Evaluation and Management of Patients with Acute Decompensated Heart Failure

Fluid Overload



It is recommended that patients admitted with ADHF and evidence of fluid overload be treated initially with loop diuretics - usually given intravenously rather than orally. (Strength of Evidence = B)

Ultrafiltration may be considered in lieu of diuretics. (Strength of Evidence = B)


Diuretic Therapy for Decompensated HF. Although their safety and efficacy have not been established in randomized, controlled trials, extensive observational experience has demonstrated that loop diuretics, generally alone but at times in combination with non-loop diuretics, effectively relieve congestive symptoms in patients admitted with volume overload. These agents remain first line therapy for the management of congested patients with ADHF (see Section 7 Tables 7.2 and 7.3).

Observational experience also suggests that loop diuretics should be administered intravenously for best effect in the setting of worsening HF. The bioavailability of oral furosemide is highly variable from patient to patient and even from day to day in the same patient and may be considerably lower in patients with decompensated HF. Furosemide, a commonly used loop diuretic, has a short duration of action, with a peak effect at 1 to 2 hours, which resolves approximately 6 hours after dosing. Administration 2 or more times a day may be necessary and is often the best approach when these agents are initially ineffective. Increasing the dose also improves response to diuretics if the current dose is insufficient to achieve maximal delivery of drug to the tubules. Alternatively, a continuous infusion of loop diuretic may help to maintain constant drug levels at target sites in the renal tubules.

Intravenous loop diuretics can produce significant acute reductions in left and right ventricular filling pressures as rapidly as 15 minutes after administration. This helps explain why some patients experience improvement in symptoms prior to the onset of the diuretic effect of these drugs.48 In contrast, administration of intravenous furosemide has been associated with neurohormonal activation, which may result in worsening of hemodynamics secondary to systemic vasoconstriction in the early stages of therapy.49 However, as sodium excretion increases and diuresis ensues, volume loss leads to a reduction in cardiac filling pressures and improvement in symptoms.49

Ultrafiltration. Mechanical methods of fluid removal are being actively investigated as potential alternatives to pharmacologic diuresis.50 Small uncontrolled studies have long suggested the utility of this approach using not only traditional dialysis but hemofiltration methods.51 Initial studies supporting the use of a venovenous system, or ultrafiltration, were small and had limited outcomes.52,53 But they did provide evidence supporting ultrafiltration as an option that may be considered for the reduction of fluid overload in acute decompensated HF. In addition, a single session of ultrafiltration was shown to reduce neurohormones and increase subsequent diuretic responsiveness.

The most extensive study of 200 patients hospitalized with HF and hypervolemia showed no effect on dyspnea at 48 hours, but did show a significant reduction in weight compared to bolus or continuous diuretics at 48 hours and an improvement in rehospitalization rates at 90 days.54 Despite its apparent effectiveness, cost, need for venous access, and nursing support are concerns, and more study is necessary.



It is recommended that diuretics be administered at doses needed to produce a rate of diuresis sufficient to achieve optimal volume status with relief of signs and symptoms of congestion (edema, elevated JVP, dyspnea), without inducing an excessively rapid reduction in 1) intravascular volume, which may result in symptomatic hypotension and/or worsening renal function, or 2) serum electrolytes, which may precipitate arrhythmias or muscle cramps. (Strength of Evidence = C)


Careful repeated assessment of signs and symptoms of congestion and changes in body weight is recommended, because clinical experience suggests it is difficult to determine that congestion has been adequately treated in many patients. (Strength of Evidence = C)


Monitoring of daily weights, intake, and output is recommended to assess clinical efficacy of diuretic therapy. Routine use of a Foley catheter is not recommended for monitoring volume status. However, placement of a catheter is recommended when close monitoring of urine output is needed or if a bladder outlet obstruction is suspected of contributing to worsening renal function. (Strength of Evidence = C)


Relief of congestion is a self-evident goal of diuretic therapy in congested patients admitted with worsening HF. Achieving this result, while avoiding hypotension and worsening renal function, often requires close observation and careful titration of these agents. Excessively rapid diuresis may result in symptomatic declines in blood pressure and reduced renal function, even while some degree of congestion persists.

Clinical experience suggests it may be difficult to identify persistent congestion. In contrast, even modest relief of congestion may be associated with substantial improvement in dyspnea and sense of well being in many patients despite ongoing volume overload, which may result in premature discharge. The care of patients admitted with worsening HF requires careful physical and symptom assessment and monitoring of vital signs, body weight, and laboratory results to optimize fluid status. Reduction in body weight during hospitalization should be anticipated in patients presenting with significant congestion. Careful history will often document a clear weight gain and suggest a target weight that may be desirable to achieve before discharge. However, accurate determinations of body weight and, even more so, intake and output are not easy to achieve, even in the hospital environment. These measurements should be correlated with other evidence of resolving congestion to achieve the best assessment of an adequate therapeutic response.



Careful observation for development of a variety of side effects, including renal dysfunction, electrolyte abnormalities, symptomatic hypotension, and gout is recommended in patients treated with diuretics, especially when used at high doses and in combination. Patients should undergo routine laboratory studies and clinical examination as dictated by their clinical response. (Strength of Evidence = C)

It is recommended that serum potassium and magnesium levels be monitored at least daily and maintained in the normal range. More frequent monitoring may be necessary when diuresis is rapid. (Strength of Evidence = C)

Overly rapid diuresis may be associated with severe muscle cramps. If indicated, treatment with potassium replacement is recommended. (Strength of Evidence = C)


Overview of the Adverse Effects of Diuretics. Despite beneficial effects in acute HF, diuretics may be associated with a variety of adverse effects that often require alterations in their use or the use of concomitant medications.55 Patients treated with diuretics should be monitored carefully for excessive urine output, development of hypotension, rise in serum BUN and creatinine levels and reductions in serum potassium, and magnesium levels. Serial determinations of creatinine and BUN are particularly important when these side effects are present or anticipated. Diuretic therapy must be highly individualized based on the degree of fluid overload present and the degree of volume loss produced to minimize these side effects.

Hypokalemia. Potassium must be monitored closely, especially during the period when diuresis is most pronounced, with supplementation given as needed. Patients with reduced serum potassium need immediate replacement before diuretic therapy for worsening HF. Aldosterone antagonists may be used cautiously in the setting of marked potassium wasting.

Hypotension. In patients with reduced LVEF and ventricular dilation, the effect of loop diuretics on cardiac output and blood pressure often seems counterintuitive. Despite decreasing filling pressures, loop diuretics usually do not produce clinically significant reductions in cardiac output or blood pressure in patients with worsening HF and LV systolic dysfunction. In patients with ventricular dilation and volume overload, total stroke volume is relatively independent of filling pressures.56 Diuretic-induced reductions in left and right heart filling pressures are frequently accompanied by augmented forward stroke volume and cardiac output, related to (1) diminution in functional mitral regurgitation; (2) diminution in functional tricuspid regurgitation; and (3) reduction in right ventricular volume, associated with relief of ventricular-interdependent LV compression and improved effective LV distensibility.

In contrast, some patients do experience symptomatic hypotension with decreasing cardiac output and blood pressure during therapy. Intravascular volume must be maintained by reequilibration as interstitial fluid moves into the vascular bed to maintain blood pressure even as diuresis proceeds. The time course of this phenomenon varies among patients and, especially during periods of brisk diuresis, may lag behind the decline in intravascular volume, resulting in hypotension despite persistent total body fluid overload.

Diuresis accompanied by a reduction in filling pressure may make patients more sensitive to the hypotensive effects of drugs with vasodilator properties. Diuretics may significantly enhance the hypotensive effects of ACE inhibitors, even when volume overload is still present. Patients with HF with preserved LVEF or restrictive, hypertrophic, or infiltrative cardiomyopathies may be more sensitive to diuresis and may decrease their blood pressure during diuretic therapy despite continued volume expansion. All patients receiving diuretic therapy need careful monitoring to prevent adverse hemodynamic effects from excessive volume loss.

Neurohormonal Activation. Older studies demonstrated that increased activity of the renin-angiotensin and sympathetic nervous systems may occur with intravenous diuretics, and result in secondary increases in systemic vascular resistance.57 It has been hypothesized that this acute vasoconstrictor response may play a role in the development of worsening renal function during treatment of ADHF. However, more recent studies in patients with ADHF have shown a reduction in plasma neurohormones, including norepinephrine, endothelin-1, and BNP, with parenteral diuretic and vasodilator therapy,58 as well as following ultrafiltration.54,59 Furthermore, the reduction in neurohormones appears to correlate with urine output and sodium excretion.60 Whether changes in circulating neurohormones have beneficial or adverse long-term effects in patients with ADHF or alter the responsiveness to diuretic therapy requires further study.

Other Side Effects. Diuretic agents may increase the incidence of digitalis toxicity, either by decreasing glomerular filtration rate or by inducing hypokalemia and hypomagnesemia. Electrolyte disturbances induced by diuretics may result in arrhythmia. Hyponatremia may occur as a result of diuretic therapy, in part because of increases in circulating vasopressin, which can further reduce renal clearance of free water, plus an increase in free water intake in turn impeding restoration of euvolemia.61,62 Diuretic therapy can also precipitate exacerbations of gout and at high doses cause reversible hearing loss.



Careful observation for the development of renal dysfunction is recommended in patients treated with diuretics. Patients with moderate to severe renal dysfunction and evidence of fluid retention should continue to be treated with diuretics. In the presence of severe fluid overload, renal dysfunction may improve with diuresis. (Strength of Evidence = C)


Diuretic therapy may further worsen renal function in patients with baseline renal insufficiency. Loop diuretics may produce intrarenal regulatory changes, related in part to neurohormonal activation, which can compromise glomerular filtration rate. Excessive diuresis or overly rapid diuresis may lower preload so that systemic blood pressure is compromised, especially in patients with marked HF with preserved LVEF and significant LV hypertrophy or restrictive physiology.

Despite these physiologic disadvantages, the net effect of diuretic therapy in individual patients with ADHF is difficult to predict. In some patients with reduced renal function at baseline, decongestion may improve serum creatinine and BUN, even as intravascular volume and filling pressures decline. Improved renal blood flow in response to relief of abdominal fluid overload is postulated as one physiologic mediator of this beneficial effect. Reduction of central venous pressure is another potential mechanism contributing to increases in glomerular filtration rate.



When congestion fails to improve in response to diuretic therapy, the following options should be considered:

  • Re-evaluating presence/absence of congestion
  • Sodium and fluid restriction,
  • Increasing doses of loop diuretic,
  • Continuous infusion of a loop diuretic, or
  • Addition of a second type of diuretic orally (metolazone or spironolactone) or intravenously (chlorothiazide).

Another option, ultrafiltration, may be considered. (Strength of Evidence = C)


Most patients admitted with worsened HF and congestion will respond adequately to loop diuretics with resolution of volume overload; however, a minority will experience some resistance to diuretic therapy. Increasing the frequency and then the dose of loop diuretic is recommended in these cases to restore volume status. Distal tubular diuretics augment the natriuretic effect of loop diuretics. These agents should be considered as adjunctive therapy in patients with diuretic resistance who do not respond to more frequent administration or escalating doses of loop diuretics. However, these agents can exacerbate adverse effects of loop diuretics, such as hyponatremia and hypokalemia.

Continuous infusion of a loop diuretic may produce higher and more sustained concentrations of furosemide within the renal tubule than repeated bolus injection. Continuous infusion may be associated with less prerenal azotemia and fewer other side effects compared with bolus administration, possibly because this method avoids the high peak concentrations associated with bolus dosing. 63



A low sodium diet (2 g daily) is recommended for most hospitalized patients. (Strength of Evidence = C)

In patients with recurrent or refractory volume overload, stricter sodium restriction may be considered. (Strength of Evidence = C)


Restricting fluid intake to 2 L/day is usually adequate for most hospitalized patients. Dietary sodium restriction is important, even short-term in the hospital setting, to help restore euvolemia. The level of sodium restriction prescribed during hospitalization may be greater than typically feasible in the outpatient setting. Education regarding sodium and fluid restriction may be initiated during an admission.64



Fluid restriction (<2 L/day) is recommended in patients with moderate hyponatremia (serum sodium <130 mEq/L) and should be considered to assist in treatment of fluid overload in other patients. (Strength of Evidence = C)

In patients with severe (serum sodium <125 mEq/L) or worsening hyponatremia, stricter fluid restriction may be considered. (Strength of Evidence = C)


Severe hyponatremia is not a common manifestation of ADHF, but is an ominous sign. However, recent results suggest that even reductions in serum sodium traditionally considered mild (<137 mEq/L) are associated with prolonged hospitalization and increased in-hospital mortality.65 Patients whose reduction in serum sodium is related to volume depletion as a result of diuretic therapy or environmental conditions will respond to administration of sodium and water. However, the great majority of hyponatremia in HF patients occurs in the setting of volume overload and cannot be corrected by the administration of sodium, which will only compound volume expansion.

Fluid restriction may produce some improvement in serum sodium concentration and may be transiently effective in mild hyponatremia. Fluid restriction can be difficult to maintain, because thirst is a common symptom in patients with HF. Patients may feel a certain amount of fluid ingestion is necessary for good health and that restriction will be harmful. Education concerning the benefits and lack of adverse effect of fluid restriction may help promote adherence. In patients with HF, hyponatremia is associated with a higher risk of clinical deterioration, including renal and hepatic dysfunction, longer hospital stays and high rehospitalization and mortality rates.66-69 The degree of hyponatremia is inversely associated with mortality.69 Hyponatremia in patients with HF is due to an inability to excrete free water, primarily due to neurohormonal activation. Increases in norepinephrine and angiotensin II result in decreased delivery of sodium to the distal tubule by causing decreased renal perfusion, while arginine vasopressin increases water absorption from the distal tubule. In addition, angiotensin II directly promotes thirst. Thus serum sodium is a marker for poor cardiac output and neurohormonal activation.

Recently it has been suggested that hyponatremia may be associated with more neurocognitive symptoms than previously recognized. In a case-control study of 122 patients (none with HF) with hyponatremia (serum sodium 126 +/-5 mEq/L), falls and attention deficits were far more common in the hyponatremic patients.70 Treatment of hypervolemic hyponatremia with a V2-selective vasopressin antagonist (tolvaptan) was associated with a significant improvement in The Mental Component of the Medical Outcomes Study 12 item Short Form General Health Survey.71

Treatment of hyponatremia consists of water restriction and maximization of medical therapies such as ACE-inhibitors or angiotensin receptor blockers which block or decrease angiotensin II, resulting in improved renal perfusion and decreased thirst. Vasopressin antagonists have been shown to improve serum sodium in hypervolemic, hyponatremic states with either a V2-selective or a non-selective vasopressin antagonist.71,72 Longer term therapy with a V2 selective vasopressin antagonist does not improve mortality but appears to be safe.73,74 Currently two vasopressin antagonists are available for clinical use (conivaptan and tolvaptan) and only short-term studies are available. At present it may be reasonable to utilize a non-selective vasopressin antagonist to treat hyponatremia in patients with HF who are observed to have significant cognitive symptoms due to hyponatremia. However, the long-term safety and efficacy of this approach remains unproven. In patients with refractory hyponatremia, alternative causes (eg, hypothyroidism, hypoaldosteronism, syndrome of inappropriate antidiuretic hormone) should be excluded.



Routine administration of supplemental oxygen in the presence of hypoxia is recommended. (Strength of Evidence = C)

Routine administration of supplemental oxygen in the absence of hypoxia is not recommended. (Strength of Evidence = C)


Routine oxygen administration in patients with acute HF is recommended to improve oxygen delivery to vital organs, including the myocardium. While there have been no randomized trials to support this, improving systemic and myocardial hypoxemia would be expected to improve the overall clinical status of patients with acute HF.

Supplemental oxygen therapy should be individualized. The congested dyspneic patient who presents with hypoxemia requires oxygen therapy. Patients with systemic fluid overload that does not compromise oxygenation do not require oxygen therapy. Supplemental oxygen in the setting of normal oxygen saturations on room air could be problematic if the patient also has a history of obstructive lung disease. In selected patients, oxygen may decrease elevated pulmonary vascular resistance and improve right heart function. Supplemental oxygen is also recommended in patients with acute myocardial infarction (MI) complicated by HF. The role of nocturnal oxygen in patients with central sleep apnea remains unproven.



Use of non-invasive positive pressure ventilation may be considered for severely dyspneic patients with clinical evidence of pulmonary edema. (Strength of Evidence = A)


Previous small trials investigating the use of noninvasive ventilation (NIV) in emergency department patients with acute HF suggested it improved symptoms, decreased the need for subsequent intubation and reduced mortality.75-77 There were some concerns from one of the early trials that bi-level positive pressure ventilation may have led to an increase in the number of patients with myocardial infraction.78 However, a subsequent study did not encounter this association, and a review of data from the original trial suggests that a disproportionate number of patients with evolving MI may have been enrolled in the interventional arm.79,80 Several subsequent meta-analyses based on these smaller NIV trials also suggested that both intubation rates and mortality were reduced with NIV.81-83 However, results from a recent large randomized trial in the United Kingdom suggest while NIV improved patient's dyspnea and metabolic abnormalities, it did not significantly change mortality and intubation rates when compared to standard oxygen therapy.84 Further, the authors found no significant differences in efficacy between continuous positive pressure ventilation and bi-level positive pressure ventilation. While this study has rather robust results and randomized over 1100 patients, it is worth noting that concomitant therapy was not standardized, opioids were used in over 50% of patients, and over 15% of patients on standard therapy crossed over to NIV. Further, the primary endpoint was measured at 7 days, a time far removed from the time frame of use of NIV. Despite those limitations, the preponderance of evidence suggests NIV is a useful temporizing measure that improves dyspnea but likely has no impact on intubation rates or mortality.

Table 7.2: Summary of Recommendations for the Administration of Beta Blocker Therapy

  • Initiate at low doses
  • Uptitrate gradually, generally no sooner than at 2-week intervals
  • Use target doses shown to be effective in clinical trials
  • Aim to achieve target dose in 8-12 weeks
  • Maintain at maximum tolerated dose
Considerations if symptoms worsen or other side effects appear
  • Adjust dose of diuretic or other concomitant vasoactive medication
  • Continue titration to target dose after symptoms return to baseline
  • Considerations if uptitration continues to be difficult
    • Prolong titration interval
    • Reduce target dose
    • Consider referral to a HF specialist
If an acute exacerbation of chronic HF occurs
  • Maintain therapy if possible
  • Reduce dosage if necessary
  • Avoid abrupt discontinuation
  • If discontinued or reduced, reinstate gradually before discharge

See Recommendations 7.10-7.11 and accompanying text for specific recommendations

Table 7.3: Loop Diuretics

Agent Initial Daily Dose (mg) Maximum Total Daily Dose (mg) Elimination Duration of Action (hr)
Furosemide* 20-40 mg qd or bid 600 mg 65%R 35%M 4-6
Bumetanide* 0.5-1.0 mg qd or bid 10 mg 62%R 38%M 6-8
Torsemide* 10-20mg qd 200 mg 20%R 80%M 12-16
Ethacrynic acid*,+ 25-50 mg qd or bid 200 mg 67%R 33%M 6

Equivalent doses: furosemide 40 mg = bumetanide 1 mg = torsemide 20 mg = ethacrynic acid 50 mg.
R = renal; M = metabolic; B = excreted into bile; U = unknown.
*Available for oral or intravenous administration (no dosage adjustments).
+Non-sulfa containing, may be used in sulfa-allergic patients.