides a foundation for investigating the brain regions and cognitive processes that distinguish normal and abnormal eating behavior in men and women.

[1] 110w Behavioral differences between men and women are well documented, but the underlying neurobiology remains an area of intense investigation (1)(2)(3). Sex differences were found in the size and morphology of several areas of the human brain (4). In resting conditions, men have higher metabolic activity in the vicinity of the temporal-limbic areas and cerebellum and lower metabolic activity in the vicinity of the middle and posterior cin-gulate gyrus than do women (5). These observations are consistent with the hypothesis that sex differences in cognitive and emotional processing may have neuroanatomical correlates. However, it is unclear how the sex specificity of the "idling" brain extends to other aspects of human behavior.

[2] 73w Eating behavior has a major effect on human health because of the etiologic role of excessive energy intake in the development of obesity and related disorders. The prevalence of obesity is higher in women than in men (6,7), and sex differences in eating behavior have been extensively described (8,9). Substantial molecular (10), pharmacologic (11), and functional (12)(13)(14)(15) evidence attests to the importance of the central nervous system in monitoring energy intake in humans.

[3] 95w In the present study we sought to examine the effect of sex on the brain's response to eating-related stimuli. We studied changes in regional cerebral blood flow (rCBF), a marker of neuronal activity, in response to hunger (36-h fast) and satiation (in response to a liquid meal) in women and men. On the basis of our previous neuroimaging findings in men (12,13) and the sexspecific neuropharmacology of these areas of the brain (16)(17)(18), we hypothesized that the hypothalamus, paralimbic areas, and prefrontal regions would have the largest sex differences in response to hunger and satiation.

[1] 74w The metabolite and hormone concentrations and subjective ratings of hunger and satiation for all subjects before and after a liquid meal are shown in Table 1. With the exception of PP, no sex differences were detected in the postmeal changes. We observed extensive overlapping of the neuroanatomical correlates of hunger and satiation in the women and the men (Figure 1). However, despite similar overall neuroanatomical profiles, some differences in regional brain responses were measured.

[2] 41w In response to hunger (36-h fast), greater increases in rCBF in the vicinity of the dorsolateral prefrontal cortex, middle tempo-ral gyrus, and paralimbic areas, including the posterior cingulate and parahippocampal gyrus, occurred in the men than in the women (Table 2).

[3] 74w In response to satiation (after the administration of a liquid meal), greater increases in rCBF in the vicinity of the dorsolateral prefrontal cortex, precuneus, angular gyrus, and a region including the occipital cortex and some aspects of the posterior temporal lobe occurred in the women than in the men (Table 3). Satiation elicited greater increases in rCBF in the vicinity of the ventromedial prefrontal cortex in the men than in the women (Table 3).

[1] 38w Our study was designed to detect sex differences in the neuroanatomical correlates of hunger and satiation, and we had a priori expectations of observing the largest differences in the hypothalamus, paralimbic areas, and areas of the prefrontal cortex.

[2] 168w We found that differences in specific regions of the brain occurred against a background of similarities between the women and the men (Figure 1). These similarities attest to the reliability of the functional measures and suggest that the brain responses of men and women to eating-related stimuli are fundamentally more similar than different. Nevertheless, our analyses showed that the men had relatively higher neuronal activity in the temporal lobe in response to hunger and that the women had relatively higher neuronal activity in the occipital lobe in response to satiation. The lateral and ventromedial aspects of the temporal lobe were previously found to be more activated in resting conditions in men than in women, and this difference might represent a sex-specific pattern of activation independent of nutritional stimuli (5). The greater activation of the occipital cortex in response to satiation in the women than in the men suggests that visual imagery may be an important component of the cognitive processing of satiation and of related behavior in women.

[3] 58w We did not find any overall sex differences in the hypothalamic response to hunger and satiation. Although the hypothalamus is rich in estrogen and androgen receptors (4) that could modulate the activity of neuronal pathways involved in the regulation of eating behavior, technical limitations may have prevented us from identifying sex differences in the local brain activity (12).

[4] 150w We previously hypothesized that the paralimbic areas participate in a central orexigenic network modulated via feedback loops by the ventromedial prefrontal cortex (12,13). We observed that the men responded to hunger with a greater activation of the posterior cingulate, parahippocampal gyrus, and dorsolateral prefrontal cortex than did the women. The posterior cingulate was found to be the most consistently activated region in response to a variety of emotional stimuli (29). Converging evidence suggests that this region plays a role in the interaction between emotion and memory during the presentation of aversive stimuli (29). However, the direction of the change may be sex-specific because in women the posterior cingulate is deactivated in response to both anxiety and sadness (30). Therefore, the activation we observed in this region might represent a malespecific brain response elicited by the concomitant recollection and present perception of the unpleasant feelings (discomfort, pain, and anxiety) of hunger.

[1] 262w Twenty-two men and 22 women, all of whom were righthanded, were recruited from the greater Phoenix area by newspaper advertisement (some of the results from the men were reported previously; 12, 13). Subjects were in good health; all Am J Clin Nutr 2002;75:1017-22. Printed in USA. © 2002 American Society for Clinical Nutrition Sex differences in the human brain's response to hunger and satiation 1,2 Angelo Del Parigi, Kewei Chen, Jean-François Gautier, Arline D Salbe, Richard E Pratley, Eric Ravussin, Eric M Reiman, and P Antonio Tataranni women were studied while they were in the follicular phase of their menstrual cycle. Exclusion criteria included a family history of obesity or diabetes and a history of substance abuse or addiction. Structured Clinical Review for DSM-III-R (19) was used to screen for behavioral or psychiatric conditions (claustrophobia, major depression, the presence of psychotic symptoms, anorexia nervosa, or bulimia nervosa) incompatible with safe and successful participation in the study. The absence of dietary restraint was assessed by using a 3-factor eating questionnaire (20). Subjects took no medications for ≥ 1 mo before admission. All subjects were admitted for Ϸ1 wk to the Clinical Diabetes and Nutrition Section of the National Institutes of Health in Phoenix. Subjects were restricted to the metabolic ward and were limited to sedentary activity for the duration of the study. The protocol was approved by the institutional review boards of the National Institute of Diabetes and Digestive and Kidney Diseases (National Institutes of Health) and the Good Samaritan Regional Medical Center. Informed, written consent was obtained from all subjects before participation.

[2] 83w The experimental procedures were described previously (12). In brief, on admission all subjects consumed a weight-maintaining diet (50% carbohydrate, 30% fat, and 20% protein). Body composition was assessed with the use of dual-energy X-ray absorptiometry (DPX-l; Lunar Corp, Madison, WI), and daily resting energy expenditure was measured for 45 min with the use of a ventilated-hood system (DeltaTrac; SensorMedics, Yorba Linda, CA). Subjects fasted for 36 h before the brain-imaging procedures. Water and nonenergetic, noncaffeinated beverages were provided ad libitum during the fast.

[3] 196w Positron emission tomography (PET) and magnetic resonance imaging procedures were conducted at Good Samaritan Regional Medical Center (Phoenix, AZ). A magnetic resonance image of the brain was performed with the use of a 1.5 Tesla Sigma system (General Electric, Milwaukee) to rule out gross anatomical abnormalities and to facilitate comparisons between brain structure and function as described below. For the PET procedure, a transmission scan with a 68 Ge/ 68 Ga-ring source was performed to correct subsequent emission images for radiation attenuation. During each scan (1 min), subjects rested quietly in the supine position without movement and were asked to keep their eyes closed and pointing forward. PET images of regional brain activity (counts • voxel Ϫ1 • min Ϫ1 ) in each subject were obtained with the use of an ECAT 951/31 scanner (Siemens, Knoxville, TN). For each scan, a 1850-MBq (50-mCi) intravenous bolus of [ 15 O]H 2 O was injected. Two scans were obtained at baseline and 2 after feeding, with intervals of Ϸ10 min between scans. Blood samples were collected immediately after each scan to measure concentrations of glucose, free fatty acids, insulin, leptin, glucagon-like peptide 1, pancreatic polypeptide (PP), and gastrin.

[4] 174w A liquid formula meal (53% carbohydrate, 32% fat, and 15% protein, 6.3 MJ/L; Ensure-Plus; Ross-Abbott Laboratories, Columbus, OH) was orally administered to induce satiation. With the patient supine on the PET table, a plastic extension tube was inserted into the mouth to the middle of the tongue. To elim-inate possible confounding factors of regional brain activation, such as tactile stimulation of the tongue and motor neuron activity, 2 mL of water were orally administered (through the plastic tube) to induce swallowing before each of the 4 PET scans. To induce satiation after the PET scans in the baseline condition, a peristaltic pump (IMED 980, Imed Inc, San Diego) was set to orally deliver (through the plastic tube for 25 min) a liquid meal providing 50% of the previously measured daily energy expenditure. Subjective ratings of hunger and satiation were recorded after each PET scan (21). To familiarize each subject with the experimental setting and to minimize the risk of learning-related artifacts, the feeding procedure was repeated twice on the research ward before PET scanning.

[5] 89w Plasma glucose concentrations were measured with the use of the glucose oxidase method (Beckman Instruments, Fullerton, CA) and plasma insulin concentrations with the use of an automated radioimmunoassay (Concept 4; ICN Biomedicals Inc, Costa Mesa, CA). Serum fatty acid concentrations were measured with the use of an enzymatic colorimetric method (Wako Chemicals, Richmond, VA). Leptin concentrations were measured with the use of a solid-phase sandwich enzyme immunoassay (Amgen, Thousand Oaks, CA). Concentrations of glucagon-like peptide 1, gastrin, and PP were measured with the use of a commercially available radioimmunoassay.

[6] 350w Automated algorithms were used to align each subject's sequential PET images (22), to transform PET images into spatial coordinates of a standard brain atlas (23), to investigate increases in rCBF independent of variations in whole-brain measurements (analysis of covariance), and to generate normalized t value (ie, z score) (24) maps of increases in rCBF during hunger (average of the 2 images in the hunger condition minus the average of the 2 images in the satiation condition) and satiation (average of the 2 images in the satiation condition minus the average of the 2 images in the hunger condition) with the use of statistical parametric mapping (SPM99; Wellcome Department of Cognitive Neurology, University College, London; 25). After evaluating the effects of hunger and satiation within each sex group, we tested sex differences in hunger and satiation by using Student's t test and transformed the t values to z scores. The z score is the multiplier of SDs away from the null hypothesis (no sex differences) (24). The z score maps were then superimposed onto the magnetic resonance imaging template in the standard atlas coordinates (26) to allow visual inspection of the composite images. To reduce type I errors, a critical z score ≥ 2.58 (P < 0.005) was used to characterize statistical significance. The choice of this critical z score was based on experiments to analyze changes in rCBF that were conducted in our laboratory on 8 healthy women (studied twice) during a well-characterized behavioral task (resting quietly, eyes fixed on a cross hair, opening and closing the right fist). In those experiments we used the same imaging system, radiotracer technique, attenuation-correction and image reconstruction methods, Gaussian filter, and statistical parametric method to characterize true signals, false signals, and true signal-to-noise ratios as those used in the present study. At the critical z score of 2.58, not corrected for multiple comparisons, we consistently detected each of the true signals and detected 0-1 false signals for each analysis of the composite images. Therefore, we assumed a z score of 2.58 as the best trade-off between type I and type II errors (27,28).

[1] 21w The metabolite and hormone concentrations and subjective ratings of hunger and satiation for all subjects before and after a liquid meal

[2] 220w foot_1 Men Women P (sex ϫ meal (n = 22) (n = 22) interaction) Age (y) 31 ± 8foot_2 31 ± 9 0.9 Body fat (%) 28 ± 12 34 ± 9 0.1 Metabolite Glucose (mmol/L) Before meal 4.4 ± 0.5 4.5 ± 0.5 After meal 6.6 ± 0.9foot_3 7.0 ± 1.2 3 0.4 FAA (mmol/L) Before meal 0.90 ± 0.3 0.92 ± 0.20 After meal 0.59 ± 0.2 3 0.64 ± 0.19 3 0.7 Hormone Insulin (pmol/L)foot_4 Before meal 33 ± 22 22 ± 20 After meal 474 ± 543 3 354 ± 446 3 0.6 Leptin (g/L) Before meal 4.6 ± 4.6 25.7 ± 24.3 After meal 4.6 ± 4.6 26.6 ± 25.8 0.1 Gastrin (ng/L) Before meal 52 ± 43 39 ± 10 After meal 59 ± 35 51 ± 24 3 0.4 PP (pmol/L) Before meal 21 ± 23 15 ± 16 After meal 85 ± 85 3 37 ± 34 3 0.008 GLP-1 (ng/L) Before meal 74.1 ± 12.2 121.1 ± 15.8 After meal 75.0 ± 12.5 122.7 ± 15.3 0.8 Subjective ratingfoot_5 Hunger (mm) Before meal 75 ± 23 75 ± 25 After meal 30 ± 24 3 17 ± 23 3 0.2 Satiation (mm) Before meal 20 ± 18 10 ± 11 After meal 74 ± 20 3 79 ± 26 3 0.1

[3] 69w We previously proposed (12,13) that the prefrontal areas exert central control over the hedonic effects of food ingestion. Interestingly, the ventrolateral prefrontal areas that were found to be activated in both sexes in response to satiation (Figure 1) are the areas that have been suggested to be involved in behavioral planning when the decision is based on previous experience of the likely reward value of stimuli and responses (31).

[4] 320w On a background of extensive similarities, satiation induced preferential activation of dorsal prefrontal areas [Brodmann areas (BAs) 8, 9, and 10] and visual association areas in the women and preferential activation of ventral prefrontal areas (BAs 10 and 11) in the men. The upper dorsolateral prefrontal cortex (BAs 8 and partially 9) has been implicated in the maintenance of working memory (32). The meaning of the preferential 1020 DEL PARIGI ET AL Images of brain responses in the 22 women and the 22 men to hunger and satiation at +8 (hunger) and Ϫ4 mm (satiation) from a horizontal plane between the anterior and posterior commissures (coordinates of the Talairach and Tournoux brain atlas; 23). The images on the right are overlaps of the images for the women and the men. The right hemisphere in each section is on the reader's right. Brain areas with a significant increase in regional cerebral blood flow in response to hunger or satiation are shown in blue for the women and in yellow for the men. Bold colors represent the statistical level of P < 0.005, and transparent colors represent the level of 0.005 ≤ P < 0.05. Brain areas with a significant increase in regional cerebral blood flow in response to hunger or satiation in both sexes are shown in red. Images were generated with the use of positron emission tomography and magnetic resonance imaging data. Color-coded images were superimposed on an average of the subject's brain magnetic resonance images (gray-scale image). The figure is intended only for visual inspection of some regions of the brain, including the putamen (PU), insula (INS), caudate nucleus (CN), frontal operculum (FROP), thalamus (TH), superior temporal gyrus (STG), middle temporal gyrus (MTG), ventrolateral prefrontal cortex (VLPFC), and inferior occipital gyrus (IOG). Sex differences in brain responses to hunger and satiation and their relative significance were assessed by statistical parametric mapping and are shown in Tables 2 and 3.

[5] 73w activation of this area in response to satiation is unclear. Also unclear is whether the preferential activation of occipitotemporal areas, especially those involved in object recognition, signifies retrieval of visual information to process the sensation of satiation. Interestingly, the angular gyrus, which is believed to give temporal dimension to thoughts (33), is considered to be one of the neuroanatomical markers of evolutionary separation between the sexes and is more developed in women (33).

[6] 158w The ventromedial prefrontal cortex (BAs 10 and 11) receives dopaminergic afferents from the ventral tegmental area through the mesocortical pathway (34) and projects back to the inferior temporal cortex, the hippocampus, and the cingulate cortex (35). It also receives afferent projections from the amygdala and the cingulate cortex (36), which themselves receive afferent dopaminergic projections from the ventral tegmental area through the mesolimbic pathway (34). The ventromedial prefrontal cortex has been implicated in processing associations between stimuli, responses, and outcomes under changing circumstances (31). In our experimental setting, the meal assumed an important reward value after the 36-h fast. Because the ventromedial prefrontal cortex projects back to the areas that we found to be more activated in hunger in the men than in the women, we speculate that men may have a brain response that is consistent with deriving a greater hedonic effect from eating and a more rewarding feeling of satiation at the end of the meal.

[7] 89w As expected, we found no sex differences in the postmeal changes in the concentrations of plasma metabolites and hormones and the subjective ratings of hunger and satiation, with the exception of PP. This hormone is a marker of the parasympathetic efferent discharge rate to the pancreas and is thought to have mild anorectic properties (37). Lower concentrations of PP in women than in men have been described in basal (38) and hypoglycemic conditions (39). The potential implications of our observation of sex-specific brain control of eating behavior remain unclear.

[8] 152w The limitations of our experimental approach have been addressed previously (12). Briefly, they include 1) spatial resolution, contrast resolution, and the accuracy of the image deformation algorithm used to compute statistical maps, which prevent detection of significant state-dependent changes in regional brain activity in small regions such as hypothalamic, thalamic, and brainstem nuclei; 2) the choice of the critical z score of 2.58 (ie, P < 0.005 uncorrected for multiple comparisons), as discussed in Subjects and Methods; and 3) the scan order effect because the satiation condition always followed the hunger condition. Because our results were not corrected for the potential number of independent comparisons, they also should be considered preliminary. Moreover, sex differences in gastric distension cannot be excluded, although meal size was a function of individual resting energy expenditure (ie, meal size was sex-and body weight-related) and the women had satiation ratings similar to those of the men (Table 1).

[9] 113w In conclusion, we observed extensive similarities in the brain responses to hunger and satiation between the men and the women. Nonetheless, the results of our study indicate that 1) hunger elicits greater activation of brain regions mainly involved in processing emotion in men than in women, 2) satiation elicits more extensive activation of neocortical areas involved in sensorial association and behavioral planning in women than in men, and 3) satiation elicits greater activation of cortical areas involved in processing association between stimuli and responses in men than in women. This study provides a foundation for investigating the brain regions and cognitive processes that distinguish normal and abnormal eating behavior in men and women.

[10] 30w Greater increases in regional cerebral blood flow in response to satiation (after the administration of a liquid meal) in the 22 women than in the 22 men or vice versa

[11] 70w foot_9 Atlas coordinatesfoot_10 Region Brodmann area x y z z Scorefoot_11 mm Women compared with men Frontal: dorsolateral prefrontal cortex 8, 9, 10 Ϫ12 +56 +36 3.44 Precuneus 19 +12 Ϫ92 +40 2.61 Parietal: angular gyrus 39 Ϫ62 Ϫ60 +24 2.81 Occipito-temporal cortex 18, 19 +56 Ϫ74 Ϫ4 3.85 37, 38, 39 +58 Ϫ72 +4 4.24 Men compared with women Frontal: ventromedial prefrontal cortex 10, 11 +14 +52 Ϫ24 3.26