|
INTRODUCTION
year, hospitalization accounts for over two-thirds of all health expenditures for heart failure Heart failure leads to over 1 million hospitalizain Western nations. As a marker of disease protions annually and is the most common diagnogression, worsening functional status, and poor sis for hospitalization in patients over the age of prognosis, hospitalization can be considered a 65 years. Readmission rates are 25–40% during failure of the medical regimen. The major purthe next 6 months. Death occurs in only 4% of poses of hospitalization are to establish stability hospital admissions, but in almost half of and to design a regimen that will maintain it patients during the year after discharge.
REASONS AND GOALS FOR HEART FAILURE HOSPITALIZATION
The most common precipitant for hospitalization with heart failure is worsening of congestive symptoms due to accumulation of excess circulating and total body volume (Table 12-1). However, decompensated heart failure may also be discovered during a hospitalization for other presenting symptoms such as tachyarrhythmias or angina in patients with low left ventricular ejection fraction (LVEF). For any patient presenting with new-onset heart failure, it is critical to review the primary etiology and potential exacerbating factors in order to address conditions that can be remedied, such as active ischemia and primary valve disease.
This chapter will focus on the approach to patients with chronic heart failure that has previously been evaluated. Hospitalization for patients with heart failure should address the fundamental goals of symptom relief, attention to potential exacerbating factors, clinical stabilization, and design of the discharge regimen, including education in connection with subsequent outpatient management (Table 12-2).
PROFILES OF HOSPITALIZED PATIENTS
The typical patients hospitalized with heart failure have been well-defined recently by large registries. Previous information was gleaned from randomized controlled trials in which patients were relatively young with few comorbidities and were felt sufficiently stable to undergo an active or placebo therapy. Centralized databases in Scotland and Canada have revealed the older age and almost 50% 1-year mortality of most populations hospitalized with heart failure.1,2 Details of the initial clinical profiles and hospital events have been provided in the United States by the Acute Decompensated Heart Failure National Registry (ADHERE), including over 100,000 hospitalizations voluntarily reported by community and academic centers.3 Recognizing the different hospitalized populations from communities and trials (Table 12-3), application of trial findings and the parallel guidelines should be carefully adapted for the profile of the individual patient.
Heart failure with a dilated ventricle and low ejection fraction (EF) is the basis for the major clinical trials and current heart failure guidelines. It has been increasingly recognized, however, that about half of heart failure hospitalizations occur in patients with an EF >40%, usually without significant dilation of the ventricle, often with left ventricular hypertrophy. Although often referred to as “diastolic dysfunction,” accepted echocardiographic criteria for diastolic dysfunction are often found in patients with dilated heart failure and low EF and are not always present in patients with heart failure and preserved EF. The most typical clinical history for such a patient includes hypertension, diabetes, and advanced age, and about half of these patients are women, as discussed in Chap. 11.
Although much has been made of the distinctions between the pathophysiology of heart failure with low EF and with preserved EF, the presentation at the time of hospitalization is usually with similar symptoms. Blood pressure and cardiac output are less often low in the preserved EF patient, and ventricular tachyarrhythmias are rare, but renal dysfunction and atrial fibrillation are equally common. Furthermore, our current acute therapies are remarkably similar for the two populations, although some agents may be selected for different reasons. This chapter will focus on those considerations and therapies developed for patients with low LVEF, but will indicate those considerations that are appropriate to all patients hospitalized with heart failure regardless of EF (all EF).
Initial Evidence of Congestion (All Ejection Fraction)
The majority of patients hospitalized with symptoms of heart failure at rest or minimal exertion have elevation of both right- and left-sided filling pressures. Elevation of left-sided filling pressures can be reflected in symptoms of orthopnea and dyspnea on minimal exertion, such as walking to the bathroom or getting dressed. Shortness of breath relates primarily to stiffness of the lung interstitium limiting comfortable respiratory excursion. This is attributed most often to high hydrostatic pressure in the pulmonary veins, although may be exacerbated by impaired pulmonary lymphatic drainage into the right side, and by low serum proteins leading to reduced intravascular oncotic pressure. Elevation of right-sided systemic venous pressures can be associated with the discomfort of peripheral edema, hepatic congestion, and ascites, although can also cause anorexia and early satiety without detectable intra-abdominal fluid retention. Dominance of right-sided or left-sided symptoms does not necessarily define the relative elevation of pressure in these two venous circulations.
The relationship of right- to left-sided filling pressures has been delineated in chronic advanced heart failure with low EF by Drazner, indicating that right atrial pressure > or <10 mm Hg correlates with pulmonary wedge pressure > or <22 mm Hg in 80% of patients.4 In addition, for this population of patients from which severe intrinsic pulmonary disease was excluded, pulmonary artery systolic pressure was highly correlated with pulmonary capillary wedge pressure, pulmonary artery systolic pressure being about twice the wedge pressure, once elevated. These relationships have not been established for heart failure with preserved EF.
The clinical detection of high pulmonary capillary wedge pressures has been assessed specifically only for patients with low EF, but in the absence of other information, similar strategies should be used for the patient with heart failure and preserved EF. Rales and edema are relatively insensitive to the presence of chronic fluid elevation, and therapies targeted only for clear lungs and ankles will result in undertreatment.5 In young patients, edema is very rare even with severe volume overload, while elderly patients often develop peripheral edema from local venous disease without elevated central filling pressures.
Jugular venous pressure remains the most important component of physical assessment for elevated filling pressures, both at baseline and through the course of changing therapies. Other measures that can be useful for experienced examiners include the Valsalva maneuver (which can be done approximately with a stethoscope and bedside blood pressure cuff or more precisely with a simple manometer device), and the leftward radiation of the pulmonic component of the second sound as indicative of elevated pulmonary artery systolic pressure in heart failure, most commonly due to elevated left-sided filling pressures.
Four Basic Hemodynamic Profiles
The concept of four major clinical hemodynamic profiles has been useful to triage patients both for initial therapy and for subsequent outcome (Fig. 12-1).6 Some of the utility has been in the recognition that filling pressures are often severely elevated in the absence of clinical hypoperfusion, although the converse is relatively uncommon. The four profiles are defined by the result of two binary questions regarding filling pressures and perfusion: Is there evidence of congestion? This question is answered as described above. Is there evidence of clinical hypoperfusion? Clues to hypoperfusion can be a proportional pulse pressure <25%, cool extremities, hypotension even to low-dose angiotensin-converting enzyme (ACE) inhibitors, and progressive renal dysfunction, although as discussed below, renal dysfunction is not primarily a result of low cardiac output. Although some patients with low perfusion may actually feel cold to palpation, the terms “cold and wet” or “cold and dry” are not meant to describe the actual physical temperature, but to conceptualize the circulatory status and guide initiation of therapy. Clinicians are less astute at recognizing cardiac index <2.2 L/min/m2 than recognizing pulmonary wedge pressure >22 mm Hg.
Despite the limited accuracy of our estimates of cardiac output, the clinical profiles are important for prognosis. Outcome over the next 6- and 12-month period has been clearly linked to clinician determination of clinical profile at admission. Patients assessed to be warm and wet, which is the majority of hospitalizations in most series and the large majority of community hospitalizations, had twice the rate of death or urgent transplant at 1 year as those who appeared warm and dry. The cold and wet profile was associated with 3.7 times the 1-year mortality, while patients with the cold and dry profile had mortality similar to the warm and wet.7 Patients with immediate compromise of organ perfusion due to life-threatening cardiogenic shock would fall within the cold and wet profile definition, but would require specific attention for definitive therapy beyond that for most cold and wet patients.
WHAT IS THE GOAL FOR DRY?
The major goal for most hospitalized patients is to establish optimal filling pressures to maintain stability, regardless of the EF. Both excess circulating volume and high afterloads from systemic vasoconstriction and poorly compliant great arteries contribute to elevated filling pressures. The relative contribution of the volume overload and the decreased venous compliance and increased arterial tone vary between individuals. Before chronic use of ACE inhibitors, many patients with dilated low EF heart failure presented with severe vasoconstriction, which is now much less common than dominant volume overload.8 For patients with preserved LVEF, volume overload is a major factor, but the contribution of inefficient ventriculo-vascular coupling into stiff vessels may be most important for chronic progression and acute decompensation. While the final goal for both warm-and-wet and cold-and-wet profiles is the warm and dry profile, initiation of therapy for the warm-and-wet is simplistically stated as “dry them out” while for cold-and-wet is to “warm up and then dry out,” as discussed below under addition of vasoactive intravenous agents.
What is the target filling pressure in dilated low EF heart failure? At one time it was presumed that filling pressures needed to be high in order to maintain stroke volume from the chronically dilated ventricle. This arose in part as an extension from the observation that a pulmonary wedge pressure of 18 mm was optimal early after myocardial infarction; this, however, is a state characterized by acute reduction of compliance in a nondilated ventricle. It is now well-accepted that filling pressures can be reduced to near-normal levels while maintaining or even improving stroke volume in chronic dilated heart failure.9 High filling pressures impair left ventricular function by imposing high oxygen demand, diminishing subendocardial perfusion, and increasing ventricular turgor by impairing coronary venous drainage against high right atrial pressures. They further detract from cardiac output through dynamic mitral regurgitation, which often consumes up to 75% of total stroke volume during decompensation. The regurgitant fraction is frequently reduced to only 25% of total stroke volume after effective therapy to reduce filling pressures, due to a decrease in the effective regurgitant orifice.10
Measured filling pressures remain robust predictors of survival in dilated low EF heart failure. They are much more predictive than indices of cardiac output, and the filling pressures measured after therapy tailored to reduce filling pressures are more important than those measured at admission.11 It remains unclear, however, whether the achievement of the lowest filling pressures creates survivors, or merely identifies those with more favorable physiology regardless of therapy.
The target filling pressure in preserved LVEF heart failure has been assumed to be much higher, reflecting the relatively noncompliant left ventricle often characterized by left ventricular hypertrophy. The pressure-volume curves for these ventricles have not been well characterized, however. Some patients diagnosed with heart failure and preserved LVEF may have relatively little apparent abnormality of myocardial compliance, but instead have a fluid-retaining state with a high set point beyond that required for maintenance of cardiac output. For these patients, documentation of good cardiac output after diuresis to normal filling pressures can be helpful in guiding future adjustment of volume status. For almost all patients with resting symptoms from elevated left-sided filling pressures (in the absence of outflow gradients), fluid status can be reduced without compromise of cardiac output.
How Can Therapy Be Monitored?
Because it is the elevated filling pressures that lead to the symptoms of congestion at the time of heart failure hospitalization, therapy in a hospital is aimed at reduction of these filling pressures. Therapy is associated with improved symptoms, decreased clinical evidence of elevated filling pressures, decreased filling pressures measured invasively or estimated noninvasively, and decreased correlates of filling pressures, such as echocardiographic mitral regurgitation and natriuretic peptide levels. As the importance of lower filling pressures is recognized for relieving symptoms, for titrating b-blocking agents, and for limiting disease progression, further emphasis will likely be placed on strategies by which to monitor and adjust filling pressures. However at this time, the only two strategies that have been rigorously compared are the clinical evaluation and invasive measurement during hospitalization.
Symptoms and Signs of Congestion (All Ejection Fraction)
The most immediate goal of therapy for heart failure regardless of LVEF is relief from the symptoms of congestion that lead to hospital admission. As a gauge of therapy, however, symptom improvement is most useful at the beginning of therapy, to indicate progress in the right direction. For someone who has undergone gradual decompensation, the dramatic contrast after initial rapid reduction of filling pressures often is perceived by the patient as “back to normal,” or “back to baseline,” which is often still excessive compared to normal volume status. At this level, it requires very little fluid retention for symptomatic regression. Patients admitted with clinical fluid retention are at high risk for early readmission if they go home as soon as symptoms improve.
The adequacy of diuresis is tracked using resolution of orthopnea; hepatomegaly; peripheral edema; if present, and not due to local venous disease alone; and the most sensitive sign, jugular venous pressure <8 mm Hg. As discussed in the earlier paragraphs, clinical assessment of right atrial pressure provides a reasonable gauge of therapy. It correlates with left-sided filling pressures in almost 80% of patients with chronic dilated heart failure, and is assessed with reasonable accuracy by experienced clinicians (Table 12-4).4 Although the accuracy of the physical signs discussed in the previous section have been better validated for single-time measurements than for the clinical assessment of changes in filling pressures, they appear to correlate reasonably well both at the beginning and end of therapy in the hospital.
Pulmonary Artery Catheters in Monitoring for Adjustment of Therapy
Direct measurement of hemodynamics during hospitalization may be useful for diagnosis of baseline hemodynamic profile in patients for whom the clinical assessment is ambiguous or discordant (Table 12-5). It may be particularly useful for determining the contribution of heart failure to a complex clinical picture such as sepsis, acute renal failure, or acute coronary syndrome in the setting of chronic heart failure. A common reason for determining left heart and pulmonary pressures is in the evaluation of concomitant pulmonary and cardiac disease in which the cause of dyspnea and elevation of right heart pressures could be due either to left-sided failure or intrinsic pulmonary disease.
In addition to diagnosis, hemodynamic monitoring has been used to guide ongoing adjustment of therapy for chronic decompensated heart failure.12 For patients with the usual ratio of right atrial to wedge pressure <two-thirds, recommended hemodynamic goals of therapy tailored to reduce filling pressures are pulmonary capillary wedge pressure ≤16 mm Hg and right atrial pressure ≤8 mm Hg. Systemic vascular resistance is a target of therapy only as necessary to reduce the filling pressures, with a goal being 1100–1200 in normal size individuals in whom the wedge pressure is still high. If the filling pressures are not high, aggressive reduction of systemic vascular resistance is generally associated with symptomatic hypotension. Cardiac output and mixed venous saturation are useful for trending general circulatory status and usually improve with effective reduction of filling pressures. Beyond that, however, therapy targeted specifically to improve cardiac output has not been beneficial as part of a strategy for chronic management. (As mentioned above, hemodynamic monitoring to direct use of inotropic therapy in pressor doses in support of The pulmonary artery catheter (PAC) was compared to clinical assessment during adjustment of therapy for a population of hospitalized patients with advanced chronic heart failure who either had one prior hospitalization during the past year or chronic high-dose diuretic therapy prior to the current admission (NHLBI-sponsored ESCAPE trial).13 Adverse events specifically related to PAC occurred in 4% of patients, and there were twice as many infections in the patients with PAC as those treated based on clinical assessment alone. Both groups of patients had marked improvement in clinical status during hospitalization (Table 12-6). PAC had no effect on in-hospital mortality or the overall endpoint of days alive out of hospital over the next 6 months, although there was a strong trend for benefit in the higher volume centers (Fig. 12-2). Despite slightly higher overall net diuresis in patients whose therapy was monitored with the PAC (average 1.9 days of invasive monitoring), there was significantly less deterioration in renal function by discharge. There was a consistent trend for better functional capacity and quality of life in the patients whose therapy was adjusted with the PAC (Fig. 12-3). Reduction in pulmonary capillary wedge pressure correlated with greater improvement in functional status, and the final pulmonary capillary wedge pressure was a strong predictor for the primary endpoint.
Based on the lack of benefit for the primary endpoint, PAC is not recommended during routine therapy of patients hospitalized with heart failure, nor is it recommended in centers that do not currently have extensive experience in the monitoring and therapy of this hospitalized population. It is reasonable, however, to consider the use of PAC monitoring to further adjust therapy in patients who demonstrate recurrent or refractory symptoms despite ongoing standard therapy adjusted according to clinical assessment (Table 12-5). Although randomized trial data do not address, nor are they likely to address in future, the potential benefit of PAC specifically for the small number of patients who appear dependent on intravenous inotropic agents, use of PAC for more precise adjustment of fluid status and vasodilator therapy seems warranted in view of the dismal prognosis if intravenous inotropic therapy is administered continuously for chronic palliation.
B-Type Natriuretic Peptide Levels
B-type natriuretic peptide (BNP) levels have been useful in the urgent evaluation of dyspnea in patients without previous diagnosis and in stratifying risk for patients in the periinfarction period, at the time of hospital admission, and at the time of hospital discharge. BNP levels between 200 and 1000 characterize the majority of patients with chronic heart failure with low EF, slightly lower values characterize patients with heart failure and preserved EF, and persistent levels over 1300 predict highest risk for patients with known chronic heart failure.14,15
BNP levels tend to run higher with older age and worse renal dysfunction, and lower with obesity. They are correlated with filling pressures, and tend to change directionally with changes in filling pressures, but may continue to decline over time despite stable fluid balance. The baseline levels and the slopes of change for BNP levels vary widely between individuals. Current trials are testing whether BNP levels can be used as a target for adjusting therapy over time in the outpatient setting. At the present time, high BNP levels in the inpatient setting may identify patients at high risk for poor outcome, but are not therapeutic targets.
SPECIFIC THERAPIES
Use of Diuretics—All Ejection Fraction
Assuming the presence of volume overload, escalating diuretic therapy is an early focus of hospitalization. Discussed in detail in Chap. 10, diuresis is often initiated with an intravenous bolus of at least the milligram equivalent of oral furosemide, with subsequent dose doubling until brisk diuresis is noted, then boluses two to three times daily in the hospital. Continuous furosemide infusion should be considered initially when the need for large volume diuresis is anticipated, or after initial boluses have not been effective. Supplementation with a thiazide diuretic, oral metolazone, or intravenous hydrodiuril frequently initiates brisk diuresis when loop diuretics alone are ineffective in a patient after chronic high-dose therapy. If diuresis remains ineffective, particularly in the setting of marginal blood pressure, it may be necessary to consider whether to reduce doses of neurohormonal antagonists until diuresis is achieved. Use of additional vasoactive infusion is discussed below.
Mechanical fluid removal may be considered when other efforts to remove excess fluid have been unsuccessful. This previously required cumbersome machinery and physical restriction of the patient by large catheters. Often fluid removal sufficient to relieve symptoms in diuretic resistance has been followed by progressive renal insufficiency. Now that fluid can be more easily removed in ambulatory patients without high volume circuits, ambulatory fluid removal devices are under investigation for use earlier in the hospital course. However, renal function and electrolytes should be monitored closely during any intervention that removes fluid rapidly.
Addition of Vasoactive Intravenous Agents
During hospitalizations for dilated low EF heart failure, intravenous vasodilators or inotropic agents are added in approximately 25% of patients (Fig. 12-4).3 The major agents currently considered for addition to diuretic therapy during heart failure hospitalization are the vasodilators nesiritide and nitroglycerin, and the intravenous inotropic agents dobutamine, low-dose dopamine, or milrinone. A brief review of these intravenous vasoactive agents will be followed by discussion of situations in which they might be used.
Diuretics only Nesiritide Nitroprusside NTG Dobutamine Dopamine Milrinone
Vasodilators
During chronic decompensation of dilated low EF heart failure, vasoconstriction is frequently present. In the previous era prior to the chronic use of ACE inhibitors, vasoconstriction was a more prominent part of decompensation, with systemic vascular resistance often above 1500–2000 dynes/s/cm.5,16 Nitroprusside is a potent balanced vasodilator resulting in immediate reductions in systemic vascular resistance and venoconstriction. Pulmonary capillary wedge pressure falls and cardiac output increases, often by 30% or more. Nitroprusside remains the most effective and reliable vasodilator when systemic vascular resistances are high, but is limited by cyanide toxicity and diminishing physician familiarity with its use. The other nitrosovasodilator, nitroglycerin, has slightly less arterial vasodilation but is also effective when titrated to reduce vasoconstriction. The need for monitored titration of these nitrosovasodilators limited their use in favor of the more convenient inotropic agents except at experienced heart failure centers.
More recent hemodynamic studies indicate less vasoconstriction at the time of decompensation, even in advanced heart failure trial populations, perhaps related in part to the effects of chronic inhibition of the renin-angiotensin system in earlier stages of heart failure.13 At the same time, the identification and recombinant technology for manufacture of the human BNP created enthusiasm for use of this endogenous vasodilator. Used in pharmacologic doses in heart failure patients, nesiritide is a modest arterial vasodilator that causes reductions of pulmonary capillary wedge and right atrial pressures, which were linked to improvement in heart failure symptoms. Nesiritide caused slightly greater reduction of pulmonary artery and pulmonary capillary wedge pressures with slightly less reduction in blood pressure than nitroglycerin titrated in a blinded protocol to relatively low doses.17
In addition to systemic vasodilation, the natriuretic peptides are renal vasodilators that increase renal blood flow. Their natriuretic effect is modest, but may allow a decrease in total diuretic dose. The clinical significance of these cardiorenal effects during heart failure hospitalization is controversial, and remains under investigation.
Administration of either nitroglycerin or nesiritide has been shown to accelerate improvement of heart failure symptoms after hospital admission.17 Nitroglycerin causes headaches more often than nesiritide. All vasodilators can cause hypotension, which is generally well tolerated in supine patients, and responds to withdrawal of the vasodilator, although resolution takes longer with nesiritide due to the longer half-life. Both nitroglycerin and nesiritide occasionally cause hypotension associated with bradycardia. The biggest concern for the use of vasodilators is in patients who have been incorrectly assessed to have elevated volume status but actually are volume depleted or excessively vasodilated prior to administration.
Inotropic Agents
The most common intravenous inotropic agent used during heart failure hospitalization is dobutamine, which acts through b-adrenergic receptors to increase cyclic adenosine monophosphate (AMP) production.18 Cardiac output is increased, often with a slight increase in blood pressure. Stimulation of peripheral b-receptors without significant a-receptor stimulation leads to a slight decrease in systemic vascular resistance in most patients, but to a lesser degree than seen with intravenous vasodilators. Pulmonary capillary wedge pressure is modestly decreased. Heart rate usually increases, particularly in atrial fibrillation. Dobutamine increases the occurrence of atrial fibrillation and ventricular tachyarrhythmias. Clinical ischemic episodes are increased with dobutamine, which has also been associated with asymptomatic troponin leak. Due to these risks, and to the difficulty of weaning after intravenous inotropic therapy, doses used should be the lowest that provide the desired effect. Patients with chronic heart failure often respond well to doses as low as 2 µg/kg/min of dobutamine, and rarely need the 5 µg/kg/min that has often been used as a starting dose.
Dopamine binds to b-receptors but also stimulates a-adrenergic receptors and dopaminergic receptors located primarily in the kidney. Low-dose dopamine (≤3 mcg/kg/min) sufficient to activate renal dopaminergic receptors is frequently said to be “renal-dose dopamine.” At doses of 1–3 µg/kg/min for patients with heart failure, however, responses are very similar to those observed with dobutamine in terms of blood pressure and urine output. While there is little information regarding any selective renal effects, all doses of dopamine have detectable hemodynamic effects to increase cardiac output, heart rate, and potentially ischemia and tachyarrhythmias when used in patients with heart failure. As dopamine can release norepinephrine from nerve terminals, initiation may theoretically be associated more with tachycardia and ischemia than with dobutamine, but clinical events appear similar. Dopamine at a low dose is a reasonable drug to initiate in a patient in whom declining perfusion may necessitate escalation to pressor doses if early response is not favorable. Doses ≥5 mcg/kg/min usually increase systemic vascular resistance (Fig. 12-5), with increasing inotropy and vasoconstriction up to doses of 15–20 mcg/kg/min, above which there is little further clinical effect.
In the rare cases of acute deterioration where blood pressure cannot be supported with escalating doses of dopamine, further inotropic and vasoconstrictor effect can be gained from the full agonist epinephrine, starting at doses of 1µg/min (not usually dosed per kilogram). The most common time this is employed is in patients with acute fulminant myocarditis or shock post-infarction or cardiotomy. Norepinephrine is occasionally employed when abnormal vasodilation is suspected, because it has little effect on the vasodilatory b2-adrenergic receptors and is thus an even more potent vasoconstrictor than epinephrine. It is also considered more likely to cause renal and peripheral ischemic injury. These two agents carry high risks of tachyarrhythmias and ischemia (Fig. 12-5), and their use is reserved only for imminent life-threatening situations while more definitive intervention is arranged.
Particularly for the vasoplegia that occasionally develops in severe circulatory compromise, vasopressin may provide additional support for blood pressure and potentiate the actions of the catecholamines. All of these agents have half-lives in minutes, and can be rapidly titrated and weaned.
Milrinone, often termed an “ino-dilator,” is a phosphodiesterase inhibitor, the only one currently approved. It acts directly to inhibit the breakdown of cyclic AMP, bypassing the b-receptors that may become downregulated after prolonged stimulation. The phosphodiesterase inhibitors, however, do appear to cause other downregulatory adaptations. As with other inotropic agents, the risks of tachycardia and ischemia are increased by milrinone. Compared to the other inotropic agents, the phosphodiesterase inhibitors cause more vasodilation, which in some patients may be the dominant effect. Milrinone should not be used when the primary concern is hypotension, which can be aggravated. Clinical hypotensive episodes occurred more commonly with milrinone than with placebo infusion in one trial.19 This is particularly concerning as the pharmacologic half-life of milrinone is 2–4 hours, and is prolonged by impaired renal excretion. The physiological offset may be further prolonged, often lasting up to a day after discontinuation of chronic use.
Because milrinone bypasses the b-receptors, it is often considered for patients felt to need inotropic support in the presence of b-adrenergic blockade. If the major concern is hypotension, milrinone should still not be the first choice for the reasons above, as blood pressure may fall further. b-Adrenergic antagonism is rarely complete in these patients, who will generally respond to dobutamine and dopamine, although higher doses may be required. In general, there is no consistent rationale for patients requiring urgent inotropic support to be maintained on previous
rdiac output increased to a similar degree with all three agents, there were greater decreases in systemic
Figure 12-6 Hemodynamic effects of nitrosovasodilators (NTP and NTG) and current intravenous inotropic therapy as described in different studies. The populations in which nitroprusside, dobutamine, and milrinone were compared had more severe hemodynamic compromise and more striking hemodynamic changes during therapy than the patients receiving low-dose NTG and nesiritide in the VMAC trial, as described by Monrad et al. The high-dose nitroglycerin study was reported by Elkayam et al. NTP—nitroprusside; NTG—nitroglycerin; VMAC— Vasodilation in the Management of Acute Congestive Heart Failure. (VMAC Investigators. Intravenous nesiritide vs nitroglycerin for treatment of decompensated congestive heart failure: a randomized controlled trial. JAMA. 2002;287:1531–1540. Monrad ES, Baim DS, Smith HS, et al. Milrinone, dobutamine, and nitroprusside: comparative effects on hemodynamics and myocardial energetics in patients with severe congestive heart failure. Circulation. 1986;73:III168–III174. Elkayam U. Nitrates in the treatment of congestive heart failure. Am J Cardiol. 1996;77:C41–C51.)
Initiation of inotropes is often considered more convenient, as there is less concern about initial responses than with intravenous vasodilators. However, after initial stabilization, continued use of these infusions may mask inadequacy or intolerability of the oral regimen. It is thus recommended that these infusions be stopped at least 24–48 hours prior to discharge to determine stability on oral therapy.22 Although requiring more supervision to initiate, vasodilator infusions are more convenient once it is time to wean onto oral therapies. Failure of weaning is much less common with vasodilator than with inotropic therapy, unless diuresis has been inadequate prior to weaning. The effects of intravenous vasodilators can more easily be matched with available oral therapies.
Randomized trial data are very limited regarding these agents in hospitalized populations. In a randomized trial of patients without baseline hypotension, the addition of milrinone was associated with more hypotension, tachyarrhythmias, and other cardiac events than placebo infusion, with no benefit for subsequent outcomes.19 Nitroglycerin and nesiritide caused earlier symptom relief and lower wedge pressures than placebo infusions.17 Most information comparing inotropes and vasodilator agents derives from retrospective review, in which it is not possible to capture all of the reasons leading to use of inotropic therapy, vasodilator therapy, or neither. Attempts to determine the factors leading to selection of inotropic therapy suggest that the practice at a given site dominates over physiologic variables. Nonetheless, a consistent theme emerges of worse in-hospital and subsequent outcomes in patients who have received intravenous inotropic therapy while in hospital. These differences persist when adjusted for all recognized clinical factors contributing to outcome, such as renal function, blood pressure, serum sodium, and diuretic dose. Patients receiving intravenous vasodilators in actual practice have baseline profiles indicative of more compromise than patients not receiving any intravenous therapy except diuretics. Outcomes with intravenous vasodilator therapy in these analyses have not been significantly different from outcomes with no vasoactive therapy, with or without adjustment for baseline characteristics.
COMBINING THERAPIES FOR PROFILES
Wet and Warm—Diuretics Only? (All Ejection Fraction)
The first decision regarding intravenous vasoactive therapy is made at the time of admission. Most patients will demonstrate moderate decompensation with congestive symptoms without evidence of acute circulatory compromise. Diuretic therapy would be initiated during continuation of the usual outpatient heart failure regimen, with consideration of additional vasoactive therapy if the response to escalating diuretic doses was inadequate over the next 48–72 hours. Patients who have required additional intravenous therapy for adequate response on previous admissions, or those in whom effective therapy was previously limited by poor renal function, might be considered for earlier use of these adjunctive intravenous agents but should also undergo frank discussions about prognosis and the appropriate goals of subsequent care.
Occasionally patients will present with frank pulmonary edema, often due to sudden severe elevation of filling pressures, particularly in the presence of severe hypertension or relatively low plasma oncotic pressure. This presentation is more common in patients with preserved than reduced LVEF. In this emergency setting of impending respiratory failure, intravenous vasodilators should be started along with diuretics for rapid relief of symptoms, improved oxygenation, and hopefully avoidance of intubation. Assuming that systolic blood pressure is adequate, intravenous nitroglycerin or nesiritide may be used. Nitroprusside is generally avoided in the emergency setting where myocardial ischemia may be present, due to concern that nitroprusside may produce coronary steal.
Intravenous vasoactive therapy is not currently considered necessary for most wet and warm patients, but it may allow lesser total diuretic dose. There is increasing concern that high doses of diuretics not only identify patients with poor underlying renal compensation, but may actually aggravate renal dysfunction during diuresis. It is not known whether patients requiring high doses to overcome diuretic “resistance” would benefit from earlier use of intravenous therapies that might improve renal blood flow and decrease diuretic requirements, to “spare” the kidney.
Definition of Two Limiting Profiles—All Ejection Fraction
During therapy directed to relieve congestion as clinically assessed, some patients who initially appear to fit the above profile of warm and wet do not respond as anticipated. Two recognized patterns that limit efficacy of the usual hierarchy of diuretic therapies are disproportionate right ventricular (RV) dysfunction and the cardiorenal syndrome (Fig. 12-1). Intravenous vasoactive infusions are frequently added to facilitate diuresis for these limiting profiles.
Right Ventricular > Left Ventricular Failure
Most patients with chronic heart failure have RV filling pressures that are less than half of the left-sided ventricular filling pressures, although they generally change in parallel. In the hemodynamic study of 1000 patients by Drazner, only 6% of patients had right atrial pressures >10 if the pulmonary wedge pressure was <22 mm Hg.4 As outcomes improve with heart failure, however, there is the clinical impression that more patients are surviving to develop progressive right heart failure. It should be emphasized that the distinction is not necessarily apparent from clinical assessment, as many patients with the average right-left relationship of elevated filling pressures will nonetheless have their clinical presentation dominated by symptoms of systemic venous congestion rather than dyspnea.
For patients in whom the right-sided pressures are more than two-thirds the left, it is more difficult to gauge optimal volume status. If the jugular venous pressure is reduced to the usual near-normal targets, the left-sided filling pressures could be excessively reduced, leading to a fall in cardiac output, hypotension, and renal dysfunction. More commonly, it is not possible to reduce the right atrial pressures, leading to escalating interventions with their own risks. Patients in whom diuresis is ineffective or limited by hypotension while jugular venous pressures are still elevated may benefit from invasive measurement of hemodynamics in order to establish modified filling pressure targets.
For the RV profile, vasodilation can be helpful if left-sided filling pressures are also markedly elevated, as the reduction of systemic vascular resistance, mitral regurgitation, and pulmonary pressures should allow better RV performance. Often, however, inotropic therapy is used to increase cardiac output and improve hemodynamic status. It is generally difficult to maintain improvement once inotropic therapy is weaned. If left ventricular assist device support is being considered, these patients need thoughtful evaluation regarding the potential need for added RV support.
The Cardiorenal Syndrome (All Ejection Fraction) Whether LVEF is low or preserved, renal function becomes the major limiting factor in effective therapy during at least 25% of heart failure hospitalizations. The cardiorenal syndrome is variously defined, but might be considered the worsening of renal function (e.g., >0.3 mg or 25% increase of creatinine) during diuresis for symptomatic heart failure, despite persistence of clinical volume overload.23 While little is known about the specific causes and therapeutic targets of the cardiorenal syndrome, the major advance that has been made in understanding is the growing recognition that the acute decline in renal function is not usually the result of an acute decline in cardiac output.24 This has been appreciated from direct hemodynamic studies, but also from the prevalence of the same clinical syndrome in patients with preserved LVEF in whom resting cardiac output is not reduced. For patients with low EF, the contribution of chronically impaired renal perfusion is assumed. For patients with preserved EF, additional components may be the greater prevalence of diabetes and baseline renal dysfunction, often in the setting of hypertension. Additional risk factors for all LVEF groups include chronic high diuretic dosage, duration of heart failure, and baseline renal dysfunction.25 Although the data are not well-established, there may be a major overlap between the disproportionate RV profile and the cardiorenal syndrome.
The pathophysiology is now agreed to include direct cardiorenal connections beyond those provided by central cardiac output (http:// www.nhlbi.nih.gov/meetings/workshops/cardiore nal-hf-hd.htm). Low pressure baroreceptors within the atria and pulmonary circuit may become desensitized by chronic distention, such that beneficial volume reduction is transduced as volume depletion. Changes in vasopressin and other circulating neurohormones are tightly influenced by cardiac distention, vascular baroreceptors, and intrarenal hemodynamics. Multiple responses in the glomeruli, afferent and efferent arterioles, and tubules likely contribute to the diminution in effective renal blood flow and increase in fluid retention in chronic heart failure. There is increasing concern that the high doses of diuretics used to achieve clinical targets may themselves be contributing to progressive renal dysfunction. Nonetheless, the robust relationship between elevated filling pressures and adverse outcomes mandate continued focus on volume reduction, hopefully with newer strategies.
The cardiorenal syndrome is generally treated with agents that could improve renal blood flow.
One method is to increase renal blood flow by increasing total cardiac output. This can be accomplished by low-dose inotropic therapy with dobutamine or dopamine, without evidence that any selective dopaminergic effect of dopamine is clinically useful. For some patients with marked volume overload, the progressive improvement that can occur in cardiac performance, peripheral perfusion, nutrition, and activity with diuresis may be adequate to maintain better renal function even after the inotropic therapy is discontinued.
The development of selective agents to enhance renal vasodilation remains a focus of new investigation. The natriuretic peptides can enhance renal blood flow as well as inhibit tubular reabsorption in multiple experimental settings. It has been difficult to show these effects in clinical practice for patients with the cardiorenal syndrome. BNP, nesiritide, has been associated with improved renal function in acute renal failure and is currently being studied in postoperative settings as well as in patients hospitalized with risk factors for the cardiorenal syndrome. Combinations of natriuretic peptides are also under investigation for this condition. Specific inhibitors of vasopressin and adenosine are also being evaluated.
Wet and Cold—How Acute?
The patient presenting with chronic decompensation and a cold and wet profile would generally be considered early for initiation of additional vasoactive therapy “to warm up” in order to “dry out.”22 As the “cold and wet” profile is an imprecise clinical definition, it would also be reasonable in some cases to observe the initial response to intravenous diuretic therapy before adding other therapy, particularly if there is an obvious factor to be addressed, such as recent increase in b-blocker dose or anemia requiring transfusion. The choice of vasodilators or inotropic agents in this population depends upon the adequacy of blood pressure and the assumption regarding systemic vascular resistance. When perfusion appears inadequate, it is important to reevaluate the level of neurohormonal antagonist therapy. Patients who have recently had initiation or escalation of b-blocker therapy should return to the previous level until stabilized. Inhibition of the renin-angiotensin system can also decrease perfusion in patients whose clinical compromise is so severe that angiotensin is potentiating maintenance of systolic blood pressure. These inhibitors may occasionally need to be stopped or decreased when adequate perfusion cannot be maintained. It remains controversial when or whether it is preferable to continue neurohormonal antagonism when the price includes the addition of inotropic therapy with its inherent risks.
Patients with dilated low EF heart failure who are initially without evidence of hypoperfusion may occasionally move into the cold and wet profile during observation, particularly in conjunction with the cardiorenal syndrome. Inotropic therapy is generally chosen over vasodilator therapy in the presence of severe hypotension (systolic blood pressure <80 mm Hg), for which vasodilator therapy should be administered only with extreme caution, usually with invasive monitoring.
It is critical to identify the patient in whom hypoperfusion is acutely progressive, with imminent compromise of organ function. This is the patient who presents a hemodynamic cold and wet profile together with additional components of life-threatening circulatory compromise such as lactic acidosis, anuria, declining mental status, or systolic blood pressure <70 mm Hg. The initial therapy for this patient will include inotropic stimulation with blood pressure support, usually with dopamine at pressor (vasoconstriction) doses. In the occasional patient who fails to respond, addition of epinephrine provides additional inotropic stimulation, although at high cost of tachyarrhythmias and ischemia.18 At this level of circulatory compromise, mechanical support should be considered immediately if appropriate in terms of the longer term outlook. It should be recognized that only a small minority of patients are appropriate candidates for urgent support with current mechanical devices. For patients in whom circulatory compromise is not rapidly resolving, decisions will need to be made expeditiously regarding the appropriate escalation of intervention or shift to emphasis on comfort over life-sustaining measures.
DESIGN FOR DISCHARGE
Design of the regimen for discharge should be a topic of consideration as soon as the patient is admitted (Table 12-2). At that time, consideration of the factors leading to hospital admission should include not only the medication regimen, but patient understanding and compliance, and potential exacerbating factors to be addressed during hospitalization, such as hypothyroidism or atrial fibrillation with rapid ventricular rates.
Neurohormonal Antagonists and Vasodilators
Once the patient has reached optimal volume status, neurohormonal antagonist therapy should be adjusted. Neurohormonal inhibition is generally not increased while volume overload persists, as transient decreases in cardiac output and blood pressure may impair effective diuresis. (For patients admitted with heart failure and hypertension, neurohormonal antagonists may be titrated up immediately as needed for blood pressure control, with close monitoring of renal function.) On the other hand, doses of renin-angiotensin system antagonists tolerated during volume overload may cause too much vasodilation when volume status is restored to normal and the peripheral vasculature is more responsive.
Initiation of b-blockade in hospital has been shown to be well-tolerated and effective in many patients with good baseline blood pressure who have responded well to diuretic therapy without evidence of hypoperfusion or the need for inotropic support (Table 12-3).26 If such patients were previously stable on b-blocker therapy before admission, most can be discharged on the same doses. Patients with recent escalation in dose prior to admission should in general be discharged on the earlier dose. For patients with recent need for inotropic support, severely reduced renal function, or systolic blood pressures <90 mm Hg, b-blockers would generally not be started until stability was demonstrated in the first few weeks after discharge. The striking benefit of b-blockers to decrease mortality, hospitalizations, and disease progression mandates that they be considered for all patients with heart failure, including the elderly. Despite vigorous efforts, however, b-blocker therapy is successfully initiated in only 65–80% of patients requiring hospitalizations for advanced heart failure. Patients admitted on b-blockers in whom they have to be discontinued by the time of hospital discharge have a 6-month mortality over twice as high as those who can tolerate b-blocker therapy.
The majority of patients tolerate ACE inhibitors or angiotensin receptor blockers even in late-stage heart failure. However, some patients develop circulatory-renal limitations of hypotension, progressive renal dysfunction, or hyperkalemia that lead to discontinuation of ACE inhibitors. While the relative risks and benefits of ACE inhibitors in such patients have not been established, those who discontinue ACE inhibitors for these reasons have a 1-year mortality that exceeds 50%.27
Spironolactone has been shown to improve survival and decrease hospitalizations when patients are carefully selected and monitored to reduce the risk of life-threatening hyperkalemia.28,29 Particular care is required when initiating this potassium-sparing diuretic during periods of changing volume status and renal function. Of particular concern is the patient with fluctuating renal function who tolerates spironolactone initiated in the hospital during active diuresis and kaliuresis, in whom hyperkalemia may not manifest until after discharge.
Combinations of hydralazine and nitrates, or, in some patients, high-dose nitrates alone, represent alternative therapy for patients no longer able to tolerate ACE inhibitors. Nitrates have often been added to ACE inhibitors in patients to treat persistent exertional dyspnea or marked vasoconstriction. Recent information on the benefit of adding the hydralazine-nitrate combination to ACE inhibitors in moderate-severe ambulatory heart failure indicates that these three agents in combination can produce further benefit in survival and quality of life.30
Risk Assessment During Hospitalization
Multiple parameters predict clinical outcome during and after hospitalization with heart failure. At the time of admission, risk for long length of stay and in-hospital mortality has been analyzed from the ADHERE database of routine clinical information for over 100,000 hospitalizations, both with reduced and preserved EFs.3 The strongest adverse predictive factor was admission blood urea nitrogen of >43 mg/dL. Once knowing that, the next stratification was systolic blood pressure > or <115 mm Hg. For populations with more advanced disease and rehospitalizations (Table 12-3), high blood urea nitrogen or creatinine and low systolic blood pressure also predict worse outcome.3,13 Additional information is provided in that population by elevated right atrial filling or pulmonary capillary wedge pressures, with the most predictive measurements being those obtained after best efforts to optimize therapy. High BNP levels at admission predict longer stay and higher in-hospital mortality. Discharge BNP levels are even more predictive, indicating that those patients whose BNP decreases during in-hospital therapy have a better outcome than those whose BNP remains higher.15 In the advanced heart failure population, short 6-minute walk distance at discharge, or the inability to do the 6-minute walk test, additionally predicts rehospitalization and mortality. The discharge regimen itself is predictive, with those patients on b-blockers, ACE inhibitors, and low diuretic doses having the best outcome. Patients at high risk for rehospitalization and death should be carefully evaluated for further intervention.
A few selected patients may have options for cardiac transplantation or left ventricular assist devices. Others may benefit from enrollment in intensive heart failure management programs, some with options for home monitoring in addition to daily weights and frequent phone calls. For patients without definitive replacement options, discussions should take place regarding patient preferences and increasing emphasis on palliation of symptoms.31
Patients with LVEF ≤30% have a significant risk of sudden death that increases with the clinical severity of disease. Patients with a prior history of sudden death, ventricular tachycardia, or syncope are at highest risk. Patients with this history and otherwise good prognosis for 1–2-year survival should undergo defibrillator implantation during the related hospitalization or as soon as possible thereafter. For patients who have not had prior events, the risk for sudden death increases in parallel with the risk for terminal hemodynamic decompensation, which is now the dominant mode of death for patients with heart failure. The risk factors discussed above identify patients who are most likely to have recurrent heart failure events. A robust predictor of poor outcome remains Class IV symptoms, defined as symptoms at rest or with minimal exertion. These are the indications for most heart failure hospitalizations, and constitute a contraindication to defibrillator implantation.
Stability at a better functional class cannot be determined until patients return for follow-up after hospitalization. Furthermore, there is some concern that defibrillator implantation after a recent event such as myocardial infarction may itself be associated with increased nonsudden death.32 With these considerations, it is reasonable to defer decisions regarding implantable cardioverter defibrillator (ICD) implantation for primary prevention to the outpatient setting.
Criteria for Discharge
With rising concern for hospital costs associated with length of stay, there is frequently pressure to discharge patients as soon as symptoms have improved. Failure to address all of the goals of hospitalization is a common cause for hospital readmission. It is frequently said that the first day of readmission is a more costly one than an extra day to ensure stability at the end of the initial hospitalization. Most crucial are the goals of stability (Table 12-7). Specifically, patients should demonstrate stability of fluid balance and blood pressure for at least 24 hours on the regimen planned for discharge.22 If intravenous therapy has been used, patients should not be discharged until at least 24 hours from the last intravenous therapy, or at least 48 hours after discontinuing an agent with prolonged physiological effects such as milrinone.
Education should begin early during hospitalization, with involvement of both patient and family. Key components include the major symptoms of heart failure and the elements of stable fluid balance regarding salt, fluid intake, daily weights, and the flexible diuretic regimen. Connection to follow-up care is critical, both for scheduled follow-up and for early changes noted at home. The benefits of specific heart failure management after heart failure hospitalization have been clearly shown, reducing rehospitalization and mortality, within those programs where a specialized nurse or nurse practitioner with ongoing knowledge of the patient can implement treatment changes within specified ranges.33 They have not been realized with centralized call centers isolated from patients and clinical decision makers.
Heart failure is a chronic undulating disease with periods of good clinical stability and periods of decompensation. Presentation to the hospitalization serves as a warning that the clinical condition is no longer stable. Heart failure hospitalization must be viewed not as a challenge for rapid discharge, but as an opportunity to reevaluate, revise the medical regimen, and improve the clinical course. The hospital plays a central role in the overall continuum of inpatient and outpatient care, which must be coordinated and integrated in order to maximize the quality and length of life for patients with heart failure.
REFERENCES
1. Lee DS, Austin PC, Rouleau JL, et al. Predicting mortality among patients hospitalized for heart failure: derivation and validation of a clinical model. JAMA. 2003;290:2581–2587.
2. MacIntyre K, Capewell S, Stewart S, et al. Evidence of improving prognosis in heart failure: trends in case fatality in 66 547 patients hospitalized between 1986 and 1995. Circulation. 2000;102:1126–1131.
3. Fonarow GC, Abraham WT, Yancy CW, et al. Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis. JAMA. 2005;293:572–580.
1. Drazner MH, Yancy C, Shah MR, et al. Utility of the history and physical examination in assessing hemodynamics in patients with advanced heart failure: the ESCAPE trial (abstract). Circulation. 2005;112:11–640.
2. Stevenson LW, Perloff JK. The limited reliability of physical signs for estimating hemodynamics in chronic heart failure. JAMA. 1989;261:884–888.
3. Nohria A, Lewis E, Stevenson LW. Medical management of advanced heart failure. JAMA. 2002;287:628–640.
4. Nohria A, Tsang SW, Fang JC, et al. Clinical assessment identifies hemodynamic profiles that predict outcomes in patients admitted with heart failure. J Am Coll Cardiol. 2003;41:1797–1804.
5. Stevenson L, Bellil D, Grover-McKay M, et al. Effects of afterload reduction (diuretics and vasodilators) on left ventricular volume and mitral regurgitation in severe congestive heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol. 1987;60: 654–658.
6. Stevenson LW, Tillisch JH. Maintenance of cardiac output with normal filling pressures in patients with dilated heart failure. Circulation. 1986;74:1303–1308.
7. Rosario LB, Stevenson LW, Solomon SD, et al. The mechanism of decrease in dynamic mitral regurgitation during heart failure treatment: importance of reduction in the regurgitant orifice size. J Am Coll Cardiol. 1998;32:1819–1824.
8. Stevenson LW, Tillisch JH, Hamilton M, et al. Importance of hemodynamic response to therapy in predicting survival with ejection fraction less than or equal to 20% secondary to ischemic or nonischemic dilated cardiomyopathy. Am J Cardiol. 1990;66:1348–1354.
9. Stevenson LW. Tailored therapy to hemodynamic goals for advanced heart failure. Eur J Heart Fail. 1999;1:251–257.
10. ESCAPE Investigators and Coordinators. Evaluation study of congestive heart failure and pulmonary artery catheterization. JAMA. 2005; 294:1625–1633.
11. Johnson W, Omland T, Hall C, et al. Neurohormonal activation rapidly decreases after intravenous therapy with diuretics and vasodilators for class IV heart failure. J Am Coll Cardiol. 2002;39:1623–1629.
12. Logeart D, Thabut G, Jourdain P, et al. Predischarge B-type natriuretic peptide assay for identifying patients at high risk of re-admission after
decompensated heart failure. J Am Coll Cardiol. 2004;43:635–641.
1. Fonarow GC, Chelimsky-Fallick C, Stevenson LW, et al. Effect of direct vasodilation with hydralazine versus angiotensin-converting enzyme inhibition with captopril on mortality in advanced heart failure: the Hy-C trial. J Am Coll Cardiol. 1992;19:842–850.
2. VMAC Investigators. Intravenous nesiritide vs nitroglycerin for treatment of decompensated congestive heart failure: a randomized controlled trial. JAMA. 2002;287:1531–1540.
3. Stevenson LW. Clinical use of inotropic therapy for heart failure: looking backward or forward? Part I: inotropic infusions during hospitalization. Circulation. 2003;108:367–372.
4. Cuffe MS, Califf RM, Adams KF, Jr, et al. Short-term intravenous milrinone for acute exacerbation of chronic heart failure: a randomized controlled trial. JAMA. 2002;287:1541–1547.
5. Follath F, Cleland JG, Just H, et al. Efficacy and safety of intravenous levosimendan compared with dobutamine in severe low-output heart failure (the LIDO study): a randomised double-blind trial. Lancet. 2002;360:196–202.
6. Monrad ES, Baim DS, Smith HS, et al. Milrinone, dobutamine, and nitroprusside: comparative effects on hemodynamics and myocardial energetics in patients with severe congestive heart failure. Circulation. 1986;73: III168–III174.
7. Stevenson LW, Massie BM, Francis GS. Optimizing therapy for complex or refractory heart failure: a management algorithm. Am Heart J. 1998;135:S293–S309.
8. Gottlieb SS, Abraham W, Butler J, et al. The prognostic importance of different definitions of worsening renal function in congestive heart failure. J Card Fail. 2002;8:136–141.
9. Weinfeld MS, Chertow GM, Stevenson LW. Aggravated renal dysfunction during intensive therapy for advanced chronic heart failure. Am Heart J. 1999;138:285–290.
1. Forman DE, Butler J, Wang Y, et al. Incidence, predictors at admission, and impact of worsening renal function among patients hospitalized with heart failure. J Am Coll Cardiol. 2004;43: 61–67.
2. Gattis WA, O’Connor CM, Gallup DS, et al. Predischarge initiation of carvedilol in patients hospitalized for decompensated heart failure: results of the Initiation Management Predischarge: Process for Assessment of Carvedilol Therapy in Heart Failure (IMPACT-HF) trial. J Am Coll Cardiol. 2004;43:1534–1541.
3. Kittleson M, Hurwitz S, Shah MR, et al. Development of circulatory-renal limitations to angiotensin-converting enzyme inhibitors identifies patients with severe heart failure and early mortality. J Am Coll Cardiol. 2003;41:2029–2035.
4. Pitt B, Zannad F, Remme WJ, Randomized Aldactone Evaluation Study Investigators. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med. 1999;341:709–717.
5. Bozkurt B, Agoston I, Knowlton AA. Complications of inappropriate use of spironolactone in heart failure: when an old medicine spirals out of new guidelines. J Am Coll Cardiol. 2003;41:211–214.
6. Taylor AL, Ziesche S, Yancy C, et al. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med. 2004;351: 2049–2057.
7. Stevenson LW. Rites and responsibility for resuscitation in heart failure: tread gently on the thin places. Circulation. 1998;98:619–622.
8. Hohnloser SH, Kuck KH, Dorian P, et al. Prophylactic use of an implantable cardioverterdefibrillator after acute myocardial infarction. N Engl J Med. 2004;351:2481–2488.
9. McAlister FA, Stewart S, Ferrua S, et al. Multidisciplinary strategies for the management of heart failure patients at high risk for admission: a systematic review of randomized trials. J Am Coll Cardiol. 2004;44:810–819.
|
