Use this URL to cite or link to this record in EThOS:
Title: Eating behaviour and neural representations of hunger and satiety in patients with acquired structural hypothalamic damage : a clinical and functional neuroimaging study
Author: Steele, C. A.
ISNI:       0000 0004 7428 6558
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Hypothalamic obesity (HO) is a relatively rare cause of obesity within the population as a whole, but studies of patients with hypothalamic damage show there is a significant prevalence of obesity within the patient group. Additionally, the obesity is difficult to prevent and manage and increases morbidity and mortality in an already at risk patient group. The work of this thesis had two main objectives. Firstly, to examine the prevalence of obesity and associated morbidities in patients with acquired, structural hypothalamic damage with a descriptive cohort study (n=110). A separate descriptive study quantified obesity and metabolic risk factors in young patients with pituitary tumours, with or without hypothalamic damage (n=41), as they were also identified as a possible at risk group during clinical observations. The second objective was to investigate the underlying pathophysiology of HO using various complementary techniques to study patients with hypothalamic damage who remained weight-stable (HWS), patients with HO and age- and BMI-matched controls. These cross-sectional case-control studies used functional magnetic resonance imaging (fMRI; 9 HO, 7 HWS, 20 controls), the universal eating monitor (UEM; 6 HO, 6 HWS, 9 obese controls [OC], 10 non-obese controls [NOC]), Three-Factor Eating Questionnaire (TFEQ; 8 HO, 6 HWS, 9 OC, 11 NOC) and three-day food diaries (6 HO, 7 HWS, 8 OC, 11 NOC) to assess eating behaviour. The first descriptive study included 110 adults with tumours causing hypothalamic damage attending a specialist neuroendocrine clinic. There was a significant prevalence of weight gain and obesity; 81.8% were overweight/heavier, 56.4% obese and 13.6% morbidly obese, despite proactive assessment and treatment during routine clinic visits. Hypertension (30.9%), dyslipidaemia (54.5%), type 2 diabetes mellitus (T2DM) (14.5%) and cardiovascular disease (9.1%) were also prevalent. In 41 patients with childhood/adolescent-onset pituitary adenomas there was also a relatively high prevalence of obesity (39.0%) and cardiovascular risk factors (2 receiving antihypertensive medications, 2 with T2DM and 4 with treated dyslipidaemia), despite the majority of tumours being microadenomas (i.e. too small to cause hypothalamic damage and indicating the need for long-term follow-up of these patients. The first study into the pathophysiology of HO involved the use of functional MRI in 9 patients with HO, 7 HWS and 20 age- and BMI-matched controls. Participants underwent fMRI scans in a fasted state, as well as one hour and three hours following a fixed-load breakfast (25% of their calculated basal metabolic rate [BMR]). At each scan session participants viewed alternating blocks of photographs of high- or low-calorie food, with non-food photographs also viewed to use as a baseline comparison, to allow purely visual activation to be subtracted from any BOLD signal differences which occurred. Whole-brain statistical analysis revealed significantly lower BOLD signal in HWS participants compared to HO (and to controls) in the food motivation and reward-related brain regions of the posterior insula and lingual gyrus (p=0.001) when viewing high-calorie food photographs (compared to non-food photographs). These differences in reward-related brain regions may be implicated in the development of HO/the ability to remain weight-stable despite hypothalamic damage. Eating behaviour studies were undertaken on a separate study day where participants were asked to eat an unlimited pasta meal until adequately full, while seated at the UEM. This allowed monitoring of total intake, eating duration, eating rate and intra-meal on-screen ratings of hunger, fullness and meal pleasantness using visual analogue scales. Additionally Three Factor Eating Questionnaires (TFEQs) and three-day MRC-Human Nutrition Research diaries were completed at home to assess more long-term real-world eating habits. None of the eating behaviour studies identified significant statistical differences between HO and HWS, but this may have been due to lack of statistical power. There was however an unusual pattern of eating rate in those with HO on visual ascription. This involved an initial tendency towards a higher eating rate, followed by a reduction in rate, with a further increase towards the end of the meal. Further investigation of this pattern with larger numbers of participants would be important to determine whether it is a significant finding or merely an anomaly due to the small group size. Interestingly controls ate significantly more at the UEM (even when intake was adjusted according to fat-free mass or as a proportion of the estimated BMR) and for significantly longer than participants with hypothalamic damage, who reported lower hunger at the start of the meal, but there was no difference between obese and non-obese participants. In keeping with previously published research disinhibition scores measured using the TFEQ were higher in participants with simple obesity [1], with no evidence of increased disinhibition in HO participants [2], although the small numbers studied should be noted. Finally, during both of the study days blood sampling was undertaken to look for biochemical/hormonal variances between the groups. Fasting and area under the curve (AUC) active ghrelin concentrations were significantly higher in controls than in patients, in keeping with some (but not all) previous studies [2-4]. Consistent with previously published research leptin concentrations were significantly higher in obese compared to non-obese participants The small size of this study was a significant limitation and limits the generalisability of the findings, particularly in the eating behaviour studies. This was due in part to the rarity of the condition - the occurrence of acquired, structural hypothalamic damage is relatively rare and the number of individuals with hypothalamic damage who remain weight stable is small, leading to particular difficulty in recruiting patients for the HWS group. Whilst there were some interesting findings further larger studies should enable greater clarification and would allow correlation between clinical and biochemical findings. Further studies in larger cohorts could explore the unusual eating pattern seen in those with HO when studied using the UEM and the apparent lack of disinhibition seen in this group. Although this research provides some preliminary novel evidence to support differences in BOLD signal in reward-related regions of the brain as a possible driver of HO, the mechanisms through which these differences exert their effects to contribute to the development of HO require further elucidation and further study with larger numbers of patients is needed.
Supervisor: Daousi, Christina ; Stancak, Andrej ; Wilding, John Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral