impact of hypertension and renin-angiotensin system inhibitors in aortic stenosis

Impact of hypertension and renin–angiotensin systeminhibitors in aortic stenosisRomain Capoulade, Marie-Annick Clavel, Patrick Mathieu, Nancy Cot�e, Jean G. Dumesnil, Marie Arsenault,�Elisabeth B�edard and Philippe Pibarot

Institut Universitaire de Cardiologie et de Pneumologie de Qu�ebec/Qu�ebec Heart & Lung Institute, Laval University, Qu�ebeccity, QC, Canada

ABSTRACT

Background Experimental studies revealed that renin–angiotensin system (RAS) could play a crucial role in thepathophysiology of aortic stenosis (AS). The objectives of this study were to examine (i) the impact of hyper-tension on AS progression and clinical events and (ii) the effect of angiotensin-converting enzyme inhibitors(ACEIs) and angiotensin-receptor blockers (ARBs).

Materials and methods In this observational study, we retrospectively analysed clinical and Doppler echo-cardiographic data prospectively collected in 338 patients with AS. Patients were separated into four groups:patients without hypertension and not treated by RAS medication (Ctrl group), patients with hypertension butnot treated by RAS medication (HTN group), patients treated with ACEIs, and patients treated with ARBs. ASprogression rate was assessed by the annualized increase in peak aortic jet velocity.

Results Compared with Ctrl group, patients in HTN group had faster stenosis progression (P = 0�01). Patients onARBs had slower AS progression compared with Ctrl (trend P = 0�10) and HTN (P = 0�002) groups, whereaspatients on ACEIs had similar progression rate compared with Ctrl group (P = NS) but lower compared with HTNgroup (P = 0�02). On multivariable analysis, compared with Ctrl group, HTN group was associated with faster ASprogression rate (P = 0�002), whereas ARBs with slower progression (P = 0�0008). During a mean follow-up of6�2 � 2�4 years, HTN (hazard ratio [HR] = 2�45; P = 0�006) and ACEI (HR = 2�30; P = 0�01) groups were associ-ated with a significant increase in all-cause mortality compared with Ctrl group, whereas ARB group (HR: 0�89;P = 0�80) not. In multivariable analysis, HTN and ACEI groups remained associated with increased mortality.

Conclusions Hypertension is associated with significantly faster stenosis progression and higher incidence ofclinical events in patients with AS. ARBs but not ACEs were found to abolish the increased risk of mortalityassociated with hypertension.

Keywords Angiotensin-receptor blocker, aortic stenosis, Doppler echocardiography, hypertension, renin–angiotensin system.

Eur J Clin Invest 2013; 43 (12): 1262–1272

Calcific aortic stenosis (AS) is the most common cardiovascular

disease in developed countries after coronary artery disease

(CAD) and systemic arterial hypertension [1,2]. Recent studies

suggest that AS is not a degenerative disease resulting from dec-

ades of repetitive mechanical stress, but rather an active disease

related to atherosclerosis [3–5] and mediated by a complex inter-

action between chronic inflammation, lipid deposition and oxi-

dation, osteoblastic processes and calcification [3,6–10]. However,

no medical therapies have been proven to slow the progression of

AS or reduce valve-related events rate in patients with AS, and

surgical or transcatheter aortic valve replacement (AVR) remains

the sole efficient form of treatment for symptomatic severe AS.

A high proportion (30–70%) of patients with calcific AS

concomitantly have hypertension [11,12], and a recent study

[13] reported that this frequent comorbidity is associated

with increased risk of cardiovascular events in the AS

population. However, little is known about the impact of

hypertension on AS progression and about the potential

protective effects of antihypertensive medications. The objec-

tive of this study was to examine the impact of hypertension

on AS progression rate and clinical events and to assess

the effect of angiotensin-converting enzyme inhibitors

(ACEIs) and angiotensin-receptor blockers (ARBs) on these

outcomes.

1262 ª 2013 Stichting European Society for Clinical Investigation Journal Foundation. Published by John Wiley & Sons Ltd

DOI: 10.1111/eci.12169

ORIGINAL ARTICLE

Materials and methods

Patient populationPatients with aortic valve sclerosis and peak aortic jet velocity

>2�0 m/s, and at least two Doppler echocardiography exams

separated by at least 6 months were prospectively entered in a

computerized database between 1999 and 2007. Patients were

excluded if they had symptomatic AS, moderate to severe

aortic regurgitation or significant mitral valve disease, left

ventricular (LV) ejection fraction <50% and incomplete infor-

mation about the medications. Two patients had both ACEIs

and ARBs and were excluded. According to these criteria, 338

patients were included in this study, and the clinical and

echocardiographic data of these patients were retrospectively

analysed. The study was approved by the Ethics Committee of

the Quebec Heart and Lung Institute.

Clinical dataClinical data were recorded during baseline exam and

included age, gender, height, weight, systolic and diastolic

blood pressure, diagnosis of hypertension, diabetes, hyper-

lipidemia, CAD, and chronic obstructive pulmonary disease

(COPD).

Patients were separated into four groups according to the

diagnosis of hypertension and the use of medication target-

ing the RAS: (1) no hypertension group (i.e. Ctrl group)

composed of 92 patients (27%) with no diagnosis of hyper-

tension and no RAS inhibiting medication; (2) hypertension

group (HTN group) composed of 77 patients (23%) with

diagnosis of hypertension but not treated with RAS medica-

tion; (3) ACEI group composed of 113 patients (33%) treated

with ACEIs; (4) ARB group composed of 56 patients (17%)

treated with ARBs.

In the ACEI group, 38 (33%) patients were under ramipril,

27 (24%) under lisinopril, 17 (15%) under quinapril, 11 (10%)

under fosinopril, 9 (8%) under perindopril, and the 11 (10%)

remaining patients were under enalapril, cilazapril, captopril,

benazepril or trandolapril.

In the ARB group, 22 (39%) patients were under losartan,

11 (20%) under irbesartan, 11 (20%) under calsartan, 8 (14%)

under cardesartan and 4 (7%) under telmisartan.

Doppler echocardiographic data

Aortic valve morphology and function. The aortic valve

morphology (i.e. bicuspid or tricuspid) was recorded. The

indices of AS severity included peak aortic jet velocity (VPeak),

peak and mean transvalvular gradients, and the aortic valve

area (AVA). The AVA could not be determined in 18% (n = 62)

of patients due to subvalvular flow acceleration or inadequate

measure of LV outflow tract diameter.

Left ventricular geometry and function. left ventricular

minor axis internal dimension, posterior wall thickness and

interventricular septal thickness were measured at end-

diastole according to the recommendations of the American

Society of Echocardiography [14]. The relative wall thickness

was calculated by dividing the sum of the LV posterior wall

and interventricular septal thicknesses by the LV internal

dimension. Left ventricular mass was calculated with the

corrected formula of the American Society of Echocardiog-

raphy and was indexed to body surface area (BSA) [15,16].

The LV ejection fraction was calculated with the biplane

Simpson method.

Global LV hemodynamic load. As a measure of global LV

hemodynamic load, we calculated the valvulo-arterial imped-

ance (Zva) [12,16]: Zva = (SBP + DPmean)/SVi, where SBP is the

systolic blood pressure, ΔPmean is the mean transvalvular

gradient and SVi is the stroke volume indexed to BSA.

Study end-pointsThe primary end-point for this study was the progression rate

of valve stenosis (i.e. progression rate of peak aortic jet velocity)

measured by Doppler echocardiography. To account for inter-

individual differences in follow-up duration, annualized

change in peak aortic jet velocity was calculated by dividing the

difference between first and last measurements by the time

between examinations. The secondary end-points were all-

cause mortality, cardiovascular mortality and the composite of

death or AVR motivated by the development of symptoms or

LV systolic dysfunction. The secondary end-points were

obtained from patient’s medical record or Qu�ebec provincial

death registry.

Cardiovascular mortality was classified according to the

definitions of the American Heart Association [17]. The last

update of the clinical events was performed in July 2011.

Statistical analysisContinuous data were expressed as mean � standard devia-

tion and compared between the four study groups using a one-

way analysis of variance (ANOVA) followed by Tukey’s post hoc

test. Categorical data were expressed as a percentage and

compared with the chi-square test or Fischer exact tests when

appropriate. Multivariable linear regression analysis was per-

formed to identify the independent predictors of the primary

outcome variable: that is, AS progression rate. We entered in

the multivariable model: (i) the variables with P value <0�10 in

individual analysis and (ii) the other traditional cardiovascular

risk factors (i.e. age and gender).

To adjust for covariate and number of patient differences

between groups that may lead to biased estimates of treatment

European Journal of Clinical Investigation Vol 43 1263

IMPACT OF HTN AND RAS INHIBITORS IN AS

effect, we computed, for each patient, an inverse-probability-

of-treatment-weight estimator. This weighted variable, unlike

traditional propensity score, can be implemented when there

are more than two treatment groups, such as in the present

study. Similarly to the propensity score method for two treat-

ment groups, the weighted variable method evaluates the

probability of patients being included in one of the four groups

and calculates weights based on factors hypothesized to influ-

ence group selection (i.e. age, CAD, hyperlipidemia, diabetes,

use of calcium channel blockers and b-blockers). This variable

was then included in weighted linear and Cox multivariable

models.

Kaplan–Meier curves and log-rank tests of the time-to-event

data were used to assess the effect of groups on the composite

end-point of AVR or death, all-cause mortality and cardiovas-

cular mortality. The predictors of all-cause and cardiovascular

mortality were assessed with the use of individual and multi-

variable Cox proportional hazard analyses. The traditional risk

factors of mortality and variables with P value <0�10 were

entered in the multivariate Cox model (i.e. age, CAD, COPD,

hyperlipidemia, diabetes and Zva). The impact of AVR during

follow-up was analysed with AVR entered as a time-dependent

covariate in the Cox proportional hazard models for all-cause

and cardiovascular mortality. A P value <0�05 was considered

to be statistically significant.

Reporting of the study conforms to STROBE statement along

with references to STROBE and the broader EQUATOR

guidelines [18].

Results

Baseline characteristics of the study groupsBaseline characteristics of the 338 patients included in this

study were presented in Table 1.

Compared with Ctrl group, patients of the three others

groups (i.e. HTN, ACEI and ARB groups) were older and had

significant higher prevalence of CAD, COPD, hyperlipidemia

and diabetes (Table 1). As expected, diagnosis of hypertension

and systolic blood pressure was higher in these three groups

compared with the Ctrl group. Furthermore, patients with

medication targeting the RAS (i.e. ACEI and ARB groups) had

higher prevalence of hyperlipidemia and diabetes compared

with HTN group.

After adjustment for the weighted variable, the differences

between groups were no longer significant except for the

presence of hypertension, as expected by study design

(Table 1).

With regards to baseline Doppler echocardiographic data,

indices of AS severity and LV ejection fraction were similar in

all groups, but patients in Ctrl group had higher prevalence of

bicuspid aortic valve phenotype, whereas patients in HTN and

ACEI groups had higher Zva compared with Ctrl group

(Table 2). LV mass index was significantly higher in HTN and

ACEI groups compared with Ctrl group (Table 2).

Comparison of AS progression rate between groupsWhen compared to patients of Ctrl group (annualized pro-

gression of VPeak: +0�17 � 0�20 m/s/year), AS progression rate

was significantly (P = 0�01) faster in patients of HTN group

(+0�26 � 0�23 m/s/year) and tended (P = 0�10) to be slower in

ARB group (+0�12 � 0�20 m/s/year) (Table 2 and Fig. 1).

Similar results were found with the other indices of AS severity

(Table 2).

After adjustment for the weighted variable, patients in HTN

group had significantly faster AS progression (P = 0�001) andpatients on ARBs had slower progression (P = 0�0006) com-

pared with the Ctrl group (Fig. 1). Patients on ACEIs had

similar progression rate compared with those in Ctrl group

(P = 0�96).

Predictors of AS progression rateIn multivariable analysis, after adjustment for variables asso-

ciated with faster AS progression, baseline factors indepen-

dently associated with faster AS progression rate were presence

of diabetes (P = 0�03), higher VPeak (P = 0�002) and HTN group

(P = 0�002), whereas ARB medication was associated with

slower stenosis progression rate (P = 0�0008) compared with

Ctrl group (Table 3, Model #1). After further adjustment for the

weighted variable, being in HTN group was independently

associated with faster AS progression (P = 0�002), whereas

being in ARB group was associated with slower AS progression

(P = 0�001) compared with Ctrl group (Table 3, Model #2).

Predictors of clinical outcomesDuring a mean follow-up time of 6�2 � 2�4 years, 73 deaths

occurred of which 34 were related to cardiovascular causes

(heart failure: 16 patients; acute pulmonary oedema: three

patients; stroke: seven patients; myocardial infarction: five

patients, cardiac arrhythmia: three patients). Moreover, 153

AVRs motivated by the development of symptoms or LV

systolic dysfunction were performed during this period. The

30-day mortality following AVR was 2% (three patients).

When analysing the combined end-point of death or AVR

motivated by the development of symptoms or LV systolic

dysfunction, the 8-year event-free survival was 41 � 6% in the

HTN group, 52 � 7% in the Ctrl and ACEI groups, and

63 � 8% in the ARB group, respectively (Fig. 2, Panel A).

Compared with HTN group, ARB group had a twofold

reduction in the occurrence of AVR or death (HR: 0�51; 95% CI

0�28–0�89; P = 0�01).Eight-year overall survival was lower in the HTN (66 � 6%)

and ACEI (63 � 8%) groups compared with that in the Ctrl


R. CAPOULADE ET AL. www.ejci-online.com

(84 � 5%) and ARB (86 � 5%) groups (P = 0�006, Fig. 2,Panel B). Compared with the Ctrl group, ARB group had

similar mortality risk (HR: 0�89; 95% CI 0�11–2�25; P = 0�80),whereas patients in the HTN group had a 2�45-fold (HR: 2�45;95% CI 1�28–4�92; P = 0�006) and those in the ACEI group a

2�30-fold (HR: 2�30; 95% CI 1�21–4�62; P = 0�01) increase in

all-cause mortality (Fig. 3, Panel A).

After adjustment for the others baseline variables associated

with all-causes mortality, being in the HTN group (HR: 2�27;95% CI 1�09–4�71; P = 0�03) or ACEI group (HR: 2�46; 95% CI

1�16–5�20; P = 0�02) was associated with increased risk of

all-cause mortality compared with the Ctrl group. On the other

hand, being in the ARB group was not associated with

increased mortality (HR: 0�57; 95% CI 0�27–2�06; P = 0�57)(Fig. 3, Panel B). After further adjustment for the weighted

variable, there was a trend for an association between ARB

medication and reduced mortality compared with the Ctrl

group (HR: 0�43; 95% CI 0�17–1�12; P = 0�08), whereas HTN and

ACEI groups were no longer significantly associated with

increased risk of mortality (Fig. 3, Panel C). Similar results

were obtained when we added AVR as a time-dependant

variable in these multivariable models.

Table 1 Baseline clinical characteristics of the study population

Whole cohort

(n = 338)

Ctrl group

(n = 92)

27%

HTN group

(n = 77)

23%

ACEI group

(n = 113)

33%

ARB group

(n = 56)

17% P value†Weighted

P value‡

Clinical data

Age, years 69 � 14 61 � 18 73 � 12* 71 � 11* 72 � 11* <0�0001 NS

Male gender, % 52% 55% 52% 52% 46% NS NS

Height, cm 164 � 10 165 � 10 163 � 10 164 � 10 164 � 10 NS NS

Weight, kg 73 � 14 71 � 12 74 � 14 75 � 15 74 � 16 NS NS

Body surface area, m2 1�80 � 0�20 1�78 � 0�18 1�79 � 0�20 1�81 � 0�21 1�81 � 0�24 NS NS

Body mass index, kg/m2 27 � 4 26 � 4 28 � 5 28 � 5* 27 � 4 0�04 NS

History of hypertension, % 73 0 100* 100* 100* <0�0001 <0�0001Systolic blood pressure, mm Hg 137 � 23 128 � 19 143 � 24* 138 � 24* 139 � 22* 0�0002 0�01Diastolic blood pressure, mm Hg 73 � 11 75 � 8 75 � 11 71 � 11 73 � 11 NS NS

Coronary artery disease, % 36 17 40* 48* 37* <0�0001 NS

Chronic obstructive pulmonary

disease, %

17 9 25* 17 20 0�05 NS

Hyperlipidemia, % 56 39 53 69*,¶ 61* 0�0003 NS

Diabetes, % 22 10 14 34*,¶ 30*,¶ <0�0001 NS

Medication data

b-blockers, % 34 12 36* 51*,¶ 34*,§ <0�0001 NS

Calcium channel blockers, % 36 13 44* 38* 55*,§ <0�0001 NS

Diuretics, % 32 0 45* 41* 48* <0�0001 <0�0001Statin, % 49 35 39 65*,¶ 52* 0�0001 NS

Antidiabetics, % 18 8 10 29*,¶ 21* 0�0002 NS

Antiplatelet, % 53 30 53* 68*,¶ 59* <0�0001 NS

Nitrates, % 19 5 22* 23* 27* 0�001 NS

Values are mean � SD.

Ctrl, control; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; HTN, hypertension.†P value of the one-way ANOVA.‡P value of the one-way ANOVA after adjustment for the weighted variable. The following symbols indicate the significance of the Tukey’s post hoc test.

*P < 0�05 from ‘Ctrl Group’, ¶P < 0�05 from ‘HTN group’, §P < 0�05 from ‘ACEI group’.



When using the ARB group as the referent group in the

multivariable analysis, patients in the HTN group (HR: 3�04;95% CI 1�23–7�57; P = 0�02) and those in the ACEI group (HR:

3�30; 95% CI 1�33–8�19; P = 0�01) were at increased risk of

mortality, whereas the patients in the Ctrl group were not (HR:

1�34; 95% CI 0�49–3�71; P = 0�57; Fig. 3, Panel B). Similar results

were obtained after adjustment for the weighted variable

(Fig. 3, Panel C).

The analysis of cardiovascular mortality showed that patients

in the Ctrl and ARB groups have similar 8-year survival

(94 � 3% and 95 � 3%, respectively), and this survival rate

was significantly higher than that of patients of the ACEI and

HTN groups (79 � 6% and 83 � 5%, respectively) (P = 0�03,Fig. 2, Panel C). When using the Ctrl group as the reference,

being in the ACEI (HR: 3�19; 95% CI 1�22–9�92; P = 0�02) orHTN (HR: 2�99; 95% CI 1�11–9�39; P = 0�03) groups was

Table 2 Doppler echocardiographic data

Whole cohort

(n = 338)

Ctrl group

(n = 92)

27%

HTN group

(n = 77)

23%

ACEI group

(n = 113)

33%

ARB group

(n = 56)

17% P value†Weighted

P value‡

Baseline Doppler echocardiographic data

Bicuspid aortic valve, % 16 37 5* 7* 13* <0�0001 0�09Peak aortic jet velocity, m/s 2�85 � 0�47 2�88 � 0�47 2�91 � 0�47 2�81 � 0�48 2�76 � 0�44 NS NS

Peak transvalvular gradient,

mm Hg

33 � 11 34 � 11 35 � 11 33 � 11 31 � 10 NS NS

Mean transvalvular gradient,

mm Hg

19 � 7 19 � 7 20 � 7 19 � 7 17 � 6 NS NS

Aortic valve area, cm2 1�15 � 0�22 1�17 � 0�22 1�11 � 0�21 1�15 � 0�24 1�17 � 0�22 NS NS

Indexed aortic valve area,

cm2/m2

0�64 � 0�13 0�66 � 0�13 0�62 � 0�12 0�64 � 0�14 0�66 � 0�12 NS NS

Degree of aortic stenosis severity

Mild, % 34 34 27 35 37 NS NS

Moderate, % 44 50 46 40 41

Severe, % 22 16 27 25 22

Relative wall thickness ratio 0�48 � 0�10 0�45 � 0�10 0�49 � 0�10 0�48 � 0�11 0�49 � 0�09 0�09 NS

LV mass index, g/m2 104 � 26 97 � 25 109 � 28* 107 � 27* 102 � 23 0�02 NS

Valvulo-arterial impedance

mm Hg/ml/m2

3�9 � 0�8 3�6 � 0�7 4�2 � 0�9* 4�0 � 0�9* 3�9 � 0�7 <0�0001 0�001

LV ejection fraction, % 65 � 7 66 � 6 64 � 7 64 � 7 65 � 7 NS NS

Parameters of stenosis progression rate

Peak aortic jet velocity, m/s/year 0�19 � 0�21 0�17 � 0�20 0�26 � 0�23* 0�18 � 0�18¶ 0�12 � 0�20¶ 0�0008 0�0003Peak transvalvular gradient,

mm Hg/year

5�2 � 5�9 5�0 � 6�6 7�0 � 6�1* 4�9 � 5�3¶ 3�4 � 5�4¶ 0�006 0�001

Mean transvalvular gradient,

mm Hg/year

3�3 � 3�5 3�3 � 3�5 4�2 � 3�5 3�2 � 3�4 2�4 � 3�2¶ 0�03 0�02

Aortic valve area, cm2/year �0�07 � 0�07 �0�07 � 0�06 �0�09 � 0�07 �0�07 � 0�07 �0�04 � 0�08¶ 0�03 0�004

Values are mean � SD.

Ctrl, control; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blockers; HTN, hypertension.†P value of the one-way ANOVA.‡P value of the one-way ANOVA after adjustment for the weighted variable. The following indicate the significance of the Tukey’s post hoc test.

*P < 0�05 from ‘Ctrl Group’, ¶P < 0�05 from ‘HTN group’, §P < 0�05 from ‘ACEI group’.



associated with increased risk of cardiovascular mortality,

whereas being in the ARB group was not (HR: 0�77; 95% CI

0�11–3�58; P = 0�75). In multivariable analysis including the

others predictors of cardiovascular mortality, being in the HTN

(HR: 3�45; 95% CI 1�07–15�37; P = 0�04) or ACEI (HR: 3�57; 95%CI 1�08–16�19; P = 0�04) group was associated with increased

mortality compared with the Ctrl group, whereas there was no

significant increase in mortality risk in the ARB group (HR:

0�76; 95% CI 0�10–4�73; P = 0�77). Further adjustment for AVR

provided similar results.

Discussion

The main findings of this study are as follows: (i) hypertension

is associated with faster stenosis progression and reduced

survival in patients with AS, (ii) ARBs but not ACEs are inde-

pendently associated with reduced progression rate of the ste-

nosis when compared to hypertensive or normotensive patients

not treated with such medication, (iii) patients on ARBs had

better survival compared with hypertensive patients not trea-

ted by RAS medication or treated with ACEIs and they had

similar survival compared with patients with no hypertension

despite substantially older age and worse risk profile.

Impact of hypertension and renin–angiotensinsystem on progression of ASConcomitant arterial hypertension is observed to be present in

a large proportion (30–70%) of patients with calcific AS

[11,12,16,19,20]. Several studies reported an association

between hypertension and increased prevalence of aortic scle-

rosis and AS [21,22]. In the present study that includes a patient

population with a large spectrum of calcific aortic valve disease

(i.e. from aortic sclerosis to severe AS), we found that hyper-

tension is associated with faster disease progression. The acti-

vation of the RAS as well as the repetitive mechanical stress

imposed on aortic valve cusps [23,24] may contribute to accel-

erate the fibro-calcific remodelling of the valve and the pro-

gression of the stenosis in patients with hypertension. The RAS

is a crucial component of hypertension and has intricate links

with obesity and inflammation [25]. On that account, several

studies reported that ACE, chymase and their enzymatic

product angiotensin II are present and active within valves

explanted from patients with AS [4,26]. Angiotensin II has

several pro-inflammatory and pro-fibrotic effects that may

contribute to the pathogenesis of AS [4,26,27]. In this regard, we

have previously reported that increased plasma level of

angiotensin II correlates with higher degree of inflammation in

AS valves [28] and several studies have shown that inflamma-

tion promotes mineralization of valvular tissues [27,29,30].

Moreover, in a previous study of 208 patients who underwent

AVR, we found that ARBs, but not ACEIs, were associated with

a lower aortic valve weight and with less severe fibro-calcific

remodelling of valvular tissues [31]. Furthermore, in an animal

model of AS where abnormalities of plasma lipid and glycemic

profile were reversed by a genetic switch, valvular calcium

burden decreased significantly but markers of pro-fibrotic sig-

nalling remained elevated and valve function remained

impaired [32]. When taken together, these findings suggest that

besides tissue mineralization, fibrosis may also have an

important contribution to the thickening and stiffening of valve

cusps and thus to the progression of the stenosis. Hence, the RAS

may have role not only in the calcification but also and, to a

greater extent, in the fibrosis of the valvular tissues.

Impact of hypertension and renin–angiotensinsystem on outcomes of ASIn a recent substudy from the SEAS trial including patients

with mild to moderate AS [13], hypertension was associated

with a twofold increase in mortality. The present study

corroborates and expands these previous findings. Indeed,

hypertension was independently associated with 2�27-foldincrease in all-cause mortality and a 3�45-fold increase in

cardiovascular mortality in patients with mild to severe AS.

Furthermore, ARBs but not ACEs were able to abolish the

increased risk of mortality associated with HTN in this series.

The pathophysiology of adverse outcomes in AS is, in large

part, determined by the imbalance between the increase in the

global hemodynamic load and the capacity of the left ventricle

to overcome this increase in load [33]. Several studies reported

Figure 1 Comparison between groups of the AS progressionrate. Comparison of the annualized progression rate of peakaortic jet velocity in the following groups: Ctrl group (patientswithout hypertension and nontreated with RAS medication);HTN group (patients with diagnosis of hypertension but nottreated with RAS medication); ACEI group (patients treatedwith ACEIs); and ARB group (patients treated with ARBs). Thenumbers of the top of the bar are mean annualized peak aorticjet velocity �SEM and, between brackets, the number ofpatients in each group. The symbols indicate the significance ofthe Tukey’s post hoc test: *P < 0�05 vs. Ctrl group; ¶P < 0�05 vs.HTN group; †P < 0�05 vs. Ctrl group after adjustment for theweighted variable. Error bars represent SEM.



that the valvulo-arterial impedance, which reflects the global

(i.e. valvular + arterial) hemodynamic load imposed on the left

ventricle, is an independent predictor of mortality and cardiac

events in AS [16,34]. Nevertheless, in the present study,

hypertension remained independently associated with

increased mortality even after adjusting for the valvulo-arterial

impedance, thus suggesting that other factors may underlie the

association between hypertension and worse outcomes in AS.

In particular, the activation of the RAS may contribute to the

acceleration of the valve stenosis per se as discussed above, as

well as to the faster deterioration of the ventricular and arterial

structure and function. To this effect, it has been shown that

RAS and transforming growth factor-beta play a pivotal role in

the development of myocardial fibrosis, hypertrophy and

dysfunction [35]. Hence, patients with AS and concomitant

hypertension may be at higher risk to develop more pro-

nounced LV hypertrophy and myocardial fibrosis, irrespective

of the increased hemodynamic load. In turn, severe LV hyper-

trophy and/or fibrosis have been linked to increased risk of

mortality and morbidity in AS patients [36–38].

Impact of RAS medications on AS progression andoutcomesOne retrospective study showed a strong association between

the use of ACEIs and decreased rate of valve calcification

measured by CT [39], whereas another one found no significant

effect of such medication on hemodynamic progression rate of

AS measured by Doppler echocardiography [40]. The

Table 3 Predictors of AS progression rate

Univariate Multivariable Model #1* Multivariable Model #2*

b coeff. � SE P value b coeff. � SE P value b coeff. � SE P value

Clinical data

Age NS NS NS

Female gender NS NS 2�70 � 1�26 0�03Body Mass Index NS – – – –

History of hypertension NS – – – –

Systolic blood pressure 0�10 � 0�05 0�04 NS NS

Diastolic blood pressure 0�21 � 0�11 0�05 NS NS

Coronary artery disease NS – – – –

Chronic obstructive pulmonary disease NS – – – –

Hyperlipidemia NS – – – –

Diabetes 2�43 � 1�46 0�09 3�72 � 1�65 0�03 5�57 � 1�68 0�001Group

Ctrl group (Referent) – – – – – –

HTN group 7�70 � 1�98 0�0001 7�12 � 2�28 0�002 7�23 � 2�28 0�002ACEI group �0�40 � 1�76 0�82 �0�93 � 1�99 0�64 0�16 � 1�94 0�93ARB group �6�45 � 2�22 0�004 �8�58 � 2�52 0�0008 �8�17 � 2�54 0�001

Doppler echocardiographic data

Bicuspid aortic valve �2�60 � 1�53 0�09 NS NS

Peak aortic jet velocity 0�08 � 0�02 0�0004 0�08 � 0�03 0�002 0�09 � 0�03 0�0005Relative wall thickness ratio 23�5 � 12�0 0�05 NS NS

Valvulo-arterial impedance 2�46 � 0�85 0�004 NS NS

LV ejection fraction NS – – – –

Ctrl, control; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; HTN, hypertension; AS, aortic stenosis; LV, left ventricular.

*Model #1 included variables with P < 0�10 in univariate analysis plus age and gender; Model #2 was Model #1 weighted by propensity score. b coeff. is the raw-

score regression coefficient � SE.



discrepancies between these previous studies [39,40] may be

explained by the fact that the former measured the progression

of valve calcification, whereas the latter assessed the valve

stenosis progression, which may be influenced not only by the

mineralization but also by the fibrosis of the valvular tissues.

Panel (a)

Panel (b)

Panel (c)

Figure 2 Comparison between groups of the time-to-eventcurves for the clinical end-points. This figure shows the event-free survival curves for the composite end-point of aortic valvereplacement or death (Panel A), all-cause mortality (Panel B)and cardiovascular mortality (Panel C). The groups aredescribed in the legend of Fig. 1. The symbols indicate thesignificant difference between groups: *P < 0�05 vs. Ctrl group;†P < 0�05 vs. ARB group. The numbers at the bottom of thegraph represent the number of patients at risk at each follow-upyear. The P value is that of the log-rank test.

Panel (a)

Panel (b)

Panel (c)



In a recent study of 204 randomly selected subjects, the use of

ACEIs or ARBs predicted the lack of progression of aortic valve

sclerosis estimated with the use of the ultrasonic backscatter of

the aortic valve [41]. In a retrospective analysis of a large single-

institution registry [42], the use of ACEIs or ARBs was also

associated with improved survival and lower risk of cardio-

vascular events in patients with AS. However in these previous

studies [41,42], there was no separate analysis to determine the

respective effect of ARBs vs. ACEIs. In the present study, we

found that ARBs but not ACEIs are associated with reduced

stenosis progression and improved survival compared with

patients with hypertension not treated with these medications.

The superiority of ARBs over ACEIs in this context may be due,

at least in part, to the fact that the latter do not block the

angiotensin II-producing enzyme, chymase, which is abun-

dantly expressed in AS valves [26]. Blocking the RAS down-

stream of the cascade with ARBs may thus provide a more

effective mean to inhibit the RAS-mediated effects on the aortic

valve. More complete blockade of the cardiac RAS with ARBs

may also have incremental protective effects on myocardial

structure and function [43]. Hence, whereas ARBs and ACEIs

provided similar benefits in prevention or postmyocardial

infarction trials [44,45], the survival benefit observed in AS

patients treated with ARBs may be due to more efficient inhi-

bition of the progression of both the valve stenosis and the

myocardial fibrosis and dysfunction.

Study limitations

The main limitation of this study is the nonrandomized design.

Furthermore, the data were prospectively collected but retro-

spectively queried. Nevertheless, the associations between ARB

therapy and AS progression and outcome were further rein-

forced following adjustment for treatment selection bias with

the use of the weighted variable method.

The data about the indication, dose, onset and duration of

ACEI or ARB therapy were not available. This limited our

ability to discriminate the class vs. drug vs. dose effects.

Furthermore, selection and follow-up biases may have

affected the results of the study given that only patients with

two consecutive echocardiography exams and available medi-

cation were analysed in this study and that the stenosis pro-

gression may not be linear over time. Nonetheless, patients on

ARBs had slower progression of stenosis and similar survival

compared with patients with no hypertension despite sub-

stantially older age and much worse risk profile. Finally, due to

the small number of patients, especially in ARB group, these

results should be considered hypothesis generating and will

need to be evaluated in future trials.

Conclusion

Hypertension is associated with faster stenosis progression and

reduced survival in patients with AS. ARBs but not ACEs may

provide significant reduction in AS progression rate and mor-

tality. These findings open news avenues for the treatment of

AS and provide an impetus for the elaboration of randomized

trials focused on renin–angiotensin system inhibitors on AS

population. According to the results of this study, ARBs would

appear to be the best candidate for such trial.

AcknowledgementsWe thank Isabelle Fortin, Jocelyn Beauchemin, Jacinthe Aube

and Martine Parent for their help in data collection and man-

agement, and Serge Simard and Anne-Sophie Julien for their

assistance in the statistical analyses.

Sources of fundingThis work was supported by Grant MOP-114997 from Cana-

dian Institutes of Health Research (CIHR), Ottawa, Ontario,

Canada, and a grant from the Foundation of the Qu�ebec Heart

and Lung Institute. R.C. was supported by a studentship grant

of International Chair of Cardiometabolic Risk, Qu�ebec, Qu�e-

bec, Canada. M-A.C. was supported by a Vanier studentship

grant of CIHR. N.C. was supported by a studentship grant of

Fonds de Recherche en Sant�e du Qu�ebec (FRSQ), Montreal,

Qu�ebec, Canada. P.M. and M.A. are research scholars from the

FRSQ, Montreal, Qu�ebec, Canada. P.P. holds the Canada

Research Chair in Valvular Heart Diseases, CIHR.

DisclosuresNone.

AddressInstitut Universitaire de Cardiologie et de Pneumologie de

Qu�ebec/Qu�ebec Heart & Lung Institute, Laval University,

Qu�ebec city, 2725 Chemin Sainte-Foy, Qu�ebec city, QC, G1V-

4G5, Canada (R. Capoulade, M.-A. Clavel, P. Mathieu, N. Cot�e,

J. G. Dumesnil, M. Arsenault, �E. B�edard, P. Pibarot).

Figure 3 Comparison between groups of the individual andadjusted hazard ratio for all-causes mortality. Panel A showsthe hazard ratio of individual Cox proportional hazard modelfor the different groups (see legend of Fig. 1), Panel B showsthe hazard ratio of multivariate Cox proportional hazard modeland Panel C shows the hazard ratio after further adjustment forthe weighted variable. The numbers at the top of the bar arehazard ratio and the 95% confidence interval between brackets.Error bars represent 95% confidence interval of the hazardratio. The symbols indicate the significant difference betweengroups: *P < 0�05 vs. Ctrl group in analysis where Ctrl groupwas defined as the referent group; †P < 0�05 vs. ARB group inanalysis where ARB group was defined as the referent group.



Correspondence to: Dr Philippe Pibarot, Institut Universitaire

de Cardiologie et de Pneumologie de Qu�ebec, 2725 Chemin

Sainte-Foy, Qu�ebec city, QC, G1V-4G5, Canada.

Tel.: 418 656 8711 (ext. 5938); fax: 418 656 4602;

e-mail: [email protected]

Received 21 May 2013; accepted 29 August 2013

References1 Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG,Enriquez-Sarano M. Burden of valvular heart diseases: a population-based study. Lancet 2006;368:1005–11.

2 Chaliki HP, Brown ML, Sundt TM, Tajik AJ. Timing of operation inasymptomatic severe aortic stenosis. Expert Rev Cardiovasc Ther2007;5:1065–71.

3 Freeman RV, Otto CM. Spectrum of calcific aortic valve disease:pathogenesis, disease progression, and treatment strategies.Circulation 2005;111:3316–26.

4 O’Brien KD. Pathogenesis of calcific aortic valve disease: a diseaseprocess comes of age (and a good deal more). Arterioscler ThrombVasc Biol 2006;26:1721–8.

5 Rajamannan NM, Bonow RO, Rahimtoola SH. Calcific aorticstenosis: an update. Nat Clin Pract Cardiovasc Med 2007;4:254–62.

6 Boss�e Y, Miqdad A, Fournier D, P�epin A, Pibarot P, Mathieu P.Refining molecular pathways leading to calcific aortic valve stenosisby studying gene expression profile of normal and calcified stenotichuman aortic valves. Circ Cardiovasc Genet 2009;2:489–98.

7 Otto CM, Kuusisto J, Reichenbach DD, Gown AM, O’Brien KD.Characterization of the early lesion of ‘degenerative’ valvular aorticstenosis. Histological and immunohistochemical studies. Circulation1994;90:844–53.

8 O’Brien KD, Reichenbach DD, Marcovina SM, Kuusisto J, Alpers CE,Otto CM. Apolipoproteins B, (a), and E accumulate in themorphologically early lesion of ‘degenerative’ valvular aorticstenosis. Arterioscler Thromb Vasc Biol 1996;16:523–32.

9 Olsson M, Dalsgaard CJ, Haegerstrand A, Rosenqvist M, Ryden L,Nilsson J. Accumulation of T lymphocytes and expression ofinterleukin-2 receptors in nonrheumatic stenotic aortic valves. J AmColl Cardiol 1994;23:1162–70.

10 Olsson U, Camejo G, Hurt-Camejo E, Elfsber K, Wiklund O,Bondjers G. Possible functional interactions of apolipoprotein B-100segments that associate with cell proteoglycans and the ApoB/Ereceptor. Arterioscler Thromb Vasc Biol 1997;17:149–55.

11 Antonini-Canterin F, Huang G, Cervesato E, Faggiano P, Pavan D,Piazza R et al. Symptomatic aortic stenosis: does systemichypertension play an additional role? Hypertension 2003;41:1268–72.

12 Briand M, Dumesnil JG, Kadem L, Tongue AG, Rieu R, Garcia Det al. Reduced systemic arterial compliance impacts significantly onleft ventricular afterload and function in aortic stenosis: implicationsfor diagnosis and treatment. J Am Coll Cardiol 2005;46:291–8.

13 Rieck AE, Cramariuc D, Boman K, Gohlke-Barwolf C, Staal EM,Lonnebakken MT et al. Hypertension in aortic stenosis: implicationsfor left ventricular structure and cardiovascular events. Hypertension2012;60:90–7.

14 Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E,Pellikka PA et al. Recommendations for chamber quantification: areport from the American Society of Echocardiography’s Guidelinesand Standards Committee and the Chamber Quantification WritingGroup, developed in conjunction with the European Association of

Echocardiography, a branch of the European Society of Cardiology.J Am Soc Echocardiogr 2005;18:1440–63.

15 de Simone G, Devereux RB, Daniels SR, Koren MJ, Meyer RA,Laragh JH. Effect of growth on variability of left ventricular mass:assessment of allometric signals in adults and children and theircapacity to predict cardiovascular risk. J Am Coll Cardiol1995;25:1056–62.

16 Hachicha Z, Dumesnil JG, Pibarot P. Usefulness of thevalvuloarterial impedance to predict adverse outcome inasymptomatic aortic stenosis. J Am Coll Cardiol 2009;54:1003–11.

17 Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden WBet al. Executive summary: heart disease and stroke statistics–2013update: a report from the American Heart Association. Circulation2013;127:143–52.

18 Simera I, Moher D, Hoey J, Schulz KF, Altman DG. A catalogueof reporting guidelines for health research. Eur J Clin Invest2010;40:35–53.

19 Hachicha Z, Dumesnil JG, Bogaty P, Pibarot P. Paradoxical low flow,low gradient severe aortic stenosis despite preserved ejectionfraction is associated with higher afterload and reduced survival.Circulation 2007;115:2856–64.

20 Cramariuc D, Cioffi G, Rieck AE, Devereux RB, Staal EM, Ray S et al.Low-flow aortic stenosis in asymptomatic patients: valvular arterialimpedance and systolic function from the SEAS substudy. JACCCardiovasc Imaging 2009;2:390–9.

21 Aronow WS, Schwartz KS, Koenigsberg M. Correlation of serumlipids, calcium, and phosphorus, diabetes mellitus and history ofsystemic hypertension with presence or absence of calcified orthickened aortic cusps or root in elderly patients. Am J Cardiol1987;59:998–9.

22 Stewart BF, Siscovick D, Lind BK, Gardin JM, Gottdiener JS,Smith VE et al. Clinical factors associated with calcific aortic valvedisease. Cardiovascular Health Study. J Am Coll Cardiol1997;29:630–4.

23 Chen JH, Chen WL, Sider KL, Yip CY, Simmons CA. beta-cateninmediates mechanically regulated, transforming growth factor-beta1-induced myofibroblast differentiation of aortic valve interstitialcells. Arterioscler Thromb Vasc Biol 2011;31:590–7.

24 Butcher JT, Simmons CA, Warnock JN. Mechanobiology of the aorticheart valve. J Heart Valve Dis 2008;17:62–73.

25 Mathieu P, Poirier P, Pibarot P, Lemieux I, Despres JP. Visceralobesity: the link among inflammation, hypertension, andcardiovascular disease. Hypertension 2009;53:577–84.

26 Helske S, Lindstedt KA, Laine M, Mayranpaa M, Werkkala K,Lommi J et al. Induction of local angiotensin II-producing systems instenotic aortic valves. J Am Coll Cardiol 2004;44:1859–66.

27 Akat K, Borggrefe M, Kaden JJ. Aortic valve calcification: basicscience to clinical practice. Heart 2009;95:616–23.

28 Cote N, Pibarot P, Pepin A, Fournier D, Audet A, Arsenault B et al.Oxidized low-density lipoprotein, angiotensin II and increasedwaist circumference are associated with valve inflammation inprehypertensive patients with aortic stenosis. Int J Cardiol2009;145:444–9.

29 Kaden JJ, Dempfle CE, Grobholz R, Fischer CS, Vocke DC, Kilic Ret al. Inflammatory regulation of extracellular matrixremodeling in calcific aortic valve stenosis. Cardiovasc Pathol2005;14:80–7.

30 Mohty D, Pibarot P, Despres JP, Cote C, Arsenault B, Cartier A et al.Association between plasma LDL particle size, valvularaccumulation of oxidized LDL, and inflammation in patients withaortic stenosis. Arterioscler Thromb Vasc Biol 2008;28:187–93.



31 Cot�e N, Couture C, Pibarot P, Despr�es JP, Mathieu P. Angiotensinreceptor blockers are associated with lower remodelling of stenoticaortic valves. Eur J Clin Invest 2011;41:1172–9.

32 Miller JD, Weiss RM, Serrano KM, Castaneda LE, Brooks RM,Zimmerman K et al. Evidence for active regulation of pro-osteogenicsignaling in advanced aortic valve disease. Arterioscler Thromb VascBiol 2010;30:2482–6.

33 Pibarot P, Dumesnil JG. Improving assessment of aortic stenosis.J Am Coll Cardiol 2012;60:169–80.

34 Lancellotti P, Donal E, Magne J, Moonen M, O’Connor K,Daubert JC et al. Risk stratification in asymptomatic moderateto severe aortic stenosis: the importance of the valvular,arterial and ventricular interplay. Heart 2010;96:1364–71.

35 Hein S, Arnon E, Kostin S, Schonburg M, Elsasser A,Polyakova V et al. Progression from compensated hypertrophy tofailure in the pressure-overloaded human heart: structuraldeterioration and compensatory mechanisms. Circulation2003;107:984–91.

36 Milano AD, Faggian G, Dodonov M, Golia G, Tomezzoli A,Bortolotti U et al. Prognostic value of myocardial fibrosis in patientswith severe aortic valve stenosis. J Thorac Cardiovasc Surg2012;144:830–7.

37 Weidemann F, Herrmann S, Stork S, Niemann M, Frantz S, LangeV et al. Impact of myocardial fibrosis in patientswith symptomatic severe aortic stenosis. Circulation 2009;120:577–84.

38 Cioffi G, Faggiano P, Vizzardi E, Tarantini L, Cramariuc D, Gerdts Eet al. Prognostic value of inappropriately high left ventricular massin asymptomatic severe aortic stenosis. Heart 2011;97:301–7.

39 O’Brien KD, Probstfield JL, Caulfield MT, Nasir K, Takasu J,Shavelle DM et al. Angiotensin-converting enzyme inhibitors andchange in aortic valve calcium. Arch Intern Med 2005;165:858–62.

40 Rosenhek R, Rader F, Loho N, Gabriel H, Heger M, Klaar U et al.Statins but not angiotensin-converting enzyme inhibitors delayprogression of aortic stenosis. Circulation 2004;110:1291–5.

41 Sverdlov AL, Ngo DT, Chan WP, Chirkov YY, Gersh BJ, McNeil JJet al. Determinants of aortic sclerosis progression: implicationsregarding impairment of nitric oxide signalling and potentialtherapeutics. Eur Heart J 2012;33:2419–25.

42 Nadir MA, Wei L, Elder DH, Libianto R, Lim TK, Pauriah M et al.Impact of Renin-Angiotensin system blockade therapy on outcomein aortic stenosis. J Am Coll Cardiol 2011;58:570–6.

43 Routledge HC, Townend JN. ACE inhibition in aortic stenosis:dangerous medicine or golden opportunity? J Hum Hypertens2001;15:659–67.

44 Pfeffer MA, McMurray JJ, Velazquez EJ, Rouleau JL, Kober L,Maggioni AP et al. Valsartan, captopril, or both in myocardialinfarction complicated by heart failure, left ventricular dysfunction,or both. N Engl J Med 2003;349:1893–906.

45 Yusuf S, Teo KK, Pogue J, Dyal L, Copland I, Schumacher H et al.Telmisartan, ramipril, or both in patients at high risk for vascularevents. N Engl J Med 2008;358:1547–59.



impact of hypertension and renin-angiotensin system inhibitors in aortic stenosis

Documents