SHORT COMMUNICATION Can birth weight predict later body composition in anorexia nervosa? L Mattar 1,2,3 , C Pichard 4 , N Godart 1,2,3 , J-C Melchior 5,6 and EVAN group 7 BACKGROUND/OBJECTIVES: The relationship between birth weight and body composition at later stages in life was not studied previously in anorexia nervosa (AN). The aim of the following brief report is to present results concerning the relationship between birth weight and later body composition specifically in AN, and to check if the programming of body composition from birth weight is still detected in severely emaciated AN patients. SUBJECTS/METHODS: One hundred and fifty-one female AN patients aged between 13 and 44 were recruited from 11 inpatient treatment facilities in France. Birth weight, body weight and height were obtained. Body composition was measured using bioelectrical impedance. Birth weight was significantly correlated to lifetime maximum body mass index (BMI; r ¼ 0.211, P ¼ 0.009) and significantly correlated to fat-free mass index (r ¼ 0.190, P ¼ 0.027) but not to fat mass index (FMI). RESULTS: This report confirms that even in AN when patients are severely emaciated and where fat-free mass (FFM) and fat mass (FM) are low, a link between birth weight and FFM and BMI can still be identified, independently from age. CONCLUSION: Further studies are needed on larger samples exploring other factors, such as gender, puberty and ethnicity. European Journal of Clinical Nutrition (2012) 66, 964--967; doi:10.1038/ejcn.2012.21; published online 29 February 2012 Keywords: anorexia nervosa; birth weight; fat mass; fat-free mass; body mass index body composition INTRODUCTION Several studies investigated the link between birth weight and body composition components (fat mass (FM) and fat-free mass (FFM)) in early adolescence and adulthood in the context of predicting obesity and cardiovascular risk in adulthood. 1-7 In fact, there is a large corpus of literature on the relationships between birth weight and body composition at later stages in life, and on birth weight being one of the perinatal predictors of body composition, predicting obesity and cardiovascular disease in childhood or early adulthood. 1-7 Singhal et al. 6 showed that heavier birth weight was associated with greater FFM, but not with greater FM, in children and adolescents. This association was independent from age, gender, pubertal stage, socioeconomic status and physical activity. Another study performed on female twins also showed a positive correlation between birth weight and FFM, and also height in adulthood (3.26 cm adult height/kg increase in birth weight). 1,8 In the context of anorexia nervosa (AN), to our knowledge, no work has been done on birth weight and the programming of body composition in later stages of life in the condition of severe weight loss. AN is one of the most common chronic diseases in pediatrics with an evolution into adulthood for some patients; it is characterized by very low body mass index (BMI) and by an altered perception of weight and body image. 9 Few studies have investigated perinatal factors and AN later in life; one study found that in twins the one with the lighter birth weight was more likely to have a history of AN or other psychiatric disorders 10 whereas another found that birth weight had no link with AN. 11 AN is characterized by drastic weight loss affecting to different extent lean and FM depending on various factors such as physical activity, vomiting, laxative abuse and diet. Generally, weight loss is due to an extensive loss of FM, 12 thus body composition might be more susceptible to environmental influences such as weight loss rather than perinatal programming. The aim of the following brief report is, first, to investigate on the relationship between birth weight and later body composition in AN and, second, to check if the programming of body composition from birth weight is still detectable in severely emaciated AN patients. SUBJECTS AND METHODS Subjects One hundred and fifty-one female AN patients were recruited conse- cutively and prospectively from 11 inpatient treatment facilities in France (CHU-Bordeaux Bouvard, CHU-Bordeaux Pommereau, Cochin--Maison des Adolescents, Institut Mutualiste Montsouris, MGEN--La Verrie ` re, CHU- Nantes, CHU-Rouen, Robert Debre hospital, Sainte-Anne hospital, Saint- Etienne hospital, Villejuif -Paul Brousse). AN diagnosis was based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) criteria, obtained using the Eating Disorder Examination 13 and the Composite International Diagnostic Interview 3.0 with the following BMI criteria: BMI o10th percentile up to age 17, and BMIo17.5 for 17 and older. 14 Four patients out of the 151 did not have a BMIo17.5 at admission, but 2 of them had moved from a BMI above the 97th percentile to a BMI on the 10th percentile relative to their age in the year before hospitalization, and 2 of them had had a BMIo17.5 in the preceding 3 months, but were admitted to a medicine unit and had gained weight just Received 25 October 2011; revised 2 February 2012; accepted 2 February 2012; published online 29 February 2012 1 INSERM U669, Maison de Solenn, Paris, France; 2 Universite ´ Paris-Sud and Universite ´ Paris Descartes, Paris, France; 3 Psychiatry Unit, Institut Mutualiste Montsouris 42, Faculte ´ de me ´ decine Paris Descartes, Paris, France; 4 Unite ´ de Nutrition, Ho ˆ pitaux Universitaires de Gene ` ve, Rue Gabrielle-Perret-Gentil, Gene ` ve, Switzerland; 5 Unite ´ de Nutrition clinique- Maladies infectieuses, De ´partement de Me ´decine aigue ¨, Ho ˆ pital Raymond Poincare ´ , Garches, France and 6 Universite ´ Versailles St-Quentin-en-Yvelines, France. Correspondence: Dr L Mattar, Eating Disorders, INSERM U669, Maison de Solenn, 97 Boulevard De Port Royal, Paris 75014, France. E-mail: lamamattar@gmail.com 7 See Appendix. European Journal of Clinical Nutrition (2012) 66, 964 - 967 & 2012 Macmillan Publishers Limited All rights reserved 0954-3007/12 www.nature.com/ejcn