Genetic Polymorphisms of the Renin-Angiotensin System and Atheromatous Renal Artery Stenosis

Methods

We studied 58 consecutive patients referred to our Hypertension Clinic from May 1996 to September 1998 and diagnosed as having RAS. All but 2 patients having a significant RAS (at least 70% lumen reduction) were enrolled; the 2 remaining patients refused to give blood for DNA evaluation. Patients referred to our Hypertension Clinic ranged from 250 to 300 patients per year, and RAS patients represented 10% to 12% of the total. Patients with renal artery lumen narrowing <70% were excluded in order to compare only phenotypically extreme conditions.

All the patients included in the present study underwent angiographic evaluation and were found to have significant monolateral or bilateral RAS (at least 70% lumen reduction). All of them were examined for severe hypertension requiring >2 drugs and/or associated with clinically evident vascular disease, mainly affecting the lower extremities. Angiographic study was performed when a moderate or high index of clinical suspicion, defined according to Mann and Pickering,⁶ was recognized.

Routine biochemical and clinical data were collected on the occasion of the first examination, venous blood samples for genetic analysis were taken after renal angiography, and the patient’s informed consent was obtained.

Because renal angiography was regarded as unacceptable in control subjects for ethical reasons, we exploited the possibility offered by our recently performed large case-control study in subjects with an angiographically documented normal or pathological coronary bed.¹⁸ One hundred two subjects with valvular heart disease and normal coronary arteries who proved normotensive at 3 consecutive evaluations over a 3-month period were considered as a control group. Hypertension was defined as systolic blood pressure >140 mm Hg or diastolic blood pressure >95 mm Hg. Moreover, control subjects were enrolled providing that they had both a normal coronary angiogram at cardiac catheterization and no history or clinical or instrumental evidence of atherosclerosis in vascular districts other than the coronary bed. With these selection methods, the prevalence of renal artery disease can be considered extremely low or close to zero.^{1 5 6 7 8 9}

Results

The relevant characteristics of the population studied are summarized in Table 1⇓. As expected, the RAS patients presented more conventional cardiovascular risk factors (higher levels of glucose, triglycerides, and LDL cholesterol and lower HDL cholesterol levels) and higher (although not yet statistically significant) serum creatinine levels than did the control subjects.

The distribution of the genotypes for the 3 polymorphisms in the control population was compatible with a Hardy-Weinberg equilibrium. Table 2⇓ indicates the allele frequencies and genotype distributions in RAS patients and control subjects. The allele frequencies of AGT M235T and the ATR A1166C polymorphisms were not statistically different between patients and controls. The frequency of the D allele in the RAS group was significantly higher than that in the control group (0.70 versus 0.55; χ² 6.88, P=0.01; odds ratio [OR] 1.9, 95% CI 1.17 to 3.07). An increased frequency of the DD genotype was also evident (DD 28 of 58 patients in the RAS group versus 32 of 102 subjects in the control group; χ² 6.62, P<0.05; OR 2.04, 95% CI 1.05 to 3.95). The II genotype was associated with a lower relative risk of RAS (OR 0.34, 95% CI 0.19 to 1.47). There was no preferential association between ACE and AGT or ATR genotypes in the patients (Table 3⇓).

The distribution of the main vascular risk factors (smoking and levels of total, LDL, and HDL cholesterol, triglycerides, uric acid, and glucose) and serum creatinine levels were unaffected by I/D polymorphism (data not shown).

Discussion

The main result of the present study was the demonstration that patients affected by atheromatous RAS had a significantly higher D allele frequency than did the control population; about half of them were DD homozygotes and only 8.6% were II homozygotes compared with 31% and 22% in the control population, respectively (Tables 2⇑ and 3⇑). In contrast, AGT and ATR polymorphism alleles and genotypes were similarly distributed between the 2 populations and were not associated with the occurrence of renal vascular pathology (Table 2⇑).

Despite the current interest in the genetics of the renin-angiotensin system in relation to the vascular and renal risk profile, RAS, a clinical condition characterized by persistent activation of this system,^{23 24 25 26} has not been extensively studied. Because of the relative scarcity of reports, comparison of our results with previously published data are objectively limited. To the best of our knowledge, the only article on this topic is the study by Missouris et al,¹⁰ who evaluated ACE polymorphism in 56 English RAS patients; AGT and ATR polymorphisms were not analyzed. Besides the geographical origin, a relevant difference vis-à-vis the present study was that Missouris et al used subjects from the general population as a control group, without any objective angiographic information about their coronary arteries. With our approach, however, we were able to rule out the possibility that the controls might have had substantial (though not clinically manifest) coronary atherosclerosis; thus, the occurrence of atheromatous RAS would also be highly unlikely.^{1 5 6 7 8 9} Having considered objective conditions, we therefore feel confident that we reduced the likelihood of spurious results. Apart from this difference, however, the D allele frequency observed in the English population was similar to that found in our Italian patients.

Reported frequencies of the DD genotype in “control” populations vary from 16% to 28% in Japanese subjects^{27 28} to ≈22% to 32% in Caucasians.^{19 27 28 29 30} Our data confirm that the range of frequency of DD polymorphism in Northern Italy is of the same order of magnitude as that generally reported for Caucasian populations (homozygosity in our control subjects was 31.37%). At this level, any possible selection bias deriving from a “spurious” low frequency in the control population and leading to an erroneous statistical difference in favor of the RAS group should be excluded. In addition, the distribution of the ACE genotypes in the control population was compatible with a Hardy-Weinberg equilibrium.

Another possible source of bias lies in the power of the study. The present study analyzing 320 alleles had a predicted 80% power (5% probability) of detecting an allelic difference such as that observed in D allele frequency between the RAS and control groups (Table 2⇑); thus, the results should be reliable.

The problem of the power of the study is also of major relevance with regard to the negative results that we observed for AGT and ATR polymorphisms. For this reason, our findings do not allow us to completely rule out an association between AGT and ATR polymorphisms and RAS risk; therefore, firm conclusions should be avoided. However, it should be borne in mind that the allele frequency of AGT was so balanced in the 2 groups (Table 2⇑) that a positive association with this polymorphism appears highly improbable.

The predisposing role of ACE I/D polymorphism in cardiovascular disease has been a subject of controversy and debate in recent years: both positive and negative associations with coronary atherosclerotic disease (CAD) and myocardial infarction (MI) have been reported (for review see References 27, 28, 31, and 32^{27 28 31 32} ). Controversial findings have also been reported for several other pathological conditions such as cerebrovascular disease (for a review see Reference 33³³ ), left ventricular hypertrophy,^{34 35 36} and diabetic or nondiabetic nephropathy (for review see References 37 and 38^{37 38} ). Note that very few of the numerous published studies reported an elevated DD genotype frequency such as that found in the present study. If we also take into account the results reported by Missouris et al,¹⁰ an estimated frequency of DD in RAS patients should range from 41% to 48.3% (mean 44.7%, 51 of 114 patients), and the estimated D allele frequency should range from 0.64 to 0.70 (mean 0.67).

As already stated, unsuspected RAS is commonly found in patients with angiographically documented CAD, with a prevalence ranging from 11% to 30% (the mean for 11 published reports was 23%).⁴ If ≈50% of patients affected by RAS carry a DD genotype, then 5.5% to 15% (mean 11.5%) of DD genotypes observed in a population of CAD patients may vary in relation to the relative proportion of unsuspected RAS patients included in the study. Because patients affected by CAD or MI were not angiographically checked for RAS in the previous studies, negative or positive statistical associations with I/D polymorphism might depend on this unrecognized confounding factor. Similarly, one can speculate regarding the reported increase in D allele frequency in cerebrovascular disease or myocardial hypertrophy, which commonly complicate the clinical history of the patients with RAS.^{2 39}

In conclusion, several lines of evidence suggest a close link between ACE polymorphism, activation of the renin-angiotensin system, and the causes of RAS. Reported controversial associations between D allele frequency and various forms of cardiovascular pathology (eg, coronary artery disease) should be reconsidered in light of this potentially confounding factor.