Early recognition and treatment of hypertensive heart disease

Jason A. Mitchell, Hector O. Ventura and Mandeep R. Mehra

Purpose of review Introduction

Hypertension leads to left ventricular hypertrophy, diastolic Hypertensive heart disease (HHD) is a constellation of
dysfunction, and eventually clinical heart failure structural and morphologic changes in the heart due to
(hypertensive heart disease). There is an important need chronic increased afterload (hypertension), chronic vol-
for physicians to recognize this entity, understand its ume overload (increased preload), or both. Although the
pathophysiology, and become fluent in treatment options number of patients with chronic heart failure is increasing
available. This review of recent basic science and clinical due to improved anti-ischemic and anti-arrhythmic ther-
data serves to accomplish this task. apies, a large number of patients are presenting with
Recent findings symptoms of heart failure despite echocardiographically
In the past year, a number of exciting concepts have come normal systolic function. These patients with heart failure
to the forefront. First, data on the use of brain natriuretic with preserved systolic function likely represent an early
peptide as a screening biomarker has shown promise in form of hypertensive heart disease and deserve aggressive
patients with symptoms of heart failure. For the earliest treatment. The discussion that follows focuses on identi-
detection of structural changes, serum propeptide of fying these patients, the pathophysiology of hypertensive
procollagen type I, a marker of the deposition of type I heart disease, and recent clinical trial data supporting
collagen in the cardiac interstitium provides a noninvasive treatment modalities.
way to quantify myocardial fibrosis. Treatment options in the
past few years have focused heavily on the anti-fibrotic
effects of inhibitors of the renin-angiotensin-aldosterone Pathophysiology
system, perhaps supplanting beta blockers as In the cardiac muscle, collagen synthesis and degradation
first-line agents to regress left ventricular hypertrophy. The are ongoing and the equilibrium of these processes is con-
concept of aldosterone escape is discussed, highlighting trolled by a variety of hormones and cytokines. Pro-fibrotic
the importance of aldosterone inhibitors in these patients. substances include the renin-angiotensin-aldosterone
Lastly, we provide a comprehensive review of all system (RAAS), inflammatory cytokines (TGF-b for ex-
antihypertensive classes and their effects on hypertensive ample), and growth factors such as insulin-like growth fac-
heart disease. tor. In contrast, bradykinin, nitric oxide, natriuretic
Summary peptides, and glucocorticoids have collagenolytic effects
The incidence of hypertensive heart disease is increasing. [1]. In the setting of constant elevations in blood pres-
Earlier detection may be possible with newly studied serum sure, this delicate balance is disrupted, leading to in-
biomarkers. All anti-hypertensive treatments have shown creased collagen deposition in the interstitial space [2].
improvement in regressing left ventricular hypertrophy, but This fibrosis can also take several forms. When there is pa-
inhibitors of the renin-angiotensin-aldosterone system renchymal cell loss, a reparative (replacement) fibrosis
possess the most potent anti-fibrotic effects. It is occurs, as might occur microscopically due to a hyperten-
increasingly important for clinicians to identify and manage sive crisis or macroscopically from a myocardial infarction.
hypertensive heart disease to prevent increased morbidity Additionally, a perivascular-interstitial fibrosis occurs
and mortality. without parenchymal cell loss due to chronic elevations
in the effector hormones of the RAAS system [3].
Keywords
diastolic dysfunction, hypertension, hypertensive heart Additionally, as with any muscle exposed to constant in-
disease, left ventricular hypertension, myocardial fibrosis creased stretch, the response to elevated blood pressure
is cardiac myocyte hypertrophy. The pattern of hypertro-
—289. ª 2005 Lippincott Williams & Wilkins.
Curr Opin Cardiol 20:282— phy differs with the stimulus as described below [4]. In-
creased cardiac preload causes individual myofibrils to
Ochsner Clinic Foundation, Department of Cardiology, New Orleans,
Louisiana, USA
lengthen as a measure to increase end-diastolic volume.
This adaptation uses Starling forces to maintain a normal
Correspondence to Hector O. Ventura, Ochsner Clinic Foundation, Department of cardiac stoke volume and ejection fraction. Over time this
Cardiology, 1514 Jefferson Highway, New Orleans, LA 70121, USA leads to eccentric hypertrophy, as is seen physiologically in
Tel: 504 842 6281; fax: 504 842 5960; e-mail: hventura@ochsner.org
the setting of the athletic heart, and pathologically in set-
Current Opinion in Cardiology 2005, 20:282—
—289
tings of chronic volume overload (e.g., chronic kidney dis-
ª 2005 Lippincott Williams & Wilkins.
ease). Continuous volume overload can lead to further
0268-4705 maladaptive ventricular dilatation.
282
 Early recognition and treatment of hypertensive heart disease Mitchell et al. 283

Increased afterload, as seen in systolic hypertension, leads sion (‘risers’), who are less likely to be under antihyper-
to an increase in the circumferential diameters of individ- tensive therapy, and carry up to a six-fold increased risk
ual myofibrils and eventually concentric hypertrophy. At of cardiac events [13]. A study by Baguet et al. [14] of
first, the wall thickening allows ventricular function and 59 patients with obstructive sleep apnea demonstrated
end-diastolic pressures to remain normal, but over time LVH rates of 15–20% as well as nighttime hypertension
the end-diastolic pressures rise, the ventricular chamber in 76%. Many of these nighttime hypertensives were previ-
dilates, and the ejection fraction diminishes. ously unrecognized as having hypertension [14]. This study
highlights obstructive sleep apnea as an under-diagnosed
And thus, hypertension induces maladaptive processes in etiology of hypertension and likely hypertensive heart
the myofibril and interstitial space that each contribute to disease as well.
impaired relaxation, filling, and performance of the left
ventricle. Untreated, this leads to clinical heart failure.
Diastolic dysfunction
Left ventricular hypertrophy Diastolic dysfunction represents an abnormality of diastolic
To intervene in this maladaptive process, we must identify filling and relaxation of the left ventricle, commonly due
changes at their earliest. Noninvasive methods include to LVH, myocardial interstitial fibrosis, or both. In many
ECG, echocardiography, and more recently advanced im- patients it can predate any clinical symptoms and fre-
aging such as cardiac magnetic resonance (CMR). The quently is present in the absence of other structural abnor-
causal link between left ventricular hypertrophy (LVH) malities or systolic dysfunction. The common method of
and hypertension was first described by Johnson at autopsy detecting diastolic dysfunction is through echocardio-
over 130 years ago [5]. With the advent of EKG, the first graphic analysis of the pulse-wave Doppler signal of the
noninvasive method was available. Epidemiologically, mitral inflow during diastole and the waveform analysis
30-year follow-up data from the Framingham Heart Study of pulmonary vein flow. There are four stages of diastolic
showed hypertensive subjects (BP $ 160/95) when com- dysfunction.
pared with normotensive controls, had a 10-fold increase
in the incidence of ECG LVH [6]. The first stage, impaired relaxation, is characterized by
a reversal of the normal E/A ratio (<1.0) with a prolonged
Although the specificities of these EKG criteria approach deceleration time (>200 milliseconds) of the E wave. Left
90%, their sensitivities range from 10–60% [7]. With quan- atrial pressure is still typically normal, as it is estimated
titative echocardiography, we have a more reliable tool to by a peak systolic greater than peak diastolic velocity in
measure LV chamber mass, wall thickness, and configura- the pulmonary veins. In the second stage termed pseudo-
tion to assess the ventricular structure. Recent evaluation normalization, the E/A ratio and deceleration time are
of the Framingham data demonstrated 19% and 24% rates within normal limits (1.0–2.0 and 150–200 milliseconds,
of LVH by echocardiography in men and women, respec- respectively), but the pulmonary venous flows are re-
tively, compared with 1.3% of subjects using ECG criteria versed, suggesting elevated left atrial pressures. The third
[8]. There is even evidence that the changes in LV mass stage termed reversibly restrictive is characterized by an
may precede the development of overt hypertension. In increased E/A ratio (>2.0) and a short deceleration time
another study from the Framingham cohort, the risk of de- (<150 milliseconds) that normalizes with provocation,
veloping hypertension in follow-up increased by 20% for such as a Valsalva maneuver, along with elevated left atrial
each 26.5 g/m in LV mass index, and by 16% for every pressures. The final stage is irreversibly restrictive in
2.5-mm increase in LV wall thickness [9]. which provocative maneuvers do not normalize the mitral
inflow pattern.
This paradox of hypertension causing LVH, yet LV struc-
tural changes may predate clinically diagnosed hyperten- One criticism of pulse-wave mitral inflow patterns in the
sion can be explained by several factors. First, the detection of diastolic dysfunction is that these findings are
activation of the renin-angiotensin system and its effector extremely load (volume) dependent. With the recent ad-
hormones, most notably aldosterone, has been shown to vances in tissue Doppler, several minimally load-dependant
independently cause hypertension, myocardial fibrosis, measurements are available to increase the sensitivity and
and LVH [10]. Even when BP was controlled in several specificity of diastolic dysfunction detection. Myocardial
small human experiments, only medications inhibiting tissue Doppler at the base of the mitral valve annulus
the RAAS system decreased myocardial fibrosis when can reveal the early and post-atrial contraction filling ve-
compared with diuretics or calcium channel blockers alone locities, termed E# and A# respectively. An E/E# ratio <8
[11,12]. Second, 24-hour ambulatory blood pressure mon- predicts a normal LVEDP, while a ratio >15 implies an
itoring has demonstrated diurnal variations in blood pres- LVEDP > 15 mm Hg. Moreover, an E#/A# ratio >1 indi-
sure, delaying diagnosis due to normal daytime blood cates restrictive diastolic dysfunction. Lastly, a difference
pressures. Up to 10% of patients have nocturnal hyperten- between the A-wave duration and A#-wave duration of >20
284 Hypertension

milliseconds suggests at least pseudo-normal diastolic performed endomyocardial biopsies and measured the col-
dysfunction [15]. lagen volume fraction (CVF) as well as coronary sinus and
peripheral serum concentrations of carboxy-terminal pro-
Identification of these patients with the earliest of diastolic peptide of procollagen type I (PIP). They elegantly dem-
abnormalities could prove very important, as treating this onstrated positive linear correlations between coronary
patient population will likely impact survival and will be sinus and peripheral PIP and CVF as well as a negative cor-
discussed below. relation between CVF and ejection fraction. Compared
with controls, the coronary sinus and peripheral PIP con-
Serum biomarkers centrations in the hypertensive groups were elevated.
One of the earliest serum biomarkers tested in the setting More importantly, there was a statistically significant in-
of LVH and diastolic dysfunction was brain natriuretic crease in PIP when comparing hypertensive patients with
peptide (BNP). Kohno et al. [16] published data showing heart failure to those without [20•]. Further work is needed,
a positive correlation of BNP to degree of LVH in 31 but it appears possible for the earliest detection of fibrosis
asymptomatic hypertensive patients. Furthermore, with and cardiac remodeling in hypertensive patients.
treatment and reductions in LVH, the levels of BNP
decreased [16]. In the trial, however, the levels of BNP Prognosis
measured were small (10–30 pg/ml) compared with con- Patients are likely to be identified as having LVH, diastolic
temporary levels seen with chronic systolic heart failure. dysfunction, or both upon routine ECG or echocardio-
In elucidating a cutoff value for BNP in diastolic dysfunc- graphic evaluation and many of these patients are asymp-
tion, several recent studies came to different conclusions tomatic. It is important to understand the risk of diastolic
regarding the usefulness of serum BNP. First, Bibbins- dysfunction, LVH, or symptoms of heart failure despite
Domingo et al. [17] evaluated 293 outpatients with stable normal systolic function. Vasan et al. [21] published 6-year
coronary disease via echocardiography and serum BNP follow-up data on the Framingham cohort in 73 patients
measurements. They found poor screening test character- with CHF. The annual mortality rate of the 37 patients
istics for BNP and diastolic dysfunction at both 30 pg/ml with normal LVEF (>50%) was 8.7% versus 3.0% for
and 100 pg/ml thresholds. They concluded that BNP was matched controls [21]. On a large scale, Varadarajan et al.
not useful as a screening test. One criticism of the trial is [22] observed 2258 patients hospitalized for CHF for
the definition of diastolic dysfunction used was reversal of a mean of 2 years. Of these, 963 patients had a normal
flow in the pulmonary veins, which is considered stage II ejection fraction (>55%) and had a 5-year survival of
diastolic disease (pseudo-normal), and excludes those with 22% [22]. Drazner et al. [23] recruited 5201 asymptomatic
a relaxation abnormality. In a second trial, Atisha et al. [18] patients in the Cardiovascular Health Study to be ob-
evaluated a BNP cutoff value of 20 pg/ml in 202 patients served longitudinally. Of the 4683 patients with a normal
with symptoms suggestive of heart failure presenting at ejection fraction (>55%), 3042 had baseline and follow-up
the San Diego VA. They demonstrate that a BNP value measures of left ventricular mass (LVM). Over the 5 years
of 20 pg/ml has a sensitivity of 75% and specificity of of follow-up, the risk ratio of developing a depressed LV
38% with a negative predictive value of 73% in identifying ejection fraction in quartile 4 compared with quartile 1 of
patients with diastolic dysfunction. LVM index was 4.3. The increased LVM index group
(those above the median) had a roughly two-fold increase in
Incidentally, the sensitivity and negative predictive value myocardial infarction, heart failure, or angina at follow-up [23].
of either systolic or diastolic dysfunction were both 100%
[18]. Another study at the San Diego VA in 294 patients Treatment has been shown to impact this increased car-
referred for echocardiography showed a BNP value of diovascular event rate. Devereux et al. [24] published
62 pg/ml had a sensitivity of 85% and specificity of 83% a sub-study of the Losartan Intervention For Endpoint Re-
for detecting any diastolic dysfunction. Additionally, those duction in Hypertension (LIFE) trial assessing the reduc-
patients with restrictive diastolic dysfunction had the tions in LVH as well as cardiovascular clinical outcomes.
highest BNP levels, with a mean BNP of 408 ± 66 pg/ml The trial showed significant relative risk reductions in
[19]. Therefore, BNP may have some use in the detection all-cause mortality (26%), CV mortality (34%), and the
of diastolic dysfunction in the general population. composite (16%) between those patients in whom LVH
regressed by at least 25.3 g/m2 compared with those
Exciting research into the extent of myocardial fibrosis by who did not. This was independent of blood pressure low-
endomyocardial biopsies and serum biomarkers has made ering, age, smoking, diabetes, prior stroke, prior myocardial
possible even earlier detection of hypertensive heart dis- infarction, and baseline heart failure [24].
ease. Querejeta et al. [20•] recently published their work
on collagen type I synthesis in 65 patients with hyperten- Pharmacologic therapy
sion. All patients had LVH by echocardiography and With the above methods of non-invasively identifying
roughly half had clinical symptoms of heart failure. They patients at risk for hypertensive heart disease as well as
 Early recognition and treatment of hypertensive heart disease Mitchell et al. 285

the increased morbidity and mortality that is reduced with ume fraction (CVF), pro-matrix metalloproteinases 2 and
therapy, identification of patients early is imperative. As 9 (proMMPs), and left heart catheterization-based mea-
noted above, the stages of hypertensive heart disease sures of myocardial stiffness. Their findings demonstrated
are predictable, beginning with fibrosis, LVH, and diastolic no difference in MyoD between treatment groups, although
dysfunction and if untreated, lead to left ventricular dila- both were elevated compared with controls. There was
tation, elevated left end-diastolic pressure, and systolic a significant reduction in the CVF in the ACE-inhibitor–
dysfunction. From 1964 to 1969, the VA Cooperative treated patients. Also, the ACE-inhibitor–treated group
Study Group chaired by Dr. Ed Fries demonstrated that had a two-fold increase in proMMP2 compared with non
management of hypertension alone dramatically reduced ACE-inhibitor–treated patients, accounting for the re-
the incidence of morbid events, namely stroke, heart fail- duced fibrosis due to MMP2’s collagenolytic activity. A
ure, renal failure, coronary artery disease, and death [25]. positive correlation was found between myocardial stiff-
We will next review the various drug classes and their re- ness and CVF while no correlations were found between
versal of hypertensive heart disease, both through blood stiffness and MyoD, or MyoD and CVF. These findings
pressure lowering and other effects. demonstrate the role of myocardial fibrosis in myocardial
stiffness, and ultimately diastolic dysfunction as well as
Angiotensin converting enzyme inhibitors the inhibitory role of this fibrosis by ACE-inhibitors [29].
Angiotensin converting enzyme inhibitors (ACE-I) have
been studied for over a decade in the setting of hyperten- Angiotensin receptor blockers
sion and LVH. An early trial investigating LVH regression Antagonism of the RAAS system is clearly beneficial in
with ACE-inhibitors, was published by Esper et al. [26] in preventing the progression of myocardial hypertrophy
1993. They evaluated 30 patients with essential hyperten- and interstitial fibrosis. Angiotensin-converting enzyme
sion and a mean left ventricular mass index (LVMI) of inhibitors alone do not prevent the production of angiotensin
133 g/m2. After 16 weeks of treatment with lisinopril, II. Therefore, angiotensin receptor blockers (ARBs) are
the LVMI decreased by 13 g/m2, their E-wave velocity another therapeutic target of the RAAS. Research is quite
rose from 0.6 m/s to 0.68, and their E/A ratio reversed from active studying either ARBs alone or in combination with
0.98 to 1.3. In a larger trial of 235 patients, Devereux et al. ACE-inhibitors.
published the PRESERVE trial comparing enalapril to ni-
fedipine in patients with ejection fractions >40% and The CATCH trial (Candesartan Assessment in the Treat-
echocardiographic evidence of increased LVMI. Patients ment of Cardiac Hypertrophy), was a multicenter, prospec-
were followed-up with echocardiography at 6 and 12 tive, randomized double-blind trial comparing the effects
months. Both groups demonstrated a mean reduction in of candesartan (8–16 mg/d) and enalapril (10–20 mg/d)
LVMI by 16 g/m2. Careful inspection of the data showed on left ventricular mass index (LVMI) in patients with
a significant reduction of systolic and diastolic blood pres- hypertension. Both drugs reduced LVMI, by 15.0 and
sure with nifedipine early in the trial. Despite this there 13.1 g/m2 respectively [30].
was a trend toward improved posterior wall thickness re-
gression with enalapril, raising the question of blood pres- The Losartan Intervention For Endpoint Reduction in
sure control alone in LVH regression [27]. Hypertension (LIFE) trial examined treatment with los-
artan or atenolol in 9193 patients with hypertension and
To explore the mechanisms of LVH reduction with ACE- ECG-derived LVH. Despite the same degree of blood
inhibition further, Dahlöf et al. [28] compared enalapril to pressure lowering, losartan reduced ECG-derived LVH
hydrochlorothiazide in 28 men with hypertension and by a factor of two as well as had a 25% relative risk reduc-
echocardiographically confirmed LVH. The ACE-inhibitor tion of stroke and a 13% relative risk reduction in the com-
reduced the LVMI to a greater degree than the diuretic posite endpoint of CV mortality, stroke, and myocardial
(21 g/m2 vs 13 g/m2) despite the same degree of blood infarction when compared with atenolol [31].
pressure lowering. Also, hydrochlorothiazide had no im-
pact on the posterior and septal wall thickness, while ena- In the Candesartan in Heart failure Assessment of Reduc-
lapril decreased both by a significant degree [28]. Thus, tion in Mortality and morbidity (CHARM)-preserved trial,
there appears to be additional benefit of ACE-inhibition Yusuf et al. [32] evaluated the impact of candesartan ver-
above blood pressure control alone. sus placebo in patients with NYHA functional class II-IV
CHF and an LVEF greater than 40%. Compared with pla-
Brilla et al. [29] recently published an elegant study eval- cebo, there was a borderline significant reduction in CV
uating 97 patients with hypertension treated with either death and hospitalizations for heart failure of 16% with
ACE-inhibition–based regimens or those without. They candesartan compared with placebo [32].
performed endomyocardial biopsies of the subjects and
compared them to normals (heart transplant donor con- In the past year, Müller-Brunotte et al. [33] evaluated 115
trols) in terms of myocyte diameter (MyoD), collagen vol- patients with hypertension and echo-derived LVH treated
286 Hypertension

with either irbesartan or atenolol and observed for 48 crease of left ventricular mass index, the canrenone
weeks. Both groups had equal blood pressure reductions group showed a significantly greater increase in left ven-
and although both had reductions in LVMI, irbesartan tricular diastolic indices when compared with the control
did so to a larger degree (16% vs 9%). group. They conclude that in essential hypertension, a low
dose of aldosterone antagonist added to anti-hypertensive
These studies highlight the added LVH reversing benefit treatment significantly improved left ventricular diastolic
of inhibiting the RAAS system over blood pressure control function. They postulated that this improvement could be
alone. therefore ascribed to a direct action of the drug on the
myocardium [39].

Aldosterone antagonists In comparing aldosterone inhibitors alone or in combina-

Although targeting the angiotensin converting enzyme tion with ACE-inhibitors, Pitt et al. [40] studied 202
and its receptor initially inhibits the RAAS, the Random- patients with LVH and hypertension treated for 9 months
ized Evaluation of Strategies for Left Ventricular Dysfunc- with either eplerenone, enalapril, or both. LVMI was de-
tion (RESOLVD) trial clearly demonstrated the concept termined by cardiac MRI. Despite all groups achieving the
of aldosterone escape. The authors studied a cohort of same degree of blood pressure lowering, the combination
768 patients with a depressed ejection fraction (EF <40%) group had the largest degree of LVH reversal [40].
and NYHA II-IV heart failure treated with candesartan,
enalapril, or both for 43 weeks. At 17 weeks there was a sig- Therefore, the addition of an aldosterone inhibitor to an-
nificant reduction in aldosterone levels from baseline, but other inhibitor of the RAAS system appears to be benefi-
by 43 weeks this improvement disappeared [34•]. cial in reversing LVH and interstitial fibrosis.

Mechanisms of aldosterone escape are speculative and in-

clude ACTH as a secretagogue and chymase generation of b-blockers
angiotensin II by a non-ACE inhibitor pathway [35•]. This For many years, a common treatment for hypertension and
increase in aldosterone despite ACE-inhibition was shown LVH was b-adrenergic receptor blockade. Owing much to
to have prognostic value in a neurohumoral sub-study of the morbidity and mortality efficacy of certain b- blockers
the Survival and Ventricular Enlargement (SAVE) trial. in the treatment of chronic systolic heart failure, as well as
Vantrimpont et al. [36] showed that of the 534 survivors the JNC VII recommendations of b-blockers and thiazide-
of an incident myocardial infarction, the patients with based diuretics for initial treatment of hypertension, they
lower aldosterone levels had a reduced 2-year cardiovascu- have become a widely prescribed class of anti-hyperten-
lar event rate. This highlights the benefit of aldosterone sives [41]. As identification of diastolic filling abnormali-
receptor blockade in addition to the other anti-RAAS ther- ties is more widely performed, b-blockade is used to slow
apies listed above. the heart rate and allow more time for LV filling to occur.
Especially in certain subgroups, such as post myocardial
Regarding fibrosis and remodeling, several recent animal infarction, early initiation of b-blockers has a favorable
studies have proven benefit with aldosterone blockade. outcome on LV remodeling [42]. With respect to LVH
First, Suzuki et al. [37] studied 14 cardiomyopathic dogs and its regression, many studies in the 1980s demonstrated
treated with eplerenone. They demonstrated reductions efficacy of b-blockers in reversing LVH. One early paper
in end-systolic and end-diastolic volumes as well as stabi- by Cherchi et al. [43] showed LVH regression on echocar-
lization of the ejection fraction. Also, compared with controls, diography in 78 hypertensive patients treated with ate-
there was a 28% reduction in cardiomyocyte cross-sectional nolol, timolol, or chlorthalidone. It has subsequently
area and a 37% reduction of collagen volume fraction [37]. become clear of the many beneficial roles of b-blockers
A second animal study by Endemann et al. [38] evaluated in today’s cardiology population. Next we will focus on
spontaneously hypertensive rats fed a low-, normal-, or b-blockers in patients with hypertensive heart disease,
high-salt diet with or without eplerenone therapy. A group a normal ejection fraction, and presence of LVH or diastolic
of high-salt diet rats were also treated with hydralazine. dysfunction.
Despite equal blood pressure improvement, rats fed
a high-salt diet treated with hydralazine exhibited In terms of echo-derived diastolic dysfunction, Bergstrom
increased heart weight and ventricular fibrosis, while et al. [44] treated 97 patients with diastolic dysfunction
eplerenone-treated rats did not [38]. with escalating doses of carvedilol versus placebo. Echo-
cardiograms were performed at baseline and 6 months.
A clinical study by Grandi et al. [39] tested the hypothesis They showed a small increase in E/A ratio in the carvedilol
that the aldosterone antagonist canrenone can improve group; however, there was no other statistically significant
left ventricular diastolic function in essential hyperten- change in an echo-derived measure compared with placebo
sion. Despite unchanged blood pressure and similar de- [44].
 Early recognition and treatment of hypertensive heart disease Mitchell et al. 287

Beta blockers, as have most antihypertensives, have been They assessed echocardiographic findings, exercise duration,
shown to decrease LVH due to its antihypertensive effect. and QOL scores using the Minnesota Living with Heart
In a recent comparative trial between irbesartan and ate- Failure questionnaire at follow-up. The authors demon-
nolol, Schneider et al. [45] treated 240 patients with hy- strated an improvement in exercise duration and QOL
pertension and EKG-derived LVH (by Sokolow, Cornell, scores in both groups, but to a greater degree with cande-
and CornellQRS criteria) with either irbesartan or atenolol sartan despite identical blood pressure improvements
for 18 months. Although neither group had a significant dif- [49].
ference in LVMI by echocardiography, the irbesartan group
did have a significant decrease in EKG-derived LVH Regarding dihydropteridine calcium channel blockers,
whereas the atenolol group did not [45]. One criticism DeRosa et al. [50] studied 84 patients with hypertension
of this study is that the degree of LVH by echocardiogra- and echo-derived LVH. After 2 years of treatment with
phy at baseline was mild and neither treatment arm felodipine, the patients’ LVMI was decreased by an aver-
showed a benefit in echo-derived LVH. age of 18% [50]. Islim et al. [51] treated 33 patients with
amlodipine for 20 weeks and demonstrated a significant
Concerning reversal of interstitial fibrosis, Ciulla et al. reduction in LVMI (from 169 g/m2 to 141 g/m2) and im-
[46•] recently published a study comparing the effects proved diastolic filling. Comparing calcium channel
of losartan versus atenolol in 106 hypertensive patients blockers to other antihypertensives, Havranek et al. [52]
with echo-derived LVH. The authors used a method, pre- published the Appropriate Blood Pressure Control in Di-
viously validated, of evaluating echoreflectivity and spread abetes (ABCD) trial, looking at 468 patients with diabetes
of echoes about the distribution (BB) to estimate collagen and hypertension treated with either enalapril or nisoldi-
volume fraction [47]. In addition, they measured serum pine. All patients underwent 5 years of EKG follow-up us-
markers of collagen type I and III synthesis (PIP and ing the Cornell voltage criteria for LVH. The ACE-inhibitor
PIIIP, respectively) or degradation (CITP). The BB was group had a greater decline in LVH compared with the
significantly reduced in the losartan-treated group while nisoldipine group. Incidentally, the use of nisoldipine in
it increased in the atenolol-treated group, suggesting op- this group was associated with higher rates of MI, CVA,
posite effects on cardiac fibrillar collagen turnover. The ratios or cardiovascular death compared with the ACE-inhibitor
of PIP and PIIIP to CITP suggested a decrease in synthe- group [52].
sis of both collagen types I and III in the losartan-treated
group. Conversely, patients in the atenolol group had un- Arterial vasodilators and diuretics
changed synthesis of type I collagen and increased synthe- There is limited recent data on other antihypertensives in
sis of type III collagen. the setting of hypertensive heart disease, LVH, and dia-
stolic dysfunction. Leenen et al. [53] studied 19 patients
These recent findings demonstrate that beta blockade is with hypertension and LVH by echocardiography treated
partially effective in reversing LVH and ineffective in re- with either hydralazine or prazosin. After 12 months, only
versing myocardial fibrosis of hypertensive heart disease the prazosin group had a significant reduction in LVH,
compared with the RAAS inhibitors. while the hydralazine group did not [53]. In a comparative
trial between lisinopril and hydralazine in 30 hypertensive
Calcium channel blockers patients, Fogari et al. [54] added each drug onto a baseline
Calcium channel blockers have several main properties therapy of b-blockers and diuretics. Baseline and 6-month
felt to be beneficial in hypertensive heart disease. First assessments of LVMI showed a significant decrease in
they have obvious anti-hypertensive effects. Second, they the LVMI of the lisinopril-treated group, but not in the
have lusitropic properties, enhancing relaxation of the hydralazine-treated group. Because of an increase in circu-
myocardium by attenuating calcium homeostasis. Lastly, lating norepinephrine and ambulatory heart rates, the
centrally acting calcium channel blockers slow the heart authors postulated that this difference was due to reflex
rate, allowing more diastolic filling time. sympathetic activation secondary to hydralazine [54].

An early clinical study by Setaro et al. [48], evaluated the One recent and very provocative study examined the
use of verapamil in 20 men with hypertension, normal LV effects of loop diuretics on myocardial fibrosis. López
function (>45%), and abnormal diastolic function. Verap- et al. [55•] treated 36 patients with NYHA class II to IV
amil increased exercise capacity by 33% and improved CHF and mean ejection fraction of 38% with either torse-
ventricular filling rate by 30%. They concluded that verap- mide or furosemide. They demonstrated, by right septal
amil may have efficacy in these patients with CHF and endomyocardial biopsies at baseline and 8 months fol-
a normal ejection fraction [48]. Recently, Little et al. low-up, a significant reduction in myocardial fibrosis in
[49] compared verapamil to candesartan in 21 subjects the torsemide-treated patients, which was not seen with
with hypertension, preserved ejection fraction (>50%), furosemide. Moreover, in concert with the endomyocar-
and diastolic dysfunction in a two-week crossover manner. dial biopsies, the serum marker PIP correlated well with
288 Hypertension

the measured collagen volume fraction [55•]. This study 15 Otto C. Textbook of Clinical Echocardiography. 3rd ed. 2004. Elsevier
Saunders.
strengthens the concept of PIP as a serum marker of col-
16 Kohno M, Horio T, Yokokawa K, et al. Brain natriuretic peptide as a marker for
lagen deposition and also raises the question of which loop hypertensive left ventricular hypertrophy: changes during 1-year antihyperten-
diuretic should be used in patients with volume overload sive therapy with angiotensin-converting enzyme inhibitor. Am J Med 1995;
98:257— —265.
and heart failure.
17 Bibbins-Domingo K, Ansari M, Schiller N, et al. Is B-type natriuretic peptide
a useful screening test for systolic or diastolic dysfunction in patients with
Conclusion coronary disease? Data from the heart and soul study. Am J Med 2004;
116:509— —516.
The pathogenesis of heart failure from hypertension likely
18 Atisha D, Bhalla M, Morrison LK, et al. A prospective study in search of an
involves myocardial hypertrophy, interstitial fibrosis, and optimal B-natriuretic peptide level to screen patients for cardiac dysfunction.
diastolic dysfunction. There appears to be benefit of all Am Heart J 2004; 148:518— —523.
antihypertensive classes mentioned here with the excep- 19 Lubien E, DeMaria A, Krishnaswamy P, et al. Utility of B-natriuretic peptide in
detecting diastolic dysfunction — comparison with Doppler velocity record-
tion of hydralazine in reducing LVH. However, antagonists ings. Circulation 2002; 105:595— —601.
of the RAAS also inhibit or reverse myocardial interstitial 20 Querejeta R, López B, Gonzalez A, et al. Increased collagen type I synthesis in
fibrosis, adding to their benefit. As with any disease pro- • patients with heart failure of hypertensive origin: relation to myocardial fibro-
sis. Circulation 2004; 110:1263— —1268.
cess, prevention is paramount. The new recommendations Excellent study introducing PIP as a serum marker of myocardial fibrosis.
from JNC VII about stricter blood pressure control cou- 21 Vasan R, Larson M, Benjamin E, et al. Congestive heart failure in subjects with
pled with improved public awareness of the risks of hyper- normal versus reduced left ventricular ejection fraction. JACC 1999; 33:
1948—
—1955.
tension and early recognition of cardiac abnormalities,
22 Varadarajan P, Pai RG. Prognosis of congestive heart failure in patients with
should translate into less patients with hypertensive heart normal versus reduced ejection fractions: results from a cohort of 2,258 hos-
disease in the future. pitalized patients. J Card Fail 2003; 9:107——112.
23 Drazner M, Rame J, Marino EK, et al. Increased left ventricular mass is a risk
factor for the development of a depressed left ventricular ejection fraction
within five years. JACC 2004; 43:2207— —2215.
References and recommended reading 24 Devereux R, Wachtell K, Gerdts E, et al. Prognostic significance of left ven-
Papers of particular interest, published within the annual period of review, have tricular mass change during treatment of hypertension. JAMA 2004;
been highlighted as: 292:2350— —2356.
• of special interest
•• of outstanding interest 25 Veterans Administration Cooperative Study on Antihypertensive Agents.
Effect of treatment on morbidity in hypertension. Results in patients with diastolic
blood pressure averaging 90 through 114 mmHg. JAMA 1970; 213:
1 Laurent D. Dynamic state of collagen: pathways of collagen degradation 1143— —1152.
in vivo and their possible role in regulation of collagen mass. Am J Physiol
1987; 252:C1— —C9. 26 Esper R, Burrieza OH, Cacharron JL, et al. Left ventricular mass regression
and diastolic function improvement in mild and moderate hypertensive pa-
2 Moncrieff J, Lindsay M, Dunn F. Hypertensive heart disease and fibrosis. Curr tients treated with lisinopril. Cardiology 1993; 83:76—
—81.
Opin Cardiol 2004; 19:326— —331.
27 Devereaux R, Palmieri V, Share N, et al. Effects of once-daily angiotensin-
3 González A, López B, Diez J. Fibrosis in hypertensive heart disease: role of the converting enzyme inhibition and calcium channel blockade— —based antihy-
renin-angiotensin-aldosterone system. Med Clin N Am 2004; 88:83— —97. pertensive treatment regimens on left ventricular hypertrophy and diastolic
4 Frohlich ED. Pathogenesis of hypertensive left ventricular hypertrophy and di- filling in hypertension. Circulation 2001; 104:1248—
—1254.
astolic dysfunction. In: Hypertension Primer. Izzo JL, Black HR, editors. 3rd 28 Dahlöf B, Hansson L. Regression of left ventricular hypertrophy in previously
ed. Dallas, TX: American Heart Association; 2003: 175— —177. untreated essential hypertension: different effects of enalapril and hydrochlo-
5 Johnson S. On certain points in the anatomy and pathology of Bright’s dis- rothiazide. J Hypertens 1992; 10:1513— —1524.
ease and the kidney, II; on the influence of the minute blood-vessels upon 29 Brilla C, Rupp H, Maisch B. Effects of ACE inhibition versus non-ACE inhib-
the circulation. Med Chir Trans 1868; 51:57— —80. itor antihypertensive treatment on myocardial fibrosis in patients with arterial
6 Levy D. Left ventricular hypertrophy: epidemiological insights from the hypertension. Herz 2003; 28:744— —753.
Framingham Heart Study. Drugs 1988;35(suppl 5):1— —5. 30 Cuspidi C, Muiesan ML, Valagussa L, et al. Comparative effects of candesar-
7 Levy D, Labib SB, Anderson KM, et al. Determinants of sensitivity and spec- tan and enalapril on left ventricular hypertrophy in patients with essential
ificity of electrocardiographic criteria for left ventricular hypertrophy. Circula- hypertension: the candesartan assessment in the treatment of cardiac hyper-
tion 1990; 81:815— —820. trophy (CATCH) study. J Hypertens 2002; 20:2293— —2300.
8 Haider AW, Larson MG, Benjamin EJ, Levy D. Increased left ventricular mass 31 Dahlöf B, Devereux R, Kjeldsen SE, et al. Cardiovascular morbidity and
and hypertrophy are associated with increased risk for sudden death. JACC mortality in the losartan intervention for endpoint reduction in hyperten-
1998; 32:1454— —1459. sion study (LIFE): a randomized trial against atenolol. Lancet 2002;
359:995— — 1003.
9 Post WS, Larson MG, Levy D. Impact of left ventricular structure on the in-
cidence of hypertension: the Framingham Heart Study. Circulation 1994; 32 Yusuf S, Pfeffer MA, Swedberg K. Effects of candesartan in patients with
90:179——185. chronic heart failure and preserved left-ventricular ejection fraction: the
CHARM-Preserved Trial. Lancet 2003; 362:777— —781.
10 Frishman W, Stier C. Aldosterone and aldosterone antagonism in systemic
hypertension. Curr Hypertens Rep 2004; 6:195——200. 33 Müller-Brunotte R, Edner M, et al. Irbesartan and atenolol improve diastolic
function in patients with hypertensive left ventricular hypertrophy. J Hypertens
11 Brilla CG, Funck RC, Rupp H. Lisinopril-mediated regression of fibrosis in patients 2005; 23:633— —640.
with hypertensive heart disease. Circulation 2000; 102:1388— —1393.
34 McKelvie RS, Yusuf S, et al. Comparison of candesartan, enalapril, and their
12 Lopez B, Querejeta R, Varo N, et al. Usefulness of serum carboxy-terminal propep- • combination in congestive heart failure: randomized evaluation of strategies
tide of procollagen type I in assessment of the cardioreparative ability of antihyper-
for left ventricular dysfunction (RESOLVD) pilot study. Circulation 1999;
tensive treatment in hypertensive patients. Circulation 2001; 104:286— —291.
100:1056— —1064.
13 Kario K, Shimada K. Risers and extreme-dippers of nocturnal blood pressure Excellent — earliest trial to identify the concept of aldosterone escape in ACE-
in hypertension: antihypertensive strategy for nocturnal blood pressure. Clin inhibitor treated patients.
Exp Hypertens 2004; 26:177— —189.
35 Struthers A. The clinical implications of aldosterone escape in congestive
14 Baguet J, Hammer L, Levy P, et al. Night-time and diastolic hypertension are • heart failure. Eur J Heart Fail 2004; 6:539—
—545.
common and underestimated conditions in newly diagnosed apnoeic pa- Excellent — reviews the concept and implications of aldosterone escape in clinical
tients. J Hypertens 2005; 23:521——527. practice.
 Early recognition and treatment of hypertensive heart disease Mitchell et al. 289

36 Vantrimpont P, Rouleau J, et al. Two-year time course and significance of neu- 46 Ciulla M, Paliotti R, Esposito A, et al. Different effects of antihypertensive ther-
rohumoral activation in the Survival and Ventricular Enlargement (SAVE) • apies based on losartan or atenolol on ultrasound and biochemical markers of
study. Eur Heart J 1998; 19:1552— —1563. myocardial fibrosis. Circulation 2004; 110:552— —557.
Excellent trial bringing basic science to clinical practice.
37 Suzuki G, Morita H, et al. Effects of long-term monotherapy with eplere-
none, a novel aldosterone blocker, on progression of left ventricular dys- 47 Ciulla M, Paliotti R, Hess D, et al. Echocardiographic patterns of myocardial
function and remodeling in dogs with heart failure. Circulation 2002; fibrosis in hypertensive patients: endomyocardial biopsy versus ultrasonic tis-
106:2967— — 2972. sue characterization. J Am Soc Echocardiogr 1997; 10:657— —664.
38 Endemann D, Touyz RM, Iglarz M, et al. Eplerenone prevents salt-induced vas- 48 Setaro J, Zaret B, Schulman D, et al. Usefulness of verapamil for congestive
cular remodeling and cardiac fibrosis in stroke-prone spontaneously hyper- heart failure associated with abnormal left ventricular diastolic filling and nor-
tensive rats. Hypertension 2004; 43:1252— —1257. mal left ventricular systolic performance. Am J Cardiol 1990; 66:981— —986.
39 Grandi AM, Imperiale D, Santillo R, et al. Aldosterone antagonist improves 49 Little WC, Wesley-Farrington DJ, Hoyle J, et al. Effects of candesartan and
diastolic function in essential hypertension. Hypertension 2002; 40:647— — verapamil on exercise tolerance in diastolic dysfunction. J Cardiovasc Phar-
652. macol 2004; 43:288— —293.
40 Pitt B, Reichek N, Willenbrock R, et al. Effects of eplerenone, enalapril, and 50 DeRosa M, Giordano A, Della Guardia D, et al. Reversal of left ventricular
eplerenone/enalapril in patients with essential hypertension and left ventric- hypertrophy following once daily administration of felodipine for two years to elderly
ular hypertrophy. Circulation 2003; 108:1831— —1838. subjects with isolated systolic hypertension. Cardiology 1999; 92:39— —44.
41 The Seventh Report of the Joint National Committee on Prevention. Detec- 51 Islim I, Watson R, Ihenacho H, et al. Amlodipine: effective for treatment of mild
tion, evaluation, and treatment of high blood pressure. JAMA 2003; 289: to moderate essential hypertension and left ventricular hypertrophy. Cardiol-
2560— —2570. ogy 2001; 96(suppl 1):10— —18.
42 Doughty R, Whalley G, et al. Effects of carvedilol on left ventricular remodel- 52 Havranek E, Esler A, Estacio R, et al. Differential effects of antihypertensive
ing after acute myocardial infarction: The CAPRICORN echo substudy. Cir- agents on electrocardiographic voltage: results from the Appropriate Blood
culation 2004; 109:201— —206. Pressure Control in Diabetes (ABCD) trial. Am Heart J 2003; 145:993— —998.
43 Cherchi A, Sau F, Seguro C. Possible regression of left ventricular hypertro- 53 Leenen F, Smith D, Farkas R, et al. Vasodilators and regression of left ven-
phy during antihypertensive treatment with diuretics and/or beta blockers. tricular hypertrophy. Am J Med 1987; 82:969— —978.
J Clin Hypertens 1987; 3:216— —225.
54 Fogari R, Zoppi A, Mugellini A, et al. Effects of lisinopril vs. hydralazine on left
44 Bergstrom A, Andersson B, Edner M, et al. Effect of carvedilol on diastolic ventricular hypertrophy and ambulatory blood pressure monitoring in essen-
function with diastolic heart failure and preserved systolic function. Results tial hypertension. Eur Heart J 1995; 16:1120— —1125.
of the Swedish Doppler-Echocardiography Study (SWEDIC). Eur J Heart Fail
55 López B, Querejeta R, Gonzalez A, et al. Effects of loop diuretics on myocar-
2004; 6:453— —461. • dial fibrosis and collagen type I turnover in chronic heart failure. JACC 2004;
45 Schneider M, Klingbeil A, Delles C, et al. Effect of irbesartan versus atenolol 43:2028— —2035.
on left ventricular mass and voltage — results of the cardiovascular irbesartan Excellent — elegant study with serious potential ramifications in the treatment of
project. Hypertension 2004; 44:61— —66. patients with CHF.