α-Adducin Gly460Trp Variant Increases the Risk of Stroke in Hypertensive Dutch Women
The Gly460Trp variant of the α-adducin gene has been associated with renal sodium retention and salt-sensitive hypertension. Independent of blood pressure, salt sensitivity has been related to cerebrovascular events. We studied the risk of stroke, coronary heart disease (CHD), and myocardial infarction (MI) associated with the α-adducin variant and examined the extent to which this risk is modified by the presence of hypertension. We performed a case-cohort study in a prospective cohort of 15 236 initially healthy Dutch women. We applied a Cox proportional hazards model with an estimation procedure adapted for case-cohort designs to study the relation of the polymorphism and CHD (n=210), MI (n=71), any stroke (n=74), and ischemic stroke (n=49). Subjects with the Gly460Trp variant had a 2.8 times higher risk of stroke (95% confidence intervals [CI] 1.3 to 5.8) under the dominant genetic model, which did not attenuate after adjustment. The same pattern was found under per-allele comparison. Risk of ischemic stroke in the variant allele carriers was 3.9 times higher than in subjects with the common genotype (95% CI 1.7 to 8.6) using dominant inheritance model. The same patterns were found under per-allele comparison. CHD and MI were not related to the variant. The risk of ischemic stroke was more pronounced among women with systolic hypertension (10.9; 95% CI 3.6 to 31.5). The findings in this prospective study in a population based cohort of Dutch women strongly suggest that presence of the α-adducin Gly460Trp polymorphism increases the risk of stroke. This risk is particularly elevated in the presence of systolic hypertension.
- cerebrovascular disorders
- ischemic stroke
- coronary artery disease
- myocardial infarction
- case-cohort studies
The functional Gly460Trp (G460W) variant of the α-adducin gene, accompanied by an amino acid substitution of tryptophan in place of glycine at residue 460 (National Center for Biotechnology Information single nucleotide polymorphism cluster ID rs4961), has been associated with renal sodium retention and salt-sensitive hypertension, through enhancement of the activity of the sodium pump. This leads to increased renal tubular sodium reabsorbtion and may eventually increase blood pressure. Several studies showed an increased prevalence of hypertension in the presence of the Gly460Trp variant of the α-adducin gene,1,2 but no relation with blood pressure has been reported.1 One reason may be that several other mechanisms compensate changes in blood pressure leading to hypertension in some but not in all carriers of the mutation.
Independent of blood pressure, salt sensitivity has been related to cardiovascular events.3 Preliminary data suggest that presence of the Gly460Trp variant may be atherogenic, and some recent studies have suggested that carriers of the mutation are at an increased risk of stroke4,5 and myocardial infarction5 although reports are scarce and not consistent.1,6,7 In addition, Psaty and coworkers reported that hypertensive carriers of the mutation may particularly benefit from treatment with diuretics, rather than other antihypertensive drugs.4
We prospectively examined the relation of the α-adducin Gly460Trp polymorphism with risk of stroke and coronary heart disease and the interaction between the Gly460Trp variant and hypertension with respect to the events in Dutch women.
Materials and Methods
Participants were recruited between 1993 and 1997 among women living in Utrecht and vicinity who attended the regional breast cancer-screening program.8 17 357 women aged 49 to 70 were included. At enrollment women underwent a physical examination and completed a general questionnaire on lifestyle and medical factors. Data on morbidity were obtained from the Dutch Centre for Health Care Information, which holds a standardized computerized register of all hospital discharge diagnoses in the Netherlands since 1990. Using the International Classification of Diseases, ninth Revision (ICD-9), we categorized cardiovascular disease (codes 390 to 459), CHD (codes 410 to 414), including acute myocardial infarction (code 410), stroke (codes 430 to 438), including subarachnoidal hemorrhage (ICD 430), intracerbral hemorrhage (ICD 431 to 432), ischemic (ICD 433 to 435), and other cardiovascular disease. The first diagnosis was taken as end point of interest. The database was linked to the cohort on the basis of birth date, gender, postal code, and general practitioner with a validated probabilistic method.9
Information on vital status was gained through linkage with the national municipal administration database. Causes of death were obtained from the women’s general practitioners. All women signed an informed consent form before study inclusion. The study was approved by the Institutional Review Board of the University Medical Center Utrecht.
We applied the case-cohort design introduced by Prentice.10 In this design, data are collected on all subjects, but the data would only be analyzed on cases and subcohort members. The subcohort is a randomly selected sample of 10% (n=1736) from the 17 357 women in the total cohort. Women who did not consent to linkage with vital status registries or who were not traceable (cases n=3/subcohort n=38) were not included. Women who reported a diagnosis of cardiovascular disease (ICD-9; 390 to 459) at baseline who had missing questionnaires or blood or DNA samples were excluded. This resulted 15 236 women in the total cohort and 1522 women in the subcohort at baseline (as the control group). The entire cohort 17 357 women had been followed up to January 1, 2000 for the occurrence of cardiovascular events as defined earlier. For all subjects that had a cardiovascular event, follow-up ended at the date of diagnosis or at the date of death. Moving out of the Netherlands (n=2) and death attributable to causes other than cardiovascular disease were considered censoring events. All others were censored on January 1st, 2000. During follow-up 210 CHD cases (71 subjects with AMI) and 74 strokes (49 subjects with ischemic stroke) occurred.
Women were classified as current, past, or never smokers. Systolic and diastolic blood pressures were measured in duplicate at the left arm with the subjects in sitting position after 10 minutes of rest with an automated and calibrated oscillomat (Bosch & Son, Jungingen, Germany). Subsequently, the mean systolic and diastolic blood pressure was calculated. Height and weight were measured without shoes and wearing light indoor clothing to compute body mass index (BMI), which was defined as weight divided by height squared (kg/m2). Hypercholesterolemia or diabetes was defined as a self-reported physician diagnosis. Presence of hypertension was defined as a measured systolic blood pressure (SBP)≥140 mm Hg or a diastolic blood pressure (DBP)≥90 mm Hg or a self-reported physician diagnosis. In addition, patients with stage 2 systolic hypertension (SBP≥160 mm Hg) were considered as systolic hypertensive subjects, because SBP is a better predictor of major cardiovascular adverse events than is DBP, particularly in older individuals.11
Genetic analysis was performed at the Cardiovascular Genotyping (CAGT) laboratory of the department of internal medicine of the University Hospital Maastricht. Genomic DNA was extracted from buffy coats with the use of the QIAamp Blood Kit (Qiagen Inc). Genotyping was performed using a multilocus genotyping assay for candidate markers of cardiovascular disease risk (Roche Molecular Systems Inc).12 Briefly, each DNA sample was amplified using 2 multiplex polymerase chain reactions, and the alleles were genotyped simultaneously using an array of immobilized sequence-specific oligonucleotide probes. This array of probes is blotted on plastic strips and, after staining, genotypes can be scored based on blue (positive) and white (negative) bands. Each blue band, representing a specific genotype, was scored by specific software (counting the pixel intensity of each band) and checked manually. Investigators involved in genotyping were blinded to the case-control status of the participants. A random double-check was performed to detect potential genotyping errors.
To assess the relation of Gly460Trp polymorphism with outcomes, we used a Cox proportional hazards model with an estimation procedure adapted for case-cohort designs (unweighted method by Prentice, incorporated in a SAS macro at http://lib.stat.cmu.edu/general/robphreg). It computes weighted estimates with robust standard errors from which 95% confidence intervals were estimated. This approach provides association estimates that resemble those from full-cohort analysis, in particular when the sampling fraction is ≥10%.13
Baseline characteristics of the subcohort were examined by α-Adducin genotype. Hardy-Weinberg equilibrium (HWE) was tested with the χ2 exact test (the expected count under the HWE was less than 5 for 1 of the 3 genotypes among the controls). The relation between genotypes and potential risk factors was studied using t test for continuous variables and χ2 test for dichotomous variables. We assessed the association between the polymorphism and events under the dominant genetic model (combined GT and TT genotypes versus GG as the baseline group) and per-allele comparison. The per-allele hazard ratio of the rare allele was compared between cases and controls by assigning scores of 0, 1, and 2 to common homozygote, heterozygote, and rare homozygote, respectively. This is equivalent to a comparison of the T allele versus G allele as the baseline. To deal with possible nonlinearity, continuous predictors (age, SBP, DBP, BMI, Waist to hip ratio, HDL, LDL, and total cholesterol) were modeled by using fractional polynomials.14
All analyses were performed for any stroke, ischemic stroke, CHD, and MI. A value of P<0.05 (2-sided) was considered significant. In addition, interactions between genotype and hypertension on the risk of events were tested on both the multiplicative and additive scales. First, the multiplicative interaction between genotype and hypertension was assessed by the use of a product term and the Wald χ2 test. Furthermore, a 2-by-4 table used for presenting and analyzing data on gene-environmental interaction as proposed previously15 and synergy index of multiplicativity (SIM) was calculated [SIM=HRADD1×HTN/(HRADD1×HRHTN)]. A synergy index of multiplicativity of one means no interaction on a multiplicative scale, whereas a synergy index of multiplicativity greater than 1 means that the joint effect of gene and hypertension is larger than expected from the product of their individual effects. An additive interaction was assessed using the synergy index of additivity (SIA) by the delta method described by Hosmer and Lemeshow.16 The SIA equal to 1 means no interaction on additive scale. The modeling strategies included the assessment of the interaction without and with adjustment for age. As it has been shown previously, biological interaction better reflected by an additive interaction than a multiplicative interaction.17–19
In Table 1 general characteristics of the subcohort (n=1522) are given. 1452 (95.4%) of participants had the common genotype (GG), 68 (4.5%) were heterozygous (GT), and 2 (0.1%) were homozygous (TT) for the tryptophan allele. The genotype distribution was in Hardy-Weinberg equilibrium (χ2 exact test=1.63; P=0.20). Characteristics of stroke and CHD cases and subcohort are shown in Table 1.
None of the conventional risk factors was significantly related to the Gly460Trp polymorphism (Table 1). There was a tendency for those subjects with tryptophan allele to have a higher weight and systolic and diastolic blood pressure. Age, hypertension, hypercholesterolemia, systolic blood pressure, diastolic blood pressure, waist to hip ratio, and HDL and LDL were significantly related with future stroke. All risk factors (except weight) were statistically significantly related with future CHD (Table 1). Median follow-up time for the subcohort was 4.3 years, with a total of 6525 person years. The actual follow-up in the baseline cohort of 15 236 women was 64 768 person years.
The comparison between the models using linear and fractional polynomials of age, SBP, DBP, BMI, waist to hip ratio, HDL, LDL, and total cholesterol provided no evidence for a departure from linearity (P>0.21 for all; null hypothesis: the association is linear) for different events. In the final multivariable fractional polynomial model for any stroke risk, the variant (Table 2), age (HR=1.08; 95% CI, 1.03 to 1.12; P=0.0004), hypertension (HR=1.87; 95% CI, 1.14 to 3.07; P=0.0139), and hypercholesterolemia (HR=2.73; 95% CI, 1.42 to 5.26; P=0.0026), remained in the model. We considered these 3 variables for further adjustments in the analyses (model 1).
Table 2 presents hazard ratios (HR) of events in tryptophan allele carriers compared with noncarriers, under dominant and per-allele genetic models. Genotype distribution among any stroke cases and ischemic stroke subjects were 65 (88%), 9 (12%), 0 (0%), and 41 (84%), 8 (16%), 0 (0%), respectively for GG, GT, and TT genotypes. Variant carriers were found to have an increased risk of any stroke under dominant model (HR=2.76; 95% CI, 1.32 to 5.77; P=0.007), and under per-allele comparison (HR=2.49; 95% CI, 1.28 to 4.86; P=0.007) compared with common type homozygotes, which remained statistically significant after adjustment for age, hypertension, and hypercholesterolemia (model 1) or after cumulative adjustment for age, systolic blood pressure, body mass index, serum total cholesterol, current smoking, alcohol intake, use of antihypertensive drugs, and diabetes (model 2; Table 2). Higher risks were found for ischemic stroke among variant carriers under dominant model (HR=3.86; 95% CI, 1.74 to 8.57; P=0.0009) and per-allele comparison (HR=3.35; 95% CI, 1.67 to 6.72; P=0.0007), which did not attenuate after adjustments (Table 2).
Genotype distribution among CHD and MI cases were 199 (94.8%), 11 (5.2%), 0 (0%), and 70 (98.6%), 1 (1.4%), 0 (0%), respectively for GG, GT, and TT genotypes. There was no statistically significant association between the variant and CHD under dominant genetic model (HR=1.10; 95% CI, 0.58 to 2.09; P=0.77), which did not change after further adjustments (Table 2). A same result was found for per-allele comparison (Table 2). Likewise, no association was found for MI under different genetic models (Table 2). When we considered all stroke or CHD cases, a higher risk was seen for the variant carriers, although it did not reach significant level under both genetic models (Table 2).
We investigated the interaction between hypertension and the Gly460Trp polymorphism, first, in relation to any stroke and ischemic stroke. When we considered common definition of hypertension (BP≥140/90 or use of antihypertensive drugs) we did not find a significant interaction for any stroke under multiplicative scale (age-adjusted SIM=1.03; 95% CI, 0.24 to 4.49) or additive scale (age-adjusted SIA=1.53; 95% CI, 0.01 to 3.05), and for ischemic stroke under multiplicative scale (age-adjusted SIM=1.43; 95% CI, 0.25 to 8.04) or additive scale (age-adjusted SIA=2.37; 95% CI, 0.75 to 4.00). Using stage 2 of systolic hypertension, we found a higher risk of ischemic stroke for the variant carriers with systolic hypertension (age-adjusted HR=10.88; 95% CI, 3.75 to 31.50; P<0.001) compared with common-type homozygotes without hypertension (Table 3). We found a significant synergy index of additivity equal to 3.8 (1.90 to 4.66), which also remained significant after adjustment for age (SIA=3.04; 95% CI, 1.55 to 4.53). The synergy index of 3 indicated a departure from an additive relation. For any stroke we also found an increased risk for the variant carriers with systolic hypertension (age-adjusted HR=6.69; 95% CI, 2.38 to 18.77; P<0.001) compared with common-type homozygotes without hypertension (Table 3). Under the additive scale we found a borderline significant synergy index equal to 2.03 (0.63 to 3.43) adjusted for age. These results indicated that there was an additive interaction between the α-adducin Gly460Trp variant and systolic hypertension on the risk of ischemic stroke. Before and after adjustment for age, however, we did not find a modifying effect for SBP at baseline as a continuous variable on the association of the variant and the risk of ischemic stroke (P=0.81) and any stroke (P=0.85). Evaluating interaction between the variant and SBP, hypertension (common definition), and systolic hypertension under additive and multiplicative scales for the risk of CHD and MI did not show any modification effect (0.9<P<0.5).
The results of this prospective study among women aged 49 to 70 years, using a case-cohort approach in a cohort of 15 236 initially healthy women, show increased risks of any stroke and ischemic stroke in women carrying the α-adducin Gly460Trp variant. The findings suggest that the Gly460Trp mutation in the α-adducin gene affects stroke risk independent from blood pressure levels. In addition, presence of the mutation amplifies the risk of ischemic stroke associated with systolic hypertension.
To appreciate the findings some issues need to be addressed. Data collection was prospective, before the diagnosis of cerebrovascular diseases and equal for all participants. This assures the comparability of the cases and the randomly selected controls.10 For a multifactorial trait, like stroke, this provides a valid approach to evaluate the relation between genetic factors and the risk of stroke while taking into account coexisting and risk modifying factors. In this study, prevalent cases of cardiovascular disease (ICD-9; 390 to 459) at baseline were excluded from the analyses to prevent introducing bias attributable to potentially selective survival. Additional strengths are the comprehensive data and sample collection, the virtually complete hospital admission and mortality follow-up that we have at our disposals for the entire cohort, and the design of the study which is a case-cohort study which combines the advantages of cohort studies (multiple outcomes and time-dependent covariates) with those of case-control analyses (fewer subjects), thus being more efficient than cohort studies. It should be noted that the case-control studies might have been affected by selection bias because only nonfatal cases were included, which is not the case in this study, because of our end point definitions. Moreover, we did not have misclassification of exposure (genotypes), which when present in general leads to bias toward the null, by using standard laboratory protocols and performing a random double-check to detect potential genotyping errors and the α-adducin Gly460Trp genotypes were in the Hardy-Weinberg equilibrium. Limitations of this study are the relatively short period of follow-up and the modest number of cases. Moreover, because this cohort exclusively comprises Dutch women, these results cannot be generalized to men or other ethnic groups, for whom rates of the events or the allele frequency are known to differ.
Our finding of an increased stroke risk in the presence of the α-adducin Gly460Trp polymorphism agrees with the recent findings, in which the risk of cerebrovascular disease was significantly increased within 6471 subjects of the Rotterdam study,5 a prospective cohort of both men and women aged 55 years or above, for any stroke (498 cases) or ischemic stroke (291 cases) but not for hemorrhagic stroke (47 cases). In the Rotterdam study, in the whole population after adjustment for age and sex, higher risks for any stroke and ischemic stroke were found for the variant carriers, but did not provide a gender stratification.5 Also in a case control study of 117 stroke subjects among hypertensives that were pharmacologically treated, a modest association was found for the variant allele carriers versus GG genotypes.4 Our result on the risk of stroke in women expands these findings. However, contradictory results have been published.6,7 In the ARIC study, a prospective cohort of both men and women aged 45 to 64 years which was carried out in almost 50% of African Americans, no association was found between the variant and risk of ischemic stroke (231 ischemic strokes).7 No report exists regarding potential men and women differences in relation to the risk of stroke among the variant carriers and noncarriers. Our findings with respect to MI and CHD are in accordance with results from previous studies.4,6 In a large prospective study, the α-adducin Gly460Trp polymorphism was not associated with the risk of MI, coronary events, stroke, and all cardiovascular events.6 However, in this population-based study, Li et al found for the first time a significant interaction between the variant and hypertension in relation to total and cardiovascular mortality and all cardiovascular, cardiac, and coronary events.6 Our study, as well as the Rotterdam study, provided evidence for a significant interaction between the α-adducin Gly460Trp variant and hypertension in relation to risk of ischemic stroke.5
The mutation in the α-adducin gene has been most widely studied in relation to blood pressure levels.1,2,20 Overall, the published data seem to suggest the carriers of the mutation have higher blood pressure levels and increased rates of hypertension, although not all studies were able to confirm an effect on blood pressure1,2,21 attributed partially to confounders of ethnicity, lifestyle and other environmental factors or biases vary between studies. As a significant correlation exists between SBP elevation and the increased risk of stroke, adequate control of SBP has important clinical benefits, particularly in the aging population.22,23 Therefore, our finding, as well as recent findings,5,6 on presence of a significant interaction between the variant and systolic hypertension in relation to risk of ischemic stroke could benefit the clinical setting, if confirmed in other ethnicities.
We studied a single nucleotide polymorphism in the α-adducin gene where the amino acid sequence is changed from glycine to tryptophan at residue 460. Adducin is a heterodimeric cell membrane skeleton protein composed of α and β subunits. Genetic variation in the α-adducin protein may affect ion transport through modification of actin cytoskeleton assembly and changes in sodium pump activity.24 The role of genes in susceptibility for stroke has particularly been examined in rats, in particularly spontaneously hypertensive (SHR) and stroke-prone strains. Findings in these models have suggested that stroke is attributable to both blood pressure–dependent and independent genetic factors. With similar blood pressure levels, SHR and stroke-prone SHR have a different risk of stroke, indicating a role for blood pressure independent genetic risk factors.25 In our study, we found an association between the variant and stroke after adjustment for blood pressure and hypertension. This could amplify the role of this variant on stroke, independent of blood pressure. The possible underlying mechanism of the prediction of this genetic variant for stroke, as it was proposed, is salt sensitivity which has been related to cardiovascular events, independent of blood pressure.3
In conclusion, this study shows that in healthy Dutch women free from previous cardiovascular diseases, the α-adducin Gly460Trp polymorphism was related to later in life risk of stroke and ischemic stroke. The risk of ischemic stroke associated with the α-adducin Gly460Trp polymorphism was higher among women with systolic hypertension. Conversely, the mutation appeared to modify the risk relation between elevated blood pressure and stroke.
The present study provides further evidence that the Gly460Trp polymorphism as a functional polymorphism within a candidate gene, the α-adducin, acts in the etiology of stroke. Because demonstration of gene environment interaction in the pathogenesis of complex phenotypes, such as stroke, is increasingly prevalent, especially with the strength of prospective studies, these findings with appropriate pharmacogenetic evaluation could benefit therapeutic indices. As it has been found that in carriers of the α-adducin variant, diuretic therapy was associated with a lower risk of combined MI and stroke than other antihypertensive therapies,4 those hypertensive subjects who are carriers of the variant, may be especially likely to benefit more from diuretic therapy.
We are grateful to the participants of the Prospect-EPIC study. We thank all field workers, laboratory technicians, and skillful contributions to the data collection and all those who contributed to this study. The first author also acknowledges Iranian Ministry of Health and Medical Education (FN12265) for the support of a PhD program at the Julius Center for Health Sciences and Primary Care.
Sources of Funding
The first author was supported by Iranian Ministry of Health and Medical Education (FN12265) for a PhD program at the Julius Center for Health Sciences and Primary Care. The Prospect-EPIC study was funded by “Europe Against Cancer” Programme of the European Commission (SANCO).
- Received February 22, 2008.
- Revision received March 10, 2008.
- Accepted April 8, 2008.
Bianchi G, Ferrari P, Staessen JA. Adducin polymorphism: detection and impact on hypertension and related disorders. Hypertension. 2005; 45: 331–340.
Cusi D, Barlassina C, Azzani T, Casari G, Citterio L, Devoto M, Glorioso N, Lanzani C, Manunta P, Righetti M, Rivera R, Stella P, Troffa C, Zagato L, Bianchi G. Polymorphisms of alpha-adducin and salt sensitivity in patients with essential hypertension. Lancet. 1997; 349: 1353–1357.
Psaty BM, Smith NL, Heckbert SR, Vos HL, Lemaitre RN, Reiner AP, Siscovick DS, Bis J, Lumley T, Longstreth WT Jr, Rosendaal FR. Diuretic therapy, the alpha-adducin gene variant, and the risk of myocardial infarction or stroke in persons with treated hypertension. JAMA. 2002; 287: 1680–1689.
van Rijn MJ, Bos MJ, Yazdanpanah M, Isaacs A, rias-Vasquez A, Koudstaal PJ, Hofman A, Witteman JC, van Duijn CM, Breteler MM. Alpha-adducin polymorphism, atherosclerosis, and cardiovascular and cerebrovascular risk. Stroke. 2006; 37: 2930–2934.
Li Y, Thijs L, Kuznetsova T, Zagato L, Struijker-Boudier H, Bianchi G, Staessen JA. Cardiovascular risk in relation to alpha-adducin Gly460Trp polymorphism and systolic pressure: a prospective population study. Hypertension. 2005; 46: 527–532.
Morrison AC, Doris PA, Folsom AR, Nieto FJ, Boerwinkle E. G-protein beta3 subunit and alpha-adducin polymorphisms and risk of subclinical and clinical stroke. Stroke. 2001; 32: 822–829.
Boker LK, van Noord PA, van der Schouw YT, Koot NV, Bueno de Mesquita HB, Riboli E, Grobbee DE, Peeters PH. Prospect-EPIC Utrecht: study design and characteristics of the cohort population. European Prospective Investigation into Cancer and Nutrition. Eur J Epidemiol. 2001; 17: 1047–1053.
Herings RM, Bakker A, Stricker BH, Nap G. Pharmaco-morbidity linkage: a feasibility study comparing morbidity in two pharmacy based exposure cohorts. J Epidemiol Community Health. 1992; 46: 136–140.
Prentice RL. A case-cohort design for epidemiologic cohort studies and disease prevention trials. Biometrika. 1986; 73: 1–11.
Izzo JL Jr, Levy D, Black HR. Clinical Advisory Statement. Importance of systolic blood pressure in older Americans. Hypertension. 2000; 35: 1021–1024.
Cheng S, Grow MA, Pallaud C, Klitz W, Erlich HA, Visvikis S, Chen JJ, Pullinger CR, Malloy MJ, Siest G, Kane JP. A multilocus genotyping assay for candidate markers of cardiovascular disease risk. Genome Res. 1999; 9: 936–949.
Botto LD, Khoury MJ. Commentary: facing the challenge of gene-environment interaction: the two-by-four table and beyond. Am J Epidemiol. 2001; 153: 1016–1020.
Knol MJ, van dT, I, Grobbee DE, Numans ME, Geerlings MI. Estimating interaction on an additive scale between continuous determinants in a logistic regression model. Int J Epidemiol. 2007; 36: 1111–1118.
Rothman KJ, Greenland S, Walker AM. Concepts of interaction. Am J Epidemiol. 1980; 112: 467–470.
Niu T, Xu X, Cordell HJ, Rogus J, Zhou Y, Fang Z, Lindpaintner K. Linkage analysis of candidate genes and gene-gene interactions in Chinese hypertensive sib pairs. Hypertension. 1999; 33: 1332–1337.
Staessen JA, Fagard R, Thijs L, Celis H, Arabidze GG, Birkenhager WH, Bulpitt CJ, de Leeuw PW, Dollery CT, Fletcher AE, Forette F, Leonetti G, Nachev C, O'Brien ET, Rosenfeld J, Rodicio JL, Tuomilehto J, Zanchetti A. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. The Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Lancet. 1997; 350: 757–764.
Volpe M, Iaccarino G, Vecchione C, Rizzoni D, Russo R, Rubattu S, Condorelli G, Ganten U, Ganten D, Trimarco B, Lindpaintner K. Association and cosegregation of stroke with impaired endothelium-dependent vasorelaxation in stroke prone, spontaneously hypertensive rats. J Clin Invest. 1996; 98: 256–261.