Hyperlipidemia and Pancreatitis during Pregnancy in Two Sisters with a Mutation in the Lipoprotein Lipase Gene

Case Report

Patient 1 was a 29-year-old gravida III para II at 24 weeks of gestation who was admitted to the Maine Medical Center with abdominal pain, vomiting, and a temperature of 38 °C.

Her medical history included two previous episodes of pancreatitis requiring hospitalization and a family history of elevated cholesterol levels. The first episode of pancreatitis occurred at 17 years of age. The patient's only medication was birth control pills (name of preparation unclear), which were started at 15 years of age. The patient also noted a 13.6-kg weight gain from 15 to 17 years of age. The second episode of pancreatitis occurred at 22 years of age with no clear cause, although birth control therapy was continued from 15 to 25 years of age. There was no history of alcohol abuse. The patient had not sought medical attention during two previous pregnancies, despite persistent nausea and abdominal pain. Both pregnancies resulted in healthy infants. Coincidentally, at the time of the patient's admission, an older sister (patient 2) was also pregnant and having similar episodes of abdominal pain.

Notable physical findings at the time of admission included a temperature of 38 °C, a respiratory rate of 25, and a normal gravid uterus. Neither splenomegaly, lipemia retinalis, nor eruptive xanthomas was present. Abdominal ultrasonography showed dilated pancreatic ducts and no stones. A radiograph showed a small left pleural effusion. An abdominal computed tomographic scan showed findings consistent with hemorrhagic pancreatitis. Laboratory tests with abnormal results included those for triglyceride level, 53.51 mmol/dL; total cholesterol level, 25.86 mmol/dL; high-density lipoprotein (HDL) cholesterol level, 11.4 mmol/dL; amylase level, 880 IU/L (normal, 25 to 115 IU/L); lipase level, 156 IU/L (normal, 23 to 208 IU/L); leukocyte count, 18 000 cells/mm³; calcium concentration, 1.97 mmol/dL (normal, 2.2 to 2.6 mmol/dL); albumin level, 3.2 g/dL (normal, 3.5 to 5.0 g/dL); and fasting blood sugar level, 7.21 mmol/dL.

The patient's clinical course was uneventful after treatment with intravenous fluids and a 20-g fat diet. She was discharged after 14 days.

Two months later, patient 2 was transferred from another hospital to Maine Medical Center with a diagnosis of hemorrhagic pancreatitis and the acute respiratory distress syndrome.

Three weeks before her transfer, patient 2 had delivered a full-term, 3.64-kg infant. At the time of delivery, the placental tissue was described as having a “strawberry milkshake” appearance. The patient's triglyceride level at delivery was 25.40 mmol/dL, and her total cholesterol level was 15.02 mmol/dL. Her physician prescribed cholestyramine resin, and she was discharged from the hospital. Three weeks later, she was readmitted to the same hospital with acute pancreatitis. Her lipid profile showed a triglyceride level of 54.47 mmol/dL and a total cholesterol level of 19.91 mmol/dL. She remained at the community hospital for 3 days, receiving intravenous fluids and antibiotics until she became obtunded, at which point she was transferred to the Maine Medical Center. On transfer, the patient was found to have the acute respiratory distress syndrome with respiratory failure, sepsis, perforation of the small bowel, and rupture of the splenic artery. The patient was repaired surgically, and necrotic tissue and phlegmon were removed operatively on six occasions. She was treated medically with antibiotics and total parental nutrition, restricted to less than 10 g of fat. She was eventually discharged after a 119-day hospitalization.

On discharge from the hospital, both patients began receiving 25-g fat-restricted diets and gemfibrozil, 600 mg twice a day.

One year after discharge, both patients were clinically stable with normal functional status. A nutritionist confirmed dietary compliance by using a 3-day food record review. Patients 1 and 2 had lost 9 kg and 10 kg of body weight, respectively, compared with their prepartum body weights. Patient 1 weighed 62 kg, and patient 2 weighed 56 kg. The lipid profile of patient 1 was the following: total cholesterol level, 4.99 mmol/dL; triglyceride level, 2.57 mmol/dL; HDL cholesterol level, 0.65 mmol/dL; and low-density lipoprotein (LDL) cholesterol level, 3.15 mmol/dL. The lipid profile of patient 2 was the following: total cholesterol level, 4.08 mmol/dL; triglyceride level, 3.25 mmol/dL; HDL cholesterol level, 0.65 mmol/dL; and LDL cholesterol level, 1.96 mmol/dL.

In both patients, DNA sequencing [10] showed a G→A substitution at base pair 818, modifying the codon GGG to GAG. This results in a glutamic acid residue substituted for glycine at residue 188 (gly188→glu) of LPL (Figure 1). Both patients were heterozygous for this defect and were apoprotein E genotype 3/3 [11].

View larger version:

• In this window
• In a new window

Figure 1. Lanes are G, A, T, C, from the left in all samples. Samples are normal (N) sequence, patient 1, and patient 2. Heterozygous patients have bands in both G and A lanes, showing the presence of both alleles. Base pair 818, the second nucleotide of the residue 188 codon, is indicated with an arrow. Double-stranded DNA cycle sequencing of exon 5 of lipoprotein lipase.

Discussion

These cases show a possible cause of the pancreatitis of pregnancy—a disorder of triglyceride clearance related to defective LPL. Lipoprotein lipase is a key lipolytic enzyme that may be down- and up-regulated by various influences, including insulin, estrogen, and medications [12, 13]. More than 20 mutations in the LPL gene have been previously described [14]. The LPL gene appears on chromosome 8, spans 35 kb, has 10 exons, and codes for 474 amino acids in a functional enzyme. The catalytic domain is centrally located among exons 4, 5, and 6, a highly conserved region. The LPL amino acid residue 188 mutation is located in exon 5 [15].

Both patients showed a defect at amino acid residue 188 of LPL Figure 1 and are heterozygous for the mutation. Patient 1 had a more classic variety during the last trimester and subsequently responded well to fat restriction and conservative management. Patient 2 had a more severe and protracted course, possibly because of the use of cholestyramine resin, a drug that may have elevated rather than decreased her triglyceride levels [16, 17]. Patient 2 had multiple medical and surgical complications related to the severity of her pancreatitis. Patient 1 had mild glucose intolerance that may have exacerbated the triglyceride level. Years before developing pregnancy-induced pancreatitis, patient 1 may have had hyperlipidemia and pancreatitis as a result of the use of birth control pills. As early as 1972, clinicians were alerted to cases of marked hyperlipidemia and pancreatitis associated with the use of birth control pills, although no genetic testing was available at that time [18]. Apoprotein E (apo E), particularly apo E allele 2, has been found in association with higher triglyceride levels in pregnant patients with cholymicronemia [19]. We did not find apo E alleles 2 or 4 in our patients; both had apo E genotype 3/3. Therefore, we concluded that apo E was not a primary cause of the hypertriglyceridemia in these patients.

Although it is known that pregnancy results in an increase in plasma triglyceride levels [4], how this increase occurs is not totally understood. In addition, certain persons express profound elevations in triglyceride values during pregnancy [6, 7]. No physiologic explanation for this unique population of pregnant patients with hypertriglyceridemia has been found. We found that the profound increase in triglyceride level in the presence of a lipolytic defect does not resolve whether triglyceride synthesis or LPL-induced catabolism is primarily affected. However, given the role of LPL in the triglyceride-rich particle catabolism, the finding in these sisters indicates a primary cause of a defect in LPL. In late gestation in animals, LPL activity decreases in adipose tissue but increases in placental and mammary tissues [3]. In obese women, LPL levels are inversely related to plasma estradiol levels [13], whereas in normal postmenopausal women, oral conjugated estrogen therapy elevates triglyceride levels by as much as 20% [20]. In a study involving 28 pregnant and 32 nonpregnant women in the United Kingdom and Brazil, triglyceride levels approximately doubled during the third trimester of pregnancy [4]. These data suggest that the effect of estrogen on LPL function is at least partially responsible for the increase in triglyceride levels in the third trimester—particularly in the defective state in our two patients. Ma and colleagues [9] recently reported an LPL mutation (Ser172→Cys) responsible for a 10-fold increase in plasma triglyceride levels in an East Indian pregnant woman with pancreatitis. The patient was homozygous for this mutation. We propose that the mutation found in our patients (gly188→glu), in the heterozygous form, results in substantial alteration to the integrity of the LPL dimer, which in turn causes significant reduction in catalytic function. This, together with elevated estrogen levels, could produce a significant increase in triglyceride levels.

Mutation at codon 188 appears throughout the world, perhaps most often in the French Canadian population. Both patients in our study are of French Canadian and Native American ancestry. In Quebec, the carrier rate has been estimated to be as high as 1 in 169, with a total of 19 600 persons affected [15, 21]. We believe that the rarity of the clinical syndrome of hyperlipidemia and pancreatitis in pregnancy in patients with this specific genetic defect may be due to the presence of an undetected additional mutation in these patients.

The hyperlipidemia of pregnancy has previously been treated by dietary fat restriction [5-7]. Both of our patients were placed on a 25-g, fat-restricted diet, with attempts to achieve weights close to ideal body weight.

In summary, clinicians should be alert to pregnant patients with a history of pancreatitis, disorders of triglyceride metabolism, or a family history of hyperlipidemia. Patients of French Canadian ancestry may be at increased risk. Lipid testing for elevated triglyceride levels may avert pancreatitis and complications of dietary fat restriction. Screening for genetic disorders of LPL may prove beneficial as DNA screening tests become available.

Dr. Sprecher: Cleveland Clinic, 9500 Euclid Avenue, Desk P-19, Cleveland, OH 44195.

Dr. Renfrew: Reddington Medical Associates, 46 South Factory Street, Skowhegan, ME 04976.