Incidence of pancreatitis, secondary causes, and treatment of patients referred to a specialty lipid clinic with severe hypertriglyceridemia: a retrospective cohort study
© Sandhu et al; licensee BioMed Central Ltd. 2011
Received: 18 July 2011
Accepted: 11 September 2011
Published: 11 September 2011
Severe hypertriglyceridemia (HTG) is one cause of acute pancreatitis, yet the level of plasma triglycerides likely to be responsible for inducing pancreatitis has not been clearly defined.
Methods and Results
A retrospective cohort study was conducted on patients presenting non-acutely to the Healthy Heart Program Lipid Clinic at St. Paul's Hospital with a TG level > 20 mM (1772 mg/dl) between 1986 and 2007. Ninety-five patients with TG > 20 mM at the time of referral were identified, in who follow up data was available for 84. Fifteen patients (15.8%), with a mean outpatient TG level of 38.1 mM, had a history of acute pancreatitis. Among 91 additional patients with less severe HTG, none had a history of pancreatitis when TG were between 10 and 20 mM. Among patients with TG > 20 mM on presentation, 8 (8.5%), with a mean TG level of 67.8 mM, exhibited eruptive xanthomata. A diet high in carbohydrates and fats (79%) and obesity (47.6%) were the two most frequent secondary causes of HTG at initial visit. By 2009, among patients with follow up data 53% exhibited either pre-diabetes or overt Type 2 diabetes mellitus. Upon referral only 23 patients (24%) were receiving a fibrate as either monotherapy or part of combination lipid-lowering therapy. Following initial assessment by a lipid specialist this rose to 84%, and remained at 67% at the last follow up visit.
These results suggest hypertriglyceridemia is unlikely to be the primary cause of acute pancreatitis unless TG levels are > 20 mM, that dysglycemia, a diet high in carbohydrates and fats, and obesity are the main secondary causes of HTG, and that fibrates are frequently overlooked as the drug of first choice for severe HTG.
KeywordsHypertriglyceridemia triglycerides pancreatitis dysglycemia diabetes fibrates
Hypertriglyceridemia (HTG), classically defined as fasting plasma triacylglycerols (triglycerides, TG) > 2.3 mM or 200 mg/dl, or 1.7 mM (150 mg/dl) in the definition of metabolic syndrome , is a common laboratory finding. Severe hypertriglyceridemia, e.g., TG > 20 mM (1772 mg/dl), is much rarer and almost always caused by a combination of inherited and secondary factors [2, 3]. Genetic disorders leading to hypertriglyceridemia include familial combined hyperlipidemia, familial hypertriglyceridemia, remnant removal disease (Type 3 dyslipidemia), deficiencies of lipoprotein lipase or apolipoprotein CII, and more recently characterized mutations including variants of apolipoprotein A5 [4, 5]. The most common secondary contributors to severe hypertriglyceridemia include poorly controlled diabetes mellitus, obesity, high fat and simple carbohydrate diet, excess alcohol consumption, hypothyroidism, and medications including thiazide diuretics, β-blockers, oral estrogen, retinoids, and anti-retroviral agents [2, 3].
Patients with severe HTG may present with classic findings such as abdominal pain or overt pancreatitis, eruptive or palmar xanthomas, lipemia retinalis, or they may be asymptomatic [2, 3]. The most significant complication of severe HTG is acute pancreatitis, which may lead to pancreatic necrosis and death [2, 6]. The incidence of classic signs and symptoms of HTG, including pancreatitis, however, has not been determined in patients presenting with severe HTG. In addition, the level of plasma triglycerides at which acute pancreatitis can be ascribed specifically to the presence of HTG has not been reported.
The purpose of this study was to determine the frequency of physical signs and symptoms of HTG including pancreatitis among patients with severe HTG referred to a specialty lipid disorders clinic over a 21-year period. Specifically, our study attempted to determine: (a) the frequency of classical signs and symptoms associated with severe HTG; (b) the most common secondary factors contributing to TG > 20 mM (1772 mg/dl); (c) differences in treatment for severe HTG between referring physicians and lipid clinic specialists; and (d) changes in the lipid profile of patients with severe HTG followed at a specialty lipid clinic. Overall, we found an absence of pancreatitis unless TG were > 20 mM, a relatively low incidence of classic clinical findings of HTG such as eruptive xanthomas, the presence of diabetes or pre-diabetes in the majority of HTG subjects, and a tendency of non-lipid specialists to overutilize statins and underutilize fibrates as their first-line treatment for severe HTG.
Patient Demographics (Table 1)
Demographic Information for HTG Patients At Initial Clinic Visit
Mean Age (yrs)
54.2 +/- 11.9
History of Pancreatitis
Fifteen patients (15.8%) had a history of pancreatitis prior to referral to the clinic. Of these, the mean TG level at the time of non-acute presentation to the clinic was 38.13 mM [median 30.91 mM (IQ 25.6 - 52.2)], with the lowest referral TG level associated with prior pancreatitis being 20.5 mM (1815 mg/dl). Peak TG levels at the time of acute pancreatitis were not available for this analysis. Analysis of an additional cohort of 91 patients with TG levels between 10 and 20 mM (886 - 1771 mg/dl) at time of presentation to clinic revealed a history of pancreatitis in only 3 patients. In these 3, levels of TG at the time of acute pancreatitis were available, and all were > 20 mM (1771 mg/dl). As such, we conclude that pancreatitis is unlikely to occur as a result of hypertriglyceridemia unless TG are > 20 mM acutely.
Prevalence of dysglycemia (Table 2)
Frequency of Diabetes and Elevated Fasting Blood Sugar (FBS) in HTG Clinic Patients
At Initial Visit:
Past DM Diagnosis
FBS > 7 mM
New DM Diagnosis
New DM Diagnosis
At Latest Visit:
Impaired Fasting Glucose
Total with Dysglycemia
Other risk factors for hypertriglyceridemia and personal/family history of CVD (Table 3)
Frequency of Risk Factors in HTG Patients at Initial Clinic Visit
High fat/sugar diet:
Physical findings (Table 4)
Frequency of Clinical Findings in HTG Patients
Body Mass Index:
> 30 kg/m2
< 25 kg/m2
Pre-Clinic and In-Clinic Treatment (Table 5)
Prevalence of Lipid-lowering Medications in Clinic HTG Patients
Laboratory and Anthropometric Values of Patients from Initial to Latest Visit (Table 6)
Laboratory Values and Anthropometry of HTG Patients from Initial to Latest Visit
Initial Visit (mM)
Latest Visit (mM)
Paired t - test
35.04 ± 21.89
8.07 ± 8.7
p < 0.0001
12.8 ± 6.37
6.72 ± 2.73
p < 0.0001
22.7 ± 25.59
6.70 ± 3.19
p < 0.0001
0.87 ± 1.05
1.05 ± 0.31
p = 0.224
126.91 ± 18.12
128.11 ± 19.96
p = 0.634
80.46 ± 10.05
80.46 ± 10.05
p < 0.05
29.61 ± 5.17
29.55 ± 6.27
p = 0.917
Clinical findings in patients with severe HTG
The incidence of clinical findings including pancreatitis and physical stigmata in patients with severe HTG have not previously been documented. Other studies have reported average acute levels of plasma TG in patients with TG-induced pancreatitis of 51.8 mM  and 50.5 mM , but did not define a lower limit of plasma TG at which HTG could be identified as the likely cause of the pancreatitis. A key finding of this study is that acute pancreatitis as a consequence of HTG occurs relatively infrequently, and rarely if ever unless TG levels are greater than 20 mM (1772 mg/dl). In this study a history of pancreatitis was present in 15.8% of individuals referred with TG > 20 mM, occurring in patients with a minimum non-acute TG level of 20.5 mM (1816 mg/dl) and a mean non-acute TG level of 38.1 mM (3376 mg/dl). Of 91 patients seen in our clinic with non-acute TG levels between 10-20 mM, none had a history of pancreatitis unless their TG were > 20 mM acutely. Eruptive xanthomas were found in only 8.5% of patients, at a minimum TG level of 20.5 mM (1816 mg/dl) and an average TG level of 67.8 mM (6007 mg/dl), even higher than that required to induce pancreatitis. However, the level of TG did not accurately predict whether patients would develop these signs. For instance, pancreatitis was observed in 2 patients with TG levels of 20-21 mM, and only in 1 out of 5 patients who had TG>70 mM. It is likely that factors other than the TG level may contribute to the development of pancreatitis and other physical characteristics observed in HTG patients. Further studies with a greater sample size and possibly knowledge of the underlying genetic traits are needed to better elucidate such relationships. Lipemia retinalis was an infrequent finding in our study. While it was uncertain whether this was assessed at the time of non-acute presentation in all patients, it appears that this finding is present mainly in very severe HTG and more likely to be seen in the acute versus non-acute HTG setting.
Chylomicronemia syndrome, defined as TG > 1000 mg/dl (11.3 mM) plus one of either eruptive xanthomas, lipemia retinalis, or abdominal pain/pancreatitis, had a previously quoted incidence rate of 1.7/10000 patients (< 0.02%) . Recent increases in obesity and DM rates, however, have led to a potential increase in this incidence. In our high risk patient population, 23 patients (24.2%) met the criteria for this syndrome.
Secondary contributors to severe HTG
In addition to inherited causes of HTG that were undoubtedly present in most if not all the patients in this cohort, the most common secondary factors predisposing to HTG were a high fat and carbohydrate diet, physical inactivity and obesity. In 53% of the cohort, and concomitant with these other factors in many cases, dysglycemia was present in the form of pre-diabetes or overt Type 2 diabetes mellitus. Two previous studies found similarly high levels of diabetes, 72%  and 43% , in acute HTG-induced pancreatitis. The negative impact of metabolic syndrome, dysglycemia and abdominal obesity on triglyceride levels is well documented [1, 3, 11]. Insulin resistance and diabetes are associated with an increase in plasma TG for multiple reasons, including reduced insulin-dependent inhibition of lipolysis in adipocytes, increased TG and VLDL production by the liver, and impaired insulin-dependent activation of lipoprotein lipase and hydrolysis of TG-rich lipoproteins [3, 12].
Although known to contribute to HTG, the incidence of uncontrolled DM and hypothyroidism has not been well characterized in previous studies. The majority of patients (77%) with DM at first presentation to the clinic had poor glycemic control. Diagnosed hypothyroidism was present in 17.1% of patients, with a significant number of these (33.3%) having evidence of inadequate thyroid hormone replacement (TSH > 5 uIU/mL). The high incidence of such comorbidities highlights the importance of controlling these factors to prevent severe HTG. All told, 100% of our patients had at least one secondary factor contributing to their HTG. Although genetic testing is not currently available for familial combined hyperlipidemia and is not routinely available for many other inherited causes of HTG, our assumption is that in most if not all of our patients a combination of primary genetic plus secondary causes contributed to their severe HTG. In any patient with TG levels >3-4 mM, it should be assumed there is a likely underlying inherited cause of HTG present, aggravated further by one or more secondary factors.
Lipid Lowering Therapy
At initial visit, almost as many patients were being treated by their referring physician with statin monotherapy as fibrate monotherapy, the recommended first line treatment for severe HTG. Lipid specialists, on the other hand, treated most patients with fibrate monotherapy. It has been well documented that fibrates are the most efficacious first line pharmacotherapy for HTG, and typically lower TG very rapidly and effectively [3, 11, 13]. Adjuncts to this include lifestyle modifications, such as following a low fat and low simple carbohydrate diet, avoiding alcohol, control of blood sugars, and fish oil Ω-3 fatty acid) supplementation [3, 14]. Fish oil doses of at least 2 g in doses divided two or three times daily are needed to have any major TG-lowering effect, and doses of up to 12 g daily can be used. Niacin (nicotinic acid) is also a potent TG-lowering agent. It is not practical to use in the acute setting, however, where reducing the risk of pancreatitis is the main priority, given the time needed to titrate the niacin dose to therapeutic levels (≥ 1000 mg daily). In most cases, apart from homozygous lipoprotein lipase or apolipoprotein CII deficiency, use of a fibrate in combination with a low fat and carbohydrate diet and improvement in blood sugar control is highly efficacious in reducing severe HTG to more moderate levels, as seen in our cohort. While statins have some TG-lowering effect, ranging from 20 - 28% at higher doses of newer statins , they are not effective enough in this regard to remove the risk of pancreatitis in patients with severe HTG, and should not be used on their own as first line agents. The discrepancy between treatments initiated by referring physicians and lipid specialists needs to be addressed, perhaps by an education series focused on lipid treatment. By doing so, it is possible that more patients would be able to achieve marked TG reduction without lipid clinic referrals.
In patients with severe HTG and persistent elevation of apolipoprotein B100 or LDL-C following initiation of TG-lowering therapy, treatment with combined fibrate/statin may be indicated. Although the recent ACCORD Lipid trial did not find additional benefit of adding fenofibrate to a statin in terms of cardiovascular event rate in patients with type 2 diabetes , it may be necessary to continue fibrate therapy long term in severe HTG patients who also require a statin, in order to maintain TG levels out of the range of risk for pancreatitis. While we did not observe pancreatitis in anyone with TG levels < 20 mM, due to the volatile nature of TG elevation when secondary factors are not well controlled, we recommend attempting to lower TG levels to not more than a maximum of 6-8 mM in order to reduce the risk of TG rising into pancreatitis range acutely.
This is the first cohort in which incidence rates of clinical findings in patients with severe HTG have been quantified. The classic findings of pancreatitis and eruptive xanthomas occurred in a relatively small percentage of patients, and not unless TG were > 20 mM (1772 mg/dl). Patients with extreme HTG have a combination of primary and secondary factors contributing to their HTG. Lifestyle changes (low fat and simple carbohydrate diet, increased exercise) and a reduction in comorbidities (uncontrolled DM and hypothyroidism) are critical aspects of managing HTG in conjunction with pharmacotherapy using fibrates and omega-3 fatty acid supplements. Fibrates remain the treatment of choice for severe HTG, with statins lacking sufficient TG-lowering effect to remove the risk of pancreatitis.
A retrospective chart review was conducted in the Healthy Heart Program Lipid (now Prevention) Clinic at St. Paul's Hospital, Vancouver, Canada. Patients seen in the clinic between the years of 1986 and 2007 with plasma triglyceride levels greater than 20 mM upon referral were included in the study. Patients were identified either by individual chart review, or after 1999 by scan of an electronic clinic database. In total, 95 patients met this inclusion criterion; of those, 84 patients returned for at least one follow up visit, and many patients have been followed intermittently for several years. Data extracted from the chart from the patient's initial visit and most recent follow up visit for the current study included: demographics (age, gender, ethnic background); history of pancreatitis; presence of increased fasting glucose or diagnosis of diabetes at initial or follow up visit; other historical features including secondary factors potentially contributing to the presence of HTG (dietary carbohydrates and fats, level of physical activity, alcohol consumption, hypothyroidism, current smoking, and medications potentially raising TG including thiazide diuretics, β-blockers, oral estrogen, and anti-retroviral agents), history of cardiovascular disease, and family history of premature vascular disease or known dyslipidemia; physical findings including body mass index, dermatologic findings including eruptive or palmar xanthomas, eye findings including corneal arcus and lipemia retinalis, and abdominal findings including epigastric tenderness and hepatomegaly. Use of lipid-lowering therapies including fibrates (gemfibrozil or fenofibrate), statins (simvastatin, pravastatin, atorvastatin or rosuvastatin), niacin, omega-3 fatty acids (fish oils), and bile acid binding resins were recorded. Total cholesterol, triglyceride levels, HDL-cholesterol (HDL-C) and total cholesterol:HDL-C ratio were recorded for the initial and follow up visits. This study was approved by the Institutional Ethics Review Board for Human Studies, Providence Health Care Research Institute, Vancouver, BC, Canada.
The cohort's characteristics were expressed as mean and standard deviation for continuous variables. Frequency of findings was used for dichotomous traits. Means were compared using a paired Students t test, when normality was confirmed, two tailed - 95% CI with a p value < 0.05 considered statistically significant. Data analysis was performed using SPSS v12.0 (SPSS Inc., Chicago, IL).
This study was funded by a Faculty of Medicine Summer Studentship Award to SS, and by CIHR operating grant MOP-12660 to GAF.
- Alberti KG, Zimmet P, Shaw J: The metabolic syndrome--a new worldwide definition. Lancet. 2005, 366: 1059-1062. 10.1016/S0140-6736(05)67402-8View ArticlePubMedGoogle Scholar
- Brunzell JD, Deeb SS: Familial lipoprotein lipase deficiency, apo CII deficiency and hepatic lipase deficiency. The Metabolic and Molecular Bases of Inherited Disease. Edited by: Scriver CR, Beaudet AL, Sly WS, Valle D. 2001, 2789-2816. New York: McGraw-Hill, 8.Google Scholar
- Yuan G, Al-Shali KZ, Hegele RA: Hypertriglyceridemia: its etiology, effects and treatment. CMAJ. 2007, 176: 1113-1120. 10.1503/cmaj.060963PubMed CentralView ArticlePubMedGoogle Scholar
- Wang J, Ban MR, Kennedy BA, Anand S, Yusuf S, Huff MW, Pollex RL, Hegele RA: APOA5 genetic variants are markers for classic hyperlipoproteinemia phenotypes and hypertriglyceridemia. Nat Clin Pract Cardiovasc Med. 2008, 5: 730-737. 10.1038/ncpcardio1326View ArticlePubMedGoogle Scholar
- Johansen CT, Wang J, Lanktree MB, Cao H, McIntyre AD, Ban MR, Martins RA, Kennedy BA, Hassell RG, Visser ME: Excess of rare variants in genes identified by genome-wide association study of hypertriglyceridemia. Nat Genet. 2010, 42: 684-687. 10.1038/ng.628PubMed CentralView ArticlePubMedGoogle Scholar
- Truninger K, Schmid PA, Hoffmann MM, Bertschinger P, Ammann RW: Recurrent acute and chronic pancreatitis in two brothers with familial chylomicronemia syndrome. Pancreas. 2006, 32: 215-219. 10.1097/01.mpa.0000202942.93578.ddView ArticlePubMedGoogle Scholar
- Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA. 2001, 285: 2486-2497.
- Fortson MR, Freedman SN, Webster PD: Clinical assessment of hyperlipidemic pancreatitis. Am J Gastroenterol. 1995, 90: 2134-2139.PubMedGoogle Scholar
- Lloret Linares C, Pelletier AL, Czernichow S, Vergnaud AC, Bonnefont-Rousselot D, Levy P, Ruszniewski P, Bruckert E: Acute pancreatitis in a cohort of 129 patients referred for severe hypertriglyceridemia. Pancreas. 2008, 37: 13-12. 10.1097/MPA.0b013e31816074a1View ArticlePubMedGoogle Scholar
- Leaf DA: Chylomicronemia and the chylomicronemia syndrome: a practical approach to management. Am J Med. 2008, 121. 10.12.View ArticlePubMedGoogle Scholar
- McPherson R, Frohlich J, Fodor G, Genest J, Canadian Cardiovascular S: Canadian Cardiovascular Society position statement--recommendations for the diagnosis and treatment of dyslipidemia and prevention of cardiovascular disease. Can J Cardiol. 2006, 22: 913-927. 10.1016/S0828-282X(06)70310-5PubMed CentralView ArticlePubMedGoogle Scholar
- Verges BL: Dyslipidaemia in diabetes mellitus. Review of the main lipoprotein abnormalities and their consequences on the development of atherogenesis. Diabetes Metab. 1999, 25 (Suppl 3): 32-40.PubMedGoogle Scholar
- Barter PJ, Rye KA: Cardioprotective properties of fibrates: which fibrate, which patients, what mechanism?. Circulation. 2006, 113: 1553-1555. 10.1161/CIRCULATIONAHA.105.620450View ArticlePubMedGoogle Scholar
- Oh R: Practical applications of fish oil (Omega-3 fatty acids) in primary care. J Am Board Fam Pract. 2005, 18: 28-36. 10.3122/jabfm.18.1.28View ArticlePubMedGoogle Scholar
- Jones PH, Davidson MH, Stein EA, Bays HE, McKenney JM, Miller E, Cain VA, Blasetto JW: Comparison of the efficacy and safety of rosuvastatin versus atorvastatin, simvastatin, and pravastatin across doses (STELLAR* Trial). Am J Cardiol. 2003, 92: 152-160.View ArticlePubMedGoogle Scholar
- Ginsberg HN, Elam MB, Lovato LC, Crouse JR, Leiter LA, Linz P, Friedewald WT, Buse JB, Gerstein HC, Probstfield J: Effects of combination lipid therapy in type 2 diabetes mellitus. N Engl J Med. 2010, 362: 1563-1574.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.