A Common Mutation in Lipoprotein Lipase Confers a 2-Fold Increase in Risk of Ischemic Cerebrovascular Disease in Women but Not in Men
Background—We previously showed that the common Asn291Ser substitution in lipoprotein lipase is associated with elevated plasma triglyceride levels and a 2-fold increase in the risk of ischemic heart disease in women but not men. In this study, we tested the hypothesis that this substitution is also associated with an increased risk of ischemic cerebrovascular disease (ICVD).
Methods and Results—We compared 260 patients who had nonfatal ICVD and carotid stenosis ≥50% with 1560 age-matched controls and also compared 205 Copenhagen City Heart Study cases who had nonfatal ICVD with 1210 age-matched controls. All subjects were white and from Denmark. Overall, no significant difference was observed between carrier frequencies among those with and without ICVD; however, sex interacted with genotype in predicting ICVD in the ICVD and carotid stenosis cases (P=0.02). In Copenhagen City Heart Study cases, sex was not significant (P=0.18). Odds ratios for ICVD in female mutation carriers were 2.9 (95% confidence interval [CI], 1.3 to 6.4) and 1.9 (95% CI, 0.8 to 4.6) in ICVD plus carotid stenosis cases and Copenhagen City Heart Study cases, respectively. Equivalent values in male mutation carriers were 0.8 (95% CI, 0.3 to 1.8) and 0.8 (95% CI, 0.3 to 2.0), respectively. These results were similar in analyses that also allowed for other conventional cardiovascular risk factors.
Conclusions—These results suggest that the Asn291Ser substitution in lipoprotein lipase is not associated with nonfatal ICVD in men but that it possibly confers a 2-fold risk in women.
The risk of ischemic heart disease1 and ischemic cerebrovascular disease (ICVD)2 increases with increasing plasma triglyceride levels. Lipoprotein lipase hydrolyses triglycerides contained in the core of chylomicrons and very-low-density lipoproteins.3 Thus, an impaired function of lipoprotein lipase may be associated with an increased risk of cardiovascular disease.
We recently showed that the common Asn291Ser substitution in lipoprotein lipase in the heterozygous state (heterozygote frequency is ≈5% in the general population) is associated with an increase in plasma triglycerides and a 2-fold increase in the risk of ischemic heart disease in women but not men.4 In the present study, we tested the hypothesis that this substitution is also associated with an increased risk of nonfatal ICVD.
A general population sample, the Copenhagen City Heart Study, which consisted of 9215 individuals, was collected from 1991 to 1994 and studied cross-sectionally.4 5 A total of 83 women and 122 men who had previous sudden onset of focal neurological symptoms due to ICVD were identified by an experienced neurologist5 who was unaware of the results of genotyping. ICVD was defined as ischemic stroke (focal neurological symptoms lasting >24 hours when hemorrhages were excluded on computed tomography scan; n=131), a transient ischemic attack (symptoms lasting <24 hours; n=66), or amaurosis fugax (transient blindness in 1 eye only; n=8). Patients with hemorrhagic stroke (n=10) or other nonischemic cerebral events (n=32), patients using cholesterol-lowering drugs (n=79), individuals among the 9215 subjects with plasma triglyceride levels ≥10 mmol/L (n=35; none of whom had ICVD), and participants with missing information (n=203) were excluded from the analysis, which left 205 ICVD patients and 8651 controls. From this pool, age- and sex-matched controls were drawn for all analyses.
A second study population consisted of 97 women and 163 men who had a sudden onset of focal neurological symptoms due to ICVD. These patients were referred by neurologists for carotid ultrasound examination on an outpatient basis at the National University Hospital in Copenhagen from 1994 to 1996 because of their neurological symptoms. These patients were consecutively included in the present study if they fulfilled the entry criteria of a carotid artery stenosis ≥50% on the symptomatic or most stenotic side; thus, neurological symptoms most likely were referable to large vessel atherothrombotic disease. At least 1 computed tomography scan was performed to exclude hemorrhage. These 260 patients either had ischemic stroke (n=157), transient ischemic attack (n=75), or amaurosis fugax (n=28). Furthermore, they were all residents of the same geographical area as the individuals from the Copenhagen City Heart Study, and none had plasma triglyceride levels ≥10 mmol/L.
In both study populations, <1% were non-whites, and >98% were of Danish descent. Diabetes mellitus and hypertension were diagnosed as described previously.5
We compared the 260 patients with ICVD and carotid artery stenosis ≥50% with 1560 age- and sex-matched controls (6 controls per case when possible); all controls were sampled from the general population and did not have ICVD. Furthermore, we compared the 205 individuals from the Copenhagen City Heart Study who had ICVD with 1210 age- and sex-matched controls. The study was approved by the ethical committee of the City of Copenhagen and Frederiksberg.
Screening for the Asn291Ser substitution, collection of basic characteristics, and biochemical measurements were performed as described previously.4 5 Blood was drawn in the nonfasting state among participants of the Copenhagen City Heart Study but in the fasting state among the hospital-based ICVD patients.
We used Student’s t test, ANOVA, Pearson’s χ2 test, and logistic regression analyses, as previously described, for statistical analyses.4 5 Data were analyzed for each sex, both separately and combined, after tests for interaction between sex and genotype. Logistic regression analyses were performed as bifactorial analyses allowing for age only and as multifactorial backward stepwise analyses allowing for age and other cardiovascular risk factors. The models did not allow for matching. After age and other significant predictors had entered the model, genotype was forced into the model. Homogeneity of the association of genotype and conventional cardiovascular risk factors in the prediction of ICVD were tested by the introduction, one at a time, of all possible 2-factor interaction terms between genotype and conventional cardiovascular risk factors in the logistic regression analysis. The likelihood ratio test determined significance.
Carrier frequencies and characteristics are shown in Table 1⇓⇓. Genotype frequencies in the general population (including 6 homozygous individuals) were not different from values predicted by the Hardy-Weinberg equilibrium. The analyses described below did not include 35 individuals with triglyceride levels ≥10 mmol/L; however, the inclusion of these individuals had no major influence on study outcomes.
Lipids and Lipoproteins
Genotype and sex interacted on triglyceride (P=0.03) and high-density lipoprotein (HDL) cholesterol (P=0.03) levels (Table 2⇓⇓). The Asn291Ser substitution was significantly associated with increased plasma triglyceride levels in women (P<0.001) but not men (P=0.98). HDL cholesterol was reduced 2.5 times more in female carriers than in male carriers.
Risk of Ischemic Cerebrovascular Disease
Genotype and sex also interacted in predicting ICVD in cases with ICVD plus carotid stenosis (P=0.02); in Copenhagen City Heart Study cases, this interaction was not significant (P=0.18). No interaction occurred between genotype and other covariates entered in the multifactorial logistic regression models predicting ICVD (data not shown).
On logistic regression analyses allowing for age only, Asn291Ser heterozygosity was associated with a 2- to 3-fold increase in the risk of ICVD in women (Table 3⇑); in Copenhagen City Heart Study cases, this was not significant. In men, genotype did not predict ICVD.
On multifactorial stepwise logistic regression analysis, odds ratios for ICVD associated with Asn291Ser heterozygosity were similar to those seen in the bifactorial analyses (Tables 3⇑ and 4⇑). In women, the odds ratio was 2 to 3, whereas in men, the odds ratio was close to 1.
When only ischemic stroke (excluding transient ischemic attack and amaurosis fugax) was considered, the odds ratios for ICVD in women carrying the Asn291Ser substitution were 4.1 (95% confidence intervals [CI], 1.6 to 10.8) and 1.7 (95% CI, 0.6 to 5.1) in cases with ICVD plus carotid stenosis and Copenhagen City Heart Study cases, respectively. The equivalent odds ratios for men were 1.1 (95% CI, 0.4 to 2.9) and 0.6 (95% CI, 0.2 to 2.1), respectively.
Overall, we found no significant difference in Asn291Ser carrier frequencies between subjects with and without ICVD; no effect was observed among men, either in cases with ICVD plus carotid stenosis or in Copenhagen City Heart Study cases. Furthermore, no significant effect existed in Copenhagen City Heart Study female cases. In contrast, a roughly 2-fold risk of ICVD was observed in female Asn291Ser carriers among cases with ICVD plus carotid stenosis (n=10 carriers). It was also only in this study group, and not among Copenhagen City Heart Study cases, that significant sex interaction was seen.
We recently showed that the common Asn291Ser substitution in lipoprotein lipase was associated with a significant increase in plasma triglyceride levels and a 2-fold increase in the risk of ischemic heart disease in women but not men.4 The present data indicate that the Asn291Ser substitution may also be associated with a 2-fold increase in risk of ICVD in women but not men. These sex-specific effects are supported by the following points. (1) Triglyceride level is a better predictor of the risk of cardiovascular disease in women than in men.1 (2) Triglyceride levels were significantly increased in female but not male carriers in the Copenhagen City Heart Study in 2 separate measurements taken in 1991 to 1994 and 1976 to 1978.4 (3) HDL cholesterol was 2.5 times more reduced in female carriers than in male carriers.4 (4) Odds ratios for ischemic heart disease4 and ICVD in carriers were ≈2.0 in women and ≈1.0 in men.
Lipoprotein lipase degrades the triglycerides contained in chylomicrons, and very-low-density lipoproteins thus converting these particles into smaller remnants.3 The Asn291Ser substitution does not block the formation of remnants completely; it only reduces enzyme activity by ≈40%.6 Thus, this substitution probably increases the time between the secretion of intact chylomicrons and very-low-density lipoproteins until the chylomicrons and very-low-density lipoproteins have been sufficiently depleted of triglycerides to be removed from plasma. In other words, at a given time point, more remnants of different sizes will be present in plasma than if the enzyme was fully functional, and such remnants may be atherogenic7 ; importantly, it is the cholesterol content in these triglyceride-rich lipoproteins rather than the triglyceride content that is thought to be atherogenic. Dysfunctional lipoprotein lipase also leads to the reduced production of HDL particles,3 which may also be atherogenic.
Because the inclusion of plasma triglycerides and HDL cholesterol in the logistic regression analysis did not completely abolish the association between the mutation and the risk of ICVD, it could be hypothesized that the atherogenicity of a malfunctioning lipoprotein lipase may be due not only to a direct effect of triglyceride-rich lipoprotein and low HDL levels, but also to a pathological effect on the metabolic processes involved in their formation. This could be due to (1) deposition of surface lipid in the vessel wall, (2) local production of smaller remnant lipoproteins, and (3) lipolysis alteration of endothelial barrier function.8 However, it is perhaps more likely that a single measurement of triglyceride or HDL levels does not accurately reflect the lifelong impact of the mutation on these levels.
The Asn291Ser substitution was only significantly related to ICVD in the group of patients with carotid artery stenosis. Our patients with ICVD and ≥50% carotid stenosis probably represent the 20% of the general ICVD population with atherosclerotic cerebrovascular disease,9 whereas the individuals with ICVD identified in the general population probably represent a more broad ICVD population. Thus, the promotion of atherosclerosis through an effect of Asn291Ser on plasma levels of triglycerides or HDL cholesterol would potentially result in a higher risk of ICVD in the subgroup of patients with verified carotid atherosclerosis, thereby explaining the observed higher risk in the ICVD plus carotid stenosis cases compared with Copenhagen City Heart Study cases. This is supported by the findings in the previous study on ischemic heart disease, which showed an increase in the odds ratio for ischemic heart disease in women who also had verified stenosis on coronary angiography.4
As expected, hypertension and diabetes mellitus seemed to be the most important factors in the risk of ICVD.9 However, the models explaining ICVD risk in ICVD plus carotid stenosis cases and Copenhagen City Heart Study cases differed. In women, triglyceride levels were only important in the model for Copenhagen City Heart Study cases. This may be due to the fact that plasma triglyceride levels were measured in the fasting state in cases with ICVD plus carotid stenosis but in the nonfasting state in Copenhagen City Heart Study cases4 ; the results could therefore have underestimated the effect of plasma triglyceride levels on risk of ICVD and may help explain why a triglyceride-modulating mutation would still be associated with an increased risk of ICVD, despite the seemingly small-to-no difference between hospital-based ICVD cases and controls.
It is important to note, however, that in the present study, the risk association with genotype was unaffected by whether triglyceride levels were measured in the fasting or nonfasting state. The apparent protective effect in cases with ICVD plus carotid stenosis for subjects with a body mass index in the highest tertile may reflect the fact that stroke patients in a hospital environment generally have a lower body mass index, due to more advanced disease, than nonhospitalized individuals10 and that body mass index in this setting would be inversely related to severity of disease.
Finally, female smokers seem to have an increased ICVD risk, whereas male smokers among cases with ICVD plus carotid stenosis paradoxically seem to be protected; 89% of men in the general population sample and 81% of men among patients with ICVD and ≥50% carotid stenosis were current or former smokers. One explanation for this finding could be an underreporting of smoking among ICVD plus carotid stenosis cases. It is more likely, however, that smoking among Danish men is so prevalent that the importance of smoking as a risk factor for ICVD is difficult to determine. Importantly, it would be erroneous to conclude that smoking is a protective factor in men.
The misclassification of some ICVD cases cannot be entirely ruled out; however, the patients with ICVD and ≥50% stenosis of the relevant carotid artery were all first evaluated by a neurologist and later by an experienced vascular surgeon, and participants in the Copenhagen City Heart Study who had ICVD were all diagnosed by an experienced neurologist who reviewed all hospital admissions and diagnoses; in both cases, the evaluators were unaware of the results of genotyping. Furthermore, confounding by another mutation in linkage disequilibrium with the mutation causing the Asn291Ser substitution is a possible, although unlikely, explanation of our results. In vivo evidence suggests that this substitution reduces the activity of lipoprotein lipase by 40%,6 which is sufficient to cause the effects observed. Despite this, we cannot completely rule out the possibility that some of our findings may be chance observations.
The differences in the effects of the Asn291Ser substitution that were observed among women and men are not easily explained. Possible explanations may be found in the differences in the hormonal regulation of metabolic processes in women and men. In men, low levels of testosterone are associated with low levels of lipoprotein lipase activity11 and, in a group of hypogonadal males, testosterone substitution increased the activity of lipoprotein lipase.12 In contrast, estrogen decreases lipoprotein lipase activity,13 possibly through a posttranscriptional modification of protein levels.14 Raised plasma triglyceride levels seem to be a better predictor of ischemic cardiovascular disease in women than men,1 which suggests that any genetic factor raising plasma triglyceride levels would also have a greater effect on the development of disease in women than in men. The consistency of the sex-specific effect of the Asn291Ser substitution on plasma lipid levels, the risk of ischemic heart disease,4 the risk of ICVD (as shown in the present study), and the strength of these associations suggest that these are not chance findings.
In conclusion, the Asn291Ser substitution in lipoprotein lipase, which is present in ≈5% of Danes, is not associated with an increased risk of ICVD in men, but it is possibly associated with a 2-fold risk in women.
Supported by The Danish Heart Foundation, the Danish Medical Research Council, and Chief Physician Johan Boserup and Lise Boserup’s Fund. We are grateful for the technical assistance of Pia T. Petersen and Anne-Merete Bengtsen.
- Received August 30, 1999.
- Revision received November 23, 1999.
- Accepted December 22, 1999.
- Copyright © 2000 by American Heart Association
Lindenstrom E, Boysen G, Nyboe J. Influence of total cholesterol, high density lipoprotein cholesterol, and triglycerides on risk of cerebrovascular disease: the Copenhagen City Heart Study [published erratum appears in BMJ.. 1994;309:1619]. BMJ. 1994;309:11–15.
Brunzell J. Familial lipoprotein lipase deficiency and other causes of the chylomicronemia syndrome. In Scriver C, Beaudet A, Sly W, Valle D, eds. The Metabolic and Molecular Bases of Inherited Disease. 7th ed. New York: McGraw-Hill; 1995:1913–1932.
Wittrup HH, Tybjaerg-Hansen A, Nordestgaard BG. Lipoprotein lipase mutations, plasma lipids, and lipoproteins and risk of ischemic heart disease: a meta-analysis. Circulation.. 1999;99:2901–2907.
Nordestgaard BG, Wootton R, Lewis B. Selective retention of VLDL, IDL, and LDL in the arterial intima of genetically hyperlipidemic rabbits in vivo: molecular size as a determinant of fractional loss from the intima-inner media. Arterioscler Thromb Vasc Biol. 1995;15:534–542.
Goldberg IJ. Lipoprotein lipase and lipolysis: central roles in lipoprotein metabolism and atherogenesis. J Lipid Res. 1996;37:693–707.
Benavente O, Hart RG, Sherman DG. Primary prevention of transient ischemic attack and thromboembolic stroke. In Verstraete M, Fuster V, Opol EJ, eds. Cardiovascular Thrombosis: Thrombocardiology and Thromboneurology. Philadelphia: Lippincott-Raven; 1998:585–595.
Iverius PH, Brunzell JD. Relationship between lipoprotein lipase activity and plasma sex steroid level in obese women. J Clin Invest. 1988;82:1106–1112.