Catecholamine metabolism contributes to energy homeostasis and may differ in low-weight conditions of various origins. Free plasma metanephrines, 3-methoxytyramine (3-MT), normetanephrine (NM), and metanephrine (M), are stable end-products of dopamine, noradrenaline, and adrenaline metabolism, respectively, with potential circadian rhythmicity. However, their 24-hour profiles remain unknown in anorexia nervosa (AN) and constitutional thinness (CT). This exploratory study aimed to assess 24-hour variations of plasma free metanephrines in women with acute AN (n = 10), recovering AN (AN-Rec, n = 11), CT (n = 10), and healthy controls (n = 10), and to explore their association with biomarkers of undernutrition severity and weight regulation. Participants underwent 24-hour in-patient evaluation, including standardized meals and activity, and twelve time-point plasma sampling. Free metanephrines were measured by LC-MS/MS. 24-hour rhythmicity was analyzed using cosinor regression. AN and CT showed higher mean 24-hour levels for 3-MT, NM, and M than controls, with variable patterns depending on metabolites. AN-Rec displayed higher 3-MT levels than controls only. Cosinor analysis revealed a significant 24-hour variation consistent with a circadian pattern for 3-MT with an acrophase around 13:00 in AN, AN-Rec, and CT. NM showed a circadian pattern in AN, CT, and controls. Mean 3-MT, NM and M correlated negatively with BMI and IGF-1, and positively with ALT and vitamin B12. Midday elevations of free plasma metanephrines may reflect metabolic stress and hepatic alterations in AN. Their increase may may reflect distinct underlying physiological mechanisms in both AN and CT. Further studies are needed to confirm these exploratory findings.

BACKGROUND:Anorexia Nervosa (AN) is a severe psychiatric disorder marked by excessive restriction and automatic avoidance of fat and high-caloric foods, driven by fear of weight gain and guided by negative encoding of some foods that are essential for health. These impairments seem to reflect a conflict between willingness-to-eat and fear of negative outcomes, leading to maladaptive decisions prioritizing hypocaloric foods. METHOD:We developed an experimental paradigm to investigate the link between these cognitive disturbances by monitoring gaze during three tasks evaluating the health, taste, and preference values of three food categories: hypocaloric, hypercaloric, and intermediate calorie foods. We evaluated the reject rate and quantity wanted of each food item to compose their meal. To consider different patients profiles, we applied this protocol to 67 subjects across two subtypes: AN-Restrictive (n = 37), and AN-Binge-Purge (n = 30) and 37 healthy controls. RESULTS:Both AN subtypes showed lower hedonic ratings for high-calorie foods and strong correlations between taste and health perceptions (absent in controls), though AN-BP provided lower overall health ratings than AN-R. Eye-tracking revealed subtype-specific patterns: AN-BP exhibited reduced gaze on food images and increased non-stimulus area fixation during preference evaluations, while AN-R showed greater fixation on food names during willingness-to-eat decisions. Furthermore, presentation context modulated food selection in both patient groups: intermediate foods were preferentially chosen over hypercaloric alternatives but rejected in favor of hypocaloric options, demonstrating that choices are relative rather than absolute. CONCLUSIONS:By shedding light on mechanisms of food restriction and avoidance, this study holds promise for refining therapeutic strategies; contextual manipulation of food choices may offer intervention potential with intermediate-calorie foods potentially serving as a bridge during nutritional rehabilitation.

Constitutional thinness (CT) refers to a naturally low and lifelong stable body weight, without any identifiable pathological cause. It affects 1-2% of the population and is characterized by a low BMI in the absence of clinical or biological signs of malnutrition, together with strong heritability. The diagnosis relies on excluding nutritional, endocrine, digestive, psychiatric, or tumoral causes of low BMI. It is most often established during adolescence, when distinguishing CT from anorexia nervosa (AN) becomes crucial. Unlike AN, individuals with CT do not restrict their food intake, do not fear weight gain, and exhibit neither body-image distortion nor psychological features of eating disorders. Their weight trajectory remains consistently low, and menstrual cycles are generally preserved. Energy intake is comparable to that of the general population, although meals are fractionated due to early satiation. Despite a paradoxically positive energy balance, CT is marked by a pronounced resistance to weight gain, combining rapid lipid turnover with hypofunctional skeletal muscle. Bone mineral density is also reduced. From a therapeutic perspective, CT requires tailored educational and nutritional support, along with careful monitoring of bone fragility risk. (c) 2025 Societe Franc & cedil;aise de Nutrition (SFN). Published by Elsevier Masson SAS. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

La maigreur constitutionnelle (MC) correspond à un poids naturellement bas, stable depuis l’enfance, sans cause pathologique. Elle concerne 1 à 2 % de la population et se caractérise par un IMC faible sans signes de dénutrition et une héritabilité forte. Le diagnostic repose sur l’exclusion de causes nutritionnelles, endocriniennes, digestives, psychiatriques ou tumorales d’un IMC bas. Il est le plus souvent posé à l’adolescence, au moment où la distinction avec l’anorexie mentale (AM) devient essentielle. Contrairement à l’AM, les personnes vivant avec une MC ne restreignent pas leur alimentation, ne craignent pas de prendre du poids et ne présentent ni distorsion de l’image corporelle ni traits psychologiques de trouble du comportement alimentaire. Leur courbe de poids est constamment basse et les cycles menstruels restent généralement normaux. Leur apport énergétique est similaire à celui de la population générale, mais l’alimentation est fractionnée en raison d’un rassasiement précoce. Malgré un bilan énergétique paradoxalement positif, la MC se distingue par une résistance marquée à la prise de poids associant un turnover lipidique rapide et un muscle hypofonctionnel. La densité osseuse est également réduite. Sur le plan thérapeutique, la MC nécessite un accompagnement éducatif et nutritionnel adapté, ainsi qu’une vigilance vis-à-vis du risque de fragilité osseuse.

To investigate Anti-Müllerian Hormone (AMH) levels in women with anorexia nervosa (AN) compared to individuals with constitutional thinness (CT) and healthy controls, and to evaluate the association between AMH, body mass index (BMI), and weight recovery. This monocentric cross-sectional study included 191 women with AN, 41 with CT, and 55 controls. A subgroup of 42 patients with AN was reassessed after weight recovery. Hormonal, metabolic, and psychological parameters were analyzed. AMH was measured using a validated electrochemiluminescence immunoassay. Comparisons and correlations were performed using nonparametric tests and Pearson coefficients. Mean AMH plasma level was higher in patients with AN (30.5 ± 1.6 pmol/l) than in controls (23.3 ± 2.5 pmol/l, p = 0.0252) and CT (21.4 ± 2.7 pmol/l, p = 0.0088). AMH was positively correlated with BMI, leptin, FSH, and LH in the AN group, and inversely associated with disease duration. AMH was the lowest in extreme undernutrition state (BMI ≤ 13 kg/m²: AMH: 19.5 ± 4.4pmol/l) and the highest in moderate undernutrition state (BMI 16.5–18.5 kg/m²: AMH:35.8 ± 3.1pmol/l; p = 0.0343 vs. previous). During weight recovery, AMH remained stable overall but tended to decrease in patients who regained menstrual cycles. Mean AMH is globally elevated in AN, but modulated by nutritional status, being very low in extreme undernutrition and higher during partial recovery, remaining elevated with weight normalization in patients with persistent amenorrhea. AMH might reflect the severity of undernutrition in AN, with low AMH marking the most severe forms, and relatively higher AMH indicating some preserved or recoverable reproductive function, helping the clinician in assessing reproductive prognosis in AN.

CONTEXT:Neurohypophysis (NH) function in eating disorders (ED) remains poorly elucidated. Studies on vasopressin and oxytocin display inconclusive findings regarding their levels and associations with psychological complications in ED. The profile of opioid tone, a crucial NH activity regulator, is also unknown. OBJECTIVE:To characterise the circadian profile of NH hormones and NH opioid tone using positron emission tomography/MRI (PET/MRI) imaging in patients with ED compared to healthy controls. METHODS:Twelve-point plasma circadian profiles of copeptin and oxytocin, alongside nutritional and psychological scores, were assessed in age-matched female participants: 13 patients with anorexia nervosa restrictive-type (ANR), 12 patients recovered from AN (ANrec), 14 patients with bulimia nervosa and 12 controls. Neurohypophysis PET/MRI [11C] diprenorphin binding potential (BPND) was evaluated in AN, ANrec and controls. RESULTS:Results revealed lower copeptin circadian levels in both ANR and ANrec compared to controls, with no oxytocin differences. Bulimia nervosa exhibited elevated copeptin and low oxytocin levels. [11C] diprenorphin pituitary binding was fully localised in NH. Anorexia nervosa restrictive-type displayed lower NH [11C] diprenorphin BPND (indicating higher opioid tone) and volume than controls. In ANR, copeptin inversely correlated with osmolarity. Neurohypophysis [11C] diprenorphin BPND did not correlated with copeptin or oxytocin. CONCLUSION:Copeptin demonstrated significant group differences, highlighting its potential diagnostic and prognostic value. Oxytocin levels exhibited conflicting results, questioning the reliability of peripheral blood assessment. Increased NH opioid tone in anorexia nervosa may influence the vasopressin or oxytocin release, suggesting potential therapeutic applications.

Objectif Si les anomalies hormonales hypophysaires secondaires à la dénutrition de l’anorexie mentale (AM) sont bien décrites, comme l’hypogonadisme hypogonadotrope, la prolactine reste encore méconnue. Cette étude vise à évaluer les taux circadiens de prolactine dans l’AM comparés à des volontaires saines à partir d’une cohorte rétrospective et à partir d’une scoping review de la littérature. Méthodes Étude transversale comparant les taux de prolactine, d’autres données hormonales et anthropométriques de 129 patientes AM et 40 volontaires saines de même âge. Recherche exhaustive sur PubMed des études comparatives similaires pour définir la tendance des taux de prolactine. Résultats Scoping review : 55 études originales ont été retrouvées, dont 20 analysables (37 %) retrouvant 55 % d’études avec prolactine diminuée (aucune augmentée). Cohorte rétrospective : on retrouve des taux de prolactine diminués chez les AM, mais avec un rythme circadien préservé dans les deux groupes (AM : 267±11mU/L vs C : 350±15mU/L, p<0,0001), et des valeurs hétérogènes. La prolactine basse est revenue associée à une T3Libre diminuée et un CTX élevé, avec corrélation positive entre prolactine et IGF1. Discussion Notre étude confirme une prolactine physiologiquement basse dans l’anorexie mentale, mais note un rythme circadien préservé. La production de prolactine dans l’AM ne semble pas corrélée à l’IMC mais à l’état nutritionnel. Les mécanismes potentiels incluent la dénutrition, le déficit en œstrogènes, un dysfonctionnement de régulation de la prolactine, via la kisspeptine. La prolactine basse devient un nouveau marqueur de dénutrition et potentiellement de gravité de celle-ci, utile pour le clinicien.

INTRODUCTION: Anorexia nervosa (AN) in older adult women is primarily described through reviews or case reports focusing on psychiatric traits, with no comprehensive studies evaluating their complete nutritional and hormonal profiles. This study aimed to describe a group of women with anorexia nervosa aged above 35 years old (AN35), and compare them with young women with anorexia nervosa (ANY) and normal-weight control participants. METHOD: Anthropometric, metabolic, nutritional, and psychiatric parameters were collected and compared among three groups of women: 50 AN35, 37 ANY, and 38 controls. RESULTS: AN35 exhibited a mean disease duration of 271 +/- 19 months, with 94% chronic forms and 58% restrictive types. Despite having similar BMI as ANY, AN35 displayed more altered parameters, including higher liver enzymes (p = 0.007), free T-3 (p = 0.0046) and leptin (p < 0.0001); and lower albumin (p = 0.0029), and white cells (p < 0.0001). AN35 showed significant heterogeneity in hormonal adaptation, such as free T3. Half of the patients aged above 51 years revealed high gonadotropin levels despite being undernourished. Additionally, AN35 groups presented with 50% of bones fractures, decreased T-scores under -2.5 (p < 0.0001 for femoral), and altered micro architectural HRPQT parameters compared to ANY. CONCLUSION: Anorexia nervosa in older adult women is predominantly chronic. Nutritional parameters changes with age suggests a significant heterogeneity and possible adaptation of energy balance and bodyweight set point for others. Complications may be severe, altering the quality of life, and sometimes potentially lethal. These findings highlight the potential adaptation of energy balance with age, and should assist clinicians in clinical practice.

The 5-HT6 receptor (5-HT6R) is highly expressed in the anterior striatum, a region involved in food-intake and anxiety-related behaviours that shows abnormal activity in anorexia nervosa (AN) patients. Indeed, 5-HT6R antagonists decrease food motivation across species. Rodent studies have shown the acute anxiolytic properties of WAY-208466, a selective 5-HT6R agonist. Thus, we hypothesized that this agent could increase food motivation and decrease anxiety-related behaviours by impacting cortico-striatal circuits. We administered acute (0.1 mg/kg) and subchronic (0.03 and 0.1 mg/kg) intramuscular injections to four macaques performing a foodchoice task and expressing spontaneous behaviours. We used PET imaging with the [18F]2FNQ1P specific tracer to determine 5-HT6R occupancy in several regions that could be linked with the induced behavioural changes. Acute and subchronic activation of 5-HT6R transmission led to a reversible increase in food motivation. WAY208466 also decreased anxiety-like behaviours when given acutely but not subchronically. However, in the subchronic protocol, the higher dose led to increased activity characterized by movements and object-related behaviour expression. PET imaging revealed that these behavioural changes could be sustained by corticostriatal circuits involved in food intake. These findings demonstrate that this 5-HT6R agonist subchronic treatment successfully increases food motivation but does not fully achieve the expected anxiolytic effect in monkeys. It suggests that targeting 5-HT6Rs may be a promising approach for treating the restrictive subtype of AN, which, unlike anxiety, has not been shown to benefit from treatment with selective serotonin reuptake inhibitors.

Background and objectiveOverweight and obesity affects millions of individuals worldwide and consequently represents a major public health concern. Individuals living with overweight and obesity have difficulty maintaining a low body weight due to known physiological mechanisms which prevent further weight loss and drive weight regain. In contrast, mechanisms which promote low body weight maintenance receive less attention and are largely unknown. To uncover these intrinsic mechanisms, we investigated a human cohort of constitutionally thin (CT) individuals which maintain a low body weight and are resistant to weight gain despite exposure to an obesogenic environment.MethodsTo identify novel genes that contribute to low body weight maintenance, we performed transcriptomics on adipose tissue biopsies collected from CT and normal body weight (NBW) individuals and identified sulfotransferase 1A1 (SULT1A1) as a target for further investigation in mice. Sult1a1 knockout (KO) mice were fed a standard diet to assess the impact of Sult1a1 deletion on metabolic traits. To determine if high-fat feeding recapitulated the CT weight gain resistance phenotype, Sult1a1 KO mice were fed a high-fat diet for 13-weeks. A subset of wild-type and Sult1a1 KO mice from the standard diet were further analyzed for characterization of adipose tissue respiratory capacity.ResultsIn comparison to NBW controls, adipose tissue from CT individuals expresses less SULT1A1. Sult1a1 KO mice weigh 10% less at the end of the study period and on a high-fat diet, Sult1a1 KO mice tended to gain less weight and had reduced fat mass at 14-weeks of age. These changes were associated with reduced fasting insulin and lessened adipose tissue inflammation and fibrosis. Subcutaneous adipose tissue from Sult1a1 KO mice on a standard chow diet had elevated leak respiration, uncoupling protein 1 (UCP1) expression and increased expression of a mitochondrial marker, VDAC, associating Sult1a1 deletion to adipose tissue browning.ConclusionsOur results associate Sult1a1 deletion with a tendency for lower body weight through remodeling of white adipose tissue towards a brown phenotype. The presence of UCP1, the expression of an additional mitochondrial protein and increased respiratory capacity suggest browning of the subcutaneous adipose tissue depot of Sult1a1 KO mice.

Understanding biological mechanisms underlying anorexia nervosa (AN) is necessary to develop care strategies. Despite many articles dedicated to peptides assessment in AN, there is no systematic review. A scoping review of circulating peptides published in relation to AN, comparing their results with those of controls, was conducted. Embase and PubMed databases were search from 1966 to 2022 (PROSPERO CRD42022323716). All original English articles, assessing peptides in AN (except classical markers) were analyzed. 1151 studies for 207 peptides, in 486 published articles were selected, and evidences/trends in AN were compared to controls. Fifteen clusters of function gathering peptides covering physiopathological aspects of AN were identified. This scoping review revealed a large variety of circulating peptides explored in AN. Some peptides presented with convincing results and helped understanding pathophysiologic aspects. Other peptides presented with nuanced results, partly due to insufficient number of studies, multiple assay techniques, inadequate sampling time, and lack of phenotyping. Conversion from bench-to-bed remains difficult and may explain why peptides evaluations did not currently lead to specific international recommendations or tailored therapeutic/preventive strategies. Peptide evaluation in anorexia nervosa could explore secretion profiles, and test it in well-phenotyped patients with AN, to conclude for potential clinical use, and finally design therapeutic tests.

La maigreur constitutionnelle, stable tout au long de la vie, sans cassure, n’est pas associée à un trouble du comportement alimentaire, ni une dénutrition, mais caractérisée par une alimentation par petits repas fréquents et nombreux snacking qu’il faut respecter. La résistance à la prise de poids et la facilité à en perdre entraînent un risque d’amaigrissement en cas d’évènement intercurrent. L’altération de la masse osseuse doit être cherchée et expliquée. La souffrance psychologique est mal perçue car souvent mal interprétée.

The probabilistic topography and inter-individual variability of the pituitary gland (PG) remain undetermined. The absence of a standardized reference atlas hinders research on PG volumetrics. In this study, we aimed at creating maximum probability maps for the anterior and posterior PG in young female adults. We manually delineated the anterior and posterior parts of the pituitary glands in 26 healthy subjects using high-resolution MRI T1 images. A three-step procedure and a cost function-masking approach were employed to optimize spatial normalization for the PG. We generated probabilistic atlases and maximum probability maps, which were subsequently coregistered back to the subjects’ space and compared to manual delineations. Manual measurements led to a total pituitary volume of 705 ± 88 mm³, with the anterior and posterior volumes measuring 614 ± 82 mm³ and 91 ± 20 mm³, respectively. The mean relative volume difference between manual and atlas-based estimations was 1.3

Introduction Certains patients en hypothyroïdie nécessitent des doses de levothyroxine orale élevés pour normaliser leur TSH. Les autres alternatives d’administration, si la voie orale est impossible, sont la voie intraveineuse ou intramusculaire. Observation Nous rapportons le cas d’une patiente de 34 ans aux antécédents de thyroïdectomie totale en 2012 et chirurgie gastrique pour perforation digestive haute en 2016, qui avait une hypothyroïdie réfractaire avec une TSH à 114mui/L (0,27–4,2) malgré des posologies élevées de levothyroxine orale (350μg/j soit 5,2μg/kg).Différentes formes galéniques orales ont été testées sans succès.La patiente a eu des injections en intramusculaire de levothyroxine à raison de 300μg deux fois par semaine avec diminution de la TSH à 20mui/L.Pour améliorer sa qualité de vie, nous avons remplacé les injections intramusculaire par des injections sous-cutanée tous les deux jours à raison de 200μg/j permettant une normalisation de la TSH à 4,8mui/L.Suite à la reprise chirurgicalement en 2021 (dérivation duodénale avec antrectomie), nous avons pu reprendre la levothyroxine per os à raison de 275μg/jour soit 3,9μg/kg avec une normalisation de la TSH à 0,3mui/L. Discussion La substitution par levothyroxine par voie sous-cutanée a permis la normalisation bilan thyroïdien pendant la phase de malabsorption digestive.

BackgroundAccording to case-control studies, a multitude of factors contribute to the emergence of anorexia nervosa (AN). The present systematic review examines prospective studies specifically designed to evaluate the prediction of AN onset.MethodsAccording to the ARMSTAR 2 and PRISMA 2020 checklists, the PubMed, PsycINFO and Cochrane databases were searched. The methodological quality of the studies was assessed with the Downs and Black checklist.ResultsThree articles concerning prospective studies of the general population were ultimately included in the review. The methodological quality of these studies was not optimal. Bidirectional amplification effects were observed between risk factors, some of which could have a relative predictive force as low bodyweight or body dissatisfaction. Even if not included according to specified criteria for this systematic review 11 longitudinal studies, with retrospective analysis of AN onset' prediction, were also discussed. None of these studies asserted the predictive value of particular risk factors as low body weight, anxiety disorders or childhood aggression.ConclusionsTo date there are insufficient established data to propose predictive markers of AN onset for predictive actions in pre-adolescent or adolescent populations. Future work should further evaluate potential risk factors previously identified in case-control/retrospective studies within larger prospective investigations in preadolescent populations. It is important to clearly distinguish predisposing factors from precipitating factors in subjects at risk of developing AN. Currently health care guidelines for eating disorders do not include proposal of markers to predict the onset of anorexia nervosa. The current work provides a systematic review of the scientific literature concerning this subject. To date only three published studies were designed in a prospective longitudinal way to evaluate potential predictors of anorexia nervosa onset. When taking into account these studies only low bodyweight and body dissatisfaction in early puberty were proposed as predicting elements for further anorexia nervosa development. Meanwhile the prediction precision was calculated for none of them. No other psychological elements were retained or studied. The age range of the population entering the follow up in these studies was too large covering the peak age of anorexia nervosa itself (13-16 years of age). Larger prospective studies including prepubescent individual and evaluating more psychological markers (perfectionism, negative affectivity or negative self-evaluation) or environmental ones are needed.

Background In cases of Anorexia Nervosa (AN), achieving weight gain recovery beyond the lower limits set by the World Health Organization and normalizing classical nutritional markers appears to be essential for most patients. However, this is not always adequate to restore menstrual cycles. This discrepancy can cause concern for both patients and healthcare providers, and can impact the medical management of these individuals. Thus, the purpose of this study was to assess the ability of anthropometric and hormonal factors to predict the resumption of menstrual cycles in individuals with anorexia nervosa upon reaching a normal body weight. Method Patients with AN who had achieved a normal Body Mass Index but had not yet resumed their menstrual cycles (referred to as ANRec) were evaluated on two occasions: first at visit 1 and then again 6 months later, provided their body weight remained stable over this period (visit 2). Among the 46 ANRec patients who reached visit 2, they were categorized into two groups: 20 with persistent amenorrhea (PA-ANRec) and 26 who had regained their menstrual cycles (RM-ANRec). Anthropometric measurements, several hormone levels, Luteinizing Hormone (LH) pulsatility over a 4-h period, and LH response to gonadotropin-releasing hormone injection (LH/GnRH) were then compared between the two groups at visit 1. Results Patients in the RM-ANRec group exhibited higher levels of follicular stimulating hormone, estradiol, inhibin B, LH/GnRH, and lower levels of ghrelin compared to those in the PA-ANRec group. Analysis of Receiver Operating Characteristic curves indicated that having ≥ 2 LH pulses over a 4-h period, LH/GnRH levels ≥ 33 IU/l, and inhibin B levels > 63 pg/ml predicted the resumption of menstrual cycles with a high degree of specificity (87%, 100%, and 100%, respectively) and sensitivity (82%, 80%, and 79%, respectively). Conclusions These three hormonal tests, of which two are straightforward to perform, demonstrated a high predictive accuracy for the resumption of menstrual cycles. They could offer valuable support for the management of individuals with AN upon achieving normalized weight. Negative results from these tests could assist clinicians and patients in maintaining their efforts to attain individualized metabolic targets. Trial registration IORG0004981.

Introduction: Studies into food-related behaviors and emotions are increasingly being explored with Virtual Reality (VR). Applications of VR technologies for food science include eating disorder therapies, eating behavior studies and sensory analyzes. These applications involve 3D food stimuli intended to elicit cravings, stress, and/or emotions. However, the visual quality (i.e., the realism) of used food stimuli is heterogeneous, and this factor's influence on the results has never been isolated and evaluated. In this context, this work aims to study how the visual quality of food stimuli, exposed in a virtual reality environment, influences the resulting desire to eat.Methods: 28 subjects without eating disorders were included in this protocol, who evaluated the desire to eat induced by 10 3D food stimuli, each duplicated in 7 quality levels (for a total of 70 stimuli).Results: Results show that visual quality influences the desire to eat, and this effect depends on the type of food and users' eating habits. We found two significant thresholds for visual quality: the first provides the minimal quality necessary to elicit a significant desire to eat, while the second provides the ceiling value above which increasing the quality does not improve further the desire to eat.Discussion: These results allow us to provide useful recommendations for the design of experiments involving food stimuli.

L’hormone anti-mullérienne (AMH), produit des cellules de la granulosa des follicules (pré)antraux ne varie pas durant le cycle menstruel, c’est un bon marqueur ovarien. L’anorexie mentale (AM), responsable d’un hypogonadisme central avec aménorrhée pourrait mettre les ovaires au repos. Cette étude étudiait la réserve ovarienne via l’AMH chez les patientes atteintes d’AM. Nous avons comparé l’AMH avec les paramètres métaboliques et hormonaux entre 191 patientes AM, 41 maigreurs constitutionnelles (MC) et 55 témoins. Nous avons étudié les fluctuations de l’AMH selon l’IMC des AM, puis chez 42 patientes AM en reprise de poids (AMRec). L’AMH était significativement augmentée chez les AM (30,5 ± 1,6 pmol/l) comparée aux témoins (23,3 ± 2,5 pmol/l ; p = 0,0252) et MC (21,4 ± 2,7 pmol/l ; p = 0,0088). L’AMH était significativement abaissée chez les AM avec IMC < 13 kg/m2 (19,5 ± 4,4 pmol/l) comparée aux AM avec IMC entre 16,5 et 18,5 kg/m2 (35,8 ± 3,1 pmol/l ; p = 0,0343). L’AMH était significativement diminuée chez les AMRec avec cycles comparée aux AM (26,9 ± 2,9 pmol/l vs. 33,3 ± 3,1 pmol/l ; p = 0,0302) et comparée aux AMRec en aménorrhée (36,6 ± 3,8 pmol/l ; p = 0,0478). L’AMH est augmentée chez les AM comparée aux témoins. En dénutrition l’AMH diminue quand l’IMC diminue. Avec la renutrition, l’AMH diminue quand l’IMC augmente chez les patientes ne présentant pas naturellement une possible dystrophie ovarienne. L’AMH élevée malgré reprise pondérale mais persistance de l’aménorrhée pourrait s’expliquer par une possible dystrophie ovarienne, très présente dans la population générale (4–21 %).

Patients with Anorexia Nervosa (AN) and athletes share intense physical activity and pituitary hormonal disturbances related to absolute (AN) or relative (athletes) undernutrition. Pituitary gland (PG) structure evaluations in those conditions are scarce, and did not differentiate anterior from posterior lobe. We evaluated the structure–function relationship of anterior and posterior PG in AN and athletes, and potential reversibility of this alteration in a group of weight-recovered patients (AN_Rec). Manual delineation of anterior (AP) and posterior (PP) PG was performed on T1-weighted MR images in 17 women with AN, 15 women with AN_Rec, 18 athletes women and 25 female controls. Anthropometric, hormonal, and psychometric parameters were explored and correlated with PG volumes. AP volume (APV) was lower in AN (448 ± 82 mm 3 ), AN_Rec (505 ± 59 mm 3 ), and athletes (540 ± 101 mm 3 ) vs. Controls (615 ± 61 mm 3 , p < 0.00001, p < 0.00001 and p = 0.02, respectively); and smaller in AN vs. AN_Rec ( p = 0.007). PP volume did not show any differences between the groups. APV was positively correlated with weight ( R = 0.36, p = 0.011) in AN, and luteinizing hormone ( R = 0.35, p = 0.014) in total group. In AN, mean growth hormone (GH) was negatively correlated with global pituitary volume ( R = 0.31, p = 0.031) and APV ( R = 0.29, p = 0.037). Absolute and relative undernutrition led to a decreased anterior pituitary gland volume, which was reversible with weight gain, correlated with low bodyweight, and blockade of gonadal hypothalamic–pituitary axis. Intriguing inverse correlation between anterior pituitary gland volume and GH plasma level could suggests a low storage capacity of anterior pituitary gland and increased reactivity to low insulin-like growth factor type 1.