Chris J Gardner, Anders F Mattsson, Christina Daousi, Márta Korbonits, Maria Koltowska-Haggstrom and Daniel J Cuthbertson
Prevalence of GH deficiency (GHD) caused by traumatic brain injury (TBI) is highly variable. Short-term studies show improvement in quality of life (QoL) during GH replacement (GHR), but long-term data are lacking. The aim of this study was to analyse the clinical characteristics of post-traumatic hypopituitarism and the QoL effects of long-term GHR.
Pfizer International Metabolic Database patients with GHD caused by TBI and by non-functioning pituitary adenoma (NFPA) were compared regarding: clinical characteristics at baseline and 1-year of GHR, and QoL response up to 8-years of GHR (QoL-AGHDA total scores and dimensions) in relationship with country-specific norms.
TBI patients compared with NFPA patients were younger, diagnosed with GHD 2.4 years later after primary disease onset (P<0.0001), had a higher incidence of isolated GHD, higher GH peak, a more favourable metabolic profile and worse QoL, were shorter by 0.9 cm (1.8 cm when corrected for age and gender; P=0.004) and received higher GH dose (mean difference: 0.04 mg/day P=0.006). In TBI patients, 1-year improvement in QoL was greater than in NFPA (change in QoL-AGHDA score 5.0 vs 3.5, respectively, P=0.04) and was sustained over 8 years. In TBI patients, socialisation normalised after 1 year of GHR, self-confidence and tenseness after 6 years and no normalisation of tiredness and memory was observed.
Compared with NFPA, TBI patients presented biochemically with less severe hypopituitarism and worse QoL scores. GHR achieved clinically relevant, long-term benefit in QoL.
Chris J Gardner, Mohsen Javadpour, Catherine Stoneley, Mani Purthuran, Shubhabrata Biswas, Christina Daousi, Ian A MacFarlane and Daniel J Cuthbertson
Hypopituitarism following subarachnoid haemorrhage (SAH) has been reported to be a frequent occurrence. However, there is considerable heterogeneity between studies with differing patient populations and treatment modalities and most importantly employing differing endocrine protocols and (normal) reference ranges of GH. We aimed to examine prospectively a cohort of SAH survivors for development of hypopituitarism post-SAH using rigorous endocrine testing and compare GH response to glucagon stimulation with a cohort of healthy controls of a similar BMI.
Design and methods
Sixty-four patients were investigated for evidence of hypopituitarism 3 months post-SAH with 50 patients tested again at 12 months. Glucagon stimulation testing (GST), with confirmation of deficiencies by GHRH/arginine testing for GH deficiency (GHD) and short synacthen testing for ACTH deficiency, was used. Basal testing of other hormonal axes was undertaken.
Mean age of patients was 53±11.7 years and mean BMI was 27.5±5.7 kg/m2. After confirmatory testing, the prevalence of hypopituitarism was 12% (GHD 10%, asymptomatic hypocortisolaemia 2%). There was no association between hypopituitarism and post-SAH vasospasm, presence of cerebral infarction, Fisher grade, or clinical grading at presentation. There was a significant correlation between BMI and peak GH to glucagon stimulation in both patients and controls.
Identification of ‘true’ GHD after SAH requires confirmatory testing with an alternative stimulation test and application of BMI-specific cut-offs. Using such stringent criteria, we found a prevalence of hypopituitarism of 12% in our population.
Caroline A Steele, Ian A MacFarlane, Joanne Blair, Daniel J Cuthbertson, Mohammed Didi, Conor Mallucci, Mohsen Javadpour and Christina Daousi
To elucidate the long-term outcomes of pituitary adenomas diagnosed in childhood and adolescence, knowledge of which remains sparse.
Design and methods
A retrospective review of patients aged ≤21 years at diagnosis of pituitary adenoma, attending a neuroendocrine service in Liverpool, UK, between 1984–2009.
There were 41 patients (33 female), mean age at diagnosis 17.3 years (range 11–21) and mean follow-up 9.6 years; 29 patients had prolactinomas (15 macroprolactinomas), 6 non-functioning pituitary adenomas (NFPAs), 5 Cushing's disease (CD) and 1 acromegaly. All prolactinoma patients received dopamine agonists (DAs) and three also underwent pituitary surgery. Furthermore, ten patients underwent surgery: five with CD, one with acromegaly and four with NFPA. Four received radiotherapy after surgery. Another ten patients received hormone replacement: nine hydrocortisone, five thyroxine, seven sex steroids and five GH; another seven had severe asymptomatic GH deficiency. Three female patients were treated for infertility (two successfully). Thirteen patients gained significant weight (body mass index (BMI) increase >2 kg/m2) since diagnosis and 16 in total are now obese (BMI>30 kg/m2). Five were treated with orlistat and one attended a weight management service. Two received antihypertensive medications, two had type 2 diabetes and four were treated for dyslipidaemia.
This is one of the largest reviews of patients aged 21 or younger at diagnosis of pituitary adenoma followed up by a single service. Two-thirds had prolactinomas, all were treated with DAs and three underwent surgery. Increased cardiovascular risk factors (obesity and dyslipidaemia) and infertility are important sequelae and active identification and treatment are necessary.
Chris J Gardner, Andrew J Irwin, Christina Daousi, Ian A McFarlane, Franklin Joseph, Jimmy D Bell, E Louise Thomas, Valerie L Adams, Graham J Kemp and Daniel J Cuthbertson
Non-alcoholic fatty liver disease (NAFLD) is reported to be more common in patients with GH deficiency (GHD) than in the general population. We aimed to determine: i) liver fat in patients with GHD compared with age and body mass index (BMI)-matched controls; and ii) effect of 6 months of GH replacement (GHR) on liver fat.
Participants and methods
The study included 28 GHD patients and 24 controls. 12 patients were studied before and after 6 months of GHR. Anthropometry, liver enzymes and lipid profiles were measured, and body composition and intrahepatocellular lipid (IHCL) were determined by magnetic resonance imaging and spectroscopy.
Age and BMI (median (inter-quartile range)) of patients and controls were 52.6 (14) vs 52.6 (12) years (P=0.9) and 27.8 (24.7, 34.7) vs 27.9 (25.1, 32.1) kg/m2 (P=0.9). IGF1 was lower in the patients (11.5 vs 16.0 nmol/l, P=0.002). There was no difference in liver transaminases, lipids or IHCL between patients and controls (2.8 (1.3, 8.6) vs 5.0 (1.5, 12.7), P=0.72), despite significantly higher visceral fat in GHD patients. Thirty-two percent of patients and 50% of controls had NAFLD (defined as IHCL >5.6%), and the relationship between IHCL and BMI was the same in each group. GHR significantly reduced abdominal subcutaneous and visceral fat in all patients; however, GHR did not reduce liver fat.
NAFLD is equally common in patients with GHD and matched controls. GHR is associated with a hierarchical reduction in fat deposition (fat loss: visceral > subcutaneous > liver). Further studies involving GHD patients with NAFLD are required to conclude the role of GHR in treating NAFLD.
Caroline A Steele, Daniel J Cuthbertson, Ian A MacFarlane, Mohsen Javadpour, Kumar S V Das, Catherine Gilkes, John P Wilding and Christina Daousi
Obesity is highly prevalent among adults with acquired, structural hypothalamic damage. We aimed to determine hormonal and neuroanatomical variables associated with weight gain and obesity in patients following hypothalamic damage and to evaluate the impact of early instigation of weight loss measures to prevent or limit the severity of obesity in these patients.
Retrospective study of 110 adults with hypothalamic tumours attending a specialist neuroendocrine clinic. BMI was calculated at diagnosis and at last follow-up clinic visit. Endocrine data, procedures, treatments and weight loss measures were recorded and all available brain imaging reviewed.
At last follow-up, 82.7% of patients were overweight or heavier (BMI≥25 kg/m2), 57.2% were obese (BMI≥30 kg/m2) and 14.5% were morbidly obese (BMI≥40 kg/m2). Multivariate analysis revealed that use of desmopressin (odds ratio (OR)=3.5; P=0.026), GH (OR=2.7; P=0.031) and thyroxine (OR=3.0; P=0.03) was associated with development of new or worsened obesity. Neuroimaging features were not associated with weight gain. Despite proactive treatments offered in clinic in recent years (counselling, dietetic and physical activity advice, and anti-obesity medications), patients have continued to gain weight.
Despite increased awareness, hypothalamic obesity is difficult to prevent and to treat. Improved understanding of the underlying pathophysiologies and multicentre collaboration to examine efficacy of novel obesity interventions are warranted.
Daniel J Cuthbertson, Martin O Weickert, Daniel Lythgoe, Victoria S Sprung, Rebecca Dobson, Fariba Shoajee-Moradie, Margot Umpleby, Andreas F H Pfeiffer, E Louise Thomas, Jimmy D Bell, Helen Jones and Graham J Kemp
Background and aims
Simple clinical algorithms including the fatty liver index (FLI) and lipid accumulation product (LAP) have been developed as surrogate markers for non-alcoholic fatty liver disease (NAFLD), constructed using (semi-quantitative) ultrasonography. This study aimed to validate FLI and LAP as measures of hepatic steatosis, as determined quantitatively by proton magnetic resonance spectroscopy (1H-MRS).
Data were collected from 168 patients with NAFLD and 168 controls who had undergone clinical, biochemical and anthropometric assessment. Values of FLI and LAP were determined and assessed both as predictors of the presence of hepatic steatosis (liver fat >5.5%) and of actual liver fat content, as measured by 1H-MRS. The discriminative ability of FLI and LAP was estimated using the area under the receiver operator characteristic curve (AUROC). As FLI can also be interpreted as a predictive probability of hepatic steatosis, we assessed how well calibrated it was in our cohort. Linear regression with prediction intervals was used to assess the ability of FLI and LAP to predict liver fat content. Further validation was provided in 54 patients with type 2 diabetes mellitus.
FLI, LAP and alanine transferase discriminated between patients with and without steatosis with an AUROC of 0.79 (IQR=0.74, 0.84), 0.78 (IQR=0.72, 0.83) and 0.83 (IQR=0.79, 0.88) respectively although could not quantitatively predict liver fat. Additionally, the algorithms accurately matched the observed percentages of patients with hepatic steatosis in our cohort.
FLI and LAP may be used to identify patients with hepatic steatosis clinically or for research purposes but could not predict liver fat content.
Joseph D Maxwell, Howard H Carter, Ylva Hellsten, Gemma D Miller, Victoria S Sprung, Daniel J Cuthbertson, Dick H J Thijssen and Helen Jones
Remote ischaemic preconditioning (rIPC) may improve cardiac/cerebrovascular outcomes of ischaemic events. Ischaemic damage caused by cardiovascular/cerebrovascular disease are primary causes of mortality in type 2 diabetes mellitus (T2DM). Due to the positive effects from a bout of rIPC within the vasculature, we explored if daily rIPC could improve endothelial and cerebrovascular function. The aim of this pilot study was to obtain estimates for the change in conduit artery and cerebrovascular function following a 7-day rIPC intervention.
Twenty-one patients with T2DM were randomly allocated to either 7-day daily upper-arm rIPC (4 × 5 min 220 mmHg, interspaced by 5-min reperfusion) or control. We examined peripheral endothelial function using flow mediated dilation (FMD) before and after ischemia-reperfusion injury (IRI, 20 min forearm ischaemic-20 min reperfusion) and cerebrovascular function, assessed by dynamic cerebral autoregulation (dCA) at three time points; pre, post and 8 days post intervention.
For exploratory purposes, we performed statistical analysis on our primary comparison (pre-to-post) to provide an estimate of the change in the primary and secondary outcome variables. Using pre-intervention data as a covariate, the change from pre-post in FMD was 1.3% (95% CI: 0.69 to 3.80; P = 0.09) and 0.23 %cm/s %/mmHg mmHg/% (−0.12, 0.59; P = 0.18) in dCA normalised gain with rIPC versus control. Based upon this, a sample size of 20 and 50 for FMD and normalised gain, respectively, in each group would provide 90% power to detect statistically significant (P < 0.05) between-group difference in a randomised controlled trial.
We provide estimates of sample size for a randomised control trial exploring the impact of daily rIPC for 7 days on peripheral endothelial and cerebrovascular function. The directional changes outline from our pilot study suggest peripheral endothelial function can be enhanced by daily rIPC in patients with T2DM.
Victoria S Sprung, Kelly A Bowden Davies, Juliette A Norman, Andrew Thompson, Katie L Mitchell, John P H Wilding, Graham J Kemp and Daniel J Cuthbertson
Data suggest that metabolic health status, incorporating components of metabolic syndrome (MetS), predicts cardiovascular disease (CVD) risk better than BMI. This study explored the association of MetS and obesity with endothelial function, a prognostic risk factor for incident CVD.
Forty-four participants were phenotyped according to BMI as non-obese vs obese (<30 or >30 kg/m2) and according to the International Diabetes Federation criteria of MetS: ≤2 criteria MetS (MetS−) vs ≥3 criteria MetS (MetS+); (1.)non-obese MetS− vs (2.) non-obese MetS+ and (3.) obese MetS− vs (4.) obese MetS+. Flow-mediated dilation (FMD), body composition including liver fat (MRI and spectroscopy), dietary intake, intensities of habitual physical activity and cardio-respiratory fitness were determined. Variables were analysed using a one-factor between-groups ANOVA and linear regression; mean (95% CI) are presented.
Individuals with MetS+ displayed lower FMD than those with MetS−. For non-obese individuals mean difference between MetS+ and MetS− was 4.1% ((1.0, 7.3); P = 0.004) and obese individuals had a mean difference between MetS+ and MetS− of 6.2% ((3.1, 9.2); P < 0.001). Although there was no association between BMI and FMD (P = 0.27), an increased number of MetS components was associated with a lower FMD (P = 0.005), and after adjustment for age and sex, 19.7% of the variance of FMD was explained by MetS, whereas only 1.1% was explained by BMI.
In this study cohort, components of MetS, rather than obesity per se, contribute to reduced FMD, which suggests a reduced bioavailability of nitric oxide and thus increased risk of CVD.