Exploring the relationship between pancreatic fat and insulin secretion in overweight or obese women without type 2 diabetes mellitus: A preliminary investigation of the TOFI_Asia cohort.

OBJECTIVE: While there is an emerging role of pancreatic fat in the aetiology of type 2 diabetes mellitus (T2DM), its impact on the associated decrease in insulin secretion remains controversial. We aimed to determine whether pancreatic fat negatively affects ß-cell function and insulin secretion in women with overweight or obesity but without T2DM. METHODS: 20 women, with normo- or dysglycaemia based on fasting plasma glucose levels, and low (< 4.5%) vs high (≥ 4.5%) magnetic resonance (MR) quantified pancreatic fat, completed a 1-hr intravenous glucose tolerance test (ivGTT) which included two consecutive 30-min square-wave steps of hyperglycaemia generated by using 25% dextrose. Plasma glucose, insulin and C-peptide were measured, and insulin secretion rate (ISR) calculated using regularisation deconvolution method from C-peptide kinetics. Repeated measures linear mixed models, adjusted for ethnicity and baseline analyte concentrations, were used to compare changes during the ivGTT between high and low percentage pancreatic fat (PPF) groups. RESULTS: No ethnic differences in anthropomorphic variables, body composition, visceral adipose tissue (MR-VAT) or PPF were measured and hence data were combined. Nine women (47%) were identified as having high PPF values. PPF was significantly associated with baseline C-peptide (p = 0.04) and ISR (p = 0.04) in all. During the 1-hr ivGTT, plasma glucose (p<0.0001), insulin (p<0.0001) and ISR (p = 0.02) increased significantly from baseline in both high and low PPF groups but did not differ between the two groups at any given time during the test (PPF x time, p > 0.05). Notably, the incremental areas under the curves for both first and second phase ISR were 0.04 units lower in the high than low PPF groups, but this was not significant (p > 0.05). CONCLUSION: In women with overweight or obesity but without T2DM, PPF did not modify ß-cell function as determined by ivGTT-assessed ISR. However, the salient feature in biphasic insulin secretion in those with ≥4.5% PPF may be of clinical importance, particularly in early stages of dysglycaemia may warrant further investigation.

Pancreas Fat, an Early Marker of Metabolic Risk? A Magnetic Resonance Study of Chinese and Caucasian Women: TOFI_Asia Study.

Objective: Prevalence of type 2 diabetes (T2D) is disproportionately higher in younger outwardly lean Asian Chinese compared to matched Caucasians. Susceptibility to T2D is hypothesised due to dysfunctional adipose tissue expansion resulting in adverse abdominal visceral and organ fat accumulation. Impact on early risk, particularly in individuals characterised by the thin-on-the-outside-fat-on-the-inside (TOFI) phenotype, is undetermined. Methods: Sixty-eight women [34 Chinese, 34 Caucasian; 18-70 years; body mass index (BMI), 20-45 kg/m2] from the TOFI_Asia study underwent magnetic resonance imaging and spectroscopy to quantify visceral, pancreas, and liver fat. Total body fat was (TBF) assessed by dual-energy x-ray absorptiometry, and fasting blood biomarkers were measured. Ethnic comparisons, conducted using two-sample tests and multivariate regressions adjusted for age, % TBF and ethnicity, identified relationships between abdominal ectopic fat depots with fasting plasma glucose (FPG), insulin resistance (HOMA2-IR), and related metabolic clinical risk markers in all, and within ethnic groups. Results: Despite being younger and of lower bodyweight, Chinese women in the cohort had similar BMI and % TBF compared to their Caucasian counterparts. Protective high-density lipoprotein cholesterol, total- and high-molecular weight adiponectin were significantly lower, while glucoregulatory glucagon-like peptide-1 and glucagon significantly higher, in Chinese. There were no ethnic differences between % pancreas fat and % liver fat. However, at low BMI, % pancreas and % liver fat were ∼1 and ∼2% higher in Chinese compared to Caucasian women. In all women, % pancreas and visceral adipose tissue had the strongest correlation with FPG, independent of age and % TBF. Percentage (%) pancreas fat and age positively contributed to variance in FPG, whereas % TBF, amylin and C-peptide contributed to IR which was 0.3 units higher in Chinese. Conclusion: Pancreas fat accumulation may be an early adverse event, in TOFI individuals, with peptides highlighting pancreatic dysfunction as drivers of T2D susceptibility. Follow-up is warranted to explore causality.

Automated Processing of Plasma Samples for Lipoprotein Separation by Rate-Zonal Ultracentrifugation.

Plasma lipoproteins are the primary means of lipid transport among tissues. Defining alterations in lipid metabolism is critical to our understanding of disease processes. However, lipoprotein measurement is limited to specialized centers. Preparation for ultracentrifugation involves the formation of complex density gradients that is both laborious and subject to handling errors. We created a fully automated device capable of forming the required gradient. The design has been made freely available for download by the authors. It is inexpensive relative to commercial density gradient formers, which generally create linear gradients unsuitable for rate-zonal ultracentrifugation. The design can easily be modified to suit user requirements and any potential future improvements. Evaluation of the device showed reliable peristaltic pump accuracy and precision for fluid delivery. We also demonstrate accurate fluid layering with reduced mixing at the gradient layers when compared to usual practice by experienced laboratory personnel. Reduction in layer mixing is of critical importance, as it is crucial for reliable lipoprotein separation. The automated device significantly reduces laboratory staff input and reduces the likelihood of error. Overall, this device creates a simple and effective solution to formation of complex density gradients.

Accelerating MR Imaging Liver Steatosis Measurement Using Combined Compressed Sensing and Parallel Imaging: A Quantitative Evaluation.

PURPOSE: To determine the limits of agreement of hepatic fat fraction and R2* relaxation rate quantified with accelerated magnetic resonance (MR) imaging reconstructed with combined compressed sensing and parallel imaging compared with conventional fully sampled acquisitions. MATERIALS AND METHODS: Eleven subjects with type 2 diabetes and a healthy control subject were recruited with the approval of the Newcastle and North Tyneside 2 ethics committee and written consent. Undersampled data at ratios of 2.6×, 2.9×, 3.8×, and 4.8× were prospectively acquired in addition to fully sampled data by using five gradient echoes per repetition time at 3.0 T. Fat fraction maps were calculated by using combined compressed sensing and parallel imaging (CS-PI) reconstruction and Bland-Altman analysis performed to assess bias and 95% limits of agreement. Inter- and intrarater analysis was performed for quantitative fat fraction and R2* relaxation rate, and image quality was assessed with a four-point scale by two independent observers. RESULTS: The fat fractions from the accelerated acquisitions had 95% limits of agreement of 1.2%, 1.2%, 1.1%, and 1.5%, respectively, with no bias. When compared with the intra- and interrater 95% limits of agreement (0.7% and 0.8%), acceleration of up to 3.8× did not greatly degrade the fat fraction measurements. No or minimal artifact was detected at 2.6× and 2.9× accelerations, moderate artifact was detected at 3.8× acceleration, and substantial artifact was detected at 4.8× acceleration. CONCLUSION: Prospective undersampling and CS-PI reconstruction of liver fat fractions can be used to accelerate liver fat fraction measurements. The fat fractions and image quality produced were acceptable up to a factor of 3.8×, thereby shortening the required breath-hold duration from 17.7 seconds to 4.7 seconds.