INTRODUCTION — Generalized edema can occur in a variety of disorders, including heart failure, cirrhosis (where ascites is usually most prominent), and the nephrotic syndrome; significant edema is not always present in chronic kidney disease (CKD), where extracellular fluid volume expansion is more typically manifested by hypertension, but CKD can produce resistance to diuretic treatment. When edema is massive, and the excess fluid accumulation is widely distributed, the patient is said to have anasarca.
Edematous patients generally respond to the combination of dietary sodium restriction, treatment of the underlying disease process, and diuretic therapy, usually with a loop diuretic. However, some patients are resistant to typical doses of diuretics.
The treatment of refractory edema in adults is reviewed here. The initial therapy of edema, treatment of the different major edematous states, the clinical manifestations and diagnosis of edema in adults, and the evaluation and management of edema in children are discussed separately:
DEFINITION OF REFRACTORY EDEMA — Refractory edema is defined as edema that is refractory to typically effective doses of loop diuretics . While this definition appears to be straightforward, it is often difficult to determine the optimal extracellular fluid volume status for an edematous patient, as this process involves substantial clinical judgement and shared decision-making.
The following discussion will describe principles underlying the treatment of refractory edema; a following section will present specific recommendations for such patients.
CAUSES OF REFRACTORY EDEMA — A variety of factors can account for persistent sodium and water retention, despite treatment with typical doses of loop diuretics, including [2-6]:
●Inadequate diuretic dose or frequency (see 'Inadequate diuretic dose or frequency' below)
●Decreased intestinal absorption of oral diuretics (see 'Decreased oral diuretic absorption' below)
●Decreased diuretic secretion into tubular fluid (see 'Decreased loop diuretic tubular secretion' below)
●Increased sodium reabsorption at sites in the nephron other than those inhibited by the diuretic (see 'Enhanced tubular sodium reabsorption' below)
●Excess sodium intake (see 'High salt intake' below)
●Concomitant administration of other classes of drugs that interfere with diuretic action (see 'Drugs that interfere with diuretic action' below)
Inadequate diuretic dose or frequency — Diuretics do not produce natriuresis until a threshold rate of drug excretion is attained (figure 1). If, for example, a patient does not respond to 40 mg of furosemide, the dose may not have exceeded this threshold. Thus, the single dose should be increased to 60 or 80 mg, rather than giving the 40 mg dose twice a day. Once a single effective dose has been determined, it is most commonly administered multiple times per day at a frequency that is individualized to the diuretic needs of the patient and the pharmacokinetic profile of the agent chosen. (See "Loop diuretics: Dosing and major side effects".)
Edematous patients are typically treated with one of the sulfonamide-based loop diuretics: furosemide, bumetanide, or torsemide. Each drug has a unique pharmacokinetic profile, although all three exert their predominant effect by blocking the Na-K-2Cl cotransporter at the luminal surface of thick ascending limb cells (figure 2). (See "Loop diuretics: Dosing and major side effects" and "Mechanism of action of diuretics".)
Decreased oral diuretic absorption — In patients with heart failure, decreased intestinal perfusion, reduced intestinal motility, and perhaps also mucosal edema can reduce the rate of diuretic absorption, and therefore reduce the rate of diuretic delivery into the tubular lumen by as much as 50 to 70 percent. This reduced delivery of diuretic into the renal tubule may keep it below the threshold [7,8]. Thus, patients with acute decompensated heart failure typically require initial intravenous (IV) therapy.
The defect in diuretic absorption is often reversible, resulting in more effective oral therapy following removal of some of the edema fluid with IV diuretics and stabilization of cardiac function . Similar considerations may apply to patients with advanced cirrhosis , although some data suggest that furosemide absorption is maintained in this setting and that the risks of IV treatment outweigh the benefits. (See "Ascites in adults with cirrhosis: Initial therapy".)
Decreased loop diuretic tubular secretion — Loop diuretics must enter the tubular fluid in order to exert their diuretic effect (figure 2). Loop diuretics are highly (≥95 percent) protein bound. As a result, they primarily enter the tubule lumen by secretion by the proximal tubule, not by glomerular filtration.
Patients who do not respond to usual doses of loop diuretic therapy may be resistant because of decreased diuretic secretion into the tubular lumen. This may result from decreased kidney perfusion in patients with heart failure (due to the reduced cardiac output), cirrhosis (due to renal vasoconstriction), and kidney dysfunction (due to competitive inhibition of tubular diuretic secretion by high concentrations of organic anions that accumulate when the glomerular filtration rate is low). (See "Loop diuretics: Dosing and major side effects".)
Patients with nephrotic syndrome may also be unresponsive to diuretics due to decreased tubular secretion. Because loop diuretics are highly protein bound, severe hypoalbuminemia (<2 g/dL) associated with the nephrotic syndrome may reduce the delivery of diuretic to the renal tubule. In addition, filtered albumin in nephrotic patients may bind loop diuretics in the tubular lumen, thereby interfering with their function. Proteases that are filtered in abnormal amounts in nephrotic patients may also activate epithelial sodium channels, thereby contributing to excess sodium retention . (See "Pathophysiology and treatment of edema in adults with the nephrotic syndrome", section on 'Treatment'.)
Enhanced tubular sodium reabsorption — Some patients have partial or relatively complete resistance to a loop diuretic despite adequate secretion of the diuretic into the tubular fluid (ie, despite having the same rate of urinary diuretic excretion as normal controls) (figure 1). This problem is often due to increased tubular sodium reabsorption by nephron segments other than the loop of Henle [3,11,12]. The decreased response to the action of a diuretic that results from increased sodium reabsorption in other nephron segments has been called the diuretic braking phenomenon (figure 3) .
This phenomenon may be mediated by sodium reabsorption in various other nephron segments:
●In the proximal tubule, due to enhanced activity of angiotensin II and norepinephrine .
●In the distal tubule, due to flow-dependent hypertrophy resulting from chronic loop diuretic therapy, which increases sodium delivery and reabsorption at this thiazide-sensitive site (figure 4). The hypertrophy increases sodium reabsorption (figure 5) [13,15]. Hypokalemia may also contribute to hypertrophy and increase the activity of the thiazide-sensitive Na-Cl cotransporter [16,17].
●In the collecting tubules, due to increased aldosterone (figure 6) [4-6]. Additionally, filtered proteases in the settings of nephrotic syndrome, and perhaps heart failure, may cleave the epithelial sodium channel (ENaC), which increases its conductance for sodium [10,18,19].
The neurohumoral activation occurs both as part of the primary disease (as in heart failure or cirrhosis) and as a consequence of diuretic-induced salt and water loss. One consequence of activating compensatory sodium-retaining mechanisms as soon as the extracellular fluid volume falls is that the maximum response to an IV diuretic occurs with the first dose, even with a continuous infusion, and the natriuresis begins to diminish within the first 12 hours (figure 3) [3,20]. Similar findings are seen with oral diuretic therapy, assuming no defect in drug absorption. (See "Time course of loop and thiazide diuretic-induced electrolyte complications".)
Two strategies may be helpful in overcoming the increase in sodium reabsorption in nephron segments not inhibited by the loop diuretic; the efficacy of either of these strategies is enhanced with concurrent sodium restriction since diuretics only transiently reverse the sodium-avid state and substantial sodium retention can occur when the diuretic effect wears off :
●If the patient has a partial but inadequate diuretic response to once-daily bolus therapy, the loop diuretic dose can be repeated twice or even three times a day. (See 'Intensification of oral loop diuretic therapy' below.)
●A second or third natriuretic agent can be added. (See 'Combination oral diuretic therapy' below.)
High salt intake — Maintenance of a high sodium intake can prevent net sodium loss and a reduction in extracellular fluid volume, even if there is an appropriate natriuretic response to diuretics.
As the diuretic effect wanes, there is postdiuretic sodium retention due to the recovery of sodium reabsorption in the loop of Henle plus increased sodium reabsorption at other sites in the nephron [12,21]. If sodium intake is high, postdiuretic sodium retention can counteract part or all of the natriuresis that occurred while the diuretic was active. The postdiuretic effect can be counteracted by dietary salt restriction. (See 'Enhanced tubular sodium reabsorption' above.)
Drugs that interfere with diuretic action — Refractory edema may result from drugs that interfere with the action of diuretics. Examples include:
●Nonsteroidal antiinflammatory drugs, which reduce the synthesis of vasodilator and natriuretic prostaglandins and impair diuretic responsiveness [6,22].
●Thiazolidinediones, such as rosiglitazone, increase renal salt retention as a result of upregulation of the ENaC in the collecting ducts and also increase proximal tubule sodium reabsorption [23,24]. (See "Thiazolidinediones in the treatment of type 2 diabetes mellitus", section on 'Fluid retention/heart failure'.)
MANAGEMENT OF REFRACTORY EDEMA
Pretreatment evaluation — Before intensifying diuretic therapy, the following steps should be taken:
●Exclude excessive sodium intake – It is important to evaluate the patient's dietary sodium intake and attempt to reduce it, if excessive. To estimate sodium intake, a 24-hour urine should be collected. A value above 100 mmol in 24 hours (ie, 2.3 g of sodium) suggests that nonadherence with sodium restriction may be in part responsible for the resistance to diuretic therapy. (See 'High salt intake' above and "Patient education: Collection of a 24-hour urine specimen (Beyond the Basics)".)
A value above 100 to 120 mmol in 24 hours also suggests that the diuretic response is adequate since true diuretic resistance is manifested by intense renal sodium retention.
●Confirm that the patient requires a reduction in extracellular fluid volume – While the mere presence of residual edema or pulmonary congestion (crackles or breathlessness) is suggestive, the decision to further reduce the extracellular fluid volume requires careful clinical judgement. As an example, a patient with pulmonary hypertension and heart failure with preserved systolic function may require a right-sided filling pressure that is elevated and associated with mild pedal edema to maintain sufficient left-sided function.
By contrast, in the setting of heart failure with reduced ejection fraction, more aggressive decongestion is usually needed, possibly guided by biomarkers such as brain natriuretic peptides or ultrasonographic assessment of the inferior vena cava . (See "Natriuretic peptide measurement in heart failure".)
Stepwise approach to refractory edema — The following approach is directed primarily to patients who have heart failure or substantially impaired kidney function. The management of refractory ascites in patients with liver disease and the management of edema in patients with nephrotic syndrome are presented elsewhere. (See "Ascites in adults with cirrhosis: Diuretic-resistant ascites" and "Pathophysiology and treatment of edema in adults with the nephrotic syndrome".)
Steps are provided in a progressively escalating order, although variations in the approach will be appropriate for specific patients. As an example, starting with intravenous (IV) therapy (rather than intensification of oral diuretics) is appropriate for acutely ill patients, such as those with acute decompensated heart failure.
Our stepwise approach is similar to those successfully used in patients with acute decompensated heart failure and cardiorenal syndrome enrolled in the three randomized diuretic trials [26,27], with a goal of attaining a daily diuresis of 3 to 5 liters per day until satisfactory decongestion occurs.
Intensification of oral loop diuretic therapy — Most patients can achieve adequate control of signs and symptoms of extracellular fluid volume expansion with typical diuretic doses given orally (table 1).
The following recommendations apply when this initial approach proves insufficient:
●Patients who do not have an adequate response to oral loop diuretic therapy should have the dose increased until either a clinically significant diuresis is attained or the maximum daily dose has been reached. (See "Loop diuretics: Dosing and major side effects".)
Loop diuretics act rapidly, and therefore the patient should experience a notable increase in urine output within hours of an effective dose. If this does not occur, and especially if the body weight does not decline to indicate a reduction in extracellular volume, then doses can be escalated weekly, as long as adequate patient monitoring is available.
●When the diuretic response to oral loop diuretics is partial but inadequate (an increase in urine volume but little to no decrease in body weight), the net effect may be improved by repeating the same dose two or three times per day. Most guidelines suggest that, when initiating furosemide therapy, it be given more than once daily. Torsemide, by contrast, is typically given daily, owing to its longer half-life. (See "Loop diuretics: Dosing and major side effects".)
●Patients who have not responded to a maximum effective oral dose of furosemide may have erratic or slowed gastrointestinal absorption of the drug. (See "Loop diuretics: Dosing and major side effects".)
●Among patients who do not have an adequate response to the maximum effective oral dose of torsemide (100 to 200 mg per day in divided doses), an oral thiazide diuretic may be added to block distal reabsorption of sodium. (See 'Combination oral diuretic therapy' below.)
Combination oral diuretic therapy — When an adequate response to loop diuretics is not obtained, concurrent administration of a thiazide-like (metolazone, chlorthalidone, indapamide) or thiazide-type (hydrochlorothiazide) diuretic to block distal sodium chloride reabsorption should be employed:
●In patients without hypokalemia, we typically add metolazone (5 mg once daily initially, increased to a maximum dose of 20 mg once daily) , although hydrochlorothiazide (25 to 50 mg twice daily initially, increased to a maximum dose of 200 mg per day) is likely just as effective . These drugs can produce a substantial additional diuresis when added to loop diuretics, even if kidney function is impaired [30,31]. Chlorthalidone and indapamide are other options, although supportive data are lacking. (See "Thiazides versus loop diuretics in the treatment of hypertension", section on 'Patients with chronic kidney disease'.)
The timing of combination therapy depends upon the route by which the diuretics are given. The drugs can be administered at the same time if given by the same route (IV or oral). If, however, a thiazide diuretic such as hydrochlorothiazide is given orally in patients treated with an IV loop diuretic, the thiazide diuretic should precede the loop diuretic by two to five hours, since the peak effect of the thiazide is four to six hours after ingestion.
Combination diuretic therapy can lead to a marked diuresis in which daily sodium and potassium losses can be greater than 300 mmol and 200 mmol, respectively [31,32]. As a result, careful monitoring of fluid and electrolyte balance is essential when combination therapy is initiated (or when doses are changed). (See 'Monitoring on therapy' below.)
●However, in hypokalemic patients, we generally add a potassium-sparing diuretic first (eg, amiloride or, particularly in patients with heart failure, a mineralocorticoid receptor antagonist). Mineralocorticoid receptor antagonists were shown to be underutilized in a large study of patients who have heart failure with reduced ejection fraction, providing an opportunity for practice improvement . A thiazide-like or thiazide-type diuretic can then be added if the patient's edema remains refractory and hypokalemia has resolved.
●We typically do not add acetazolamide (a carbonic anhydrase inhibitor that blocks sodium reabsorption by the proximal tubule) unless the patient has severe metabolic alkalosis. Acetazolamide is less effective than thiazide-like and thiazide-type diuretics .
Sodium reabsorption can be blocked in the distal tubule by the concurrent administration of a thiazide-like or thiazide-type diuretic [6,30,31,35]. The distal tubule reabsorbs approximately 75 to 80 percent of the sodium delivered out of the loop of Henle. Administration of a loop diuretic increases distal sodium delivery and, therefore, sodium reabsorption in the distal tubule, an effect mediated in part by increased activity of the thiazide-sensitive Na-Cl cotransporter in the luminal membrane of the distal tubule cells .
The efficacy of adding a thiazide diuretic was illustrated in a study of five patients who had an inadequate response to furosemide in doses of 160 to 240 mg per day (given in divided doses three to four times daily) . All had significant chronic kidney disease (CKD) with a serum creatinine ranging from 2.3 to 4.9 mg/dL (203 to 433 micromol/L). Increasing the furosemide dose to 320 to 480 mg/day produced only a modest increase in urine output. By contrast, the addition of hydrochlorothiazide (25 to 50 mg twice daily) was associated with a marked diuresis and a significant reduction in weight.
Many experts use metolazone as the oral thiazide diuretic of choice in patients with refractory edema and advanced kidney disease (ie, glomerular filtration rate [GFR] below 20 mL/min) since other thiazide-type drugs were thought to be ineffective in this setting . However, the study that produced this conclusion used extremely high doses of metolazone (20 to 150 mg, the equivalent of 200 to 1500 mg of hydrochlorothiazide). Thus, there is no convincing evidence that metolazone is superior to comparable doses of other thiazides, although it, like chlorthalidone and indapamide, does have the advantage of once-daily dosing .
Potassium-sparing diuretics, such as mineralocorticoid receptor antagonists and amiloride, can help limit potassium wasting, although data demonstrating their efficacy in reducing extracellular fluid volume are limited [10,37]. In one study, for example, high-dose spironolactone (100 mg per day compared with placebo or low-dose spironolactone) did not improve clinical congestion score, increase urine output, or reduce body weight in patients hospitalized for acute heart failure who were receiving IV loop diuretics . However, these results, plus the finding that the serum potassium did not increase, may be explained by the fact that patients were followed for only 96 hours, and the onset of action of spironolactone may be slow.
Intravenous loop diuretic bolus therapy — Patients with acute decompensated heart failure and hospitalized patients with refractory edema are typically treated with IV loop diuretics. The following represents our general approach in such patients:
●The initial dose of IV loop diuretic should be approximately 2 or 2.5 times the patient's home oral dose (eg, a patient taking 40 mg of furosemide orally twice daily at home can be given 80 mg of IV furosemide as an initial bolus) .
In one large diuretic trial, IV loop diuretics given at 2.5 times the home daily dose were safe for patients with acute decompensated heart failure and exhibited modestly better outcomes than an IV dose equivalent to the home dose . In addition, the Heart Failure Association of the European Society of Cardiology (ESC) suggests initiating IV dosing at one to two times the home daily dose [40,41].
●If there is little or no response to the initial dose, the dose should be doubled at two-hour intervals, as needed, up to the maximum recommended doses (table 1) rather than repeated at the same dose. As an example, if an 80 mg IV furosemide dose does not produce an adequate diuresis within two hours, then a 160 mg IV dose should be given. Repeated administration of the same, ineffective IV dose is unlikely to increase efficacy.
It should be noted that doses higher than the "maximum effective dose" often produce further diuresis, albeit with less sodium excretion per milligram of diuretic administered. This is because the plasma diuretic concentration remains above the diuretic threshold for a longer period of time. Some evidence-based guidelines for the treatment of refractory edema recommend doses above the maximum effective dose. Yet, they may also increase the risk of side effects. (See "Loop diuretics: Dosing and major side effects", section on 'Maximum effective doses'.)
●Patients who do not have an adequate response to a maximal dose of one IV loop diuretic are unlikely to respond to another loop diuretic since their mechanisms of action are similar. In contrast to the potential benefit derived from switching oral furosemide to an alternative oral loop diuretic, there is no benefit from switching to an alternative IV loop diuretic. The higher efficacy of oral torsemide compared with oral furosemide results from differences in oral bioavailability. There is no such difference in bioavailability when the route of administration is IV. (See 'Intensification of oral loop diuretic therapy' above.)
●Loop diuretics should not be administered IV as a rapid "push." Bolus IV furosemide doses of 160 to 200 mg (and equivalent doses of bumetanide and torsemide) occasionally cause transient tinnitus. This effect can be minimized by giving the dose more slowly . Ototoxicity is most common when high doses of a loop diuretic are given rapidly at rates for furosemide above 4 mg/min.
●In patients who fail to respond to maximal IV bolus doses of a loop diuretic, a thiazide-like or thiazide-type diuretic can be coadministered. Chlorothiazide is the only thiazide diuretic that can be given IV (500 to 1000 mg once or twice daily) . However, the availability of IV chlorothiazide may be limited. An oral preparation, such as hydrochlorothiazide or metolazone, is an alternative for acute therapy and can be used chronically. In a retrospective analysis, oral metolazone appeared to be as effective as IV chlorothiazide, suggesting that this agent should be chosen unless the patient cannot take medications by mouth . As noted above, if an oral thiazide (such as metolazone) is combined with an IV loop diuretic, the metolazone should be given two to five hours before infusion of the loop diuretic.
Resistance to oral diuretics is frequently reversible, and therefore, oral diuretic therapy may become more effective following removal of some of the edema fluid with IV diuretics. Thus, patients who are treated with an IV diuretic following failure of oral therapy can be given a repeat trial of oral diuretics once the hypervolemia has improved. (See 'Conversion to oral treatment' below.)
Continuous infusion option in patients who respond to bolus therapy — In patients with refractory edema who respond to an IV bolus of a loop diuretic but need ongoing diuresis, we frequently initiate a continuous loop diuretic infusion. Continuous diuretic therapy may be less ototoxic than bolus therapy and maintains a sustained effective rate of diuretic excretion. A continuous infusion should only be used in patients who are first responsive to bolus loop diuretic therapy. Bolus therapy results in higher initial serum diuretic concentrations and, therefore, higher initial rates of urinary diuretic excretion than a continuous infusion . Thus, in patients who fail to respond to the maximal bolus doses described above, continuous infusion will be ineffective.
Our approach to continuous infusion of loop diuretics is as follows:
●An IV loading dose of 40 to 80 mg of furosemide is typically given over five minutes prior to initiating the continuous infusion. If the patient has received one or more IV boluses within the previous few hours, then an infusion can be started without a loading dose.
●After the loading dose, the starting infusion rate with furosemide varies with the level of kidney function:
•If the patient has severe kidney impairment (acute kidney injury or CKD with an estimated GFR [eGFR] less than 30 mL/min per 1.73 m2), we use an initial furosemide infusion rate of 20 mg per hour. If the diuresis is not sustained, a second bolus is given followed by a higher infusion rate of 40 mg per hour. The equivalent dose is 1 mg per hour, increasing to 2 mg per hour for bumetanide, and 10 mg per hour, increasing to 20 mg per hour for torsemide.
•If the patient has an eGFR greater than or equal to 30 mL/min per 1.73 m2, we use an initial furosemide infusion rate of 5 mg per hour. If the diuresis is not sustained, a second bolus is given followed by a higher infusion rate of 10 mg per hour. This rate may then be increased further to the maximum recommended furosemide infusion rate of 40 mg per hour if response to lower doses is poor.
●Higher infusion rates of up to 240 mg per hour are reported in the literature. However, we do not recommend these higher doses. The risk of ototoxicity and other side effects associated with these infusion rates must be weighed against alternative strategies, such as the addition of a thiazide diuretic or fluid removal via ultrafiltration, as described below. (See 'Patients unresponsive to intravenous diuretics' below.)
The physiologic rationale for a continuous IV infusion compared with bolus therapy is related to maintenance of an effective rate of drug excretion and, therefore, inhibition of sodium chloride reabsorption in the loop of Henle over time (figure 1). By contrast, bolus therapy is associated with initially higher and then lower rates of diuretic excretion; as a result, sodium excretion may be at near maximal levels for the first two hours but then gradually falls until the next dose is given (figure 3).
The efficacy of a continuous IV infusion compared with IV bolus therapy has been evaluated in randomized trials [20,26,39,45]. A continuous IV infusion is associated with less ototoxicity than bolus injections of loop diuretics, but clear evidence of greater diuretic efficacy is lacking, at least in patients with acute decompensated heart failure. The following observations illustrate the range of findings:
●In a systematic review of eight trials that included 254 patients with acute heart failure, continuous infusion led to a modestly higher urine output at 24 hours (mean difference 271 mL) . In addition, ototoxicity was less frequent in patients receiving continuous therapy. However, the apparent benefit on urine output was largely driven by a trial of 107 patients in which patients given continuous therapy also received hypertonic saline.
●The best data come from the DOSE trial, which was published after the aforementioned meta-analysis . In this trial, 308 patients with acute heart failure were randomly assigned to receive bolus IV furosemide every 12 hours or continuous infusion, and also to either low-dose (equivalent to the patient's previous oral dose) or high-dose (2.5 times the previous oral dose) diuretics. At 72 hours, the following findings were noted:
•Continuous and bolus therapy resulted in similar degrees of symptom improvement and fluid loss. However, the continuous regimen did not include an initial bolus of diuretic, which is typically recommended, and the dose of continuous infusion was not escalated in a manner similar to that recommended in most guidelines (table 2). In addition, patients assigned bolus therapy required dose escalation and, therefore, higher total furosemide doses (592 versus 480 mg, although this was not statistically different). Bolus therapy also led to nonsignificantly higher rates of ventricular tachycardia and myocardial infarction.
•High-dose as compared with low-dose furosemide produced greater net fluid loss, weight loss, and relief from dyspnea, but also more frequent worsening of kidney function (23 versus 14 percent). However, in this and some other trials, a rise in creatinine was not associated with worse long-term outcomes provided that the patients were decongested . Thus, because modest rises in creatinine are not typically associated with tubule injury , they should not lead to a reduction in dose, as long as diuresis is successful. (See 'Decongestion' below.)
Patients unresponsive to intravenous diuretics — Some patients with advanced heart failure and kidney impairment do not respond to or cannot tolerate maximal loop diuretic doses given IV in combination with thiazide diuretics. In this setting, the excess fluid can be removed by ultrafiltration during kidney replacement therapy. (See "Continuous kidney replacement therapy in acute kidney injury", section on 'Slow continuous ultrafiltration (SCUF)' and "Continuous kidney replacement therapy in acute kidney injury", section on 'Continuous venovenous hemofiltration (CVVH)'.)
Kidney replacement therapy is generally reserved for patients with substantial cardiorenal dysfunction who require concomitant dialysis or in patients enrolled in clinical trials.
Benefit from a supine posture — Diuretic responsiveness can be influenced by posture, although the effects of posture have not been specifically studied in patients with refractory edema. Patients with heart failure, for example, cannot increase cardiac output normally with assumption of an upright position. As a result, kidney perfusion and presumably urinary diuretic delivery further diminish. In addition, renal salt and water reabsorption increase.
The efficacy of assuming a supine position was evaluated in a randomized trial of six patients with heart failure and six with cirrhosis and ascites . After staying in bed overnight, the patients were treated with 1 mg of bumetanide IV and then assigned to either continued bed rest or normal daily activity for the next six hours. The procedure was repeated two days later when each patient was assigned to the opposite group. The supine position was associated with higher mean creatinine clearance (100 versus 66 mL/min) and diuretic response (1133 versus 626 mL at six hours). The upright position was associated with significant increases in plasma norepinephrine, renin, and aldosterone.
Monitoring on therapy — The need to monitor clinical and laboratory parameters varies depending upon the clinical scenario. For stable patients on maintenance diuretics, clinical status, serum electrolytes, and creatinine are assessed during routine clinical visits.
However, when diuretic therapy is being intensified to reduce the extracellular fluid volume, these parameters should be monitored frequently (weekly, for example) until stable, satisfactory doses of diuretics are attained. Frequent monitoring, at least initially, is often recommended for patients being discharged after a hospitalization for acute heart failure. In addition to daily weight monitoring, many heart failure programs provide specific advice on diuretic dosing and the need for laboratory testing.
Worsening kidney function is observed commonly during diuretic treatment and has been associated with worse outcomes . For this reason, worsening kidney function, typically defined as an increase in plasma creatinine concentration >0.3 mg/dL (26.5 micromol/L), is frequently used as a harm signal in large clinical trials . Yet, post hoc analyses of such trials have indicated that this association is not necessarily causal and may be response dependent. In the DOSE study, for example, worsening kidney function was not associated with poorer prognosis . In addition, in the ROSE-AHF trial, changes in creatinine did not indicate kidney injury, as manifested by a lack of injury biomarkers [47,48].
Some evidence suggests that the state of decongestion during diuretic treatment is a major determinant of prognosis in the setting of worsening kidney function; worsening kidney function may be viewed as an adverse effect when decongestion is incomplete (the patient remains resistant to diuretics) but not when an acceptable diuretic response has been attained.
Goals of therapy
Decongestion — The desired response to diuresis is determined by the severity of extracellular fluid volume overload and the clinical scenario. In the setting of acute decompensated heart failure, many authorities suggest relatively rapid decongestion (3 to 4 or 3 to 5 liters per day) and a goal of complete decongestion, as persistent congestion has been associated with worse outcomes [27,41]. It should be noted, however, that these associations have been determined retrospectively, and prospective data dictating the role of persistent extracellular fluid volume expansion in prognosis are lacking.
As noted, it is difficult to recommend a single approach to estimate congestion. An algorithm endorsed by the Heart Failure Association of the ESC provides some guidance (figure 7) . Yet, it should be emphasized that this approach has not been validated in prospective randomized trials.
Conversion to oral treatment — Hospitalized patients are typically considered ready for discharge once they have been stable for 24 hours on oral therapy. It is often recommended to double the furosemide dose when switching from IV to oral administration since the mean bioavailability is only approximately 50 percent. There is substantial interpatient and intrapatient variability in the bioavailability of oral furosemide, however, making it difficult to give a fixed ratio that would apply to all patients, and the Heart Failure Association of the ESC suggests starting with the same oral dose as was used IV . In our practice, we typically substitute an oral dose that is between 1 and 1.5 times the IV dose and then monitor urine output to determine its effectiveness.
●In one study of normal volunteers, a single oral dose of 40 mg produced the same natriuretic effect as an IV dose of 40 mg; in this case, the peak serum concentration was higher with the IV dose, but the duration above the diuretic threshold was longer with oral dosing . In outpatients, the efficacy of therapy can be monitored by having daily weights called into the clinician's office or monitored electronically.
●IV and oral doses are similar in patients treated with bumetanide or torsemide, which have higher rates of oral bioavailability than furosemide (70 to 95 percent and 80 to 90 percent, respectively). Careful monitoring is still required since there is substantial variability in response.
●More important than specific arbitrary dosing guidelines, however, is the need for early follow-up. A reassessment of weight, physical examination, and laboratory measurements is recommended within two weeks of discharge.
STRATEGIES WE DO NOT USE
Renal dose dopamine — Although some small, lower-quality studies suggested a benefit from low-dose dopamine, a larger, high-quality trial found no benefit and potential harm; therefore, low-dose dopamine cannot be recommended.
The addition of low-dose dopamine (or alternatively, nesiritide) in the setting of acute decompensated heart failure was examined in 360 patients in the large randomized ROSE-HF trial . The addition of low-dose dopamine did not significantly increase urine volume or improve decongestion compared with placebo, although it did increase the incidence of arrhythmias.
Albumin infusion for hypoalbuminemia — Some patients with hypoalbuminemia are relatively resistant to conventional diuretic therapy. It has been proposed that these patients may respond to the administration of 40 to 80 mg of furosemide mixed with 6.25 to 12.5 g of salt-poor albumin. Infusion of the furosemide-albumin complex is thought to increase diuretic delivery to the kidney by keeping furosemide within the vascular space. In a small initial report, this approach led to a substantial increase in sodium excretion in some patients .
However, a subsequent study in patients with nephrotic syndrome (mean plasma albumin 3.0 g/dL) found that a mixture of loop diuretic and albumin (60 mg of furosemide infused in 200 mL of a 20 percent albumin solution) produced only a modest increase in sodium excretion compared with furosemide alone, without an increase in the rate of furosemide excretion . In addition, the increase in sodium excretion was roughly equivalent to the amount of sodium contained in the colloid solution, suggesting that volume expansion was a likely explanation for the enhanced natriuresis. Notably, no net sodium loss occurred.
A similar lack of efficacy of furosemide plus albumin infusion has been demonstrated in hypoalbuminemic patients with cirrhosis (mean plasma albumin 3.0 g/dL) . Compared with furosemide alone, combination therapy did not increase the rate of either furosemide or sodium excretion.
Few patients with severe hypoalbuminemia (plasma albumin <2.0 g/dL) have been studied. It is possible that infusion of furosemide or other loop diuretics plus albumin would be effective in patients with refractory edema and severe hypoalbuminemia. However, if attempted, sodium excretion should be monitored after addition of albumin infusions, and infusions should be discontinued if they do not result in a net increase in sodium excretion. General issues related to the management of edema in patients with nephrotic syndrome are discussed in detail elsewhere. (See "Pathophysiology and treatment of edema in adults with the nephrotic syndrome".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Fluid and electrolyte disorders in adults".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topic (see "Patient education: Edema (swelling) (Beyond the Basics)").
SUMMARY AND RECOMMENDATIONS
●Refractory edema is defined as edema that is persistent despite the use of typically effective doses of loop diuretics. While this definition appears to be straightforward, it is often difficult to determine the optimal extracellular fluid volume status for an edematous patient, as this process involves substantial clinical judgement and shared decision-making. (See 'Definition of refractory edema' above.)
●A variety of factors can account for persistent sodium and water retention, despite treatment with typical doses of loop diuretics, including:
•Inadequate diuretic dose or frequency (see 'Inadequate diuretic dose or frequency' above)
•Decreased intestinal absorption of oral diuretics (see 'Decreased oral diuretic absorption' above)
•Decreased diuretic secretion into tubular fluid (see 'Decreased loop diuretic tubular secretion' above)
•Increased sodium reabsorption at sites in the nephron other than those inhibited by the diuretic (see 'Enhanced tubular sodium reabsorption' above)
•Excess sodium intake (see 'High salt intake' above)
•Concomitant administration of other classes of drugs that interfere with diuretic action (see 'Drugs that interfere with diuretic action' above)
●Before intensifying diuretic therapy in patients at steady state, excessive sodium intake (a value above 100 mmol in 24 hours) should be excluded, and it should be confirmed that the patient requires a reduction in extracellular fluid volume. (See 'Pretreatment evaluation' above.)
●The following approach is directed primarily to patients who have heart failure or substantially impaired kidney function; the management of refractory ascites in patients with liver disease and the management of edema in patients with nephrotic syndrome are presented elsewhere (see "Ascites in adults with cirrhosis: Diuretic-resistant ascites" and "Pathophysiology and treatment of edema in adults with the nephrotic syndrome"):
•Patients who do not have an adequate response to oral loop diuretic therapy should have the dose increased until either a clinically significant diuresis is attained or the maximum daily dose has been reached (table 1). In patients who have not responded to a maximum effective oral dose of furosemide, it is reasonable to switch to oral torsemide. (See 'Intensification of oral loop diuretic therapy' above.)
•When an adequate response to loop diuretics is not obtained, concurrent administration of a thiazide-like (metolazone, chlorthalidone, indapamide) or thiazide-type (hydrochlorothiazide) diuretic to block distal sodium chloride reabsorption should be employed. However, in hypokalemic patients, we generally add a potassium-sparing diuretic first. A thiazide-like or thiazide-type diuretic can then be added if the patient's edema remains refractory and hypokalemia has resolved. (See 'Combination oral diuretic therapy' above.)
•Patients with acute decompensated heart failure and hospitalized patients with refractory edema are typically treated with intravenous (IV) loop diuretics. The following represents our general approach in such patients (see 'Intravenous loop diuretic bolus therapy' above):
-The initial dose of IV loop diuretic should be approximately 2 or 2.5 times the patient's maintenance oral dose.
-If there is little or no response to the initial dose, the dose should be doubled at two-hour intervals, as needed, up to the maximum recommended doses (table 1) rather than repeated at the same dose.
-In patients who fail to respond to maximal IV bolus doses of a loop diuretic, a thiazide-like or thiazide-type diuretic can be coadministered.
•In patients with refractory edema who respond to an IV bolus of a loop diuretic but need ongoing diuresis, we frequently initiate a continuous loop diuretic infusion. Continuous diuretic therapy may be less ototoxic than bolus therapy and maintains a sustained effective rate of diuretic excretion. A continuous infusion should only be used in patients who are first responsive to bolus loop diuretic therapy. (See 'Continuous infusion option in patients who respond to bolus therapy' above.)
●Some patients with advanced heart failure and kidney impairment do not respond to or cannot tolerate maximal loop diuretic doses given IV in combination with thiazide diuretics. In this setting, the excess fluid can be removed by ultrafiltration during kidney replacement therapy. (See 'Patients unresponsive to intravenous diuretics' above.)
●Hospitalized patients are typically considered ready for discharge once they have been stable for 24 hours on oral therapy. In our practice, we typically substitute an oral dose that is between 1 and 1.5 times the IV dose and then monitor urine output to determine its effectiveness. (See 'Conversion to oral treatment' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟