Fatty pancreas disease in newly diagnosed type 2 diabetes patients: a case–control study on relationships with glycemic control and exocrine function

Abstract

Background

Fatty pancreas disease (FPD) is characterized by abnormal fat accumulation in pancreatic tissue and is often associated with obesity, metabolic syndrome, and type 2 diabetes mellitus (T2DM). While its pathophysiology and impact on pancreatic functions have been explored, the interplay between FPD, glycemic control, and exocrine dysfunction in T2DM remains inadequately defined. This study aimed to evaluate the presence of FPD, the factors affecting it, and its relationship with endocrine and exocrine pancreatic functions in newly diagnosed T2DM.

Methods

A total of 126 individuals were included in the study, comprising 63 newly diagnosed T2DM patients and 63 healthy controls matched for age, sex, body mass index and body fat distribution. Body composition, biochemical parameters (glucose, insulin, C-peptide, HbA1c), fecal elastase levels, and pancreatic/hepatic steatosis grades (evaluated using ultrasonography) were assessed.

Results

Newly diagnosed T2DM patients presented significantly higher hepatic steatosis grades (p = 0.018) and lower fecal elastase levels (p < 0.001) compared to controls. Pancreatic exocrine insufficiency was more prevalent in the T2DM group (p < 0,001). A positive correlation was observed between the FPD grade, hepatic steatosis grade, and hepatic fat fraction. A negative and statistically significant correlation (p < 0.05) was observed between FPD grade and fecal elastase level (r = -0.264). HbA1c levels demonstrated a nonlinear (inverse U-shaped) relationship with FPD, peaking at 9.8% and declining thereafter, while showing a continuous negative relationship with fecal elastase levels. HbA1c predicted low fecal elastase (< 200 μg/g) with a cutoff value of 7.4%. Patients with HbA1c levels > 9.8% presented with reduced FPD alongside persistent exocrine insufficiency.

Conclusions

Fatty pancreas disease is closely associated with hepatic steatosis, glycemic control, and exocrine pancreatic dysfunction in newly diagnosed T2DM patients. The interplay between FPD, glycemic control, and exocrine dysfunction highlights the need for comprehensive metabolic assessments in this population.

Introduction

Type 2 diabetes mellitus (T2DM) is a complex and widespread metabolic disease, characterized by chronic hyperglycemia due to insulin resistance and/or inadequate insulin secretion [1]. Diffuse excess intra-pancreatic fat deposition (IPFD), or fatty pancreas disease (FPD), is frequently observed and may even precede the development of T2DM [2]. Among the systemic complications of T2DM, metabolic dysfunction-associated steatotic liver disease (MASLD) and FPD have emerged as conditions of increasing clinical interest.

Non-alcoholic FPD refers to the accumulation of fat in the pancreas independent of excessive alcohol consumption and is increasingly common in T2DM patients, being associated with both endocrine and exocrine pancreatic dysfunction [4]. Moreover, strong evidence supports the presence of minimal, yet clinically relevant, IPFD even in the healthy human pancreas [2]. The exact pathophysiology of this clinical condition remains inadequately defined. Major risk factors for FPD include obesity and metabolic syndrome [5]. Lipid accumulation and subsequent lipotoxicity in pancreatic acinar cells can lead to acinar cell damage, resulting in exocrine insufficiency. However, the impact of FPD on exocrine pancreatic functions has not been thoroughly investigated. Furthermore, pancreatic fat cells contribute to insulin resistance, impairing islet cell function [6]. Comprehensive studies are needed to elucidate the relationships among FPD, diabetes, insulin secretion, and other influencing factors.

The relationship between FPD and T2DM appears to be bidirectional. While hyperglycemia and insulin resistance promote pancreatic fat accumulation, FPD may impair pancreatic functions, exacerbating glucose dysregulation. Risk factors such as obesity, metabolic syndrome, and advanced age are strongly associated with FPD, paralleling those of MASLD [7].

Despite advances in imaging techniques and diagnostic tools, such as fecal elastase testing and hepatic fat fraction measurements, the precise factors influencing IPFD and its impact on endocrine and exocrine pancreatic functions remain inadequately explored. Understanding these interactions is crucial for improving the clinical management of people living with T2DM.

The primary objective of this study was to evaluate the presence of FPD, the factors influencing its development, and its effects on endocrine and exocrine pancreatic functions in newly diagnosed T2DM patients compared with a control group matched for age, body weight, body mass index (BMI), and body composition. A secondary objective was to investigate the relationship between hepatic steatosis and FPD, with a focus on their presence and severity within the T2DM patient group.

Materials and methods

Study design

This study was a single-center, cross-sectional, non-interventional clinical study conducted at the Department of Endocrinology and Metabolism, Gazi University Faculty of Medicine, between August 2023 and August 2024. Written informed consent was obtained from all participants after providing detailed information about the study.

Participants

The study population consisted of two groups: 63 patients newly diagnosed with T2DM and 63 non-diabetic controls matched for age, sex, BMI, and body composition. The inclusion criteria for the patient group were individuals aged 18–80 years who were newly diagnosed with T2DM. The diagnosis of diabetes was established based on fasting plasma glucose and HbA1c measurements, following the American Diabetes Association (ADA) criteria. Individuals aged 18–80 years with newly diagnosed diabetes were included in the study if they met the diagnostic criteria of fasting plasma glucose ≥ 126 mg/dL, HbA1c ≥ 6.5%, or 2-h plasma glucose ≥ 200 mg/dL during a 75 g oral glucose tolerance test (OGTT), considering clinical history, physical examination, and laboratory findings. To confirm the classification of T2DM and differentiate it from other diabetes subtypes, additional parameters such as age at diagnosis, BMI, fasting insulin levels, C-peptide levels, HOMA-IR values, and family history were evaluated. The response to initiated treatments further supported the diagnosis. Autoantibodies associated with autoimmune diabetes were not routinely tested due to cost considerations; however, they were measured in cases where latent autoimmune diabetes in adults (LADA) or type 1 diabetes (T1DM) was clinically suspected. None of the tested patients were positive for these autoantibodies, further ruling out autoimmune diabetes. Some participants had markedly high HbA1c levels, but their fasting insulin, C-peptide, and HOMA-IR values indicated preserved insulin secretion and significant insulin resistance, making T1DM unlikely. Additionally, pancreatogenic diabetes was excluded by ensuring that individuals with a history of pancreatitis, chronic alcohol consumption, gallstones, or previous pancreatic surgery were not included in the study. These exclusion criteria helped minimize potential confounders and ensured that all included participants had a confirmed diagnosis of T2DM [5].

The control group consisted of adult volunteers who attended Gazi University Hospital for routine check-ups and met the inclusion and exclusion criteria. Controls were selected through a one-to-one matching method based on body composition, ensuring comparability with the patient group in terms of age, sex, and BMI. Diabetes was excluded based on fasting plasma glucose < 126 mg/dL and HbA1c < 6.5%, and in necessary cases, 2-h plasma glucose < 200 mg/dL following a 75 g OGTT. Exclusion criteria for both groups included a history of thyroid dysfunction, malignancy, pancreatitis, pancreatic surgery, liver, kidney, or heart failure, alcohol consumption, pregnancy, the use of medications affecting blood glucose levels (e.g., steroids), and gastrointestinal infections within one month prior to fecal examination. Control subjects were not selected from a pre-existing database of individuals with prior body composition assessments but were instead screened among volunteers until a suitable match was identified.

Data collection and clinical assessments

A detailed medical history, including the duration of hypertension, dyslipidemia, atherosclerotic cardiovascular disease (ASCVD), and medications, was obtained for all participants. Physical examination data, including height, weight, BMI, waist circumference, waist-to-hip ratio, and visceral adiposity levels, were recorded. Body composition was assessed using the Tanita BC-418MA Body Composition Analyzer (Tanita Corp., Tokyo, Japan). The patient group was treatment-naive with respect to diabetes, meaning that they were not on any antidiabetic medication. Ten patients in the patient group and one individual in the control group had a history of cardiovascular disease and were receiving statin therapy. Consequently, lipid parameters could not be matched and were excluded from the analysis.

Laboratory measurements and biochemical analysis

After at least 8 h of fasting, 25 mL of blood and 20 g of stool samples were collected from each participant. The homeostasis model assessment for insulin resistance (HOMA-IR) was calculated as follows: HOMA-IR = Fasting insulin (µIU/mL) × Fasting glucose (mg/dL) / 405 [4]. All biochemical parameters were analyzed using standardized methods. Fasting glucose was determined by the hexokinase enzymatic reference method, while HbA1c levels were assessed through high-performance liquid chromatography. Insulin and C-peptide were measured using electro-immunoassay method. Fecal elastase was analyzed using an ELISA-based method (COMBIWASH HUMAN, IDK Pancreatic Elastase ELISA, North America). Fecal elastase levels were used to diagnose exocrine pancreatic insufficiency, with values of ≤ 200 µg/g indicating mild-to-moderate exocrine pancreatic insufficiency and ≤ 100 µg/g indicating severe exocrine pancreatic insufficiency.

Imaging assessments

Various imaging modalities, including abdominal ultrasound, endoscopic ultrasound, computed tomography (CT), proton magnetic resonance spectroscopy, and magnetic resonance imaging (MRI), are commonly utilized to assess pancreatic fat content. However, ultrasound has inherent limitations in the evaluation of the pancreas, particularly in individuals with overweight or obesity [8, 9]. Nonetheless, given its low cost, wide availability, and the absence of ionizing radiation exposure, abdominal ultrasound was chosen as the primary imaging modality in our study.

Ultrasonography was performed using the RS85 Prestige system (Samsung Medison Co., Ltd.). Hepatic steatosis, hepatic fat fraction percentages and IPFD were evaluated. The hepatic steatosis grades were categorized as shown in Table 1 [10]. Hepatic fat fraction assessments (ultrasound-derived estimated fat fraction, USFF) were carried out using a convex probe with a frequency range of 1–8 MHz. The operator initially positioned a 2 × 3 cm fan-shaped region of interest (ROI) within the right liver lobe, ensuring it was located at a depth of at least 2 cm beneath the liver capsule and avoiding areas affected by reverberation artifacts, focal liver lesions, or prominent vessels. Tissue attenuation imaging (TAI) values were then automatically computed, with the reliability of each measurement indicated by an R² value. The operator aimed to secure a TAI measurement with an R² value of no less than 0.6. Using the same ROI, the tissue scatter-distribution imaging (TSI) value was subsequently calculated by activating the TSI function key. The USFF was then derived using the formula: USFF = − 44.3 + 41.9 × TAI + 0.23 × TSI. During each session, five USFF measurements were obtained, and the mean of these five acquisitions was used as the representative value for each participant, as per the vendor's recommendation [11].

Table 1 Hepatic steatosis grades

The pancreatic parenchyma was evaluated across all three segments—head, body, and tail—along with an assessment of its location, size, echogenic appearance, shape, and boundaries. The echogenicity of the pancreas was compared to that of the parenchyma in the left liver lobe. This evaluation was conducted at the same depth using a longitudinal scan near the abdominal midline. If the liver exhibited increased echogenicity, an additional comparison was made with the renal cortex. Fatty pancreas disease was diagnosed when the echogenicity of the pancreas was observed to be higher than that of the liver or the renal cortex. Since the pancreas and kidney cannot be visualized simultaneously within the same ultrasound window, the ultrasonographer assessed the differences in echogenicity between the liver and kidney and between the liver and pancreas to establish an objective contrast between the pancreas and kidney [12]. Subsequently, the pancreatic echogenicity was graded on the basis of its density. The grades of FPD are presented in Table 2.

Table 2 Fatty pancreas disease grades

Statistical analysis

The data were analyzed using IBM SPSS Statistics 26.0. Categorical variables were expressed as frequencies and percentages, while continuous variables were presented as the means and standard deviations for normally distributed data or medians and interquartile ranges for non-normally distributed data. Chi-square tests were used for comparisons of categorical variables. Independent t tests or Mann–Whitney U tests were applied for two-group comparisons of continuous variables. ANOVA or Kruskal–Wallis tests were utilized for multi-group comparisons, followed by Tukey or Bonferroni post-hoc analyses. Correlations between FPD, hepatic steatosis, fecal elastase, hepatic fat fraction, and metabolic parameters were assessed using Pearson or Spearman correlation tests. ROC analysis was performed to determine the sensitivity, specificity, and cutoff point for HbA1c and its association with hepatic steatosis, FPD, and exocrine pancreatic insufficiency. Statistical significance was set at p < 0.05.

Results

A total of 126 individuals were included in the study, comprising 63 newly diagnosed T2DM patients and 63 controls matched for age, sex, BMI and body fat distribution. No statistically significant differences were observed between the T2DM and control groups in terms of age, sex, BMI, waist-to-hip ratio or body fat distribution. However, a statistically significant difference was found between the T2DM group and the control group regarding hypertension (p < 0.001) and ASCVD (p = 0.005). A comparison of the demographic characteristics, body composition measurements, and comorbidities between the groups is presented in Table 3.

Table 3 Comparison of demographic, clinical, laboratory parameters and ultrasound findings in T2DM and control groups

No statistically significant difference was observed in FPD between the T2DM and control groups (Table 4). However, significant differences were identified among the HbA1c ≥ 10% T2DM group, the HbA1c < 10% T2DM group, and the control group in terms of FPD (p < 0.001), hepatic steatosis (p < 0.001), fecal elastase (p < 0.001), hepatic fat fraction (p < 0.001), and exocrine pancreatic insufficiency (p < 0.001). Detailed comparisons between the groups are presented in Table 4.

Table 4 Comparison of exocrine pancreatic insufficiency degrees and steatosis between T2DM (by HbA1c) and control groups

A statistically significant difference was observed between the T2DM group and the control group in terms of fasting plasma glucose (p < 0.001), fasting insulin (p < 0.001), HbA1c (p < 0.001), C-peptide (p = 0.006) and HOMA-IR (p < 0.001). A comparison of the laboratory results between the groups is presented in Table 3.

A statistically significant difference was observed between the T2DM group and the control group in terms of hepatic steatosis (p = 0.018), fecal elastase levels (p < 0.001), and exocrine pancreatic insufficiency (p < 0.001). 21 participants in the patient group had fecal elastase levels ≤ 100 µg/g, while none of the control participants had fecal elastase levels below this threshold. The comparisons between the groups are presented in Table 3.

A positive and statistically significant correlation (p < 0.05) was found between the FPD grade and hepatic steatosis (r = 0.835), hepatic fat fraction (r = 0.779), fasting insulin (r = 0.621), C-peptide (r = 0.356) and HOMA-IR (r = 0.473). A negative and statistically significant correlation (p < 0.05) was observed between FPD grade and fecal elastase level (r = − 0.264). Fecal elastase was negatively and significantly correlated (p < 0.05) with HbA1c (r = − 0.713) and HOMA-IR (r = − 0.574). The correlations between variables obtained from all participants included in the study are presented in Table 5.

Table 5 Correlation between hepatic steatosis, intra-pancreatic fat deposition, fecal elastase, hepatic fat fraction, and metabolic parameters

In patients with HbA1c < 10% and T2DM, the predictive value of HbA1c levels for the presence of FPD was calculated as AUC = 0.678 (95% CI: 0.572–0.784). The optimal cut-off value for HbA1c was determined to be 7.0. Conversely, the predictive value of HbA1c levels for the absence of FPD was calculated as AUC = 0.578 (95% CI: 0.462–0.695), with an optimal cut-off value of 9.85 (Fig. 1).

Fig. 1

The Relationship between HbA1c and fatty pancreas disease in newly diagnosed T2DM patients. This graph was created on the basis of statistically significant cut-off values derived from regression analyses and the relationships between variables assessed through correlation analyses, without the use of a specific statistical test

A statistically significant negative correlation (r = − 0,713, p < 0.01) was observed between fecal elastase and HbA1c levels (Fig. 2). The predictive value of HbA1c levels for low fecal elastase was calculated as AUC = 0.819 (95% CI:0.711–0.927). The optimal cut-off value for HbA1c was determined to be 7.4% (Fig. 3). In T2DM patients with HbA1c < 10%, the predictive value of HbA1c levels for the presence of hepatic steatosis was calculated as AUC = 0.650 (95% CI: 0.518–0.783). The optimal cut-off value for HbA1c was determined to be 6.5%.

Fig. 2

Relationship between HbA1c and fecal elastase

Fig. 3

ROC curve for predicting fecal elastase cases based on HbA1c levels

The predictive value of hepatic fat fraction levels for fecal elastase values < 200 μg/g was calculated as AUC = 0.664 (95% CI:0.560–0.768). The optimal cut-off value for hepatic fat fraction was determined to be 17.1%.

Discussion

In our study, fecal elastase levels were lower, and exocrine pancreatic insufficiency was more prevalent in the T2DM group. This phenomenon may result from pancreatic atrophy, fibrosis, and inflammation associated with diabetes. These findings may also be explained by the fact that our patients were evaluated at the time of diagnosis, during an acute hyperglycemic state, before any treatment was initiated. This acute metabolic disturbance, along with hyperglycemia and hypoinsulinemia-induced activation of pancreatic stellate cells and subsequent fibrosis, could contribute to the observed exocrine dysfunction [13]. Johnston et al. reported that poor glycemic control and elevated HbA1c levels were associated with reduced fecal elastase levels [14]. Zhang et al.’s systematic review and meta-analysis revealed a greater prevalence of exocrine pancreatic insufficiency among individuals with T2DM, especially those requiring higher insulin doses [15].

In our study, no significant difference was observed in FPD when comparing the entire T2DM group with the control group. However, significant differences were identified between the HbA1c ≥ 10% T2DM group and the HbA1c < 10% T2DM group. While the severity of FPD increased with HbA1c levels up to 9.8%, it subsequently declined beyond this threshold. A notable U-shaped correlation between HbA1c and IPFD was identified in our study. Clinical observations in our cohort revealed significant weight loss and reduced FPD in patients with HbA1c levels > 10%, suggesting the influence of insulinopenia and catabolic processes in advanced disease stages. This finding suggests that elevated HbA1c levels (> 9.8%) might correlate with advanced insulinopenia and catabolism. To our knowledge, this relationship has not been previously reported in the literature, underscoring the novelty and significance of our results. Studies have revealed a positive correlation between HbA1c levels and the severity of FPD, with average HbA1c levels increasing progressively with mild, moderate, and severe steatosis [16]. In contrast, another study reported no direct correlation between HbA1c levels and FPD in T2DM patients with an average HbA1c level of 9.5% [17].

Furthermore, we found significant positive correlations between IPFD and fasting plasma glucose, fasting insulin, C-peptide levels, and HOMA-IR. In a study by de Oliveira Andrade et al., FPD was significantly associated with elevated fasting plasma glucose, fasting insulin, and HOMA-IR values in 157 patients [18]. Another study of 158 patients reported correlations between pancreatic fat content, C-peptide levels, fasting insulin, and HOMA-IR [19]. These findings suggest that insulin resistance and beta-cell dysfunction may be key contributors to FPD, indicating a potential risk for prediabetic patients as well as those with overt diabetes. The term 'metabolic dysfunction-associated fatty pancreas disease (MAFPD)' could be considered as a more precise way to define this condition, similar to the term 'metabolic dysfunction-associated fatty liver disease (MASLD)' used for hepatic steatosis. The association between FPD and exocrine pancreatic insufficiency was also evident in our study. Fecal elastase levels showed a significant negative correlation with FPD, consistent with previous findings that linked pancreatic fat accumulation to impaired exocrine functions [20]. As fat deposition in the pancreas increases, digestive enzyme secretion decreases.

Additionally, our study highlighted the potential clinical relevance of HbA1c as a predictor of exocrine pancreatic insufficiency. A continuous negative correlation was observed between HbA1c levels and fecal elastase, even when the severity of FPD decreased at HbA1c levels > 9.8%. This trend may be attributable to advanced pancreatic atrophy and fibrosis associated with poor glycemic control, as described in previous studies [14]. Furthermore, a threshold HbA1c level of 7.4% was determined for predicting fecal elastase levels < 200 µg/g in newly diagnosed T2DM patients, contributing novel insights into the early detection of exocrine pancreatic insufficiency. Cavalot et al. reported a higher threshold of 8.4% in type 1 diabetes mellitus patients, highlighting disease-specific differences in HbA1c thresholds [21].

In our study, hepatic steatosis was significantly more prevalent in the T2DM group than in the control group, as anticipated. Insulin resistance and diabetes were identified as key risk factors contributing to hepatic steatosis. These findings are consistent with mechanisms involving dysregulation of glucose and lipid metabolism, elevated free fatty acid levels, increased triglyceride concentrations, and enhanced hepatic de novo lipogenesis. Our study also demonstrated a positive correlation between the FPD and hepatic steatosis, as well as hepatic fat fraction. A systematic review and meta-analysis by Wongtrakul et al. involving 67,803 participants identified hepatic steatosis as a significant risk factor for FPD. Conversely, FPD was found to increase the risk of hepatic steatosis by 1.7-fold, suggesting a bidirectional relationship [22]. Another study reported a strong correlation between the hepatic fat fraction and both hepatic steatosis severity and FPD, supporting our findings [23]. The shared etiological factors, including ectopic fat deposition, oxidative stress, insulin resistance, and inflammation, likely underlie this association. Our findings further revealed that hepatic steatosis and hepatic fat fraction were positively correlated with fasting plasma glucose, fasting insulin, C-peptide levels, and HOMA-IR. Consistent with our data, Cetin et al. demonstrated elevated fasting plasma glucose, fasting insulin, and HOMA-IR levels in hepatic steatosis patients compared with controls [24]. Similarly, Pacifico et al. showed a parallel relationship between hepatic fat fraction and increased insulin resistance, highlighting the significant role of hepatic fat deposition in metabolic dysregulation [19].

Interestingly, we observed a negative correlation between the fecal elastase levels and hepatic fat fraction, with an increased likelihood of fecal elastase levels < 200 µg/g when hepatic fat fraction exceeded 17%. While Mak et al. reported lower fecal elastase levels in individuals with hepatic steatosis, the association lacked statistical significance [25]. However, another study revealed exocrine pancreatic insufficiency in 85% of hepatic steatosis patients, with significantly reduced fecal elastase levels [26]. Our findings suggest that the hepatic fat fraction correlates with exocrine pancreatic dysfunction, likely due to shared mechanisms such as ectopic fat deposition and metabolic inflammation.

The main limitation of our study is the relatively small sample size, which may impact the statistical power of certain findings. Expanding the cohort size could enhance the robustness of our results. Additionally, IPFD was assessed using ultrasound rather than CT or MRI, which, although a widely accepted approach, may have inherent limitations in fat quantification. Another limitation is that lipid parameters were not included in the analysis due to a significant proportion of study participants (n = 11) using statins. Given the well-established effects of statins on lipid metabolism, including lipid levels in the analysis could have introduced a major confounding factor and potentially led to misleading interpretations. However, we acknowledge that lipid parameters could be relevant in relation to FPD, and future studies with a cohort designed to control for lipid-lowering therapy may provide further insight into this relationship. One of the strengths of our study lies in its focus on newly diagnosed, treatment-naive T2DM patients. The comparable age, sex, BMI, and body composition between the patient and control groups minimized potential confounding factors. Moreover, the detailed examination of the relationship between FPD, hepatic steatosis, and hepatic fat fraction, along with the associated metabolic parameters, adds significant value to the existing literature.

The clinical implications of our findings include the potential role of FPD as an underrecognized complication of T2DM, similar to hepatic steatosis. The evaluation of FPD during the clinical assessment of T2DM patients may aid in identifying individuals at risk for exocrine insufficiency. For patients presenting with unexplained dyspeptic symptoms, exocrine pancreatic dysfunction should be investigated using fecal elastase measurement. The observed U-shaped relationship between HbA1c levels and FPD highlights the importance of evaluating patients with moderate HbA1c elevations, particularly those exceeding 7.4%, for potential FPD and associated exocrine dysfunction. Additionally, even in patients with poorly controlled diabetes and HbA1c levels exceeding 9.8%, where FPD may be less evident due to advanced catabolic processes, exocrine pancreatic insufficiency could still be present and should be carefully assessed. This approach could facilitate early detection and intervention, ultimately improving glycemic control and digestive symptom management.

In conclusion, our findings highlight the complex interplay between IPFD, glycemic control, and exocrine pancreatic function in individuals with T2DM. While FPD is increasingly recognized in this population, its clinical significance remains to be fully understood, and further research is needed to clarify its long-term metabolic implications. The observed U-shaped relationship between HbA1c levels and FPD, alongside the associations with fecal elastase and hepatic fat fraction, suggests the importance of considering comprehensive metabolic assessments in T2DM. In contrast, exocrine pancreatic insufficiency may have more immediate clinical relevance, as it could necessitate targeted evaluation and potential therapeutic interventions. Future studies are warranted to further explore these relationships and their clinical impact.