In this study, patients with idiopathic pancreatitis had elevated TG levels at admission. Furthermore, patients with idiopathic pancreatitis had a higher rate of recurrent acute pancreatitis. To our knowledge, this is the first study to clarify TG levels in the acute phase of non-idiopathic pancreatitis, suggesting the need for early therapeutic intervention in patients with high TG levels.
In Japan, a survey conducted by the Research and Study Group of the Ministry of Health, Labour and Welfare (MHLW) in 2011 reported that the incidence rate among 63,080 patients who received treatment for acute pancreatitis in one year was 49/100,000 persons/year, with a male-to-female ratio of 1.9:1 and a mean age of 65.3±19.6 years [22]. Similarly, other epidemiological studies have shown that the incidence was higher in men and that the number of patients was higher in the 60–70-year age group [23]. Although some types of pancreatitis can be treated on an outpatient basis depending on severity, this is a disease that requires inpatient care, as per the treatment guidelines of Japan [17] and the USA [24]. In this study, only patients with pancreatitis that required inpatient care were eligible. Compared to the MHLW survey, this study may have included patients with more severe pancreatitis; however, no significant differences in the age of onset or sex were observed.
In this study, gallstone pancreatitis accounted for 26.6% of all cohorts and 88.8% of the non-idiopathic pancreatitis cohort. This may have resulted from a lower incidence rate of cholelithiasis in Japanese patients [25] compared with that in Caucasian patients [26]. Alcoholic pancreatitis accounted for 16.1% of all cohorts and 23.0% of the idiopathic cohort, which is similar to the findings of UK studies [1]. More than 50% of all patients with pancreatitis had an unknown etiology in this study. The reason for this was unclear. The idiopathic-to-non-idiopathic pancreatitis ratio was 3:7 in this study, which is similar to that found in UK studies (20–34%) [6].
There was a difference in age distribution between patients with idiopathic and those with non-idiopathic pancreatitis, which was thought to be influenced by the underlying disease. Idiopathic pancreatitis, as represented by alcoholic pancreatitis, was thought to be caused by metabolic abnormalities resulting from alcohol consumption, dietary habits, and accumulated damage to the liver and pancreas. Therefore, it is likely that pancreatitis develops at a relatively young age due to binge drinking and overeating in the late teenage years or later. In contrast, gallstone pancreatitis accounts for approximately 90% of non-idiopathic pancreatitis. As the age of onset of cholelithiasis is 60–70 years [27], the distribution may be due to the occurrence of this disease.
The average TG value within 3 days (days 0–2) after admission due to acute pancreatitis was 293.1±816.6 mg/dL for idiopathic pancreatitis. The incidence of severe HTG exceeding 500 mg/dL was less than 10%; however, approximately 30% of patients had a value that fit the definition of HTG of more than 150 mg/dL. The prevalence of dyslipidemia, which included patients with familial hyperlipidemia, was 21.6% at baseline. This value is similar to that reported previously (15–20% of all subjects referred to Lipid Clinics) [9]. Whether HTG directly triggers acute pancreatitis is unknown due to the inability to confirm a causal relationship in the DPC system and claims data and the absence of an equivalent injury or disease code for “HTG-induced acute pancreatitis.” In addition, as excessive alcohol consumption can cause HTG and affect the liver and pancreas, some patients who have been classified as having alcoholic pancreatitis may have had HTG as well. Nevertheless, the present results show that TG levels were higher in patients with idiopathic pancreatitis than in those with non-idiopathic pancreatitis, suggesting that HTG may play some role in the development of idiopathic pancreatitis.
In terms of drug therapy for pancreatitis during hospitalization, statins were used in 8.0% and 13.0% of patients with idiopathic and non-idiopathic pancreatitis, respectively. Moreover, fibrates, pemafibrate (known as a selective peroxisome proliferator-activated receptor α modulator), and EPA were infrequently used in patients with idiopathic pancreatitis. Since TG levels can be reduced rather effectively by fasting, it is possible that aggressive pharmacotherapy against HTG is not conducted during hospitalization. This may be influenced by the fact that the acute phase of pancreatitis, just after the onset, is devoted to acute treatment, and that treatment strategies for lowering TG levels, other than fasting, are not always available. A more detailed study of the risk of developing acute pancreatitis due to dyslipidemia, especially HTG, is warranted.
The recurrence rate of acute pancreatitis is known to vary depending on factors such as the cause and treatment [28]. In this study, there were no differences in the recurrence rates between the idiopathic and non-idiopathic pancreatitis cohorts up to 180 days; however, the differences between the two cohorts increased thereafter, showing significantly different prognoses. This is because the cause of non-idiopathic pancreatitis, such as gallstones, is often clearly known; the cause can be treated; or the disease can be managed continuously, such hereditary pancreatitis. In contrast, idiopathic pancreatitis is difficult to chronically manage to prevent recurrence because the cause of the disease is unclear and the patients’ lifestyle habits, such as excessive alcohol and fat intake, are involved. Although lifestyle intervention leads to a temporary improvement in diet, patients' lifestyles may revert over time. In addition, while statins are very effective in lowering LDL-C levels, they may not always effectively lower TG levels.
The incidences of CAD and stroke did not differ between non-idiopathic and idiopathic pancreatitis. As HTG is a risk factor for acute pancreatitis and atherosclerosis [2,3,29], treatment for patients with HTG should focus on preventing acute pancreatitis and CV events. Particularly, for patients with severe HTG, even if acute pancreatitis can be prevented using dietary restriction and drug therapy, the risk of atherosclerosis may not be adequately controlled because the threshold for CV events and acute pancreatitis may differ from mild-to-moderate HTG. If chronic, atherosclerosis and atherosclerotic CV events may be induced [30,31]. Although using statins to lower LDL-C levels is a standard therapy for preventing CV events in current clinical practice, there is a residual risk for atherosclerotic CV events that remains even when LDL-C levels are sufficiently lowered, and one of the causes underlying the risk is HTG [32,33]. TG-rich lipoproteins, such as intermediate density lipoproteins, very low-density lipoproteins, and remnants may promote atherosclerosis development by a different mechanism from that of LDL-C [34,35]. It has been suggested that lowering atherogenic TG-rich particles other than LDL-C may be important for preventing CV events in patients with HTG.
In this study, approximately 30% of patients with idiopathic pancreatitis had TG levels of 150 mg/dL or higher and the incidence rate of CAD or stroke was approximately 3%, which was significantly lower than that of recurrent pancreatitis. Further investigation on the incidence of CV events in patients with higher serum TG values among those with idiopathic pancreatitis is required in the future.
Limitations
First, few patients registered in the MDV database have available laboratory values, such as TG, and in the patients with available TG values, it is unknown whether the TG levels are measured immediately after the onset of acute pancreatitis. Because TG levels can easily fluctuate under the influence of diet and other factors and can decrease significantly after several hours of fasting, it was not possible to accurately determine TG levels at the onset of acute pancreatitis in this study. TG levels have been shown to be high in patients with idiopathic pancreatitis; however, we could not analyze the threshold level at which the risk of developing acute pancreatitis increased.
Second, urgent cases brought to the emergency department may have been excluded from the analysis because of insufficient available patient background information, and severe cases may also have been excluded.
Third, in this study, idiopathic pancreatitis included patients whose data included alcoholic pancreatitis as a disease name. Alcoholic pancreatitis should be excluded from idiopathic pancreatitis if alcohol is a clear cause; however, it is difficult to determine whether alcohol is the direct cause in many cases. Therefore, alcoholic pancreatitis was included in the idiopathic pancreatitis category in this study to account for this; however, it is unclear how this may have affected the results.
Last, the time from the onset of acute pancreatitis to hospitalization may vary among patients, and the exact time could not be obtained. This may have affected the study results, particularly the laboratory test values.