and quality of life, but they have not demonstrated conclusive

mortality benefits. The Comparison of Medical Therapy, Pacing,

and Defibrillator in Heart Failure (COMPANION) trial78 enrolled

1,520 patients with NYHA class II or IV (ischemic or nonischemic

cardiomyopathies, QRS interval of at least 120 milliseconds, and

LVEF ≤35%) who were treated with optimal drug therapy (ACE

inhibitors, diuretics, β-blockers, and spironolactone). Patients

were randomly assigned to receive optimal drug therapy alone,

optimal drug therapy and CRT with a pacemaker, or optimal

drug therapy and CRT with ICD (CRT-D). The primary end

point was a composite of all-cause mortality and hospitalization.

Both CRT and CRT-D groups were associated with a decreased

risk of primary end point (p = 0.014, p = 0.01, respectively)

compared with optimal drug therapy alone. All-cause mortality

at 1 year was decreased by 24% in the CRT group and 43% in the

CRT-D group; however, it was not significantly reduced in the

CRT group.

The results of the Cardiac Resynchronization in Heart Failure study (CARE-HF)77 extended the landmark findings of the

COMPANION trial. CARE-HF demonstrated a significant allcause mortality reduction for CRT pacing without defibrillator

backup (CRT) in patients with HF that was medically treated

similarly. This study was conducted in a total of 813 patients.

The inclusion criteria were NYHA class III or IV, EF of 35% or

less, and QRS duration of 120 milliseconds or longer. Approximately 35% of patients had ischemic heart disease. The primary

end point of all-cause deaths and hospitalizations for a major CV

reason occurred in fewer patients in the CRT group compared

with the optimal drug therapy group (39% vs. 55%; p <0.001).

Death or hospitalization for worsening HF was also significantly

reduced in the CRT group.

Thus, the combined results of CARE-HF and COMPANION

confirm the importance of CRT and CRT-D in improving ventricular function, HF symptoms, and exercise tolerance, while also

reducing frequency of HF hospitalizations by 37% and death by

22%.336 The role of CRT in patients with mild HF symptoms,

narrow QRS, chronic AF, and right bundle branch block need to

be explored.

According to the ACC/AHA guidelines,1,21 patients with

NYHA class III and ambulatory patients with class IV HF should

receive CRT (unless contraindicated) if they meet the following

criteria: LVEF of 35% or less, presence of electric asynchrony as

shown by a wide QRS (>120 milliseconds), and receiving optimal HF standard medical therapy. Despite optimal doses of HF

medications, C.M. continues to have HF symptoms. CRT therapy could provide incremental benefits beyond what is provided

with neurohormonal therapy.

CASE 19-6, QUESTION 2: If C.M. presented with NYHA

class I or II symptoms, would she be a candidate for CRT

therapy? What is the evidence to support CRT in NYHA class

I and II patients?

As mentioned in Case 19-7, Question 1, the CARE-HF and

COMPANION trials provide strong evidence that CRT induces

reverse modeling in patients with symptomatic NYHA class III

and ambulatory class IV HF. The next logical step was to evaluate benefits of CRT therapy in HF patients with milder symptoms (asymptomatic or mildly symptomatic). The Multicenter

InSynch ICD Randomized Clinical Evaluation (MIRACLEICDII)337 trial was the first randomized trial to enroll only class

II through IV HF patients but with separate specified end points

for class II patients. In this trial, 186 patients with NYHA class II

HF, an LVEF of less than 35%, and a QRS of more than 130 milliseconds received a CRT-ICD device. Subjects were randomly

assigned to active CRT group (ICD activated, CRT on) or control group (ICD activated and CRT off ). The primary end point

was progression of HF, defined as all-cause mortality, hospitalizations for HF, and ventricular tachycardia or ventricular fibrillation requiring device intervention. A 15% reduction in HF

progression was observed, but this was not statistically significant

(p = 0.35). At 6 months patients within the active CRT group

had improved exercise tolerance, but this was not significantly

different from the control group. However, there was a significant decrease in ventricular end-systolic volume and increased

LVEF after 6 months of therapy. Even though the study results

did not translate into improved exercise tolerance, it helped to set

the stage for future trials in patients with less symptomatic HF.

In 2008, the REVERSE trial75 enrolled 610 participants from

both the United States and Europe with NYHA class I or II HF,

484 Section 2 Cardiac and Vascular Disorders

LVEF less than 40%, and with a QRS duration of more than

120 milliseconds who received a CRT device (with or without

ICD) in combination with optimal drug therapy. Similar to the

MIRACLE ICD II trial, the patients in the active CRT group had

significant improvement in LV end-systolic volume index, LV

end-diastolic volume index, and LVEF (p <0.0001, p <0.0001,

p <0.001, respectively) compared with the control group, which

are measures of reverse remodeling. Although the primary clinical end point (the percentage of patients with worsened clinical composite score) did not meet statistical significance at

12 months in the US cohort (p = 0.1), in the European cohort338

statistical difference was seen at 24 months (p = 0.01). This difference was driven by increased time to first HF hospitalization.

Significant differences were observed in several of the secondary

end points. The aggregate data from these two clinical trials provided overwhelming evidence that linked CRT with substantial

reverse remodeling in mild HF patients.

The question of whether CRT reduces progression of HF

and reduces mortality in NYHA class I and II HF patients has

also been studied. The MADIT-CRT76 was the largest randomized trial designed to determine whether CRT-D therapy would

reduce the primary end point (all-cause mortality or HF events,

whichever occurred first) when compared with patients receiving ICD-only therapy. The study population involved cardiac

patients in NYHA functional class I or II (no or mild symptoms)

who had either ischemic or nonischemic heart disease with LVEF

of 30% or less and QRS duration of more than 130 milliseconds

on ECG. There was a 34% (p <0.001) reduction in the primary

end point, and a 44% (p <0.001) reduction in HF events when

compared with ICD therapy. Also, patients on CRT-D therapy

showed an 11% improvement in LVEF after 1 year, compared

with 3% improvement for ICD-alone patients. It is noteworthy

that both MADIT-CRT and REVERSE excluded patients with AF.

In addition, there were fewer asymptomatic patients, and some

patients had NYHA class III symptoms before enrollment, but

almost 80% of patients were classified as NYHA class II. Furthermore, both studies failed to show a benefit for CRT in patients

with QRS duration less than 150 milliseconds.

In 2010, the Resynchronization/Defibrillation for Ambulatory

Heart Failure Trial (RAFT),339 confirmed the results of previous trials (including MADIT-CRT and REVERSE), showing that

CRT had an increased benefit in patients with a QRS duration of

150 milliseconds or more and in those with left bundle branch

block. RAFT investigators randomly assigned 1,798 patients from

34 centers with NYHA class II or III HF, LVEF of 30% or less, and

a QRS duration of at least 120 milliseconds (or a paced QRS of

at least 200 milliseconds) to either ICD therapy alone or an ICD

with CRT (CRT-D). The mean follow-up time for all patients was

40 months. The primary outcome was a combination of total

mortality and HF hospitalization. The secondary outcomes

included death by any cause, death from a CV cause, and HF

hospitalizations. The primary outcome was statistically significant in the ICD group compared with the ICD-CRT group (40%

vs. 33%, respectively). These findings demonstrate that earlier

intervention with CRT-D, in addition to guideline-recommended

medical and ICD therapy, benefits this patient population.

HEART FAILURE WITH PRESERVED

LEFT VENTRICULAR EJECTION

FRACTION

CASE 19-7

QUESTION 1: D.F., a 72-year-old white woman, has a 5-

year history of HF symptoms, including decreased exercise capacity, SOB, and distended neck veins. She has minimal peripheral edema. History is suggestive of rheumatic

fever as a child, but she does not recall having any cardiac symptoms when she was younger, other than being

told she had a murmur. Her symptoms are controlled with

diuretics. She has history of HTN. She has no other medical problems, and all laboratory test findings are normal. Her BP is 155/85 mm Hg, and her HR is 90 beats/

minute. Cardiac examination reveals a prominent S4 heart

sound. Noninvasive echocardiography reveals a normal EF

of 50%. Prior treatment included furosemide, most recently

at 40 mg BID. The physician is considering adding a

β-blocker or CCB to control the BP. Why might this consideration be appropriate?

This case exemplifies a patient with HF with preserved LVEF

(HFPEF), often referred to as diastolic HF. Risk factors for HFPEF

include advanced age, female sex, HTN, and CAD. This diagnosis

can be made on the basis of LVH, clinical evidence of HF, a

normal EF, and Doppler tissue echocardiography findings. The

ideal treatment strategies for HFPEF have not been extensively

validated. A review of trials evaluating specific drug therapy in

HFPEF is listed in Table 19-13. Also, no drug selectively enhances

myocardial relaxation without having associated effects on LV

contractility or on the peripheral vasculature.14–18

Factors affecting HF control, such as dietary sodium intake,

fluid intake, compliance, and NSAID and herbal remedy use,

should be appropriately managed along with drug therapy. Symptomatic LV diastolic HF is initially treated similarly to other forms

of HF, by slow diuresis. Diuresis decreases preload and lessens

passive congestion of the ventricles. Excessive lowering of venous

and ventricular filling pressures, however, can worsen CO and

cause hypotension.

The most common cause of diastolic HF and HFPEF is HTN

that leads to LVH and decreased cardiac compliance.353 Recent

ACC/AHA guidelines recommend treating associated HTN in

accordance with the national guidelines and lower BP targets for

patients with diabetes and chronic kidney disease (<130/80 mm

Hg).21 Drugs that cause regression of LVH (e.g., ACE inhibitors,

ARBs, β-blockers) may also slow or reverse structural abnormalities associated with diastolic HF.

Although ACE inhibitors have been used with success in

some patients with HFPEF, the role of RAAS inhibition in

HF with preserved LV function has not been rigorously studied. Several recent clinical trials have attempted to address this

issue. In the Perindopril in Elderly patients with Chronic HF

(PEP-CHF) trial,342 the ACE inhibitor perindopril failed to reduce

the incidence of the primary end point (all-cause mortality or HF

hospitalizations), but did reduce symptoms and improved functional capacity in 2 years in patients with preserved LV function.

The CHARM-preserved trial180 also failed to show any difference in CV mortality, but fewer hospitalizations were seen in the

candesartan group (see Case19-1, Question 13).

The Valsartan in Diastolic Dysfunction (VALIDD)354 trial was

the first large-scale, randomized trial comparing the effects of

valsartan or placebo added to standard antihypertensive therapy

(which included diuretics, β-blockers, CCBs, or α-blockers) in

patients with mild HTN and diastolic HF. The hypothesis of this

trial was that RAAS inhibition with an ARB would be associated with greater improvement in diastolic dysfunction, possibly

because of a greater regression of LVH or myocardial fibrosis.

Patients with a history of stage 1 or 2 essential HTN were randomly assigned to receive either valsartan 160 mg, titrated up

to 320 mg, or matching placebo. Patients who did not achieve

a target BP goal of less than 135/80 mm Hg received additional

therapy starting with a diuretic followed by a CCB or a β-blocker,

485Heart Failure Chapter 19

TABLE 19-13

Clinical Trials of Pharmacotherapy in Heart Failure with Preserved Ejection Fraction

Study Patient Population

Therapy

Intervention Outcome

Treatment

Duration Results

Aronow et al. (1993)340 NYHA III;

Prior MI, EF >50%;

(n = 21)

Enalapril vs.

placebo

NYHA class, treadmill

exercise time

(seconds)

3 months Enalapril: NYHA class from 3 ± 0

to 2.4 ± 0.5 ( p = 0.005),

exercise time from 224 ± 27 to

270 ± 44 ( p <0.001) vs. no

difference in placebo

Lang et al. (1995)341 HF symptoms

>3 months;

EF >50%; (n = 12)

Lisinopril vs.

placebo

crossover

Dyspnea and fatigue 5 weeks for each

treatment

arm

No significant differences

Cleland et al. (2006)342

PEP-CHF

Diastolic dysfunction;

CV admission

within 6 months;

EF >40%; n = 850)

Perindopril vs.

placebo

Primary: composite of

all-cause mortality or

hospitalization for

HF

Mean

26.2 months

Primary: 23.6% in perindopril

group vs. 25.1% in placebo

(HR 0.92 [0.70–1.21],

p = 0.545)

Zi et al. (2003)343 NYHA class II or III;

EF ≥40%; (n = 74)

Quinapril vs.

placebo

6-minute walk test,

QoL, NYHA class

6 months No significant differences

Yusuf et al. (2003)180

CHARM-Preserved

NYHA II–IV;

hospitalization for

CV causes; EF

>40%; (n = 3,023)

Candesartan vs.

placebo

Primary: CV death or

hospital admission

for HF

Median

36.6 months

Primary: 22% in the candesartan

group vs. 24% in the placebo

group (adjusted HR 0.86

[0.74–1.00], p = 0.051).

Massie et al. (2008)344

I-PRESERVE

NYHA II–IV;

hospitalized for HF

in last 6 months; EF

≥45%; (n = 4,122

Irbesartan vs.

placebo

Primary: all-cause death

or hospitalization for

CV causes

Mean

49.5 months

Primary: 36% in the irbesartan

group vs. 37% in the placebo

group (HR 0.95 [95% CI

0.86–1.05], p = 0.35)

Yip et al. (2008)345 NYHA II–IV; history

of HF in last 2

months; EF >45%;

(n = 151)

Ramipril vs.

irbesartan

QoL, 6-minute walk

test, HF hospital

admission

12 months No significant differences

Warner et al. (1999)346 Diastolic dysfunction;

DOE; EF >50%;

SBP >150, <200

mm Hg (n = 20)

Losartan vs.

placebo

crossover

Exercise time, QoL 2 weeks for each

treatment

arm

Increase in exercise time (11.3

minutes at baseline, improved

to 12.3 ± 2.6 minutes with

losartan vs. 11.0 minutes with

placebo, p <0.05) and

improvement in QoL (25 at

baseline, improved to 18 with

losartan vs. 22 with placebo);

p <0.05 for both end points

Parthasarathy et al.

(2009)347

Diastolic dysfunction;

DOE; EF >40%;

(n = 152)

Valsartan vs.

placebo

Primary: exercise time 14 weeks No significant differences

Takeda et al. (2004)348 NYHA II–III and stage

C heart failure; EF

≥45%

Carvedilol vs.

placebo

Plasma BNP, NYHA

class, exercise

capacity

12 months NYHA class improved by 0.77

(carvedilol) vs. 0.25 (placebo)

( p <0.02), exercise capacity in

METs improved 0.69

(carvedilol) vs. worsened by

0.07 (placebo) ( p = 0.01).

Flather et al. (2005)349

SENIORS

HF hospital admission

in last year; EF

≤35%; subgroup

EF ≥35%

Nebivolol vs.

placebo

Primary: composite of

all-cause mortality or

hospitalization for a

cardiovascular cause

Mean 21 months EF >35%, primary event rate

17.6% in nebivolol and 21.9% in

placebo (HR 0.86 [95% CI

0.74–0.99], p = 0.039).

Aronow et al. (1997)350 NYHA II–III; prior

Q-wave MI; EF

>40%; n = 158)

Propranolol vs.

placebo

All-cause mortality,

all-cause mortality

plus nonfatal MI

32 months All-cause mortality (56%

propranolol group vs. 76%

placebo group, p = 0.007) and

all-cause mortality plus nonfatal

MI (59% propranolol vs. 82%

placebo, p = 0.002).

Setaro et al. (1990)351 Abnormal diastolic

filling; EF >45%;

(n = 20)

Verapamil vs.

placebo

Exercise capacity 2 weeks for each

crossover

Exercise capacity (10.7 minutes at

baseline, improved to

13.9 minutes with verapamil vs.

12.3 minutes with placebo,

p <0.05).

Ahmed et al. (2006)352

DIG

NYHA I–IV; EF

>45%; (n = 988)

Digoxin vs.

placebo

Primary: composite of

HF hospitalization or

mortality

Mean 37

months

102 (21%) in the digoxin group vs.

119 (24%) in the placebo group

(HR 0.82 [0.63–1.07],

p = 0.136).

BNP, B-type natriuretic peptide; CV, cardiovascular; DOE, dyspnea on exertion; EF, ejection fraction; HF, heart failure; HR, hazard ratio; METs, metabolic equivalents;

MI, myocardial infarction; NA, not available; NYHA, New York Heart Association; QoL; quality of life; SBP, systolic blood pressure.

486 Section 2 Cardiac and Vascular Disorders

then an α-blocker (excluding ARBs, ACE inhibitors, and aldosterone blockers). The primary end point was the change in diastolic myocardial relaxation velocity from baseline to 9 months

with a secondary end point of change in LV mass. During the

study, the placebo group received more concomitant antihypertensive therapy compared with the valsartan group. A small,

but significant, increase was seen in diastolic relaxation velocity in both groups from baseline to follow-up (p <0.001), but

there was no significant difference between the treatment groups

(p = 0.29). BPessure reduction at the end of the trial did not differ significantly between the two treatment groups (13 mm Hg

reduction in valsartan vs. 10 mm Hg in placebo), which was associated with significant improvement in diastolic function. Thus,

the authors concluded that aggressive BP control—even in mild

HTN—was associated with improvement in diastolic dysfunction, irrespective of whether BP reduction was achieved with an

RAAS inhibitor or other antihypertensive agents. Several other

trials are in progress that may provide further insight into the

role of RAAS inhibitors in HFPEF.

CHARM-Preserved180 investigated the role of candesartan in

patients with HFPEF. The trial enrolled 3,023 subjects who met

the overall CHARM trial inclusion criteria defined previously plus

one additional criterion: an EF of more than 40% (mean, 54%).

Thus, subjects would be classified as having symptomatic HF

with normal (preserved) EF. They received either an ARB alone

(n = 1,514) or placebo; only 20% of subjects in both groups

were taking an ACE inhibitor at randomization, 56% were on a

β-blocker, and 11% were on spironolactone.180 After a median

follow-up of 36.6 months, an insignificant trend was noted toward

reduction in the primary outcome of CV death or hospital admission for HF in the candesartan group (22%) compared with

placebo (24.3%; p = 0.118). CV deaths (170 vs. 170) and all-cause

mortality (244 vs. 237) were nearly identical in both groups, but

the total number of hospitalizations for HF (402 vs. 566) was

significantly reduced in the candesartan group (p = 0.014). Secondary outcomes consisting of composites of the primary outcomes plus MI, nonfatal stroke, and coronary revascularization

also showed a not significant trend in favor of candesartan. The

most common side effects with candesartan were hypotension

(2.4%), increase in creatinine (4.8%), and hyperkalemia (1.5%).

Discontinuation because of an adverse event occurred in 17.8% of

those treated with candesartan compared with 13.5% of placebo

recipients (p = 0.001) (Table 19-12). Overall, the conclusion is

that in symptomatic patients with diastolic HF, no significant

improvement in morbidity or mortality occurs with candesartan compared with placebo, other than a significant reduction in

HF-related hospitalizations.

I-PRESERVE is the largest randomized controlled trial for

the management of HFPEF performed at this time.344 The study

lasted for a mean of 49.5 months. The patient population included

was 60 years of age or older with NYHA class II through IV

symptoms, EF of at least 45%, and who were hospitalized for

HF during the last 6 months or have persistent class III or IV

symptoms (n = 4,128). Patients received irbesartan titrated to

300 mg daily or placebo. There was no difference in the primary

end point between irbesartan (36%) and placebo (37%) (hazard

ratio, 0.95; 95% CI, 0.86–1.05; p = 0.35). In addition, no significant difference in secondary outcomes (i.e., composite HF outcome, score on the Minnesota Living with HF scale at 6 months,

composite vascular-event outcome, CV death) was noted. The

irbesartan group had more patients experiencing hyperkalemia

(>6 mmol/dL) than placebo. One possible reason for the neutral results of I-PRESERVE included the high rate of dual RAAS

blockade at baseline (39% ACE inhibitor use in the irbesartan

group and 40% in the placebo group; 28% spironolactone use in

the irbesartan group and 29% in the placebo group). Based on

this high use, the study is less likely to find benefit with ARBs

in addition to other RAAS agents. Another potential limitation

of the study included the high study discontinuation rate (34%).

Overall, the study resulted in irbesartan showing no added benefit in reducing morbidity or mortality in HFPEF patients.

β-Blockers or nondihydropyridine CCBs are other classes of

drugs of interest in HFPEF. Part of their value is to control

HTN, a risk factor for all forms of HF. More specific to diastolic

HF, β-blockers and CCBs (especially verapamil) possess negative

inotropic properties that may favorably influence the pathophysiology of diastolic dysfunction by (a) slowing the HR to allow

more time for complete ventricular filling (via more complete

left atrial emptying), particularly during exercise; (b) reducing

myocardial oxygen demand; and (c) controlling BP. In addition,

negative inotropic agents decrease myocardial contractility and

can assist in overcoming the mechanical obstruction below the

aorta during systole in patients with hypertrophic cardiomyopathy. Both agents also are beneficial in decreasing ischemia in

patients with CAD.

Previous HF trials of β-blockers demonstrating decreased

morbidity and mortality have mainly focused on patients with

reduced LVEF. The Study of the Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors with Heart

Failure (SENIORS) study349 is the first major trial to evaluate

β-blocker use in elderly HF patients (70 years or older), irrespective of LV function. The trial randomly assigned patients

to nebivolol (n = 1,067) or placebo (n = 1,061). Nebivolol is a

selective β1-adernergic receptor blocker with vasodilator properties that are mediated through NO release. This effect may be

beneficial in elderly patients, who tend to have low reserves of

endothelial vasodilation.

The primary end point of the study was the combination of allcause mortality and CV hospital admissions. The end point was

significantly reduced by 14% in the nebivolol group, regardless of

the EF. Prospective subgroup analyses of the primary outcome

by LVEF (≤35% or >35%), sex, or age (≤75 years or >75 years)

showed benefits across all subgroups. Patients with EF greater

than 35%, however, appeared to benefit a little more than those

with low EF%, and all-cause mortality was lower in patients older

than 75 years treated with nebivolol compared with placebo.

The study reinforces the current recommendations that all HF

patients with reduced EF should receive β-blockers. Only 35%

of the patients had preserved LV function, however, and were

mostly men. This is not typical of patients with HFPEF, who are

generally women. Further studies are required to define the role

of β-blockers in HFPEF.

No randomized controlled trials have demonstrated mortality benefits with CCBs in patients with preserved LV function.

Nondihydropyridine CCBs can be used in patients who have a

contraindication to β-blockers to control BP and HR. Nondihydropyridine CCBs should not be used in patients with impaired

LV dysfunction.

The role of aldosterone antagonists in the management of

patients with HFPEF has not been clearly defined. The Aldosterone Antagonist Therapy for Adults with Heart Failure and

Preserved Systolic Function (TOPCAT)355 study is expected to be

completed in 2011. This trial is evaluating the impact of spironolactone versus placebo on CV morbidity and mortality during a

2-year study period in patients older than 50 years of age with

an EF greater than 45%. Once completed, we will have a clearer

answer on the role of aldosterone antagonists in the management

of patients with HFPEF.

D.F. fulfills the criteria for having diastolic dysfunction, based

on her history of long-standing HTN, which is been poorly controlled. Her BP is not at goal, and her HR is elevated. As already

discussed, uncontrolled HTN can promote LVH and myocar-

487Heart Failure Chapter 19

dial remodeling, and can adversely affect the diastolic function.

Therefore, antihypertensive therapy is warranted. Tachycardia

alone can compromise the ventricle filling time and cause

myocardial ischemia. So far, no data support the use of one agent

over another. β-Blockers and nondihydropyridine CCBs can

each reduce BP and HR. Because more experience has accrued

with β-blockers and there are some mortality data on their use

with diastolic HF patients, D.F. can be started on a β-blocker

such as metoprolol (25 mg BID).

Herbal Products and Nutritional

Supplements

CASE 19-8

QUESTION 1: W.L., a 60-year-old man with HF recently diagnosed by his naturopath, is concerned about his decreasing

exercise capacity and increasing SOB during his morning

walks in the local mall. His blood pressure is 170/85 mm

Hg, and he has 1 to 2+ ankle edema. He distrusts medical doctors and wants to treat his HF naturally. One time in

the past he was given HCTZ for BP reduction, but stopped

taking it after a few days because he did not tolerate the

urinary urgency it caused. The naturopath has prescribed

200 mg/day of hawthorn leaf and 50 mg/day of coenzyme

Q. How effective is this treatment plan likely to be?

HAWTHORN

Hawthorn extracts from the leaves and flowers of Crataegus

monogyna and Crataegus oxyacantha have been reported to have

beneficial effects in mild HF.356,357 Oligomeric procyanidins and

flavonoids are considered the key active ingredients. Hawthorn

extracts have shown positive inotropic action, weak ACE inhibition, vasodilating properties, and increased coronary blood flow

in vitro and in animal models. In short-term (8 weeks or less),

placebo-controlled trials in patients with the equivalent of NYHA

class II HF, modest improvements were noted in exercise tolerance and subjective symptoms as well as decreases in HR and

BP. Patients with more advanced HF were excluded. A systematic review by Pittler et al. also concluded that hawthorn extract

was efficacious in the treatment of HF on top of standard HF

therapy.358 Conversely, the results of the Hawthorne Extract Randomized Blinded Chronic Heart Failure (HERB-CHF) trial failed

to provide any evidence that hawthorn was beneficial in patients

with HF who were already receiving standard medical therapy.359

In clinical trials, side effects of hawthorn include nausea, vomiting, diarrhea, palpitations, chest pain, and vertigo. These side

effects are more common when doses exceed 900 mg/day, but in

some trials they have not occurred more often than with placebo.

The risks and benefits of using hawthorn and digoxin together,

both of which have positive inotropic effects, are not known.

To further investigate the longer-term benefits of hawthorn,

additive effects to conventional therapy and effect on mortality

were tested in the Survival and Prognosis: Investigation of Crataegus Extract WS 1442 in Congestive Heart Failure (SPICE) trial.360

The trial enrolled 2,681 patients with NYHA class II or III, LVEF of

35% or less, who were randomly assigned to hawthorn or placebo

for 2 years. Although the study failed to show any clear cardioprotective benefits in the treatment of chronic HF, hawthorn was

well tolerated and can be safely added to standard therapy.

COENZYME Q

Coenzyme Q, also known as ubiquinone and ubidecarenone,

is an endogenously synthesized provitamin that is structurally

similar to vitamin E, serves as a lipid-soluble electron transport

carrier in mitochondria, and aids in the