ESPE2014 Late Breaking Posters (1) (17 abstracts)
aDepartment of Obesity and Endocrinology, University of Liverpool, Liverpool, UK; bDepartment of Psychological Sciences, University of Liverpool, Liverpool, UK; cMARIARC and Department of Musculoskeletal Biology, University of Liverpool, Liverpool, UK; dDepartment of Psychology, Edge Hill University, Liverpool, UK; eDepartment of Neuroradiology, The Walton Centre NHS Foundation Trust, Liverpool, UK; fDepartment of Neurosurgery, The Walton Centre NHS Foundation Trust, Liverpool, UK
Background: Acquired hypothalamic damage frequently causes obesity (BMI ≥30 kg/m2), often refractory to treatment. The interaction of hormonal, neuronal and psychological factors underlying hypothalamic obesity (HO) remains poorly understood.
Methods: In fasted and fed states participants underwent blood sampling (GLP-1, insulin, PYY, ghrelin and glucose), fMRI scanning (viewing food/non-food photographs) and assessment of hunger and satiety (visual analogue scales). Standard biochemical analysis was undertaken. Interaction of ghrelin, PYY and insulin was analysed using ANOVA and post-prandial area under the curve (AUC) calculated. fMRI data was analysed using repeated-measures ANOVA. Analysis of covariance assessed neural activation while controlling for GLP-1, PYY, glucose and insulin.
Results: We studied nine HO (mean BMI 37.7 kg/m2), seven hypothalamic-damage weight-stable (HWS) (BMI 26.9 kg/m2), ten non-obese controls (NOC) (BMI 26 kg/m2) and ten obese controls (OC) (BMI 38 kg/m2). Pituitary hormone deficiencies were replaced; age, gender and BMI of HO/OC and HWS/NOC were similar. Fasting and post-prandial AUC insulin was no different between HO and OC or HWS and NOC. Ghrelin was similar in HO and HWS (P=0.5) and HO and OC (P=0.6). Fasting PYY was similar in HO and HWS (P=0.7), HO and OC (P=0.4), HWS and NOC (P=0.4); none had post-prandial PYY increase. Fasting GLP-1 was similar between HO and HWS (P=0.5), HO and OC (P=0.1) and HWS and NOC (P=0.7); AUC post-prandial GLP-1 was similar in HO, HWS and OC. Viewing high-calorie food photographs resulted in significantly lower insula (P=0.02) and lingual gyrus activation (P=0.025) in HWS than all others and insulin was significantly negatively correlated with insula activation. Obese participants rated hunger and desire to eat higher at all timepoints, with no difference between patients and controls.
Conclusions: Plasma insulin was strongly correlated with decreased insula activation, with PYY, GLP-1 and ghrelin unrelated. Decreased insula activation may protect HWS individuals from the weight-gain in HO.