Patients require supplemental oxygen, this should be titrated by pulse oximetry to target an arterial saturation between 92 and 96 % (see Chap. 17). Strong evidence supports the use of NPPV to treat acute cardiogenic pulmonary edema (see Chap. 20). Positive pressure ventilation is “good for the left ventricle”; it reduces the work of breathing, reduces preload and reduces LV afterload. Both CPAP and BiPAP lower intubation and mortality rates compared to conventional therapy with oxygen [30, 31]. However, CPAP should be considered as the first line intervention as it is as efficacious as BiPAP and CPAP is cheaper and easier to implement in clinical practice [30, 31].

Morphine sulfate has long been used to treat ADHF on the basis that it is a potent venodilator with additional anxiolytic properties. However, data from the ADHERE registry suggests that the use of morphine is associated with an increased risk for intubation and is an independent predictor of mortality (OR 4.84) [32]. Based on this data morphine should be avoided in patients with ADHF.

It is important to recognize that patients with “congestive heart failure” have volume overload due to sodium retention primarily as a result of activation of the renin-angiotensin-aldosterone system (RAAS). Agents which improve cardiac performance and renal blood flow, decrease activation of the RAAS and result in a diuresis. Indeed in 1776, William Withering described a patient who was “cured from dropsy” after she self-administered an old cure of a garden plant called foxglove (Digitalis purpurea). Withering conducted experiments in patients with dropsy (congestive heart failure) and showed that foxglove increased urination [33]. It was believed at the time that digitalis was a diuretic which removed “poisons” from the blood stream.

Diuretics have been the mainstay for the treatment of ADHF for the past four decades; indeed these drugs (particularly furosemide) are still widely used and recommended for this indication [11, 34]. The ADHERE study reported that 89 % of patients presented with symptoms of volume overload and that 88 % received intravenous diuretics. However, although patients are volume overloaded with features of “congestion” there is little data to support the use of loop diuretics; indeed high dose furosemide is associated with worse outcomes. It is important to note that loop diuretics are associated with a fall in GFR, this may further compromises renal function in patients with cardiac failure. High dose diuretics have a number of adverse effects in patients with heart failure which include:

It is widely (although incorrectly) believed that diuresis improves cardiac function in patients with congestive cardiac failure. It has been postulated that diuretic induced changes in preload increase ventricular performance by two mechanisms; either by shifting the ventricle to a “more optimal position on the descending limb on the Starling Curve” or by reducing left ventricular size and thereby reducing systolic wall stress (afterload) by the LaPlace effect. However, it has been clearly demonstrated that in the physiological range there is NO descending limb of the left ventricular stroke volume-pressure curve (Frank-Starlings curve) in the mammalian heart (this includes humans). Furthermore, there is currently no evidence to support the contention that diuresis increases stroke volume or cardiac output in patients with congestive cardiac failure. Braunwald and colleagues demonstrated an average fall in cardiac output of 20 % following diuresis in patients with impaired cardiac function both at rest and during exercise [35]. Nelson and coworkers compared the hemodynamic effects of intravenous furosemide (1 mg/kg) with that of intravenous isosorbide dinitrate (50–200 μg/kg/min) in patients with LV failure following myocardial infarction [36]. The pulmonary artery occlusion pressure (PAOP) fell in both groups of patients’ however, the cardiac output was maintained in the nitrate group whereas it fell by about 10 % in the furosemide group. Similarly, Hutton and colleagues compared the effects of intravenous furosemide (0.5 mg/kg) and isosorbide 5-mononitrate (15 mg) at the time of cardiac catheterization in patients with LV dysfunction [37]. In this study furosemide induced acute vasoconstriction with a reduction in cardiac output. In contrast, isosorbide 5-mononitrate maintained cardiac output while reducing the PAOP.

It should therefore be no surprise that the use of high dose loop diuretics in patients with ADHF has been associated with adverse outcomes. Analysis of the ADHERE registry has provided compelling evidence regarding the harm of high dose diuretics in patients with ADHF [38]. Patients in the registry were stratified into a low-moderate (<160 mg) and high-dose group (≥160 mg) according to the cumulative dose of intravenous furosemide (or equivalent) administrated during the first 24 h of hospitalization. Patients in the high-dose group had a significantly greater decline in renal function, a longer length of stay and a higher in-hospital mortality rate (OR 0.87; 95 % CI 0.78–0.97, p = 0.01) when compared to the low-moderate group, even after adjustment for confounding factors. Similarly, in the ESCAPE trial, high dose furosemide was an independent predictor of hospital mortality and 6-month outcome [39]. Cotter et al. randomized 110 patients with cardiogenic pulmonary edema to either high dose nitrates or high dose furosemide (80 mg iv every 15 min until improvement) after receiving an initial 40 mg dose of furosemide [40]. Mechanical ventilation was required in seven (13 %) patients in the nitrate group and 21 (40 %) in the furosemide group (p = 0.0041). Myocardial infarction occurred in 9 (17 %) and 19 (37 %) patients (p = 0.047) and death in one and three patients (p = 0.61), respectively. One or more of these endpoints occurred in 13 (25 %) and 24 (46 %) patients, respectively (p = 0.041). Worsening renal function during hospitalization is a powerful independent prognostic factor for adverse outcomes. Metra and colleagues demonstrated that the daily furosemide dose was an independent predictor for worsening renal function, which itself was a predictor of death and rehospitalization [41]. The initial daily dose of furosemide was 82 mg in the group whose renal function remained stable as compared to 160 mg in those with worsening renal function. In a nested case-control study of 382 ADHF patients, Butler and colleagues showed that higher doses of loop diuretics were associated with an increased risk of worsening renal failure independent of the amount of fluid loss [42]. In a long term follow-up study Abdel-Qadir et al. demonstrated that in elderly patients with HF a dose of furosemide ≥120 mg/day was associated with worsened outcomes and was broadly predictive of death and morbidity [43].

Diuretics however may improve renal function in patients with cardio-renal syndrome and high venous pressures (see Chap. 12, Dangers of a high CVP). A high central venous pressure is transmitted backwards where it increases renal venous pressure and renal interstitial pressure; these effects markedly reduce renal blood flow and GFR. In patients with ADHF, Mullens et al. demonstrated a near linear relationship between increasing CVP and worsening renal function [44]. In this study worsening renal function occurred significantly less frequently in patients with a CVP < 8 mmHg. Furthermore, the CVP was the only hemodynamic parameter that predicted worsening renal failure, with the CI, systolic blood pressure and PCWP being similar between those patients who maintained renal function as compared to those with worsening renal function. The effect of diuresis on renal function likely depends on the balance of reduced stroke volume (following diuresis) which decreases renal function versus reduced central venous pressure which will reduce renal venous and interstitial pressure and may improve renal function. It is difficult to predict which patients will benefit or be harmed by diuresis. However, it is possible that renal function may improve following diuresis in patients with a high CVP. Renal function should be closely monitored in all patients with ADHF receiving diuretics and the dose reduced or stopped in those with worsening renal function. Based on this data patients with ADHF should receive no more than 40–80 mg furosemide per day and that this dose should be reduced in patient with worsening renal function. Administration of the diuretic as a continuous infusion does not appear to have any advantages over giving the drug as bolus doses [45].

Diuretics are appropriate therapy in patients with symptomatic cardiogenic pulmonary edema. However, it is important to realize that patients with failing hearts (chronic) are able to tolerate high pulmonary venous pressures without developing pulmonary edema. Patients with severe chronic left heart failure frequently lack pulmonary rales on examination or alveolar edema on chest x-ay, despite high pulmonary venous pressure (and features of pulmonary venous hypertension on chest x-ray); these patients may have pulmonary venous pressures > 30 mmHg. This observation is explained by reduced pulmonary microvascular permeability as well as increased lymphatic flow in these patients.

Nitroglycerin is the most commonly used vasodilator in the setting of acute heart failure; however it should be used with caution in hypotensive patients. Nitroglycerin’s effects are mediated through the relaxation of vascular smooth muscle; it reduces preload and afterload. Nitroglycerine increase cardiac output, decreases systemic vascular resistance and improves microcirculating perfusion [46]. Nitroglycerin can be given orally, topically, or intravenously, as long as blood pressure is maintained. This is one of the few agents which has been shown to improve outcome in ADHF, [40] and it is likely that this agent is underutilized for the treatment of this condition. The IV route is recommended in patients admitted to the ICU. The dosing of nitroglycerin is often suboptimal and may need to reach doses of about 160 mg/kg/min to achieve optimal effects. Headache is a common adverse effect but is generally ameliorated with acetaminophen.

Due to its toxicity nitroprusside is best avoided. The role of angiotensin converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARB) in ADHF has not been studied. Intravenous ACE inhibitors should be avoided [47]. A low dose of an oral ACE inhibitor can be considered in patients with stable renal function but should be avoided in patients with declining renal function.

Nesiritide is identical to the endogenous BNP produced by the body. It acts as a vasodilator (arterial and venous) and antagonizes the renin-angiotensin-aldosterone and sympathetic nervous systems. Nesiritide is given as a 2 mg/kg bolus, followed by an infusion of 0.01 mg/kg/min. Pooled analyses have raised concerns about the safety of this agent, with the drug being linked to worsening renal function and increased mortality [35, 48]. Chen et al. compared low-Dose Dopamine, Low-Dose Nesiritide and placebo in patients with ADHF and renal dysfunction (The ROSE Acute Heart Failure Randomized Trial). Neither dopamine nor nesiritide improved urine output or renal function (primary end-points) nor there was there any benefit on the secondary end points reflective of decongestion, renal function, or clinical outcomes. Based on this data nesiritide cannot be recommended in patients with ADHF.

Multiple large, randomized controlled trials have demonstrated that certain beta-blockers (carvedilol, metoprolol succinate and bisoprolol) reduce mortality by about 35 % in patients with chronic HF and systolic LV dysfunction [49, 50]. Until recently it has been unclear if beta-blocker therapy should be continued, withdrawn or the dose reduced in patients admitted to hospital with ADHF. An analysis of the OPTIMIZE-HF registry and the Carvedilol or Metoporlol European Trial (COMET) demonstrated a longer hospital length of stay and a higher risk of death in patients in whom beta-blocker therapy was discontinued or the dose reduced as compared to those patients in whom the beta-blocker was continued or therapy with a beta-blocker initiated [51, 52]. In the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) study beta-blocker therapy reduced the risk of death and hospitalization in the subgroup of patients with recent or recurrent decompensation [50]. This data suggests that all patients with ADHF should continue to be treated with a beta-blocker unless specifically contraindicated. It should be noted that beta-blockers are not contraindicated in patients with concomitant COPD; indeed, evidence suggests that beta-blockers may improve survival even in those COPD patients without overt cardiovascular disease [53]. In patients not previously treated with a beta-blocker initiation of low dose treatment should be considered.

Dobutamine may have a role in patients with acute left ventricular failure due to myocardial ischemia. In this setting dobutamine may recruit hibernating myocardium and improve cardiac function. Furthermore, dobutamine should be considered in patients with sepsis induced acute systolic dysfunction. The role of dobutamine in patients with chronic heart failure is unclear. Chronic heart failure is characterized by sympathetic hyper-activation and β-receptor downregulation. Short term infusions or continuous β-stimulant therapies have not been demonstrated to be beneficial in these patients [54]. Tacon et al. performed a meta-analysis of RCT that evaluated the role of dobutamine in patients with ADHF [55]. In this study the odds ratio for mortality for patients treated with dobutamine compared with standard care or placebo was 1.47 (CI 0.98–2.21, p = 0.06). This meta-analysis demonstrated that dobutamine is not associated with improved outcome in patients with heart failure, but was associated with a strong trend towards increased mortality. β-blockers have been demonstrated to improve outcome in patients with compensated heart failure and it therefore appears counterintuitive that β-stimulant therapy would have a role in ADHF.

Milrinone acts by inhibiting the phosphodiesterase III isoenzyme, which leads to increased cyclic adenosine monophosphate (cAMP) and enhanced inotropy. It differs from dobutamine, because it elevates cAMP by preventing its degradation as opposed to dobutamine, which increases cAMP production. In the OPTIME-CHF study the use of milrinone in patients with an ischemic cardiomyopathy was associated with an increase in the composite of death or rehospitalization (42 vs. 36 % for placebo, p = 0.01) [56]. Furthermore, in the ESCAPE heart failure trial, the use of an inotrope was associated with an increased risk of death; RR of 2.14, (95 % CI 1.10–4.15) [57]. This data suggests that both dobutamine and milrinone have a limited role in the management of patients with ADHF.

Vasopressin levels are inappropriately high in both acute and chronic HF [58]. Along with activation of the sympathetic nervous system and the RAAS, non-osmotic release of vasopressin is thought to represent a maladaptive response that is central to the pathophysiology of HF. Vasopressin antagonists has been investigated in patients with both acute and chronic heart failure. Although these agents are associated with a greater net fluid loss than placebo and normalization of the serum sodium they have not been associated with improved clinical outcomes [59].

Peripheral ultrafiltration is an alternative to diuretics for sodium and water removal. The Ultrafiltration Versus Intravenous Diuretics for Patients Hospitalized for Acute Decompensated Heart Failure (UNLOAD) trial enrolled 200 patients with ADHF and showed that ultrafiltration compared with diuretics increased weight loss at 48 h (5.0 vs. 3.1 kg; p < 0.001), decreased the need for vasoactive drugs (3 vs. 13 %; p = 0.02) and reduced the rate of readmission to hospital at 90 days (18 vs. 32 %; p = 0.02) [60]. More recently, Bart et al. tested the efficacy and safety of ultrafiltration in patients with ADHF complicated by persistent congestion and worsened renal function [61]. They randomized 188 patients to a strategy of stepped pharmacologic therapy or ultrafiltration. Ultrafiltration was inferior to pharmacologic therapy with respect to the end points of the change in the serum creatinine level and body weight 96 h after enrollment. A higher percentage of patients in the ultrafiltration group had serious adverse events. The changes from baseline in the creatinine level remained significantly different between the groups up to 60 days. While these two studies provide contradictory results they suggest that ultrafiltration should be avoided in patients with cardio-renal syndrome.

AF may occur in up to 50 % of patients with severe heart failure. If AF causes sudden severe worsening of heart failure immediate cardioversion may be necessary. However, most patients can be stabilized by using amiodarone or digoxin to control heart rate. Approximately 60 % of patients with acute AF (less 1 week) will spontaneously revert back to sinus rhythm. A randomized placebo controlled study demonstrated that the rate of conversion of acute AF to sinus rhythm was similar in a group of patients treated with amiodarone compared to the group receiving placebo [62]. It had previously been assumed that the prognosis of patients with heart failure and AF was improved if these patients were converted to sinus rhythm. However a recent RCT demonstrated that a strategy of rhythm control does not reduce the rate of death or cardiovascular complications as compared to a rate-control strategy [63].

Reduction of 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 patients with left ventricular systolic dysfunction. Intravenous nicardipine or fenoldopam (see Chap. 28) followed by treatment with on oral ACE inhibitor, ARB’s or amlodipine is recommended for blood pressure control in these patients.

Routine anticoagulation is not recommended. Patients with a history of systemic or pulmonary embolism, recent atrial fibrillation, or mobile left-ventricular thrombi should be anticoagulated aiming to achieve a prothrombin time ratio of 1.2–1.8 times control (INR 2.0–3.0).

Anemia is common and a poor prognostic indicator in HF [27, 64]. Multiple factors contribute to the anemia of HF including decreased erythropoietin production due to kidney injury, ACE inhibitors (inhibit erythropoietin production), increased levels of pro-inflammatory mediators, GI blood loss due to anti-platelet drugs, as well as iron deficiency [64]. Blood transfusions, however, do not improve outcome. Blood transfusion should only be considered in patients with a Hb < 7 g/dL (see Chap. 38). Intravenous iron (200 mg ferric carboxymaltose) has been demonstrated to improve symptoms in NYHA class II or III patients who had an ejection fraction of less than 45 % and had an iron deficiency anemia (ferritin < 100 μg/L and a hemoglobin less than 13.5 g/dL) [65]. Intravenous iron should therefore be considered in patients (non-septic) with ADHF who have an iron deficiency anemia. Erythropoiesis-stimulating agents (ESA’s) appear to have limited clinical benefits in anemic patients with heart failure [66].