Angiotensin-Converting Enzyme Gene Deletion Polymorphism
A New Risk Factor for Lacunar Stroke but not Carotid Atheroma
Background and Purpose A deletion (D)/insertion (I) polymorphism in the angiotensin-converting enzyme gene has been associated with myocardial infarction. Its relations to both stroke and atheroma remain uncertain. We examined its role as a risk factor in patients with cerebrovascular disease and its relation to carotid atheroma.
Methods One hundred one patients with symptomatic carotid artery territory cerebral ischemia were compared with 137 age-matched control subjects. In the patient group, carotid atheroma was assessed by measurement of degree of carotid stenosis and intima-media thickness with high-resolution duplex ultrasound. The D/I polymorphism was examined using the polymerase chain reaction.
Results D:I allele frequency was 0.59:0.41 in case subjects and 0.48:0.52 in control subjects (P=.01). The DD genotype was more common in patients with cerebrovascular disease compared with control subjects (36/101 versus 30/137, P=.02). The DD genotype conferred a relative risk of any type of cerebrovascular disease of 1.98 (95% confidence interval [CI], 1.11 to 3.51; P=.02). However, this was largely due to a strong association in the 18 patients with lacunar stroke, in whom the D:I ratio was 0.75:0.25 (P=.0097 versus control subjects). The odds ratio for lacunar stroke associated with the DD genotype was 5.6 (95% CI, 2.0 to 15.7) and was still significant at 4.40 (95% CI, 1.45 to 12.6; P<.009) after controlling for other risk factors. There was no significant association between angiotensin-converting enzyme genotype and cerebrovascular disease due to large-vessel stenosis. There was no association between genotype and age, sex, smoking history, diabetes, or cholesterol level.
Conclusions The deletion polymorphism in the angiotensin-converting enzyme gene is a new independent risk factor for lacunar stroke but is not a risk factor for stroke associated with carotid stenosis.
Genetic factors are important in CVD.1 The genetic basis of some relatively rare metabolic and coagulation disorders predisposing to stroke is known, but the molecular basis of the genetic predisposition in the majority of patients with CVD remains unknown. Genetic influences may act either independently or by predisposing to or modulating the effect of risk factors such as hypertension. Factors involved in the pathogenesis of atherosclerosis and vasoconstriction are important in CVD. ACE is important in the production of angiotensin II and the catabolism of bradykinin, two peptides involved in the modulation of vascular tone and the proliferation of smooth muscle cells. Recently, a deletion (D)/insertion (I) polymorphism in intron 16 of the ACE gene has been identified. In an intron, or noncoding region of the gene, it is associated with the level of circulating enzyme, with higher levels in individuals homozygous for the deletion,2 and is presumably a marker for another unknown functional variation possibly located in the regulatory region of the gene. Cambien and colleagues3 have reported an association of the deletion polymorphism with myocardial infarction, independent of conventional risk factors. An association with myocardial infarction has also been found by other groups,4 5 and an increase in the DD genotype has been reported in both ischemic and idiopathic dilated cardiomyopathy.6 In an earlier study, we found a trend toward an increase in the DD genotype in patients with acute cerebral infarction, although this did not reach significance.7 The mechanism of this predisposition to vascular disease is uncertain; both a proatherogenic effect and a local effect on vascular tone are possibilities.
This study was designed with two aims. First was to determine whether the deletion polymorphism was associated with symptomatic CVD in a larger group of patients and control subjects and in particular to determine if it was associated with any pathophysiological subgroup. The second aim was to identify whether it was associated with carotid atheroma or wall thickening in patients with CVD. The latter was assessed by measuring carotid stenosis and I-M thickness on high-resolution carotid duplex ultrasound. I-M thickness appears to be a marker for atheroma. It is associated with a past history of cardiac ischemic events, angiographic coronary artery narrowing,8 and carotid plaques9 and is increased in individuals with risk factors for ischemic heart disease such as hyperlipidemia.10 It may provide a better marker for the study of potentially atherogenic risk factors in patients with CVD, in whom there is a high prevalence of carotid artery disease, because the degree of carotid stenosis is known to be dependent on local hemodynamic factors as well as systemic factors.
Subjects and Methods
One hundred one successive patients with symptomatic ischemic CVD presenting to a cerebrovascular duplex ultrasound service were enrolled. All had symptoms of carotid territory ischemia. Patients with cerebral hemorrhage were not studied. Patients with valvular heart disease, cardiac failure, and recent myocardial infarction were excluded. In view of the marked ethnic variation in the distribution of the ACE polymorphism,11 only whites were studied. Patient details are given in Table 1⇓. Presenting symptoms were transient cerebral ischemia (including amaurosis fugax) in 49 and stroke in 52 patients. CT imaging was performed in all patients. Of the stroke cases, 41 patients (78.8%) had CT infarcts in an area appropriate for the symptoms, whereas 10 (20.4%) TIA patients had an appropriate CT infarct. Stroke subtypes were divided on the basis of clinical features, CT scanning, and carotid ultrasound into four categories: (1) large-vessel disease (n=43; internal carotid artery stenosis ≥50% with symptoms in that arterial territory); (2) lacunar stroke (n=18; clinical lacunar syndrome with an appropriate CT infarct [n=16] or a typical clinical syndrome12 [1 ataxic hemiparesis, 1 clumsy hand and dysarthria] and a normal CT scan [n=2]); (3) uncertain or probable cardiac embolic source (n=30; these two categories were combined because not all patients had echocardiography); and (4) tandem pathology (n=10; more than one cause of stroke [atrial fibrillation and carotid stenosis in 5, atrial fibrillation and lacunar infarction in 1, and carotid stenosis with a typical CT lacunar infarction in 4]).
Duplex ultrasound was performed using an Acuson XP color-flow imager with a 7-MHz probe. Internal carotid artery stenosis was calculated from a combination of Doppler data (ratio of internal carotid artery systolic to common carotid artery diastolic) for stenoses >50%13 and with the B-mode modality to measure the ratio of maximum plaque thickness to luminal diameter for lesser degrees of stenosis. Maximum I-M thickness was measured on the posterior wall of the common carotid artery at least 1 cm proximal to the carotid bulb.14 Imaging was performed blinded to the result of the genetic analysis.
One hundred thirty-seven normal white volunteers were studied. Patient spouses were studied when available; in other cases, normal staff members were studied. Individuals with stroke or myocardial infarction were excluded, but subjects with risk factors for vascular disease were included. Blood was taken for genetic analysis, and blood pressure was measured. In both patient and control groups, hypertension was defined as either a systolic blood pressure >160 mm Hg, a diastolic pressure >95 mm Hg, or current treatment with antihypertensive drugs. Raised cholesterol was defined as a cholesterol level >6.5 mmol/L or if the subject was receiving cholesterol-lowering treatment.
Plasma was stored at −60°C, and ACE levels were measured in all patients by a spectrophotometric method. ACE cleaves the substrate FAPGG (furylacryloylphenylalanylglycylglycine to furylacryloylphenylalanine and glycylglycine), and the resulting fall in absorbance was measured at 340 nm using a Monarch analyzer (Instrumentation Laboratories Ltd). ACE levels in patients taking ACE inhibitors were not included in the analysis as all were <20 IU/L, thus leaving 81 ACE levels in 101 patients. ACE levels were measured in 41 control subjects of whom 5 were taking ACE inhibitors, resulting in 36 levels for analysis.
Genomic DNA was isolated from blood leukocytes using a Nucleon extraction kit (Scotlab Ltd) or a standard phenol/chloroform method. The D/I ACE polymorphism was identified using the polymerase chain reaction as previously described.3 The reaction products were analyzed with agarose gel electrophoresis, and the two alleles were identified: a 490-bp fragment I (with the insertion) and a 190-bp fragment D (without the insertion). Analysis was performed blinded to the clinical details (patients and control subjects were mixed) and blinded to the results of the carotid duplex examination.
Proportions between groups were compared using χ2 test, and the contribution of the genotype to relative risk was calculated using logistic regression analysis. Differences between degree of stenosis and I-M thickness and other continuously distributed variables between groups were analyzed using a one-way ANOVA followed by Scheffé’s multiple-comparisons test where appropriate. ACE plasma levels, I-M thickness, and systolic blood pressure were logarithmically transformed for statistical analysis; this resulted in a good approximation to a normal distribution. The mean values given in the text and tables are untransformed.
The D allele was more common in the CVD group as a whole, with a D:I allele ratio of 0.59:0.41 in the CVD group compared with 0.48:0.52 in the control group (χ2, P=.01). The DD genotype was more common in the CVD group (36/101 versus 30/137; χ2, P=.02). The OR of CVD in individuals with the DD genotype was 1.98 (95% CI, 1.11 to 3.51; P=.02). There was no significant association between the ID genotype and CVD (OR, 1.30; 95% CI, 0.67 to 2.60; P=.4). There was no relationship between genotype and sex, smoking history, diabetes, or cholesterol level.
Analysis by subtype of CVD showed that the association between genotype and CVD was largely due to an association with lacunar stroke. There was a significantly higher frequency of the DD genotype in lacunar stroke patients compared with control subjects (61.9% versus 21.3%), and the OR associated with the DD genotype for lacunar stroke was 5.60 (95% CI, 2.00 to 15.71). Logistic regression analysis demonstrated that the association between the DD genotype and lacunar stroke was significant (P=.009) independent of age, sex, hypertension, smoking status, and diabetes; after these risk factors were controlled for, the OR was 4.4 (95% CI, 1.45 to 12.6). Hypertension was also independently associated with lacunar stroke, with an OR of 3.49 (95% CI, 1.12 to 10.94) after controlling for age, sex, genotype, diabetes, and smoking.
There was a strong association between the D allele and lacunar stroke, with a D:I allele ratio of 0.75:0.25 (P=.009 versus control subjects). Logistic regression analysis demonstrated that D allele was a risk factor for lacunar stroke independent of age, sex, hypertension, smoking status, and diabetes (OR, 2.63; 95% CI, 1.08 to 6.43; P=.03). The D allele was also independently related to hypertension (OR, 3.20; 95% CI, 1.04 to 9.89; P=.04).
In contrast, there was no association between either the D allele or the DD genotype and large-vessel CVD. There was a trend toward an increase in the DD allele frequency in patients with uncertain or cardiac CVD, but this was not significant (Table 2⇓).
Within the CVD group, I-M thickness was significantly greater in patients with the II genotype compared with both DD and ID. A similar trend was found with carotid stenosis; patients with the II genotype had a greater mean percent stenosis, but this was not significant (Table 3⇓). There was a significant association between degree of stenosis and hypertension (mean±SD stenosis: hypertensive, 40.8±27.2% stenosis; normotensive, 28.1±27.8%; P=.02) and smoking history (smoker, 39.0±26.2% stenosis; never smoked, 27.3±30.8%; P=.05) but no association with diabetes or cholesterol level. There was a highly significant relationship between log I-M thickness and mean percent stenosis (r=.52, P<.0001). This relationship persisted (r=.50, P<.001) when the nine patients with a common carotid artery plaque (>1.5 mm thickness), as opposed to diffuse I-M thickening, were excluded. No relationship was found between log ACE levels and either log I-M thickness (r=.16, P=.23) or mean percent stenosis (r=.02, P=.85).
The relationship between genotype and variables was studied further in the control group. There was no relationship between DD genotype or the D allele and sex, diabetes, or smoking history. However, there was a significant association between both the DD and ID genotypes and hypertension independent of other risk factors on logistic regression analysis (P=.02 and P=.02, respectively).
There was a significant relationship between plasma ACE levels and genotype (Table 3⇑), with levels being significantly higher in the DD patients than in the II patients. A similar relationship was found in the control subjects (mean±SD levels in U/L: DD, 73.6±17.3; ID, 55.6±16.3; II, 40.0±7.3). Twenty-four percent of the variance in ACE levels in the CVD group and 43% of the variance in the control subjects were accounted for by the polymorphism.
In our white population, the D allele and the DD genotype were associated with a significantly increased risk of symptomatic CVD. This increase was due to a markedly increased risk of lacunar stroke for which both the D allele and the DD genotype were independent risk factors; an individual with the DD genotype had a 4.4-times greater risk of CVD than control subjects. In contrast, we found no association between the D allele, or DD genotype, and CVD associated with large-vessel atheroma. While the relationship with lacunar stroke was highly significant, only 18 of the 101 case patients had lacunar stroke, and therefore the association needs to be confirmed in a larger sample.
D:I allele frequencies in our normal control subjects are similar to those reported in other studies of normal whites. Analysis of our normal control population of 300 individuals aged 15 years and older shows a trend toward a lower prevalence of the DD genotype above age 50 that is consistent with the effect of a lethal gene (H. Markus, J. Barley, and N. Carter, unpublished data, 1994), and a similar finding has recently been reported in a hypertensive population.15 Our findings are similar to those of a previous pilot study from our department involving a different group of patients in the acute phase of cerebral infarction.7 An increase in the DD genotype (case patients, 30%; control subjects, 23%) was found in the patient group, although this did not reach significance. The population in this study differed in that it was larger, only whites were studied, TIA as well as stroke was included, analysis by stroke subtype was performed, all ischemic events were in the carotid territory, and all patients underwent carotid artery imaging.
Our study confirms the previously reported relationship between the D allele and increased serum ACE levels.2 However, whether the association between the deletion polymorphism and cardiovascular disease is mediated via increased ACE levels is not proven, and the mechanism of action of the DD genotype in predisposing to myocardial infarction is uncertain. Our study suggests that in the cerebral circulation it is associated with small-vessel disease. It also provides new information on the relation of ACE genotype to atheroma. It demonstrates that the DD genotype does not act by predisposing to carotid atheroma. In our CVD group, I-M thickness was significantly correlated with degree of carotid stenosis, consistent with it being a marker for atheroma. The increased I-M thickness we found in patients with the I allele may simply reflect the fact that in the DD patients factors other than large-vessel atheroma (such as small-vessel disease) were relatively more important in causing CVD, whereas in the II population factors are operating to a greater extent through atheroma.
A major risk factor for lacunar stroke is hypertension. The association with the DD genotype and lacunar stroke in our population was independent of hypertension on logistic regression analysis. However, the number of lacunar strokes was relatively small, and in the control population there was an association between the D allele and hypertension. Previous studies have shown both the presence of a relationship between the D/I ACE polymorphism and hypertension and the absence of a relationship.3 16 17 18 However, the numbers studied have usually been small, and the individuals studied were younger than subjects in this study. It is possible that the D allele increases the risk of small-vessel disease independent of hypertension; this effect may be analogous to the increased incidence of left ventricular hypertrophy, another recognized complication of hypertension, found in individuals with the D allele.18 The exact relationship between hypertension, the D allele, and lacunar stroke remains to be determined in larger case-control studies and by comparing the prevalence of the genotype in groups with and without lacunar stroke, matched for severity of hypertension.
Selected Abbreviations and Acronyms
|TIA||=||transient ischemic attack|
We thank Diana Colquhoun for assistance with ultrasound scanning.
- Received March 28, 1995.
- Revision received May 4, 1995.
- Accepted May 5, 1995.
- Copyright © 1995 by American Heart Association
Brass LM, Issacsohn JL, Merikangas KR, Robinette CD. A study of twins and stroke. Stroke. 1992;23:221-223.
Cambien F, Poirier O, Lecerf L, Evans A, Cambou J-P, Arveiller D, Luc G, Bard J-M, Bara L, Ricard S, Tiret L, Amouyel P, Alhenc-Gelas F, Soubrier F. Deletion polymorphism in the gene for angiotensin-converting enzyme is a potent risk factor for myocardial infarction. Nature. 1992;359:641-644.
Zhao Y, Higashimori K, Higaki J. Significance of the deletion polymorphism of the angiotensin converting enzyme gene as a risk factor for myocardial infarction in Japanese. Hypertens Res. 1994;17:55-57.
Blankenhorn DH, Hodis HN. Arterial imaging and atherosclerosis reversal. Arterioscler Thromb. 1994;14:177-192.
Persson J, Formgren J, Israelsson B, Berglund G. Ultrasound-determined intima-media thickness and atherosclerosis. Arterioscler Thromb. 1994;14:261-264.
Wendelhag I, Wiklund O, Wikstrand J. Atherosclerotic changes in the femoral and carotid arteries in familial hypercholesterolemia. Arterioscler Thromb. 1993;13:1404-1411.
Pullicino PM, Caplan LR, Hommel M, eds. Cerebral Small Artery Disease. New York, NY: Raven Press Publishers; 1993.
Williams MA, Deacon DFS, Szendro G, Balcaro G, Nicolaides AN. Duplex scanning in the detection and grading of internal carotid artery stenosis: improved criteria. In: Salmasi A-M, Nicolaides AN, eds. Cardiovascular Applications of Doppler Ultrasound. Edinburgh, Scotland: Churchill Livingstone Inc; 1989:247.
Morris BJ, Zee RYL, Schrader AP. Different frequencies of angiotensin-converting enzyme genotypes in older hypertensive individuals. J Clin Invest. 1994;94:1085-1089.
Harrap SB, Davidson HR, Connor JM, Soubrier F, Corvol P, Fraser R, Foy CJ, Watt GC. The angiotensin I converting enzyme gene and predisposition to high blood pressure. Hypertension. 1993;21:455-460.