Gender modulates the relationship between body weight and plasma glucose in overweight or obese subjects § Giuseppe Seghieri a, *, Federica Tesi a , Roberto Anichini a , Alessandra De Bellis a , Gianna Fabbri a , Raffaella Malagoli a , Flavia Franconi b a Department of Internal Medicine, Spedali Riuniti, Viale Matteotti 9/D, 51100 Pistoia, Italy b Centre of Excellence for Biotechnology Development and Biodiversity Research and Department of Drug Science, University of Sassari, Via Muroni 23a, Sassari 07100, Italy 1. Introduction Overweight and obesity are, as long since known, related to impaired glucose metabolism and diabetes [1–3]. Furthermore in obese or in non-obese people of both sexes, even small increases in body weight after the age of 18 years, lead to a significant increase in the relative risk to develop diabetes mellitus [4–9]. It is likewise known that fasting plasma glucose (FPG) is an important predictor for the development of diabetes, even when it is in its normal range [10–12], and it is thereby possible to hypothesise that weight change during the adult life and diabetes research and clinical practice 80 (2008) 134–138 article info Article history: Received 14 August 2007 Accepted 24 October 2007 Published on line 20 February 2008 Keywords: Gender Impaired fasting glucose Weight change Waist circumference Overweight-obesity abstract Obesity and weight increase during adult life are strong predictors of type 2 diabetes. Whether fasting plasma glucose (FPG) is likewise related to body weight as well as with its increase during the adult life in obese-overweight people and whether this relationship is different between the genders is the question asked by the present study. We measured FPG in 1063 overweight-obese subjects (395M/668F) with BMI  25 kg m 2 and classified with no history of diabetes and with a FPG < 7 mmol/l, who consequently came to the Outpatient Clinic of our Diabetes Unit to obtain dietetic advice. Weight increase was determined as the difference between actual weight and weight at 18 years (weight-diff), including only patients with weight-diff > 0. By univariate analysis age, BMI, waist circumference and weight change were loosely related to FPG in both sexes, even if the relation between plasma glucose and anthropometric variables was more consistent in females. By multivariate regression analysis, after adjusting for age, waist circumference, menopausal status and smoking habit, FPG was significantly related to both waist circumference and weight-diff only in women. Odds Ratio for fasting hyperglycaemia (FPG > 6.11 mmol/l), for each S.D. unit increase in weight-diff, after adjusting for age, waist circumference, smoking habit and menopausal status was 1.272; 95% CI: 0.863–1.901 ( p = ns) for males and 1.800; 95% CI: 1.239– 2.652 ( p = 0.002) for women. In conclusion our findings suggest that in non-diabetic over- weight-obese people, after controlling for main cofounders, anthropometric variables and in particular waist circumference and weight change after 18 years are linearly related to FPG in women, independently predicting the risk of fasting hyperglycaemia only in these latter. # 2007 Elsevier Ireland Ltd. All rights reserved. § This study was sponsored by the Fondazione Cassa di Risparmio di Pistoia e Pescia. * Corresponding author. Tel.: +39 0573 352325; fax: +39 0573 352005. E-mail address: gseghier@tin.it (G. Seghieri). available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/diabres 0168-8227/$ – see front matter # 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2007.10.025