Chapter  11:  Heart Failure: Evaluation and Care of Patients with Left-Ventricular Systolic Dysfunction: Clinical Practice Guideline Number 11

Prevention in Asymptomatic Patients

Probably the largest number of such patients will be those who have recently sustained a myocardial infarction (MI). For this reason, EF should be determined in most patients following an MI unless they are at low risk for significant systolic dysfunction-that is, unless they meet all of the following criteria:

• No previous MI.

• Inferior infarction.

• Relatively small increase in cardiac enzymes (i.e., <2-4 times normal).

• No Q waves develop on electrocardiogram (ECG).

• Uncomplicated clinical course (e.g., no arrhythmias or hypotension).

Other asymptomatic patients without MI's may be found to have reduced EF on evaluation of heart murmurs or cardiomegaly. These patients should also be treated with ACE inhibitors.

Patient Evaluation

Patient Management

Evaluation of Patients for Revascularization

Coronary artery disease is currently the most common cause of heart failure in the United States, and some heart failure patients may benefit from revascularization.

Introduction

On December 19, 1989, the Omnibus Budget Reconciliation Act (Public Law 101-239) added a new Title IX to the Public Health Service Act establishing the Agency for Health Care Policy and Research (AHCPR). AHCPR's goal is to enhance the quality, appropriateness, and effectiveness of health care services and access to such services. Section 911 of the Act establishes within AHCPR the Office of the Forum for Quality and Effectiveness in Health Care. Section 912 directs the Forum to facilitate the development and periodic review and updating of:

Following this mandate, AHCPR selected heart failure as a topic for guideline development and as one of the first three guidelines to be developed under contract. Unlike the first seven guidelines, which were developed by private-sector panels convened directly by AHCPR, these contract guidelines were created by panels of experts and consumers selected (with the concurrence of AHCPR) and staffed by a private contracting organization. In the case of the heart failure guideline, the contracting agency was RAND, a nonprofit public policy and research organization in Santa Monica, California.

Heart failure is a clinical syndrome or condition characterized by (1) signs and symptoms of intravascular and interstitial volume overload, including shortness of breath, rales, and edema, or (2) manifestations of inadequate tissue perfusion, such as fatigue or poor exercise tolerance. These signs and symptoms result when the heart is unable to generate a cardiac output sufficient to meet the body's demands. This guideline uses the term heart failure in preference to the commonly used "congestive heart failure" because many patients with heart failure do not manifest pulmonary or systemic congestion.

More than 2 million Americans suffer from the effects of heart failure, and about 400,000 new cases are diagnosed each year. Mortality is high, with 5-year mortality rates in the range of 50 percent. In addition, quality of life is often adversely affected by symptoms and reduced functional capacity.

Table 1. Estimated Total Direct Economic Cost of Heart (more...)

Table 1. Estimated Total Direct Economic Cost of Heart Failure in the United States

Type of Service	Number[a] (thousands)	Costs[b] (millions)
Hospital days	5,800[c]	$7,500
Physician office visits	3,000[d]	690
Nursing home days (1985)	17,000[e]	1,900
Drugs	Not applicable	230
Total direct costs	Not applicable	10,320

[a] Estimated from total national numbers in each category multiplied by the proportion devoted to heart failure (based on International Classification of Diseases, 9th Revision [ICD-9] codes).[1]

[b] Estimated from total national expenses in each category[2] multiplied by the proportion of expenditures devoted to heart failure (based on ICD-9 codes).[1]

[c] National Hospital Discharge Survey, National Center for Health Statistics (NCHS) 1990.[3]

[d] National Ambulatory Medical Care Survey, NCHS, 1989.[4]

[e] National Nursing Home Survey, NCHS, 1985.[5]

According to information supplied by the Health Care Financing Administration, Medicare paid $2.4 billion in 1992 for 654,000 hospital admissions for patients with a principal diagnosis of heart failure. Actual hospital charges for Medicare patients were more than double this amount, totaling $5.6 billion (mean charge: $8,500). Thus, heart failure represents an important component of uncompensated hospital care. The in-hospital mortality rate for Medicare patients in 1992 was about 7 percent, which contrasts with 30-day mortality rates of about 15 percent for heart failure patients observed in three large studies of Medicare admissions from the 1980's. [6-8] Part of this discrepancy may be accounted for by the difference between in-hospital and 30-day mortality rates, but it is also possible that in-hospital mortality rates have improved over the past few years.

Table 2. Effects of Treatment on Mortality in Patients (more...)

Table 2. Effects of Treatment on Mortality in Patients With Left-Ventricular Dysfunction or Clinical Heart Failure

Study Author, Year	Study Population	N	Duration of Followup (years)	Treatment	Mortality Rate (percent)	Absolute Mortality Difference[a] (percent)	Symptomatic Hypotension[b] (percent)
Surgical Management of Heart Failure or Left-Ventricular Dysfunction
Bounous et al., 1988[9]	First cardiac catheterization; EF ≥ 40%			CABG	14[c]	18(Operative mortality NR)
		710	3.0	Medical	32[c]
Medical Management of Left-Ventricular Dysfunction
Pfeffer et al. (SAVE), 1992[15]	3-16 days after MI; EF ≥ 40%; no current heart failure	2,231	3.5[d]	Captopril	20.4	4.2	0.7
				Placebo	24.6	4.2	0.7
SOLVD Investigators (SOLVD Prevention Trial), 1992[16]	EF ≥ 35%; no heart failure	4,228	3.1[d]	Enalapril	14.8	NS	NR
				Placebo	15.8
Medical Management of Heart Failure
Cohn et al. (VHeFT I), 1986[17]	Chronic heart failure,cardiac enlargement or EF < 45%, and MVO[sub 2] < 25 mL/kg/min	642	3.0	HYD/ISDN	36.2	10.7	NR
				Placebo	46.9
CONSENSUS Trial Study Group, 1987[10]	New York Heart Association Class IV heart failure	253	1.0	Enalapril	36.2	16.2	5.5
				Placebo	52.4
SOLVD Investigators (SOLVD Treatment Trial), 1991[18]	Chronic heart failure; EF ≥ 35%	2,569	3.5[d]	Enalapril	35.2	4.5	2
				Placebo	39.7
Cohn et al. (VHeFT II), 1991[19]	Same as VHeFT I	804	2.5[d]	Enalapril	32.8	5.4	4.5
				HYD/ISDN	38.2

[a] All mortality differences are significant at p <0.05 except where indicated (NS).

[b] Symptomatic hypotension requiring termination of the study drug.

[c] Adjusted for baseline prognostic factors.

[d] Mean duration of followup.

Notes: NR = not reported, NS = not significant, HYD/ISDN = hydralazine and isosorbide dinitrate, CABG = coronary artery bypass graft, EF = ejection fraction, MI = myocardial infarction, MVO[sub 2] = maximum oxygen uptake, SAVE = Survival and Ventricular Enlargement, SOLVD = Studies of Left-Ventricular Dysfunction, VHeFT = Veterans Affairs Vasodilator Heart Failure Trial.

Conversely, poor quality care has been shown to have a negative impact on outcomes. Patients hospitalized with heart failure who received poor quality care, as defined by explicit, validated process criteria, experienced increased death rates in the 30 days following hospital discharge. [11] Despite improvements in the quality of care in recent years, this group of investigators found that 12 percent of patients hospitalized with heart failure received poor or very poor quality care. In addition, 7 percent of patients were thought to have been discharged too soon, and those discharged in unstable condition were found to have a 16-percent mortality rate at 90 days after discharge compared with 10 percent in those who were discharged in stable condition. [12]

Deficiencies have also been found in the quality of outpatient care. Retchin and Brown [13] found that less than half the patients they studied were advised to follow a salt-restricted diet. Many patients were found to have uncontrolled hypertension, but medications were initiated or modified in only 62 percent of fee-for-service (FFS) patients and 36 percent of patients in the health maintenance organization (HMO) groups. Twenty-seven percent of patients in the FFS group had a followup visit within 1 week of discharge from the hospital; 42 percent of HMO patients were seen within this period. Improved management of outpatients could reduce morbidity and mortality and decrease hospitalizations.

Practice Variations

Fleg et al. [14] found differences in self-reported practice styles in the utilization of laboratory procedures to monitor outpatients with heart failure. In comparison with internists or family physicians, cardiologists were more likely to follow outpatients with heart failure by using echocardiography, radionuclide ventriculography, or exercise testing. The estimated yearly cost of following a patient with New York Heart Association (NYHA) Class II heart failure was $303 for the quartile of physicians with the lowest utilization of procedures compared with $1,167 for the highest quartile. If the average costs of such tests could be reduced by $250 per year, more than $500 million could be saved annually on a national basis.

Common Errors in Management

As a way to organize the panel's thinking and to help ensure that the guideline has a positive impact on quality of care, the panel identified several aspects of the care of heart failure patients that are often not managed appropriately. This list was circulated to about 20 project consultants, who concurred on almost all points and added others. The complete list follows.

Scope of Guideline

Figure 1. Overview of Evaluation and Care of Patients (more...)

   Figure 1. Overview of Evaluation and Care of Patients With Heart Failure

Figure 1

Methodology for Guideline Development

The guideline development process used during this project followed the evidence-based approach recommended by AHCPR. A 16-member expert panel was constituted to interpret and supplement the evidence available in published literature. Panelists were selected after a broad range of input was obtained from professional and health care consumer organizations and individuals. At least one professional or consumer organization nominated or endorsed each panel member. The panel consisted of five cardiologists, two cardiac surgeons, one internist-geriatrician, one general internist, one family physician, one pulmonologist, one clinical pharmacist, two nurses with expertise in cardiac care and research, and two consumer representatives. To help define and circumscribe the scope of the guideline and associated literature reviews, the panel co-chairs and project staff first developed a draft clinical algorithm. The algorithm is designed both to depict the scope of the guideline and to provide assistance with decisionmaking strategies for patients with heart failure and practitioners on a case-by-case basis. This draft algorithm was then considered by the full guideline panel at its first meeting, in February 1992.

To refine the guideline topic, the panel decided to focus the scope of the guideline on the evaluation and care of patients with left-ventricular systolic dysfunction and to exclude consideration of interventions specific to the inpatient setting (e.g., right-heart catheterization, mechanical assist devices).

To further focus the scope of the guideline and to ensure maximal impact of the guideline in actual clinical practice, the panel next developed a list of commonly seen errors in the outpatient management of heart failure, as already described. Project staff were charged with ensuring that the guideline addressed these areas of management. Because of the potentially very large number of topics available for review within the selected domain, and because time and budgetary restrictions prohibited an all-inclusive literature review, the panel rated topics to determine their priority for literature review.

Project staff then conducted comprehensive literature reviews in the selected areas, including (1) the use of echocardiography and radionuclide ventriculography for measuring left-ventricular function, (2) the use of revascularization, (3) the use of ACE inhibitors and other pharmacological agents, (4) patient education and compliance issues, and (5) followup monitoring. More than 1,000 articles were examined during this review. Of these, 237 were specifically used to develop the recommendations in this guideline and are cited in this document. A quality-assessment system (discussed below) was developed to rate methods used in these studies.

The literature reviews focused on the clinical benefits and harms (in terms of life expectancy and effects on quality of life) associated with each possible intervention and management strategy. Panelists weighed the expected benefits and harms of diagnostic and treatment strategies in order to derive their management recommendations. Costs of treatment were not explicitly taken into account when these recommendations were formulated, although the panel recognized the potential systemwide financial implications of those recommendations.

Results of the literature reviews were described in several reports prepared by project staff and distributed to panelists before the May 1992 panel meeting. Based on these reports, the panel formulated initial management recommendations for the diagnosis and treatment of heart failure patients.

On June 26, 1992, before its regular meeting, the panel convened a public open forum to give interested parties an opportunity to comment on any aspect of heart failure management of interest to them. Several organizations and individuals offered oral or written testimony that the panel considered during its subsequent deliberations. In subsequent meetings, the panel revised its management recommendations on the basis of continued evaluations of the literature, panel discussion, and input from project consultants.

During January and February 1993, a formal round of peer review was conducted in which the guideline document was circulated to about 40 professional organizations and individuals. Twenty-nine formal responses were obtained. The recommendations and suggestions contained in these responses were summarized for the panel, which considered them during its February 1993 meeting. The guideline was revised after this meeting.

Finally, a pilot review was conducted on the revised document in August 1993 to evaluate the usefulness, clarity, and feasibility of the Clinical Practice Guideline, Quick Reference Guide for Clinicians, and Patient and Family Guide. These documents were sent to 30 organizations and individuals representing primary care practitioners. Twenty responses were obtained. The strong consensus of reviewers who tried using the documents with patients was that the algorithm format provided a useful framework for applying the recommendations contained in the Clinical Practice Guideline and Quick Reference Guide for Clinicians to individual patients, particularly in complicated cases. Reviewers also believed that the structure of the documents was reasonably clear and user friendly. In addition to comments concerning the utility of the guideline in actual clinical settings, pilot reviewers also offered additional comments and recommendations. These were considered by the panel, and changes were made as appropriate.

Literature Quality-Rating System

In developing an evidence-based guideline, the quality of the studies used by the panel to guide its recommendations is important. Project staff defined seven levels of evidence for the panel's use. The levels depended on the study design, the quality of the studies (discussed below), sample sizes, and the consistency of results across centers and studies. The seven levels were:

• I.

• Evidence from large, well-conducted randomized controlled trials (RCT's).

• II.

• Evidence from small, well-conducted RCT's.

• III.

• Evidence from well-conducted cohort studies.

• IV.

• Evidence from well-conducted case-control studies.

• V.

• Evidence from uncontrolled or poorly controlled studies.

• VI.

• Conflicting evidence, but tending to favor the recommendation.

• VII.

• Expert opinion.

A quality-rating system was developed based on the large body of work done in the area of meta-analysis. [20-24] The following principles were incorporated:

• The main focus of the rating system should be on the detection of methodologic flaws that might introduce enough bias to threaten the validity of a study's results. (These were considered major flaws.)

• Those elements that are unlikely to have a major influence on validity and may represent poor reporting or editing (minor flaws) should be recorded separately rather than being incorporated into a global score of quality.

• The criteria used for grading studies should be as explicit as possible.

• If the data from a study were analyzed incorrectly by the author and enough information is given to analyze the data correctly, the corrected results should be used.

Studies were assessed on eight aspects of method: patient selection, allocation of patients to treatments, description of the therapeutic regimen, study administration, withdrawals, patient blinding, outcome measurement, and statistical analysis. Specific criteria were defined for major and minor flaws in each aspect of method.

The classification scheme for the guideline document was later simplified into a three-level system for strength of evidence:

• A

• Good evidence: Evidence from well-conducted RCT's or cohort studies (Levels I-III).

• B

• Fair evidence: Evidence from other types of studies (Levels IV-VI).

• C

• Expert opinion (Level VII).

Recommendations for Future Research

The heart failure guideline panel identified several areas related to diagnosis and treatment that were not addressed, or that were inadequately studied, in available literature. These topics constitute important areas for future research that could be addressed to the extent possible in updates of the guideline. Guideline updates are currently planned for this and other AHCPR-sponsored clinical practice guidelines.

Questions identified by the panel as in need of research were:

• 1. What are the current practice patterns for patients with heart failure?

  • How often are ACE inhibitors used?How often are heart failure patients hospitalized over the course of a year?

  • How many office visits do heart failure patients make during a year?

  • How often are various medications employed?

  • How often are echocardiograms and other noninvasive tests performed?

  • How often are heart failure patients revascularized?Are there geographic variations in management practices?

• 2. What are the costs associated with current patterns of treatment for heart failure patients?

  • Costs of office visits and hospitalizations.

  • Costs of testing.

  • Costs of medications.

  • Costs of revascularization.

  • Geographic variation in costs.

• 3. What are the sensitivity, specificity, and predictive value of noninvasive testing for ischemia in patients with heart failure? (For example, how often are positive noninvasive tests predictive of significant coronary lesions appropriate for revascularization?) Existing published data are based on patients with angina as their primary symptom.

• 4. What are the outcomes (e.g., mortality, functional status, quality of life) of revascularization (CABG versus percutaneous transluminal coronary angioplasty [PTCA]) in patients with heart failure, with or without significant angina? Again, most published outcomes data on revascularization apply to patients with angina as the predominant symptom. In addition, surgical techniques have changed since the time of most available studies.

• 5. Can patients play a more active role in their own management? Several panelists are interested in developing and testing a patient-centered self-management protocol that would incorporate patient preferences. One aspect of this protocol would probably address drug dose titration techniques according to patient-measured endpoints (e.g., daily weights). Protocol evaluation would address the outcomes (e.g, hospitalization, patient-reported health and functional status), costs, and patient satisfaction associated with the experimental intervention versus control.

• 6. What is the best way to initiate and titrate drug therapy? A study of the cost, patient acceptance, and outcomes associated with sequential versus combined administration of common pharmacological agents (e.g., diuretics, ACE inhibitors, digoxin) would be desirable.

• 7. What is the value of anticoagulation in heart failure patients?

• 8. What are patient preferences and expectations for several key areas of management?

  • Medical versus surgical management approaches.

  • Rehabilitation and exercise programs.

  • Compliance with salt- and alcohol-restricted diets.

  • Quality-of-life aspects of living with heart failure and its treatments.

• 9. How can the built-in biases of tertiary center data bases be overcome in analyzing practice patterns, outcomes, and associated costs? Panelists have suggested that a national heart failure patient registry should be established at several different practice environments. A standard set of variables would be coded, including descriptions of the presence/absence/degree of concomitant angina, management strategies pursued, costs, and outcomes.

• 10. What criteria should be used to follow patients and determine their response to treatment? Current evidence suggests that available clinical tests (e.g., echocardiogram) do not provide information pertaining to changes in patient prognosis or quality of life. Development and testing of functional health status and quality-of-life measures for use in the heart failure population could prove extremely valuable by providing endpoints for monitoring patient response to treatment.

• 11. Given the poor prognosis associated with heart failure and the lack of donor hearts, what is the role of new and innovative forms of treatment (e.g., mechanical assist devices)?

• 12. What is the impact of heart failure on patients' families? How do family members or other caregivers respond to the burden of caring for these patients?

• 13. What interventions improve compliance with low-sodium diets, exercise or activity recommendations, and medication regimens?

• 14. What nonpharmacological interventions (e.g., family and patient support groups, strategies to improve psychosocial status) improve patients' and families' ability to cope with heart failure?

• 15. What will be the impact of this and other guidelines on clinical practice and outcomes?

Symptoms Suggestive of Heart Failure

Symptoms suggestive of heart failure include the following:

• Paroxysmal nocturnal dyspnea.

• Orthopnea.

• Dyspnea on exertion.

• Lower extremity edema.

• Decreased exercise tolerance.

• Unexplained confusion, altered mental status, or fatigue in an elderly patient.

• Abdominal symptoms associated with ascites and/or hepatic engorgement (e.g., nausea or abdominal pain).

Many patients with severely impaired left-ventricular function have no symptoms of heart failure. Marantz et al. reported that 20 percent of patients with EF's less than 40 percent met no clinical criteria for heart failure, [26] and Mattleman et al. found that only 42 percent of patients with left-ventricular EF's less than 30 percent had dyspnea on exertion. [27]

When clinical heart failure develops, dyspnea on exertion appears to be the earliest symptom. In a study of patients with known coronary disease undergoing coronary artery angiography, two-thirds of patients with elevated pulmonary capillary wedge pressures had dyspnea on exertion. [28] Only 31 percent had paroxysmal nocturnal dyspnea, 23 percent had a history of leg edema, and 21 percent had orthopnea. However, many patients with normal pulmonary capillary wedge pressures also had dyspnea on exertion (specificity: 52 percent), whereas the other symptoms were unusual unless the wedge pressure was elevated (specificity: 76-81 percent).

Although the presence of any one of the symptoms listed above is sufficient to warrant consideration of heart failure as the underlying cause, orthopnea, paroxysmal nocturnal dyspnea, and progressive dyspnea on exertion are of greatest concern. The other symptoms listed above are less likely to represent heart failure if they occur in isolation. A history of previous MI, poorly controlled hypertension, or other heart disease,

It is important to note that the symptoms listed above are not always due to heart failure. Dyspnea (whether at rest or on exertion) can be caused by a wide range of conditions, including most pulmonary diseases, obesity, deconditioning, volume overload from nephrotic syndrome or renal failure, intermittent cardiac ischemia, anxiety, and acute lower respiratory infections. Patients with these conditions may not require evaluation for heart failure unless the symptoms are more severe than can be accounted for on the basis of those conditions.

Dyspnea on exertion is particularly common in patients with COPD, interstitial lung disease, chronic pulmonary thromboembolic disease, and primary pulmonary hypertension. It is usually possible to distinguish between cardiac and pulmonary causes of dyspnea on the basis of history, physical examination, and chest x-ray.

Absence of cardiomegaly on chest x-ray weighs against the diagnosis of heart failure due to systolic dysfunction except when pulmonary hyperinflation may mask an enlarged heart. [29, 30] Patients with evident pulmonary disease generally do not require a full evaluation for heart failure unless their symptoms are out of proportion to the severity of their lung disease.

In addition to obtaining information about symptoms suggestive of heart failure, it is important that a complete medical history be obtained during the initial evaluation. In particular, information should be sought concerning previous MI; angina or equivalent (e.g., flash pulmonary edema); diabetes; renal, pulmonary, thyroid, or gastrointestinal disease; and medications.

Physical Examination

The initial workup of patients with symptoms suggestive of heart failure begins with a careful physical examination.Abnormal physical findings that tend to support a diagnosis of heart failure include the following:

• Elevated jugular venous pressure or positive hepatojugular reflux.

• A third heart sound.

• Laterally displaced apical impulse.

• Pulmonary rales that do not clear with cough.

• Peripheral edema not due to venous insufficiency.

Many patients with moderate-to-severe left-ventricular systolic dysfunction or early symptoms of heart failure have no physical findings. The most sensitive physical finding is the third heart sound. A third heart sound was found in 68 percent of patients with EF's below 30 percent. [27] Rales and a displaced apical impulse were present in only 37 and 42 percent of patients, respectively. Jugular venous distention and peripheral edema appear to be the least sensitive signs. [28] Studies of patients with severe heart failure undergoing cardiac catheterization before transplantation have also shown that physical findings are frequently absent even in patients with pulmonary capillary wedge pressures as high as 35 mmHg. [31]

The specificity of physical findings is less well defined because this has not been studied in a patient population without known or suspected heart disease. Elevated jugular venous pressure (as manifested by elevated jugular vein pulsations) and third heart sound are the most specific signs of heart failure. Rales in a patient with compatible symptoms and no known pulmonary disease are highly suggestive of heart failure. Lower extremity edema is a relatively nonspecific finding common in patients without heart failure. The most common cause of peripheral edema is venous insufficiency, not heart failure. Other causes of lower leg edema include lymphatic obstruction, conditions associated with hypoalbuminemia, deep venous thrombosis, and treatment with a calcium-channel blocker. [31, 32] Leg edema is unlikely to be caused by heart failure if there is no elevation in the jugular venous pressure and no orthopnea, paroxysmal nocturnal dyspnea, or dyspnea on exertion.

As mentioned, physical examination findings may not be highly sensitive indicators of heart failure. Some patients with heart failure do not manifest any of these signs, so the absence of physical findings should not dissuade the clinician from undertaking a thorough evaluation of patients with unexplained symptoms of heart failure. [31, 32] Care must be taken in assessing clinical signs of heart failure, because several studies suggest that intra- and interrater reliability of physical signs is often lower than would be desirable. [33, 34]

Diagnostic Testing

A chest x-ray should be obtained as an early diagnostic test. Cardiomegaly in symptomatic patients is highly suggestive of heart failure, especially when accompanied by pulmonary venous congestion. [27, 29, 30] As discussed earlier, a normal chest x-ray weighs against a diagnosis of heart failure but does not rule it out. Besides helping to distinguish cardiac from pulmonary causes of dyspnea, chest x-rays can also provide information concerning pulmonary disease, calcified heart valves, left-atrial size, and right-ventricular versus left-ventricular enlargement.

An ECG should also be performed. Among the possible findings of interest are acute ischemia, arrhythmias, prior MI, left-ventricular hypertrophy, and conduction abnormalities. However, the ECG is usually nonspecific in patients with heart failure. [35]

In some patients it may be difficult to distinguish between symptoms due to cardiac and pulmonary dysfunction. Screening spirometry may be helpful in these cases, although measurement of EF may be equally (or more) illuminating.

An initial exercise test may occasionally be useful to determine a patient's exercise tolerance when the history regarding dyspnea on exertion is difficult to evaluate. This situation can occur in patients who (1) are sedentary, (2) are unable to provide an accurate history, (3) usually deny or minimize symptoms, or (4) have histories that conflict with those of family members. Patients with unexplained decreases in exercise tolerance without apparent cause should also be evaluated for heart failure.

Table 3. Recommended Tests for Patients With Signs (more...)

Table 3. Recommended Tests for Patients With Signs or Symptoms of Heart Failure

Test Recommendation	Finding	Suspected Diagnosis
Electrocardiogram	Acute ST-T wave changes Atrial fibrillation, other tachyarrhythmia	Myocardial ischemia Thyroid disease or heart failure due to rapid ventricular rate
	Bradyarrhythmias	Heart failure due to low heart rate
	Previous MI (e.g., Q waves)	Heart failure due to reduced left-ventricular performance
	Low voltage	Pericardial effusion
	Left-ventricular hypertrophy	Diastolic dysfunction
Complete blood count	Anemia	Heart failure due to or aggravated by decreased oxygen-carrying capacity
Urinalysis	Proteinuria	Nephrotic syndrome
	Red blood cells or cellular casts	Glomerulonephritis
Serum creatinine	Elevated	Volume overload due to renal failure
Serum albumin	Decreased	Increased extravascular volume due to hypoalbuminemia
T4 and TSH (obtain only if atrial fibrillation, evidence of thyroid disease, or patient age >65)	Abnormal T4 or TSH	Heart failure due to or aggravated by hypo/hyperthyroidism

Note: TSH = Thyroid-stimulating hormone, MI = myocardial infraction.

Probably the most common complicating factor for heart failure is hypertension. In patients with hypertension and clinically evident heart failure, every effort should be made to control blood pressure.

Hematocrits of 25-35 percent may aggravate underlying heart failure; patients with anemia in this range should be managed as appropriate (e.g., with iron supplements for patients with iron deficiency anemia, or with transfusion if there is evidence of ongoing ischemia) while receiving continued evaluation and management of their heart failure. Hematocrits below 25 percent can produce signs and symptoms of heart failure without underlying cardiac abnormalities; patients with anemia in this range should be reevaluated after correction of anemia to determine whether additional workup for heart failure is required. Extreme caution should be exercised if transfusion becomes necessary, as this process may exacerbate symptoms of heart failure. Use of packed red blood cells is recommended when transfusion is necessary.

Renal insufficiency or failure may cause volume overload that mimics heart failure or exacerbates underlying left-ventricular dysfunction. These patients should generally be treated with diuretics or dialysis and reassessed after correction of volume overload to determine whether further evaluation for heart failure is necessary. Decreased serum albumin may also mimic or exacerbate heart failure.

Hospital Admission and Discharge Criteria

Most of the guideline's evaluation and care recommendations can be accomplished in either an inpatient or outpatient setting. The guideline does not address strategies specific to inpatient settings (e.g., pulmonary artery catheter, cardiac assist devices) but does make some statements concerning the transition between outpatient management and inpatient care.

Specifically, practitioners should be aware of the appropriate indications for hospitalization and for discharge from hospital. The above recommendations are made in the absence of specific data, although studies of predictors of hospital mortality were useful in formulating these recommendations. [6, 8, 63] Some patients with the above findings may be managed at home or in an assisted living or nursing home setting if, in the physician's carefully considered opinion, it is safe to do so and adequate outpatient monitoring and followup visits can be arranged.

Proper discharge planning is an essential component of managing patients with heart failure. Heart failure is one of the most common causes for recurrent admission to hospitals, and many of these admissions may be avoidable. Readmission rates as high as 57 percent within 90 days have been reported in patients over the age of 70 years. [64] Vinson et al. found that the factors associated with readmission to hospitals were: failed social support systems (21 percent), inadequate followup (20 percent), failure to seek medical attention promptly when symptoms recurred (20 percent), noncompliance with diet (18 percent), noncompliance with medication (15 percent), and inadequate discharge planning (15 percent). [64] Gooding and Jette found that 36 percent of 147 patients were readmitted within 6 months. [65] Patients were more likely to be readmitted if the initial discharge site was home rather than a secondary facility (39 and 19 percent, respectively). These authors highlighted the importance of securing compliance for medication and dietary regimens and called for better home care after discharge and better coordination of care between secondary and primary care providers.

Rosenberg demonstrated that better patient education decreased hospital admissions for patients with heart failure. [66] Perlman et al. demonstrated that public health nurses could detect deterioration in heart failure at a stage early enough to allow intervention that avoided admission to the hospital. [67] Patients should usually have a followup contact within 7 days after discharge to (1) make sure that medications are being taken properly, (2) assess compliance with reduced salt diet, (3) ensure that patient's weight is stable, (4) adjust the dosage of diuretics and other medications if necessary, and (5) determine that the patient, family, and caregivers understand when and how to contact the practitioner. Electrolytes, BUN, and creatinine should also be checked about 1 week following discharge, and medications adjusted as necessary.

Interpretation of Left-Ventricular Function Testing

The majority of patients with heart failure have moderate-to-severe left-ventricular systolic dysfunction and EF's of <35-40 percent. This guideline is directed at the management of such patients. However, patients with symptoms of heart failure and EF's greater than 40 percent may still have heart failure due to left-ventricular diastolic dysfunction, valvular disease, or pericardial disease. The majority of these etiologies will be discernible with echocardiography. A full discussion of the diagnosis and treatment of these conditions is beyond the scope of this guideline, although a few comments on diastolic dysfunction are necessary because of its high prevalence.

Diastolic Dysfunction

As many as 40 percent of patients with a clinical diagnosis of heart failure have preserved left-ventricular systolic function and no evidence of valvular heart disease. [26, 29, 30, 68, 69] Most of these individuals have left-ventricular diastolic dysfunction. In these cases, the left ventricle has increased diastolic stiffness (reduced compliance) and cannot fill adequately at normal diastolic pressures. [85] The elevated pressures required for filling result in symptoms of pulmonary congestion. In addition, the reduced left-ventricular filling volume leads to lowered stroke volumes and symptoms of poor cardiac output.

Most diastolic dysfunction results from coronary artery disease or hypertension. Because ischemia can produce diastolic dysfunction before systolic dysfunction develops, physiologic testing for ischemia should be considered in patients with diastolic dysfunction, particularly when dyspnea on exertion is a prominent symptom. [85] Hypertension is also a common cause of diastolic dysfunction, [86-89] which can develop even in the absence of left-ventricular hypertrophy. [86, 90]

The optimal treatment of diastolic dysfunction is not well defined. For a full discussion, readers are referred to the recent review by Bonow and Udelson. [85] At this time, it appears that beta-blocking agents, calcium-channel blockers, and the judicious use of diuretics are the treatments of choice. Agents used to treat systolic dysfunction may be deleterious in patients with diastolic dysfunction. Excessive diuresis can reduce stroke volume and cardiac output. Digitalis may further decrease left-ventricular compliance. The role of ACE inhibitors is not clear. Vasodilators may be detrimental in diastolic dysfunction although ACE inhibitors may have beneficial effects by directly improving ventricular relaxation and causing regression of hypertrophy.

Determination of Specific Etiologies for Left-Ventricular Systolic Dysfunction

The most common causes of left-ventricular systolic dysfunction are coronary artery disease, idiopathic dilated cardiomyopathy, hypertension, and alcohol abuse. The most common potentially reversible cause of heart failure is myocardial ischemia. Therefore, patients should be carefully questioned concerning a history of chest pain or recurring episodes of sudden pulmonary edema suggestive of ischemia. Evaluation of patients with angina is discussed subsequently in the section on revascularization. Alcoholism is important to detect and treat because alcohol may aggravate cardiac dysfunction. [91-93] Where appropriate, patients should also be asked about cocaine use, and a urine test for cocaine may be helpful in selected patients. Specific treatable etiologies (e.g., sarcoidosis) should be considered when the constellation of systemic findings suggests a diagnosis. Patients with a history of liver disease, unexplained hepatomegaly, diabetes or other endocrine dysfunction, or bronze discoloration of the skin should be evaluated for hemochromatosis with a serum iron level, total iron binding capacity, and ferritin level. An exhaustive search for the etiology of heart failure in a patient without specific findings on history and physical examination is of little value.

General Counseling

Table 5. Suggested Topics for Patient, Family, and (more...)

Table 5. Suggested Topics for Patient, Family, and Caregiver Education and Counseling

General Counseling Explanation of heart failure and the reason for symptoms Cause or probable cause of heart failure Expected symptoms Symptoms of worsening heart failure What to do if symptoms worsen Self-monitoring with daily weights Explanation of treatment/care plan Clarification of patient's responsibilities Importance of cessation of tobacco use Role of family members or other caregivers in the treatment/care plan Availability and value of qualified local support group Importance of obtaining vaccinations against influenza and pneumococcal disease

Prognosis Life expectancy Advance directives Advice for family members in the event of sudden death

Activity Recommendations Recreation, leisure, and work activity Exercise Sex, sexual difficulties, and coping strategies

Dietary Recommendations Sodium restriction Avoidance of excessive fluid intake Fluid restriction (if required) Alcohol restriction

Medications Effects of medications on quality of life and survival Dosing Likely side effects and what to do if they occur Coping mechanisms for complicated medical regimens Availability of lower cost medications or financial assistance

Importance of Compliance With the Treatment/Care Plan

It is impossible for patients to absorb all necessary information in one session. For patients who have been hospitalized, it is important to discuss medications and diet before discharge from the hospital and to repeat this information in the outpatient setting. This will allow patients time to assimilate the information and to formulate more questions. Because of the large number and often specialized nature of topics that are important to discuss with heart failure patients, practitioners should seek the assistance of dieticians, nurse educators, clinical nurse specialists, pharmacists, and support groups as part of a team approach to providing patient education. Visiting or home health nurses can be especially useful.

Finally, providers should be sensitive to differences in language and culture that may impair patients' understanding and compliance. Concerted attempts should be made to mitigate the influence of these factors.

The typical symptoms of worsening heart failure (orthopnea, paroxysmal dyspnea, leg edema, or exercise intolerance) should be explained, and patients should be advised to contact their doctor or nurse if such symptoms develop. All patients should also be told to obtain a bathroom scale and to weigh themselves each morning (after urinating and before eating). They should contact their provider if their weight has changed by more than 3-5 pounds since their last clinical evaluation. If patients know the symptoms and signs of worsening heart failure, they may be able to seek care early and avoid hospitalization. In addition, in compliant patients a diuretic regimen that the patient adjusts on the basis of weights taken at home can be a useful method for preventing decompensation and hospitalization.

An explanation of what patients can expect to experience will help avoid anxiety over symptoms and prevent patients from becoming afraid to perform daily activities that might provoke shortness of breath. Patients should be advised to stay as active as possible, as described in the section on activity recommendations. However, patients should understand that they may feel tired the following day if they overexert themselves. Sexual difficulties are common in these patients, and part of this arises from fear that exertion is detrimental. Sexual practices might need to be modified to accommodate patients with limited exercise tolerance. The practitioner should not expect the patient to bring up these issues; when appropriate, these topics should be discussed directly and openly.

Patients with heart failure must understand the serious implications of this diagnosis. Patients must be provided with accurate information in order to make decisions and plans for the future (see subsequent section on discussion of prognosis). At the same time, it is important to maintain hope and morale. Psychological factors may be more important to patients' quality of life and social functioning than the degree of physiologic impairment. [94, 95] Clinical nurse specialists, educators, family members, other caregivers, and qualified local support groups can all play an important role in optimizing patients' functioning and quality of life.

Activity Recommendations

Until the past few years, reduced activities and bed rest were considered a standard part of the care of patients with heart failure. [96] Although this practice may promote diuresis in the short term, it may also carry important risks and have a long-term detrimental effect on physical functioning. [97] Even short periods of bed rest result in reduced exercise tolerance and aerobic capacity [98-100] as well as muscular atrophy and weakness. [99, 101] Recent studies show that patients with heart failure can exercise safely, and regular exercise may improve functional status and decrease symptoms. [102-106] Neurohormonal activation is also diminished by a regular home exercise program. [107] Moreover, there is no evidence that exercise negatively affects the natural history of heart failure, with the possible exception of patients with acute myocarditis or a recent MI, who may be harmed by exercise. [108, 109]

The clinical improvement that occurs with exercise training programs probably results from effects on skeletal muscle rather than changes in myocardial function. [106, 110-112] Lactate production at submaximal exercise levels is reduced after training. Bicycle ergometry and arm ergometry have both been used for training programs, and Coats et al. have shown improvements with an unsupervised home cycling program. [102] There is insufficient evidence to recommend a specific type of training program or the routine use of supervised rehabilitation programs. However, such programs may benefit patients who are anxious about exercising; are dyspneic at a low work level; or have stable angina, a recent MI, or a recent CABG. In addition, supervised rehabilitation programs can provide patient and family counseling, regular encouragement and interpersonal contact, and assistance with facilitating patient compliance with treatment recommendations.

The NYHA classification is a four-level scheme for grading the functional incapacity of patients with cardiac disease. NYHA levels can be described as follows: I-Cardiac disease without resulting limitations of physical activity; II-Slight limitation of physical activity-comfortable at rest, but ordinary physical activity results in fatigue, palpitation, dyspnea, or anginal pain; III-Marked limitation in physical activity-comfortable at rest, but less than ordinary physical activity causes fatigue, palpitation, dyspnea, or anginal pain; IV-Inability to carry on any physical activity without discomfort or symptoms at rest. Although criticized for a lack of reliability, this system is still widely used.

Dietary Recommendations

Although a reduced-sodium diet has been a mainstay in managing heart failure, no studies have evaluated a specific dietary sodium restriction. [113-121] Thus, it is not clear from the literature whether mild sodium restriction (3 grams per day) is adequate for most patients or whether moderate restriction (2 grams per day) is beneficial. A 2-gram sodium diet is unpalatable for most patients, and the cost of low-sodium foods can be a burden for some patients. [118] This may lead to noncompliance, which is a major factor precipitating hospital admission for heart failure (see below). However, if dietary salt intake is excessive, diuretic dosing may be complicated or excessive, and potassium wasting may be exacerbated. [122]

A 3-gram sodium diet may be a reasonable and realistic target for patients with mild-to-moderate heart failure. This level of sodium intake can be achieved fairly easily by not adding salt to foods and avoiding salty foods. [120] However, many elderly patients may find even a 3-gram sodium diet unpalatable. Counseling and flexibility may be needed to promote compliance and to ensure that patients do not become malnourished. Patients who require high doses of diuretics will need to reduce their daily sodium intake to 2 grams or less. Avoiding processed foods and milk products (e.g., cheese) can reduce daily sodium intake to as little as 1 gram. [114]

All heart failure patients should receive specific dietary guidelines. Referral to a dietician, clinical nurse specialist, or nurse practitioner for dietary education and counseling is recommended. Handouts and educational guides are often inadequate for patients with reading difficulties and patients from ethnic groups with different food preferences. In all cases, it is essential to involve the spouse and family in educational efforts concerning the importance of diet and salt restrictions. In some cases, family recipes and menus may need to be altered in order to facilitate compliance with salt restrictions.

Acute ingestion of alcohol depresses myocardial contractility in patients with known cardiac disease. [123, 124 This may be clinically significant in patients with heart failure, although there are no studies that address this issue. Complete abstention from alcohol is crucial for patients with alcohol-induced cardiomyopathy. 91-93] For patients without a history of alcoholism, it is unclear whether abstinence makes a difference in functional status or mortality. In general, alcohol use should be discouraged. If patients want to continue to drink, they should be strongly advised to have no more than one drink per day. One drink equals a glass of beer or wine, or a mixed drink or cocktail containing no more than 1 ounce of alcohol.

Other dietary restrictions should be discouraged unless clearly indicated (e.g., a low-fat, low-cholesterol diet for severe hypercholesterolemia). Many patients with severe heart failure suffer from a syndrome of chronic wasting, referred to as "cardiac cachexia," that may be exacerbated by unnecessary dietary restrictions. Vitamin supplementation may be advisable because of water-soluble vitamin loss associated with diuresis and problems with gastrointestinal absorption of fat-soluble vitamins.

Discussion of Prognosis

Patient counseling with respect to prognosis should be guided by recent trials [10, 18, 19, 125] and the Framingham experience, [126] which indicate that the average annual mortality rate for patients with heart failure is approximately 10 percent per year. For patient subgroups, the following estimates of mortality rates per year can be made: NYHA Class II, 5-10 percent; NYHA Class III, 10-20 percent; NYHA Class IV, 20-50 percent.

These studies may overestimate the mortality rates for the general population of patients with heart failure. In a recent study from the Mayo Clinic, it was found that the heart failure patients from a population-based cohort had a lower mortality rate than patients who were followed at the Mayo Clinic, suggesting a referral bias. [127] The population-based cohort with EF's less than 35 percent had a mortality rate of approximately 5 percent per year.

Within each NYHA class, the presence of more severe symptoms, progressive symptoms, or accompanying angina would push these estimates to the higher end of the range; conversely, milder symptoms, clinical stability, and absence of angina would shift the estimate to the lower end. As many as half of all cardiac deaths are sudden, and up to 25 percent of all deaths occur without significant worsening of heart failure. [10, 18, 126]

If a patient desires resuscitation, family members should consider learning cardiopulmonary resuscitation. Such a course should be combined with psychosocial support for patients and family members because it may otherwise have negative psychological consequences. [128] If a patient decides that resuscitation is not desired, it is crucial to discuss with family members or caretakers what to do in the event of sudden death. If paramedics are called, a chain of events is likely to take place that may be difficult to control and contrary to the patient's wishes. A call to 911 is often a reflex response to a loved one's death. Instead, when resuscitation is not desired, family members or other caregivers should be instructed to call a doctor, nurse, hospice worker, or other designated responsible individual.

The Problem of Noncompliance

Noncompliance is an important problem with any chronic disease, and heart failure is no exception. [64, 129] Noncompliance may reduce life expectancy (e.g., when patients fail to take beneficial medications) and may also be a major cause of hospitalizations. Ghali et al. found a 54-percent rate of noncompliance with diet or medications, and noncompliance was the most common factor precipitating readmission. [129] Vinson et al. found that 27 percent of 140 patients hospitalized for heart failure were rehospitalized within 90 days of discharge and 27 percent were rehospitalized at least once for recurrent heart failure. [64] Twenty-two separate admissions were subjectively judged to have resulted from medication or dietary noncompliance. Readmissions for heart failure could be substantially reduced if compliance were improved.

Although it is clear that noncompliance is a major problem in caring for patients with heart failure, it is less clear what should be done to improve this situation. Rosenberg evaluated the impact of a coordinated team approach to planning, education, and supervision on dietary compliance and readmission rates. [66] Patient-directed group meetings were at the core of the program. Dietary compliance was improved, and hospital admissions were reduced from 46 to 12 percent compared with the previous year and from 31 to 17 percent compared with a control group of patients at another hospital. These results are very encouraging because group programs can provide a more cost-effective intervention than one-on-one teaching. The group process may have other therapeutic aspects besides its educational component. Meta-analyses of educational programs for other chronic diseases also support their value in improving compliance and outcomes. [130, 131] Educational programs or qualified local support groups should be a routine part of the treatment and care of patients with heart failure. Physicians and other practitioners should be aware of the problem of noncompliance and its causes and should reinforce the importance of compliance at routine visits. In addition, although not tested in heart failure patients, effective interventions to promote compliance in hypertensives and patients with coronary artery disease have included family involvement in care and the use of patient contracts. Similar techniques may prove effective in the context of heart failure.

Initial Pharmacological Management

Diuretic Therapy

Patients with heart failure and signs of significant volume overload should be started immediately on a diuretic. Patients with mild volume overload can be managed adequately on thiazide diuretics, whereas those with more severe volume overload should be started on a loop diuretic. (Strength of Evidence = C.)

Symptoms of volume overload include orthopnea, paroxysmal nocturnal dyspnea, and dyspnea on exertion; signs include pulmonary rales, a third heart sound, jugular venous distention, hepatic engorgement, ascites, peripheral edema, and pulmonary vascular congestion or pulmonary edema on chest x-ray. There are few studies of optimal diuretic therapy for heart failure. Patients with mild heart failure can usually be managed adequately on thiazide diuretics. [132, 133] Moreover, some patients prefer a thiazide diuretic over furosemide, probably because it causes less acute diuresis than furosemide. [132] If a patient has severe volume overload at the time of presentation, severe renal insufficiency (estimated creatinine clearance <30 mL/min), or persistent edema despite thiazide diuretics, a loop diuretic should be used instead of a thiazide. Treatment of mild heart failure with ACE inhibitors alone is discussed in the next section.

Table 6. Medications Commonly Used for Heart Failure (more...)

Table 6. Medications Commonly Used for Heart Failure

Drug	Initial Dose (mg)	Target Dose (mg)	Recommended Maximal Dose (mg)	Major Adverse Reactions
Thiazide Diuretics
Hydrochlorothiazide	25 QD	As needed	50 QD	Postural hypotension, hypokalemia, hyperglycemia, hyperuricemia, rash. Rare severe reaction includes pancreatitis, bone marrow suppression, and anaphylaxis.
Chlorthalidone	25 QD	As needed	50 QD
Loop Diuretics
Furosemide	10-40 QD	As needed	240 BID	Same as thiazide diuretics.
Bumetanide	0.5- 1.0 QD	As needed	10 QD
Ethacrynic acid	50 QD	As needed	200 BID
Thiazide-Related Diuretic
Metalazone	2.5[a]	As needed	10 QD	Same as thiazide diuretics.
Potassium-Sparing Diuretics
Spironolactone	25 QD	As needed	100 BID	Hyperkalemia, especially if administered with ACE inhibitor; rash; gynecomastia (spironolactone only).
Triamterene	50 QD	As needed	100 BID
Amiloride	5 QD	As needed	40 QD
ACE Inhibitors
Enalapril	2.5 BID	10 BID	20 BID	Hypotension, hyperkalemia, renal insufficiency, cough, skin rash, angioedema, neutropenia.
Captopril	6.25-12.5 TID	50 TID	100 TID
Lisinopril	5 QD	20 QD	40 QD
Quinapril	5 BID	20 BID	20 BID
Digoxin	See pages 58-60	See pages 58-60	See pages 58-60	Cardiotoxicity, confusion, nausea, anorexia, visual disturbances.
Hydralazine	10-25 TID	75 TID	100 TID	Headache, nausea, dizziness, tachycardia, lupus-like syndrome.
Isosorbide Dinitrate	10 TID	40 TID	80 TID	Headache, hypotension, flushing.

[a] Given as a single test dose initially.

Note: ACE = angiotensin-converting enzyme,BID = twice a day, QD = once a day,TID = three times a day.

Diuretic dosing is summarized in Table 6. Proper dosing depends on the patient's size, age, renal function, compliance with dietary sodium restriction, and the amount of edema. Hydrochlorothiazide should generally be given in doses of 25-50 mg per day. Patients who have persistent volume overload on 50 mg of hydrochlorothiazide per day or the equivalent should be changed to a loop diuretic. Furosemide should be started at 20-40 mg once daily. Elderly patients may be started on 10 mg per day initially. In elderly patients with urinary incontinence or urgency, guidance should be provided concerning the timing of diuretic administration (especially loop diuretics) in relation to the patient's activities to minimize noncompliance and prevent patients from isolating themselves from society.

Patients with pulmonary edema or with marked volume overload (e.g., anasarca) should be given intravenous furosemide initially. Diuretics should then be titrated to achieve resolution or improvement of signs and symptoms of volume overload. There is no standard target dose. Occasionally, a patient with mild heart failure may require a diuretic on alternating days or as needed.

It is generally better to give hydrochlorothiazide or furosemide as a single daily dose in the outpatient setting. If the initial oral dose is inadequate, more diuresis will usually be obtained by doubling the dose rather than by giving the same dose twice daily. At high doses, furosemide can be given twice daily, but it should usually not be given more frequently. Inpatients with severe volume overload may require more frequent intravenous doses to obtain the desired brisk diuresis. Doses of furosemide as high as 240 mg twice per day may be required in the most refractory patients, but doses over about 160 mg per day usually require additional efforts to improve diuresis (discussed subsequently). It is very important to avoid excessive diuresis before starting ACE inhibitors. Volume depletion may lead to hypotension or renal insufficiency when ACE inhibitors are started. The ACE inhibitor may facilitate diuresis in some patients. In general, the dose of diuretic may be increased in parallel with the ACE inhibitor, but it is important to avoid excessive diuresis that could prevent titrating the ACE inhibitor to full therapeutic levels.

Potassium depletion commonly occurs when patients are treated chronically with diuretics. However, ACE inhibitors decrease renal potassium losses and raise serum potassium levels, so many patients with heart failure who are treated with both agents may not develop potassium depletion. All patients should have their serum potassium levels checked frequently (e.g., every 3 days until stable) during initiation, titration, or modification of diuretic or ACE inhibitor therapy and every few months thereafter. Note, however, that serum potassium can be an unreliable indicator of total body potassium stores; patients with normal serum potassium levels can have low total body potassium stores. [133, 134] Thus, patients who have a serum potassium less than 4.0 mEq/L should be given a potassium-sparing agent or oral potassium supplementation.

Potassium-sparing diuretics may be more effective than oral potassium supplements at maintaining total body potassium stores. [134] When an ACE inhibitor is used together with a potassium-sparing agent or potassium supplements, the serum potassium level must be followed closely (e.g., every 3 days until stable).

Diuretics can also cause magnesium depletion, which often accompanies potassium depletion. If high doses of diuretics are used, magnesium levels should be followed and oral supplementation given when necessary. Hypocalcemia may be a useful clue to possible significant hypomagnesemia.

Additional Pharmacological Management

Many patients with heart failure remain symptomatic despite the measures discussed earlier under "Initial Pharmacological Management." Others have comorbid conditions, such as angina or hypertension, which can adversely affect heart failure if not properly managed. This section addresses the management of these patients. Few controlled trials are available to provide guidance in these complex situations, so most recommendations derive from clinical experience. Although hypertension and angina can often be managed by primary care physicians, the treatment of patients with resistant heart failure is more complex, and the risks of treatment are greater.

Persistent Hypertension Despite ACE Inhibitors and Diuretics

Persistent hypertension should be treated with a direct vasodilator, an alpha[sub 1]-adrenergic blocker, or a centrally acting alpha blocker. Because of their negative inotropic effect, calcium-channel blockers and beta blockers should be given only under the care of practitioners experienced in the use of these agents. (Strength of Evidence = C.)

Reduction of high blood pressure may in itself have a beneficial effect on the signs and symptoms of heart failure. Hypertension is a relative concept in patients with heart failure. Although a blood pressure of 135/85 mmHg may be acceptable for a patient with a normal EF, that same blood pressure may be harmful for a patient with severe left-ventricular systolic dysfunction. Some patients with heart failure will feel better when their blood pressure is low because their cardiac output is generally higher. Most patients will have some reduction in blood pressure with ACE inhibitors and diuretics, but some will remain hypertensive or have high-normal blood pressure. Additional diuresis may achieve some blood pressure reduction, and consideration should be given to increasing the dose of ACE inhibitors.

Table 7. Medications to Treat Hypertension or High-Normal (more...)

Table 7. Medications to Treat Hypertension or High-Normal Blood Pressure in Patients With Heart Failure

Drug	Initial Dose (mg)	Maximum Dose (mg)	Major Adverse Reactions
Direct Vasodilators
Hydralazine	10 TID	75 TID	Headache, nausea, dizziness, tachycardia, lupus-like syndrome.
Minoxidil	2.5 QD	80 QD	Fluid retention, hair growth, thrombocytopenia, leukopenia.
Alpha[sub 1]-Adrenergic Blockers
Doxazosin	1 QD	16 QD	Postural hypotension, dizziness, syncope, headache.
Terazosin	1 QD	20 QD
Prazosin	1 TID	10 TID
Centrally Acting Alpha Blockers
Clonidine tablets	0.1 BID	0.6 BID	Sedation, dry mouth, blurry vision, headache, bradycardia.
Clonidine patch	0.1 weekly	0.3 weekly	Same as clonidine; contact dermatitis.
Guanabenz	4 BID	64 BID	Similar to clonidine.
Guanfacine	1 QD	3 QD	Similar to clonidine.

Note:QD = once a day,BID = twice a day,TID = three times a day.

If additional antihypertensive treatment is necessary, direct vasodilators, alpha[sub 1]-adrenergic blockers or centrally acting alpha blockers should be used ( Table 7). Nitrates have only a small antihypertensive effect when blood pressure is elevated, although they frequently decrease blood pressure substantially in patients who are already marginally hypotensive. All calcium-channel blockers and beta blockers have a negative inotropic effect. As discussed previously, there is some evidence that patients with heart failure can tolerate beta blockers and have improved symptoms or functional status (see section on beta blockers), so these agents are not absolutely contraindicated. However, they must be used with extreme caution as antihypertensive agents, and they should only be used by practitioners with experience in the use of these agents for heart failure. See preceding discussion ("Angina and Heart Failure") regarding calcium channel blockers.

Heart Failure and Coronary Artery Disease

Coronary artery disease was the second most common identifiable risk factor for heart failure (after hypertension) in the Framingham study [126] but may now well be the most common etiology. The risk of developing heart failure is increased 2-3 times in patients with angina and 4-6 times for patients with a previous MI. [190] In the SOLVD trial, 38 percent of patients had current angina, and 66 percent had suffered a previous MI. [18] Most patients with angina and some patients with a previous MI have areas of myocardial ischemia that contribute to the degree of left-ventricular systolic dysfunction. This suggests that there may be a large number of patients with heart failure and potentially reversible ischemia.

Unfortunately, there are few published data on the prevalence of clinically significant ischemia in patients with heart failure. To provide guidance to clinicians, the panel defined three subgroups of heart failure patients who are likely to have very different probabilities of significant ischemia: patients with (1) angina, (2) a history of MI but no current angina, and (3) neither angina nor a past history of MI. Although patients with heart failure and normal epicardial coronary arteries can have exertional chest pain, [127, 191, 192] the majority of patients with this combination of symptoms will have some degree of myocardial ischemia due to epicardial coronary artery disease. Patients who have suffered previous MI's may have areas of viable myocardium in the region of the infarction or ischemic regions supplied by other coronary arteries.

Kotler and Diamond have summarized studies of the detection of multivessel disease following MI. [193] In these reports, 31-56 percent (weighted average: 44 percent) of patients who suffered an MI had significant ischemia in the distribution of another coronary artery. It is unclear how many of these patients had significant angina or heart failure. This finding suggests that a substantial number of patients with heart failure and a previous MI may have other significant areas of ischemia. Whether detection and correction of this "silent" ischemia would improve prognosis is not known. This issue is discussed subsequently in more detail.

Even less is known about the prevalence of significant ischemia in patients with heart failure and no angina or prior MI. The likelihood of ischemia in a given patient is highly dependent on the presence of other risk factors for heart failure (e.g., alcoholism, severe hypertension) and risk factors for coronary artery disease (e.g., age, sex, diabetes mellitus, smoking, family history, hypertension, hypercholesterolemia). Patients with multiple risk factors for coronary artery disease and no other obvious etiology for heart failure have a much higher likelihood of having reversible ischemia as a causative or contributing factor to their heart failure. Thus, it is probably worthwhile to pursue an evaluation for ischemia in such patients. Patients with another obvious etiology for heart failure and a low risk for coronary disease are much less likely to benefit from testing.

Benefits of CABG

For patients with ischemic cardiomyopathy, the goal of myocardial revascularization is to prevent further ischemic injury to remaining functional myocardium or to restore function to so-called hibernating myocardium (i.e., nonfunctional myocardial wall segments that are underperfused but still viable). Although it has not been proven that achievement of these goals increases survival in patients with compromised left-ventricular function, some survival benefit seems likely. Conversely, patients without remaining ischemic functional or hibernating myocardium are unlikely to benefit from revascularization.

Table 8. Cohort Studies of CABG Versus Medical Management (more...)

Table 8. Cohort Studies of CABG Versus Medical Management for Patients With Heart Failure or Reduced EF and Coronary Artery Disease

Author, Date	Study Years	Inclusion Criteria	Group		Average Age (years)[a]	Percentage With Clinical Heart Failure[a]	Percentage With Limiting Angina[a]	Mean Ejection Fraction[a]	Operative Mortality (percent)	Followup (years)	Total Mortality (percent)	Absolute Mortality Difference[b] (percent)
Yatteau et al., 1974[202]	1968-72	EF < 25%	Medical	42	55	(50)	(76)	(19%)	NA	1.0	31	-19
			Surgical	24	51				33		50
Manley et al., 1976[198]	1968-71	Depressed left-ventricular function	Medical	155	55	60	100	29%	NA	6.0	68	25
			Surgical	246	54	56	100	24%	14		43
Faulkner et al., 1977[197]	1969-75	EF < 30%	Medical	70	56	66	74	20%	NA	2.0	53	30
			Surgical	46	56	43	98	21%	4		23
Vliestra et al.,1977[200]	1966-72	EF < 25%	Medical	21	(52)	NR	100	NR	NA	2.0	67	7
			Surgical	10				NR	NR		60
Pigott et al., 1982[199]	1970-77	EF < 35%	Medical	115	54	26	37	25%	NA	5.0	66	29
			Surgical	77	54	21	65	27%	1.3		37
Alderman et al.,1983[196]	1975-79	EF >= 35%	Medical	420	54	19	51	26%	NA	3.0	33	10
			Surgical	231	56	11	72	27%	7		23
Hammermeister et al.,1982[201]	1969-74	Clinical heart failure	Medical	44	(53)	100	(82)	NR	NA	6.0	Hazard ratio 0.85[c]	NA
			Surgical	83	(54)	100	(96)	NR	NR		32[d]
Bounous et al., 1988[9]	1976-83	EF >= 40%	Medical	409	54	NR	NR	29%	NA	3.0	14[d]	18
			Surgical	301	55	NR	NR	33%	NR

[a] Numbers in parentheses indicate values for a larger group of patients in the study or for the combination of medical and surgical patients. Data on subsets of patients were not given.

[b] Medical mortality minus surgical mortality.

[c]Hazard ratio for death in patients with clinical heart failure treated with surgery compared with medical management.

[d]Adjusted for baseline prognostic factors.

Note: CABG = coronary artery bypass graft surgery, EF = ejection fraction, NA = not applicable, NR = not reported.

CABG has not been shown to improve survival in patients with heart failure who do not have angina. Although individual case reports suggest that some heart failure patients with large areas of reversible ischemia and no history of angina are improved with CABG, [206, 207] several case series and cohort studies show no evidence that patients without angina benefit from surgery. [196, 202, 208, 209] Alderman et al. [196] analyzed the Coronary Artery Surgery Study registry from 1975 to 1979 and found that patients with dyspnea on exertion or fatigue as their predominant complaint had the same 3-year mortality and symptom-free survival with CABG as with medical management (both 45.2 percent). The perioperative mortality rate was 7 percent. It is not clear whether more modern surgical techniques produce fewer perioperative deaths and better outcomes from surgery than reported in these studies.

In patients without angina, revascularization is most likely to be beneficial in the subset of patients with demonstrable myocardial ischemia or large areas of residual viable but jeopardized or hibernating myocardium. It is reasonable to assume that CABG in these patients would result in somewhat less reduction in mortality than that observed for patients with angina, although patients with large areas of ischemia might obtain a degree of improvement comparable to that seen in patients with angina. Also, patients whose heart failure symptoms are felt to represent ischemia (i.e., angina equivalent), such as those with episodic, severe heart failure or exertional dyspnea out of proportion to the degree of resting left-ventricular dysfunction, may benefit from revascularization as much as patients with angina. Patients who have only small amounts of ischemia would probably receive little or no benefit. However, there are no controlled studies to support any particular viewpoint on this subject.

Risks of CABG

As recently as 1982, EF was considered the most important predictor of perioperative mortality for CABG. In 1986, improved perioperative and surgical techniques had reduced EF to the fourth most important predictor, behind repeat CABG, emergent CABG, and age. [210] However, patients with depressed left-ventricular function are still at increased mortality risk. [210-212] It is difficult to predict the actual perioperative mortality risk in patients with heart failure. Compared with patients with EF's of 40 percent or greater, patients whose EF's were below 20 percent or between 20 and 39 percent had 3.4 and 1.5 times the chance of dying in the perioperative period, respectively [211] Christakis et al. reported a perioperative mortality rate of 11.7 percent in a series of 264 patients with EF's below 20 percent who underwent CABG at the University of Toronto between 1982 and 1986. [210]

Mortality rates increase substantially when the EF is below 20 percent or the heart failure is severe (NYHA Class IV). [210-212] Important additional determinants of mortality risk with CABG include age (risk increases approximately 0.5 percent per year above age 60); sex (higher in women, although part of this difference is explained by a higher mortality rate in small individuals); comorbid conditions (e.g., diabetes, significant renal dysfunction, significant COPD, or clinical cerebrovascular disease); prior cardiothoracic surgery (substantially increases risk); need for emergent CABG (e.g., failed PTCA or acute ischemic syndrome) or concomitant valve surgery. [209, 211, 212]

These studies indicate that for patients with clinical heart failure due to left-ventricular systolic dysfunction, perioperative mortality rates in excellent centers will range from about 5 percent in patients under age 60 with mild heart failure without any comorbid conditions to more than 30 percent in patients over age 70 with severe heart failure and several comorbid conditions. In addition to mortality risk, the risk of surgical morbidity and complications must also be considered. In the New York State Cardiac Surgery Reporting System, the risk of perioperative MI was 1.5 percent, the risk of stroke was 1.6 percent, and the risk of all major nonfatal complications was 9.2 percent. [211] Centers and surgeons vary considerably in their experience with patients with severe left-ventricular dysfunction. Perioperative mortality rates for CABG also vary significantly. [213] Patients should be informed of the mortality and complication rates of the surgeon performing the operation and of the facility where the surgery will be done. The patient's primary provider should be aware of variations in operative outcomes and consider referral to a center with substantial experience with heart failure patients and demonstrated good outcomes.

Benefits and Risks of PTCA

As with CABG, the benefits of PTCA in patients with heart failure are unclear. No studies have addressed survival after PTCA compared with CABG or standard medical management. However, several case reports and case series indicate that PTCA can relieve angina and improve ventricular function or wall motion. [214-219]

Less is known about the risks of PTCA for patients with heart failure. Serota et al. reported a series of 73 patients with EF of 40 percent or less who underwent PTCA, 45 percent of whom had a history of heart failure. [218] Five patients (7 percent) had serious procedure-related complications; of the three patients with MI's, two were fatal; and two had acute coronary closure with one patient death related to acute closure. The total in-hospital mortality was 5.5 percent. Hartzler et al. reported a "procedural mortality" of 2.7 percent for a series of 664 patients with EF's of 40 percent or less who underwent PTCA. [215] The rates of nonfatal MI and emergent CABG were 0.7 and 2.0 percent, respectively. The patient populations and reporting techniques in these series are different from those reported in studies of CABG, so a direct comparison is not possible. Thus, it is not known whether PTCA has a higher, lower, or equivalent complication rate for patients with heart failure, compared with CABG.

Because PTCA has not been shown to improve survival, CABG is usually considered the procedure of choice for patients with heart failure and angina. However, numerous considerations enter into the decision regarding recommendation of PTCA versus CABG, including multiple technical factors, underlying risk of surgery, the greater morbidity associated with CABG, and patient preferences. A discussion of these factors is beyond the scope of this guideline.

Patient Counseling and Decisionmaking

There are no absolute contraindications to revascularization except if the patient refuses surgery or is unable to give informed consent. However, a number of factors may preclude intervention or raise the risk above any expected benefit:

• Unwillingness to consider surgery.

• Severe comorbid diseases, especially renal failure, pulmonary disease, or cerebrovascular disease (e.g., severe stroke).

• Very low EF (<20 percent).

• Illnesses that imply a limited life expectancy less than or equal to 1 year, including advanced cancer, severe lung or liver disease, chronic renal disease, advanced diabetes mellitus, and advanced collagen vascular disease.

• Technical factors, including previous myocardial revascularization or other cardiac procedure, inadequate vascular conduit, history of chest irradiation, and diffuse distal obstructive coronary artery atherosclerosis.

It is important to balance the expected benefits and harms associated with revascularization. Patients may consider the relatively high perioperative mortality rate more acceptable when their life expectancy is not limited by other diseases or advanced age (i.e., when there is a good chance for long-term survival if they survive surgery). However, patients with diseases that, although not terminal, might be expected to reduce life expectancy may not consider the initial risk of perioperative mortality acceptable.

Patients, family members, and caregivers should be involved in decisionmaking at all steps of the treatment and care of heart failure, but this is particularly important when revascularization is being considered. The primary practitioner should provide patients with the information necessary to begin the decisionmaking process. Later, specialists should interact directly with the patient. Patients should be advised about the relative likelihood that they have clinically significant ischemia, about the risks of angiography, and about the likely benefits and harms of revascularization if ischemia is found. Some individuals may not want to undergo evaluation if there is a low likelihood that ischemia will be found. For others, the risks of surgery will be unacceptable regardless of the possible benefits. Patients should be encouraged to ascertain the perioperative mortality rate experienced by the providers in their area (or those covered by their insurance plan).

An individual's likelihood of having ischemia and his or her chance of surviving surgery can be estimated by the history, physical examination, and EF. These factors should be discussed with the patient before any diagnostic testing for ischemia. Some patients, in consultation with their primary practitioner, may decide at this point that no studies should be done. For other patients, this decision requires a balancing between expected benefits and risks. The amount of ischemic myocardium is the best predictor of the benefit of revascularization, and testing for the presence and amount of ischemia may be necessary before a patient can reasonably decide whether the risks of revascularization are acceptable.

If a patient wishes to pursue an evaluation for ischemia, the provider must decide what test is most appropriate.

Some patients will need subsequent physiologic testing to determine the amount of ischemic myocardium or whether myocardium is viable, but it is not necessary to perform such tests routinely before angiography in this subgroup of patients.

This strategy will miss a small number of patients with false negative physiologic tests. However, in view of the lack of evidence that these patients benefit from surgery, together with a consideration of the morbidity, mortality, and cost of catheterizing all patients in this group, this drawback is considered relatively minor.

Little scientific evidence is available to guide the evaluation and care of heart failure patients who have no history of angina or MI. Some of these individuals will have silent coronary artery disease as the cause of their heart failure, although this clearly depends on the patients' risk factors for coronary artery disease and the probability that there is another etiology for the heart failure. Patients without risk factors for coronary artery disease and patients with another probable cause of heart failure (e.g., alcoholic cardiomyopathy) are unlikely to benefit from angiography or testing for ischemia.

The panel was about evenly divided between advocating no further testing and advocating a physiologic test for ischemia in patients with no angina and no history of MI. Coronary angiography was considered advisable by some panelists in the uncommon presentation of unexplained heart failure in young patients with no history of angina or prior MI, with the rationale that these patients may benefit more from early revascularization for occult coronary artery disease than from later cardiac transplantation.

No data are available that address the question of how much ischemia should be present to justify the risk of revascularization for the chance of an improvement in survival. In general, patients with severely depressed EF's (<20 percent) should undergo revascularization only if large areas of ischemia are detected. Patients with less severely depressed EF's may be willing to risk surgery for more modest-sized ischemic areas. The lack of data in this area makes it difficult to justify revascularization for small ischemic areas, except when severe angina is present.

Physiologic Testing for Ischemia

Several studies have assessed the reversibility of perfusion defects or improvement in wall motion or EF after revascularization, but it is unclear whether these changes predict improvement in symptoms, exercise tolerance, or survival. Definitive outcome studies comparing the alternative tests have not been performed.

Thus, the clinician must choose from a range of acceptable alternatives. These include:

• Exercise or pharmacological stress myocardial perfusion scintigraphy (e.g., thallium scanning).

• Exercise or pharmacological stress echocardiography.

• Stress radionuclide angiocardiography.

• Positron emission tomography (PET).

If thallium is used, reinjection imaging should be performed at rest if an abnormality is identified during the stress portion of the study. This strategy improves the accuracy for identifying the presence and extent of viable myocardium. [220] PET may be a more sensitive test than perfusion scintigraphy for detecting viable myocardium, [221] but whether increased sensitivity translates into better patient outcomes is yet to be determined. In addition, PET is not widely available and is more costly than most alternative tests. Clinicians must be familiar with the availability, quality, and cost of the different physiologic tests for ischemia at their institution and should use this information in deciding what test to order. Myocardial perfusion scintigraphy is most useful for screening patients without a history of angina to determine whether coronary angiography is necessary. If this test is normal, angiography can reasonably be forgone. Because scintigraphy relies on differences in the distribution of myocardial perfusion to determine ischemia, an occasional patient with very balanced ischemia may have a "normal" study. Increased uptake of the isotope in the lungs may be a clue to this type of false negative test.

Evaluation for Heart Transplantation

In addition to heart transplantation, studies are under way to determine the benefits and harms of innovative treatments, including new drugs and mechanical ventricular assist devices. If appropriate, patients should be informed of the possibility of taking part in such studies.

For more information on heart transplantation, the reader is referto the American College of Cardiology/American Heart Association report on this topic. [222]

Heart Failure Guideline Panel

• Marvin A. Konstam, MD, Co-chair

• Professor of Medicine

• Tufts University

• Director, Heart Failure and Cardiac Transplant Center

• New England Medical Center Hospitals

• Boston, Massachusetts

Dr. Konstam is professor of medicine and radiology at Tufts University School of Medicine. He is also director of the Adult Cardiac Catheterization Laboratory and director of the Heart Failure and Cardiac Transplant Center at New England Medical Center Hospitals, Boston. Dr. Konstam received his undergraduate and medical education at Columbia University. He received residency training in internal medicine and diagnostic radiology at the Massachusetts General Hospital and fellowship training in cardiology at the Brigham and Women's Hospital in Boston.

Dr. Konstam has been on the staff of New England Medical Center and Tufts University since 1981. His primary research interest is the pathophysiology and therapy of heart failure. He has published numerous original articles in these areas, primarily in the arenas of ventricular mechanics and function. He directed the Radionuclide-Ventricular Function Core Laboratory for the Studies of Left-Ventricular Dysfunction trials.

• Kathleen Dracup, DNSc, RN, FAAN, Co-chair

• Professor, School of Nursing

• University of California, Los Angeles

• Los Angeles, California

Dr. Dracup's nursing career includes more than 25 years of experience in cardiovascular nursing and university professorships. She received her master's degree from the University of California, Los Angeles (UCLA), and her doctoral degree from the University of California, San Francisco. The primary focus of Dr. Dracup's research has been the care of patients with cardiac disease and the effects of disease on the family. She has more than 100 publications in this area.

Dr. Dracup is professor in the UCLA School of Nursing and coeditor of the American Journal of Critical Care. She served as coeditor of Heart & Lung for more than a decade. She is currently chair of the National Heart, Lung, and Blood Institute (NHLBI) Working Group on Educational Strategies to Prevent Prehospital Delay in Patients at High Risk for Myocardial Infarction; she is also a member of several other national committees. Dr. Dracup is a fellow of the American Heart Association's Council of Cardiovascular Nursing. She has received numerous honors and awards, including a recent Fulbright Scholarship to conduct research and lecture in Australia during the 1994-95 academic year.

• Michael B. Bottorff, PharmD

• Associate Professor and Chairman Division of Clinical and Hospital Pharmacy

• University of Cincinnati College of Pharmacy

• Cincinnati, Ohio

Dr. Bottorff is associate professor and chairman, Division of Pharmacotherapy, University of Cincinnati. He received his Doctor of Pharmacy with high distinction from the University of Kentucky, Lexington, where, upon graduation in 1981, he became chief resident of the Albert B. Chandler Medical Center.

Dr. Bottorff is a member of several professional organizations, including the American College of Clinical Pharmacy of which he is currently chair, Abstract Review Committee for the Winter Meeting, and vice chair, Awards Committee; the American Pharmaceutical Association of which he was recently chairman, Clinical Section, Academy for Pharmaceutical Research and Science; and the American Society of Hospital Pharmacists. Dr. Bottorff serves on the editorial advisory board of the Journal of Applied Therapeutics and Pharmacotherapy and acts as a referee for 13 others.

• Neil H. Brooks, MD

• The American Academy of Family Physicians

• Rockville, Connecticut

Dr. Brooks is a 1968 graduate of the Hahnemann Medical College. He has been a practicing family physician in Rockville, Connecticut, since 1971. He is on the staff of Rockville General Hospital where he has chaired multiple committees and has been chair of the Department of Family Practice and chief of staff. He is the vice speaker of the American Academy of Family Physicians and serves on its Task Force on Clinical Policies for Patient Care. He is a past president of the Connecticut Academy of Family Physicians. His medical affiliations include the Connecticut State Medical Society, for which he is a member of the council, vice speaker of the House of Delegates, and a delegate to the American Medical Association. He is a founding member and president of Capital Area Independent Practice Associates (CIPA), a 1,600-physician independent practice association, and is on the Board of Directors of CHC, a 100,000-member health maintenance organization. He is also on the Medical Advisory Committee of Blue Cross/Blue Shield of Connecticut.

• Robert A. Dacey

• Consumer Representative

• Boulder, Colorado

Mr. Dacey is the past national president (1989-91) of The Mended Hearts, Inc., a 25,000-member, all-volunteer heart patient support group affiliated with the American Heart Association. Before heart surgery and a leg amputation, he was an advertising agency creative director, associate editor and publisher of Communication Arts magazine, and director of product research and development for Docere Corp., a health care and patient education publishing company. While in this position, Mr. Dacey directed the development of multimedia patient education programs and materials for orthopedic, cardiovascular, obstetrical and gynecologic, and dental implant procedures. His editorials and essays have appeared in Heartbeat, the official journal of The Mended Hearts, and Pulse, a publication of the Association of Heart Patients. His current interest is the development of patient information and education programs and materials that will help improve patient adherence to medical treatment.

• Sandra B. Dunbar, RN, DSN, FAAN

• Associate Professor, Emory University

• Nell Hodgson Woodruff School of Nursing

• Atlanta, Georgia

Dr. Dunbar is currently an associate professor and coordinator of the Critical Care Program in the Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta. She has worked extensively with cardiac patients and their families in acute and outpatient clinical settings. Her current research is related to psychosocial outcomes from treatment of recurrent ventricular dysrhythmias.

• Anne B. Jackson, MA, RN

• American Association of Retired Persons National Legislative Council

• Queens Village, New York

Ms. Jackson, who lives in Sarasota, Florida, is a member of the National Legislative Council of the American Association of Retired Persons (AARP) and serves as chair of the Health and Future Generations Committee. Ms. Jackson has testified for AARP before the U.S. House of Representatives and the U.S. Senate. She is currently a commissioner for the Physician Payment Review Commission. Ms. Jackson was a professor at City University of New York in the Department of Nursing for 25 years. She also served in the positions of medical supervisor, head nurse, and staff nurse for the Veterans Affairs Hospital for 15 years.

• Mariell Jessup, MD

• Associate Professor of Medicine

• University of Pennsylvania School of Medicine

• Director, Heart Failure Unit Risk Reduction Center

• Philadelphia Heart Institute Presbyterian Medical Center

• Philadelphia, Pennsylvania

Dr. Jessup did her cardiovascular training at the Hospital of the University of Pennsylvania, where she met her mentor in heart failure research, Dr. Karl Weber. Since then, the focus of her clinical and research activities has been patients with chronic congestive heart failure and the evaluation of treatment options for these patients. As director of the Heart Failure Unit of the Philadelphia Heart Institute, she supervises a number of investigational drug protocols. Currently, she is exploring better methods of patient triage.

• Jerry C. Johnson, MD

• Associate Professor of Medicine

• University of Pennsylvania

• Chief, Geriatric Medicine Division

• Veterans Affairs Medical Center

• Philadelphia, Pennsylvania

Dr. Johnson is chief, Geriatric Medicine Division, Philadelphia Veterans Affairs Medical Center, and director, Geriatric Medicine Fellowship Training Program, University of Pennsylvania. After training in internal medicine, Dr. Johnson was a Robert Wood Johnson Clinical Scholar before focusing on geriatrics. He is now a nationally recognized clinician and educator with extensive experience in caring for complex elderly patients with acute and chronic medical problems. He has conducted several studies of neuropsychiatric complications of acute medical problems and of quality-of-life outcomes of chronic care interventions. Dr. Johnson is now engaged in clinical studies of elderly patients with heart disease as part of the University of Pennsylvania's Clinical Research Center on Psychiatric Illness.

• Robert H. Jones, MD

• Mary and Deryl Hart Professor of Surgery

• Duke University Medical Center

• Durham, North Carolina

Dr. Jones is a cardiac surgeon with a special interest in coronary artery bypass grafting in patients with poor ventricular function and congestive heart failure. His prior research interests have focused on developing radionuclide techniques to noninvasively detect and characterize coronary artery disease. He served as project director for development of Unstable Angina: Diagnosis and Management. Clinical Practice Guideline No. 10. The guideline was sponsored by AHCPR and the National Institutes of Health (NIH).

• Robert J. Luchi, MD

• Chief, Geriatrics Section Director, Huffington Center on Aging

• Baylor College of Medicine

• ACOS Geriatrics and Extended Care

• Veterans Affairs Medical Center

• Houston, Texas

Dr. Luchi is director, Roy M. and Phyllis Gough Huffington Center on Aging at Baylor College of Medicine, and professor and chief, Geriatrics Section, Department of Medicine. He is the associate chief of staff for geriatrics and extended care at the Houston Veterans Affairs Medical Center. He received his MD degree in 1952 and was trained in internal medicine and cardiology at the University of Pennsylvania School of Medicine. Dr. Luchi is a member of the American, Central, and Southern Societies for Clinical Investigation, and a fellow of the American College of Physicians (ACP), American Geriatrics Society (AGS), Gerontological Society of America (GSA), and American College of Cardiology (ACC). He has served on national committees for the American Heart Association, ACC, the Department of Veterans Affairs, and the National Board of Medical Examiners. His major research interests include congestive heart failure in the elderly and unstable angina. He has published more than 95 scientific articles, a number of which are related to congestive heart failure.

• Barry M. Massie, MD

• Professor of Medicine

• University of California, San Francisco

• Director, Hypertension Clinic Director, Coronary Care Unit

• Veterans Affairs Medical Center

• San Francisco, California

Dr. Massie is a clinical and research cardiologist who serves as professor of medicine at the University of California, San Francisco, and Director of the Coronary Care Unit and Hypertension Clinic at the San Francisco Veterans Affairs Medical Center. His clinical research interests include the pathophysiology of heart failure, with a focus on the mechanism of exercise intolerance, and the investigation of new therapeutic agents. He also directs a research group investigating heart metabolism and hypertensive heart disease. Dr. Massie has published more than 160 articles and book chapters, mainly dealing with heart failure and hypertension. He serves on the editorial boards of several journals, including the Journal of the American College of Cardiology and Heart Failure. He is a consultant to the FDA in the evaluation of new cardiac medications.

• Bertram Pitt, MD

• Professor of Internal Medicine Associate Chairman of Academic and Industrial Programs Department of Internal Medicine

• University of Michigan Medical Center

• Ann Arbor, Michigan

Dr. Pitt is professor of internal medicine and associate chairman for academic and industrial programs in the Department of Internal Medicine, University of Michigan School of Medicine, Ann Arbor, Michigan. He received his medical degree from the University of Basel, Switzerland.

Dr. Pitt has published 245 articles in scientific journals. He is a member of the American Physiological Society-Circulation Group, the American Federation for Clinical Research, the International Society for Heart Research, and the American Heart Association. He is also president of the Michigan Chapter of the American College of Cardiology and a fellow of the American College of Cardiology. Honors he has received include Bronze Award, American Heart Association-Circulation Group; Haile Selassie Lecturer, British Heart Foundation; and Outstanding Cardiology Staff.

• Eric A. Rose, MD

• Chief, Cardiothoracic Surgery

• Columbia Presbyterian Medical Center

• New York, New York

Dr. Rose is professor of surgery and chief of the Division of Cardiothoracic Surgery at Columbia University. His interest in the surgical management of heart failure entails clinical practice, teaching, and research in the areas of reparative coronary artery and valvular procedures and cardiac replacement therapies. The latter include transplantation, mechanical cardiac assistance, and cross-species transplantation. He is the current president of the International Society for Heart and Lung Transplantation.

• Lewis J. Rubin, MD

• Professor of Physiology and Professor of Medicine

• University of Maryland School of Medicine

• Baltimore, Maryland

Dr. Rubin is professor of medicine and head of the Division of Pulmonary and Critical Care Medicine at the University of Maryland School of Medicine. His major research interests include disorders of the pulmonary circulation and cardiopulmonary interactions. He received the Vascular Disease Academic Award from NHLBI, the National Institutes of Health (NIH). Dr. Rubin serves on advisory committees for NIH, the American Heart Association, and the American Lung Association.

• Richard F. Wright, MD

• Research Director, Pacific Heart Institute

• Assistant Clinical Professor, University of California

• Los Angeles, California

Dr. Wright received his medical degree from Harvard Medical School. After an internship at the University of Southern California, he returned to Harvard, where he received internal medicine and cardiology training at the Brigham and Women's Hospital. In addition to postdoctoral research at Harvard, Dr. Wright has received research training at the Ames Research Center, National Aeronautics and Space Administration, and the Oak Ridge National Laboratory, Atomic Energy Commission. He is presently in the clinical practice of cardiology, with special expertise and ongoing research in the management of heart failure and in cardiopulmonary resuscitation. Dr. Wright has been president of the Los Angeles Society of Echocardiography and is research director at Pacific Heart Institute, assistant clinical professor at UCLA School of Medicine, and chairman of the Cardiology Section at St. John's Hospital and Health Center, Santa Monica, California.

Peer and Pilot Reviewers[a]

• William H. Barry, MD

• Salt Lake City, Utah

• Darla F. Bonham

• The Mended Hearts, Inc.

• Dallas, Texas

• Richard W. Campbell, MD

• American College of Physicians

• Indianapolis, Indiana

• Clifton R. Cleaveland, MD

• American College of Physicians

• Chattanooga, Tennessee

• Barbara J. Drew, RN, PhD

• University of California School of Nursing

• San Francisco, California

• Gerald F. Fletcher, MD

• Emory University School of Medicine

• Atlanta, Georgia

• Erika S. Froelicher, RN, PhD

• University of California School of Nursing

• San Francisco, California

• Richard Gorlin, MD

• The Mount Sinai Medical Center

• New York, New York

• John F. Hagaman, MD

• Cardiology Associates of Princeton

• Princeton, New Jersey

• Mairead L. Hickey, PhD, RN

• Brigham and Women's Hospital

• Boston, Massachusetts

• William B. Hood, Jr., MD

• University of Rochester Medical Center

• Rochester, New York

• Michael J. Horan, MD, ScM

• National Heart, Lung, and Blood Institute

• Bethesda, Maryland

• June Howland-Gradman, MSN, RN

• Emergency Nurses Association

• Park Ridge, Illinois

• Randal F. Hundley, MD

• Cardiac Diagnostic Clinic

• Little Rock, Arkansas

• Pamela Kidd, RN, PhD

• Emergency Nurses Association

• Park Ridge, Illinois

• Ronald Dee Legako, MD

• Canyon Park Family Physicians, Inc.

• Edmond, Oklahoma

• Donald F. Leon, MD

• Georgetown University Medical Center

• Washington, DC

• Patrick E. McBride, MD, MPH

• University of Wisconsin Medical School

• Madison, Wisconsin

• Debra K. Moser, DNSc, RN

• UCLA School of Nursing

• Los Angeles, California

• John B. O'Connell, MD

• The University of Mississippi Medical Center

• Jackson, Mississippi

• Milton Packer, MD

• College of Physicians and Surgeons of Columbia University

• New York, New York

• Richard B. Perry, MD

• American College of Physicians

• Potomac, Maryland

• Deeb Salem, MD

• New England Medical Center

• Boston, Massachusetts

• Robert J. Schlant, MD

• Emory University School of Medicine

• Atlanta, Georgia

• Edmund H. Sonnenblick, MD

• Albert Einstein College of Medicine

• Bronx, New York

• Stephen J. Spann, MD

• The University of Texas Medical Branch at Galveston

• Kathryn Taubert, MD

• American Heart Association

• Dallas, Texas

• George E. Thibault, MD

• Harvard Medical School

• Boston, Massachusetts

• David R. Witmer, PharmD

• American Society of Hospital Pharmacists

• Bethesda, Maryland

• James B. Young, MD

• Baylor College of Medicine

• Houston, Texas

Being listed in this section does not necessarily imply endorsement of the guideline.

AHCPR Staff

• Francis Chesley, MD

• Project officer

• Carole Hudgings, PhD

• Acting director

• Office of the Forum for Quality and Effectiveness in Health Care

• William N. LeVee

• Managing editor

• Valna Montgomery, MSW

• Product manager

Availability of Guidelines

For each clinical practice guideline developed under the sponsorship of the Agency for Health Care Policy and Research (AHCPR), several versions are produced to meet different needs.

The Guideline Technical Report contains complete supporting materials for the Clinical Practice Guideline, including background information, methodology, literature review, scientific evidence tables, and a comprehensive bibliography.

The Clinical Practice Guideline presents recommendations with brief supporting information, algorithms or flow charts, tables and figures, and pertinent references. The Quick Reference Guide for Clinicians is a distilled version of the Clinical Practice Guideline, with summary points for ready reference on a day-to-day basis.

The Consumer Version, available in English and Spanish, is an information booklet for the general public to increase patient knowledge and involvement in health care decisionmaking.

Guideline information also will be available for on-line retrieval through the National Library of Medicine, the National Technical Information Service, and some computer-based information systems of professional associations, nonprofit organizations, and commercial enterprises.

To order guideline products or to obtain further information on their availability, call the AHCPR Publications Clearinghouse toll-free at (800) 358-9295, or write to: AHCPR Publications Clearinghouse, P.O. Box 8547, Silver Spring, MD 20907.