Association between non-alcoholic fatty pancreatic disease (nafpd) and the metabolic syndrome: case–control retrospective study

Five-hundred-fifty-seven healthy and consecutive subjects not known to have hypertension or diabetes and who had a health investigation at the National Taiwan University Hospital Health Management Center were analyzed in this retrospective study during January and March, 2010. The retrospective and cross-sectional analysis study was approved by the Institutional Review Board of National Taiwan University Hospital, and this study was carried out without any linkage of personal information for all participants. All subjects received abdominal ultrasonography by gastro-hepatology specialists blinded to all information concerning these subjects during health investigation. All patients completed ultrasonographic investigation by a convex 3.5-MHz transducer (HPM2410A, Hewlett Packard, Andover, MA, USA). Usually, non-alcoholic fatty liver disease (NAFLD) diagnostic criteria included characteristic echo patterns of hepatorenal contrast, bright liver, deep attenuation, and vascular blurring. Fatty pancreas was diagnosed when there was an increase in echogenicity of the pancreatic body over that of the kidney. As the pancreas could not be compared simultaneously with the kidney in the same sonographic window, the investigators compared the difference between hepatic and renal echogenicity, and between hepatic and pancreatic echogenicity to obtain an objective pancreato-renal echo contrast [10]. Subjects were divided into the fatty pancreas group (n = 72) and the normal pancreas group (n = 485) according to the ultrasonographic diagnosis.

The height, weight, abdominal girth (AG), pulse rate, systolic blood pressure (SBP), and diastolic blood pressure (DBP) of all subjects were measured routinely in the health check-up program. AG was measured in the midsection between the lowest rib and the iliac crest of the pelvis and horizontal to the ground with the subject in an upright position. BMI and AG/height were determined for all subjects. Blood samples were tested routinely after an 8 h fast for complete blood cell count (CBC), liver function, amylase concentration, fasting blood glucose (FBG) concentration, HbA1c, total cholesterol, triglyceride (TG) concentration, high density lipoprotein cholesterol (HDL-C), and low density lipoprotein cholesterol (LDL-C). Two h postprandial blood glucose (PBG) measurements were also performed routinely.

The criteria for diagnosis of metabolic syndrome were those of the National Cholesterol Education Program Adult Treatment Panel III (ATP III). Metabolic syndrome was diagnosed when three or more of the following were present: abdominal obesity (waist circumference ≥ 90 cm for males or ≥ 80 cm for females), increased TG concentration (≥150 mg/dL), decreased HDL-C (< 40 mg/dL for males or < 50 mg/dL for females), hypertension (systolic blood pressure, SBP ≥ 130/diastolic blood pressure, DBP ≥ 85 mm Hg), and impaired FBG concentration (≥110 mg/dL).

Descriptive data are presented as means ± SD for continuous variables. Differences between the fatty pancreas and normal pancreas group were determined using the Student’s t test. The χ² test was used to evaluate the relationship between fatty pancreas and metabolic syndrome. A p value < 0.05 was considered statistically significant. All statistical analyses were performed with SPSS 12.0.

The definition criteria for metabolic syndrome are thought not ideal, and controversies continue over the validity of its predictive value for cardiovascular events and diabetes due to several factors as no inclusion of age, gender, LDL-C, family history or past history of cardiovascular disease and diabetes in the definition criteria. However, Tenenbaum et al. showed that current available evidences strongly supported that metabolic syndrome is an important clustering of cardiovascular risk factors and diabetes [19].

In the present study, fatty pancreas was found to be strongly associated with specific parameters of metabolic syndrome. For example, an association with obesity as manifested by increased BMI and AG was observed. This finding is compatible with those of an autopsy report [18] and with those of several human fatty pancreas studies involving the use of ultrasonography [13], EUS [10, 11, 17], CT [20], and MRI [12, 14, 15]. In animal study, Mathur et al. also documented that obese mice have heavier pancreas and more pancreatic fat, especially triglycerides, and concluded that obesity leads to fat infiltration of the pancreas [21]. In the present study, fatty pancreas was also found to be associated with systolic, but not diastolic, hypertension; this finding is in agreement with that of Choi et al.[17]. Additionally, fatty pancreas was found to be associated with increased FBG, PBG, and HbA1c values. In this regard it should be noted that findings of previous studies of the association between fatty pancreas and hyperglycemia are inconsistent. No association between fatty pancreas and DM and/or FBG was found in two studies [10, 13] whereas in others fatty pancreas was found to correlate inversely with insulin secretion and β–cell function and directly with impaired glycemia and DM [12, 14, 16]. In the present study, fatty pancreas was also found to be strongly associated with hyperlipidemia as manifested by increased total cholesterol, TG, and LDL-C values; however, no association of HDL-C with fatty pancreas was observed. These findings differ somewhat from those of Lee et al.[13] who reported that fatty pancreas is associated with increased total cholesterol, TG, HDL, and free fatty acid concentrations but not with increased LDL-C concentrations. Sepe et al.[10] reported that fatty pancreas was associated with hyperlipidemia but lipid profiles were not provided in this report. In the present study, fatty pancreas was found to be strongly associated with the frequency of metabolic syndrome and to correlate with the number of the parameters of metabolic syndrome. These findings are consistent with those of previous studies [10, 13].

Although central obesity had been defined as essential criteria of metabolic syndrome, Lee, et al. reported that the incidence of diabetes in subjects without central obesity was similar to that in subjects with central obesity in a Chinese hypertensive family cohort study. This indicated central obesity is not essential component of metabolic syndrome for predict diabetes [22]. In present study, fatty pancreas was found to be associated with increased FBG, PBG, and HbA1c values, and its predictive value of diabetes need to be further evaluated.

The fatty pancreas and normal pancreas groups did not differ with respect to liver function. No differences between the two groups were observed regarding albumin and globulin concentrations and AST, ALT, and γGT values. This finding is somewhat surprising in that fatty liver has been reported to be associated with fatty pancreas, and elevation of liver enzymes, especially of ALT, is often present in subjects with fatty liver. Findings of EUS studies showed an association of fatty pancreas with hepatic steatosis [10, 11, 17], and findings of one MRI study revealed a correlation of pancreatic fat content with liver fat content [15]. Furthermore, findings of a trans-abdominal ultrasonographic study showed a correlation between fatty pancreas and liver enzymes including AST, ALT, and γGT [13]. The findings of one autopsy study demonstrated that total pancreatic fat is significantly correlated with NAFLD, and fatty liver and fatty pancreas are related more significant in women [23]. In contrast, no association between pancreatic fat and liver fat was found in other MRI studies [12, 16]. The ultrasonographic study also showed that about 68% of cases with fatty pancreas concurrently had fatty liver, but most subjects (97%) with fatty liver had fatty pancreas. The positive predictive value of fatty liver in fatty pancreas was around 70%, but the negative predictive value of fatty liver in normal pancreas was high to 96%. These findings suggested that fatty pancreas could be an initial indicator of ectopic fat deposition and an earlier manifestation of metabolic syndrome than fatty liver [13].

No statistically differences in markers of pancreatic carcinoma, including CEA and CA 19–9, were observed between the fatty pancreas and control groups. These findings are consistent with those of a previous study [10] in which fatty pancreas was not found to be associated with pancreatic carcinoma.

Two additional findings of the present report are of particular interest. The first is that serum amylase values were significantly lower for the fatty pancreas as compared to normal pancreas group. This finding contrasts with that of Sepe et al.[10] who found no association between fatty pancreas and serum amylase or lipase concentrations. It is known that elevated serum amylase levels often accompany acute pancreatitis and are occasionally caused by other conditions such as pancreatic tumors. However, few clinical studies investigating clinical implications of low serum amylase. Low serum amylase was thought due to diffuse pancreas destruction secondary to advanced pancreatitis [24, 25]. Low serum amylase was also associated with insulin resistance in obese animal models [26, 27]. Nakajima et al. showed that low serum amylase is associated with increased risk of metabolic abnormalities, metabolic syndrome, and diabetes, and provided a pancreatic exocrine-endocrine relationship in certain clinical condition [28]. Lower serum amylase values in subjects with fatty pancreas possibly reflected diffuse destruction of pancreas due to fat infiltration. The second is that the platelet count was significantly higher for the fatty pancreas as compared to normal pancreas group. To our knowledge, the present study is the first to examine this relationship. Further exploration of the associations of fatty pancreas with serum amylase and platelet count is planned.

Table 4 summarizes prior findings regarding factors associated with fatty pancreas. Based on a number of currently conflicting findings, further studies of the associations of fatty pancreas with age, hyperglycemia, hyperlipidemia, fatty liver, and serum amylase are warranted. Some limitations of this study need to be noted. Owing to its cross-sectional case–control retrospective fashion, we only demonstrated the relation between fatty pancreas and metabolic syndrome, and the long-term effect of presence of fatty pancreas to metabolic syndrome was uncertain due to lack of longitudinal data. There was no fasting insulin measurement in the routine health investigation, and insulin resistance was unable to calculate.