The pathogenesis of diabetes mellitus is mainly related to impaired β-cell function and insulin resistance [8]. Excessive fat accumulation in pancreatic tissues can contribute to the development of diabetes mellitus through lipid regulation of pancreatic β-cells, insulin resistance, and defective insulin-secreting function caused by disorders of fat metabolism [1, 9]. Kim et al. [10]investigated the relationship between pancreatic steatosis and atherosclerosis in patients with T2DM, and the results showed that pancreatic steatosis was closely associated with carotid atherosclerosis in non-obese patients with T2DM. Jeong et al. [11]evaluated pancreatic steatosis in 186 patients with T2DM by CT values and showed that pancreatic steatosis was still associated with diabetic retinopathy after adjusting for age, sex, and glycated hemoglobin levels. Therefore, effective quantitative assessment of pancreatic steatosis is important for the prevention and prognosis of T2DM. Conventional CT can only assess steatosis semi-quantitatively by measuring organ CT values, which have limited accuracy. In contrast, DECT eliminates the effects of CT drift and radiographic hardening artifacts and can simultaneously obtain virtual single-energy images, base matter images, and energy spectral profiles in a single scan, providing several quantitative parameters of the energy spectrum that are not available in conventional CT. Therefore, this study was performed to quantitatively assess the fat content of the pancreas by DECT and investigate the relationship between pancreatic steatosis and the pathogenesis of diabetes mellitus.
In this study, fat and water were selected as the base matter image pair to measure the pancreatic tissue fat content. The results showed that the pancreatic tissue fat (water) content was significantly higher in the T2DM obese subgroup than in the control group and that the pancreatic tissue fat (water) content was significantly higher in the T2DM obese subgroup than in the T2DM non-obese subgroup. One research group conducted a quantitative study of fatty liver in vivo using the fat (water) content based on fat (water)-based material images in energy spectral CT, and the results correlated well with the histopathological findings [12]. All of these results suggest that quantitative analysis of substances containing paired lipid/water-based substances is well suited for visceral fat quantification.
In this study, the CT values of pancreatic tissue were measured on a single-energy map at 70 keV. The results showed that the CT values of pancreatic tissue in the T2DM non-obese group were significantly lower than those in the control group and that the CT values of pancreatic tissue in the T2DM obese subgroup were significantly lower than those in the T2DM non-obese group. The energy spectrum plots were automatically generated by the GSI viewer software based on CT values for energy levels ranging from 40 to 140 keV, allowing a more direct reflection of the differences between substances. Similarly shaped spectral curves may reflect homology of the region of interest [13]. In this study, the control group had a lower fat (water) concentration and the spectral curve showed mostly a low flat curve, while the T2DM group had a higher fat (water) concentration and the spectral curve showed mostly a rising curve with the back of the bow going upward. A higher fat content of pancreatic tissue was associated with a higher slope of the spectral curve. Some studies have shown that the CT value of single-energy images of material with high fat content shows a gradual increase with the transition from 40 to 140 keV [14], which is consistent with the results of this study.
This study also analyzed the correlation between BMI and various energy spectrum parameters in all patients in the T2DM group. The data showed a positive correlation of the BMI with the fat (water) content and slope of the spectral curve and a negative correlation of the BMI with CT values. Tirkes et al. [15]analyzed the correlation between obesity and pancreatic steatosis by measuring the pancreatic fat fraction on magnetic resonance images and found a moderate direct correlation, which is consistent with the results of the present study.
The above results indicate that pancreatic adiposity is closely related to obesity and diabetes and that obesity increases the accumulation of pancreatic adiposity in patients with diabetes. Pancreatic steatosis is an independent risk factor for the development of T2DM [16, 17], which can lead to insulin resistance, decreased pancreatic β-cell function, and ultimately T2DM through several links. Once diabetes occurs, fat replacement in damaged tissues may lead to exocrine fat secretion from the pancreas or hyperglycemia, causing decreased mitochondrial β-oxidation and intracellular accumulation of triglycerides. Insulin resistance contributes to peripheral lipolysis and an increase in circulating free fatty acids. Prolonged exposure of pancreatic islet β-cells to high glucose and higher levels of free fatty acids leads to an increase in intracellular triglycerides. Obese patients can develop hyposecretion of leptin or leptin resistance, which can lead to excessive intracellular triglyceride deposition by attenuating the inhibitory effect of lipid esterification and inducing apoptosis in pancreatic islet β-cells [9, 18].
This study had three main limitations. First, the patients in the control group were younger and the sample size will need to be increased in the future to reduce the impact of systematic errors on the study results. Second, the fat values measured by this technique only reflect the overall fat content of the pancreas and do not reflect whether the fat deposits are within the pancreatic cells or within the pancreatic adipocytes. Third, because this was a cross-sectional study, it is not possible to further determine the causal relationship between the fat (water) concentration and the progression of diabetes or between pancreatic adiposity and the progression of diabetes. Thus, a prospective study is required.
In conclusion, DECT as a noninvasive and convenient imaging technique can quantitatively assess the fat content of pancreatic tissue by multiple parameters and provides a new method for grading the subsequent clinical development of pancreatic steatosis.