A perspective on diuretic resistance in chronic congestive heart failure

Abstract

Chronic congestive heart failure (CHF) is a complex disorder characterized by inability of the heart to keep up the demands on it, followed by the progressive pump failure and fluid accumulation. Although the loop diuretics are widely used in heart failure (HF) patients, both pharmacodynamic and pharmacokinetic alterations are thought to be responsible for diuretic resistance in these patients. Strategies to overcome diuretic resistance include sodium intake restriction, changes in diuretic dose and route of administration and sequential nephron diuretic therapy. In this review, we discuss the definition, prevalence, mechanism of development and management strategies of diuretic resistance in HF patients.

Introduction

The loop diuretics are often used in many patients with chronic congestive heart failure (CHF) due to their indisputable efficacy in relieving congestive symptoms. In simple terms, diuretic resistance in heart failure (HF) patients can be explained as a failure of diuretics to control salt and water retention even when used in appropriate doses.¹ It may occur due to decrease in renal function which leads to pharmacokinetic abnormalities such as reduction or delay in peak concentrations of loop diuretics in the renal tubular fluid.¹ However, it can also occur in the absence of such pharmacokinetic abnormalities.

Definition and prevalence of diuretic resistance

The term ‘diuretic resistance’ remains inadequately defined despite its increasing frequency. It can simply be defined as either a loss of response or reduction in the response to loop diuretics.² It can develop in one out of every three HF patients.³ Generally, failure to reduce the volume of extracellular fluid despite using diuretics appropriately can be termed as ‘diuretic resistance’. More precisely, diuretic resistance can be expressed as a fractional excretion of sodium (FE_Na+) of <0.2% that represents the amount of sodium excreted (mmol/time) as a percentage of the filtered sodium load.⁴

Epstein and colleagues conducted a study to delineate the efficacy of the metolazone-furosemide regimen in patients with diuretic refractory edema.⁵ The investigators studied 24 patients who were initially considered to be refractory to large doses of conventional loop diuretics during an 18-month period. Among those 24 patients, 8 were presumed to have refractory edema and thus entered the study. In one of the study criteria, they mentioned diuretic resistance as a failure to excrete sodium (at least 90 mmol) within 72 h of giving oral furosemide (160 mg) twice a day.⁵ Only three out of those eight patients who entered the study met the study criteria for diuretic refractory edema.

Diuretic resistance is less commonly encountered in patients with mild CHF and preserved renal function, compared with patients with moderate and severe CHF.^6,7 Meanwhile, it is reported in another study that 38% of the patients with CHF have renal impairment and thus they are at an increased risk for the development of diuretic resistance.⁸ Several trials have shown that HF is commonly accompanied by a reduction in the estimated glomerular filtration rate (eGFR) and the prevalence of moderate to severe kidney impairment is approximately 30–60% in patients with HF.^9–13 The Acute Decompensated Heart Failure National Registry (ADHERE) database reported data on over 100,000 patients with HF requiring hospitalization and approximately 30% of those patients had a diagnosis of chronic kidney disease, while only 9% had a normal eGFR.^10,11

Mechanisms of diuretic resistance in heart failure

There are several theories that can explain the mechanism of diuretic resistance development. Several physiological changes in CHF can lead to alterations in drug pharmacokinetics, such as alterations in absorption, distribution, metabolism and elimination of the loop diuretics. However, diuretic resistance in CHF patients cannot be fully explained by these pharmacokinetic changes alone because if only alterations in pharmacokinetics were responsible, then diuretic resistance should be overcome by increasing the dose or changing the route of administration.³ Instead, the diuretic resistance may be better explained by parallel changes in drug pharmacodynamics and pharmacokinetics affecting the time course of drug delivery.¹⁴ When compared with healthy volunteers, CHF patients have a reduced rate of drug absorption which leads to a delay in the time to achieve a threshold dose, with the consequent development of diuretic resistance.^15,16 However, bioavailability of diuretics remains unchanged, therefore these changes can be better explained by the presence of gastrointestinal edema in patients with active HF.¹⁷

Usually, furosemide reaches the tubular fluid by its secretion from the organic anion transporter located in the proximal tubule.¹⁸ In patients with CHF, renal insufficiency leads to diuretic resistance due to insufficient intratubular concentrations of diuretics, which can be explained by decreased renal blood flow and impaired secretion by the proximal convoluted tubule (PCT).^17,19 The secretion of loop diuretics is reduced due to the accumulation of endogenic organic anions which compete with loop diuretics for binding at the receptor site on organic anion transporter.²⁰ This competitive inhibition can be overcome by increasing the dose of loop diuretics and this explains the need for higher loop diuretic doses to achieve therapeutic urinary concentration in CHF patients with renal impairment.

Another important theory behind diuretic resistance is the drug–drug interactions, such as nonsteroidal anti-inflammatory agents (NSAIDs), which may interfere with diuretics action by inhibiting prostaglandins and thus reducing renal perfusion.^17,21–23 In severe HF, prostaglandins are vital for maintaining renal perfusion and for promoting sodium and water excretion. Therefore, inhibition of prostaglandins with either aspirin or any other NSAIDs can eventually attenuate diuretic efficacy by inhibiting sodium and water excretion.²³ It has been reported that after the discontinuation of 100 mg of aspirin in patients with terminal, intravenous (IV) catecholamine-dependent HF, there was a marked improvement, stabilization and abatement of hyponatremia with an impressive reduction in diuretic requirements.²⁴ In an animal model, administration of prostaglandin E2 in indomethacin-treated rats was shown to restore the natriuretic response to furosemide and thus the model favors the fact that the use of NSAIDs is one of the major causes of apparent diuretic resistance.²⁵

In general, acute administration of loop diuretics in healthy patients can cause activation of the renin-angiotensin-aldosterone system (RAAS) reflexively, which further increases sodium and water retention and thus curtails the diuretic effect.²⁶ This should not be prominent in severe CHF, as such patients already have an activated RAAS and thus loop diuretics cannot further activate RAAS or increase the release of neurohormones.³ Additionally, most patients with CHF are receiving RAAS inhibitors, so this should counteract further activation of RAAS following acute administration of diuretics.²⁷ Haller and colleagues reported a case of a 29-year old patient having dilated cardiomyopathy refractory to high-dose furosemide and concluded that the diuretic resistance was due to the combination of heart pump failure, dietary indiscretion and hyperaldosteronism.²⁸ Thus, it is not unusual for a CHF patient to have multiple mechanisms responsible behind the development of diuretic resistance.

High sodium intake may mask the diuretic effects and thus it can cause difficulty to the clinician in determining diuretic resistance. Dietary noncompliance is not an actual form of diuretic resistance, but it may lead to diuretic failure. In patients with dietary indiscretion, loop diuretics lead to pronounced natriuresis initially but that can be followed by post-diuresis avid sodium reabsorption, thereby causing diuretic failure.²⁹ CHF patients are usually placed on long-term diuretic treatment which may be associated with pharmacological alterations within the nephron, thus resulting in an aggravated response to sodium intake.³ Data from patients with hypertension and experimental animal studies demonstrated that the chronic inhibition of sodium reabsorption in the loop of Henle due to chronic administration of loop diuretics, led to the delivery of sodium in higher amounts to the early distal convoluted tubule (DCT) causing cellular hypertrophy and thus augmented sodium reabsorption, which eventually led to a diminished natriuresis with shifting of the dose–response curve to down and right (See Figure 1). This diminished natriuretic effect due to the chronic administration of loop diuretics is known as the ‘braking phenomenon’.^{18,26,30–32}

Figure 1.

Dose–response curve in patients with CHF on chronic loop diuretics.

Critically ill patients admitted in intensive care units (ICUs) may sometimes develop another variant of diuretic resistance, which is termed as ‘occult diuretic resistance’. In this variant, such ICU patients on loop diuretics excrete only free water with no or very little sodium in their urine which ultimately lead to dehydration and hypernatremia. It is a frustrating condition as such patients initially seem to respond to the loop diuretics. This temporary diuretic failure develops due to a combination of factors in ICU patients such as: (i) the stress causes their kidneys to retain more sodium and thus loop diuretics stimulate the production of dilute urine with very low sodium; (ii) they may have excess sodium intake in the form of various medications and IV drips formulated in normal saline and (iii) an inability to drink water due to intubation which may lead to dehydration and thus increased sodium and water re-absorption from the kidneys. There may be none or very little net fluid loss when these patients are rehydrated. The ‘occult diuretic resistance’ can be managed by a traditional approach in which the lost free water is replaced by enteral water or IV 5% dextrose while continuing diuresis. However, the ultimate amount of the lost fluid will be very low. The better approach to counteract this type of diuretic resistance is by adding a thiazide diuretic to loop diuretic therapy [Please refer ‘combination diuretic therapy (sequential nephron blockade)’ below].³³

Management of diuretic resistance

Noncompliance

The first step to be taken when diuretic resistance is encountered is to exclude noncompliance with either medication intake or sodium restriction (sodium intake should be <100 mmol/day).¹ As discussed above, when sodium intake exceeds 100 mmol/day, sodium lost by diuresis can be completely compensated by postdiuretic avid sodium retention.³⁴ Sodium intake is usually assessed by measuring 24 h urine sodium excretion in the steady state.¹ Dietary noncompliance is suspected in subjects receiving diuretic therapy, when sodium excretion is as high as >100 mmol/day without any associated weight loss. Meanwhile, medication (diuretic) compliance can be assessed by measurement of the amount of diuretic excreted in urine, although this assessment would prove useful only in a very few cases.¹

Dose adjustment

The pharmacodynamic and pharmacokinetic changes of loop diuretics occurring in CHF patients can be compensated by increasing the dose of diuretics.¹ Few researchers have studied the usefulness of high-dose furosemide in the management of refractory CHF.^35,36 A study reported the safety and efficacy of high-dose furosemide (250–4000 mg/day, oral or IV) in 35 patients with significantly compromised kidney function and severe CHF unresponsive to conventional diuretic treatment.³⁶ All patients showed improvement in symptoms and weight reduction with no significant side effects.³⁶ As discussed before, patients with renal impairment have diminished renal blood flow and compromised organic anion transporter activity which interferes with furosemide secretion and thus leads to lower concentrations in the tubular fluid.¹ Many patients with CHF have some impairment in renal function that makes it necessary to increase the diuretic dose to deliver the drug at its site of action in appropriate amounts. When salt intake is not adequately restricted, postdiuretic sodium retention is an important mechanism responsible for diuretic resistance as most loop diuretics are short acting.¹ This postdiuretic salt retention can be overcome by more frequent administration of the diuretic (3 times daily or more) that results in reduction of the drug-free interval.¹

As shown in Figure 1, the urinary drug concentrations required to achieve adequate diuresis in healthy subjects may not be able to achieve the expected diuresis in CHF patients. It means that it is often required to increase the loop diuretic dose, even without any abnormalities in drug pharmacokinetics. Bumetanide and torsemide have better bioavailability than furosemide and thus some physicians consider them more effective than furosemide in CHF patients.¹ Generally, the bioavailability of furosemide shows significant intrapatient/interpatient variability and ranges from 10–100%.³⁷ In contrast, bumetanide and torsemide have a bioavailability of 80% and 80–100%, respectively.¹⁵ Both furosemide and bumetanide were found equally efficacious when equipotent doses were administered, even though bumetanide is 40-times more potent than furosemide on a weight basis.^38,39

IV bolus injection or continuous infusion of a loop diuretic

As previously discussed, one mechanism of diuretic resistance in CHF patients is impaired absorption of loop diuretics without an absolute change in their bioavailability, compared with normal subjects.^40,41 This leads to the delayed and decreased peak concentrations in the urine. This problem may be obviated by moderately increasing the dose or switching the route of drug administration to IV.³⁵ Within minutes after a bolus injection of furosemide in patients having congestive symptoms caused by acute ischemia or valvular heart disease, pulmonary artery pressure decreases and venous capacitance increases.⁴² This finding may help to explain why the patients with pulmonary edema show quick symptom relief even before achieving significant diuresis after getting IV furosemide.¹ The concern with rapid IV injection of a loop diuretic in high doses is the development of ototoxicity, especially in patients simultaneously receiving other ototoxic drugs, such as aminoglycosides.⁴³ Therefore, it is necessary to be cautious while considering a bolus infusion of high-dose furosemide and to avoid this complication, the maximum recommended rate of furosemide infusion is 4 mg/min.⁴³ It is recommended to infuse furosemide slowly when doses are higher than 80 mg to avoid an abrupt increase in peak serum concentration.⁴⁴

Continuous IV infusion of a loop diuretic may prove effective when other strategies to manage diuretic resistance have failed.¹ It has been found to be a safe and effective therapy in CHF patients who are refractory to high-dose oral as well as IV diuretic therapy, also it prevents postdiuretic salt retention completely.¹ Several studies have compared the efficacy of continuous infusion with intermittent IV bolus administration of a loop diuretic in patients with advanced HF.^45–47 The dose of furosemide for continuous infusions ranged from 3–200 mg/h, with most patients receiving 10–20 mg/h while bumetanide was administered as 0.5 mg bolus followed by a continuous infusion at 0.5 mg/h. It was reported that the similar daily dose of loop diuretics when administered as a continuous infusion caused excretion of higher amounts of urine and electrolytes. Additionally, the risk of ototoxicity was low as the maximal plasma concentration of furosemide was significantly lower.^45–48

Combination diuretic therapy (sequential nephron blockade)

When the previously discussed strategies fail to overcome diuretic resistance, combination diuretic therapy or sequential nephron inhibitors should be considered. Diuretics that act on PCT, such as carbonic anhydrase inhibitors should be avoided in patients with HF as they can cause metabolic acidosis. The efficacy of combination therapy with loop and thiazide diuretics has been studied well in CHF patients.^49–51 One of those studies showed that in 20 patients with severe HF (NYHA functional class III or IV) and impaired renal function having diuretic resistance to high-dose (at least 250 mg) of oral or IV furosemide, the addition of 25–100 mg of hydrochlorothiazide or any other thiazide diuretic proved to be very effective in the form of reduction in body weight (6.7 ± 3.3 kg per patient) and an increase in FE_Na+ (3.5 ± 3.2% to 11.5 ± 9.0%), as well as mean daily urine volume (1899 ± 958 ml to 3065 ± 925 ml).⁵⁰ Side effects were hypokalemia, hyponatremia, dehydration and worsening renal dysfunction.⁵⁰ Another study showed that addition of thiazide diuretics proved to be very effective in achieving diuresis in patients refractory to high-dose loop diuretics. This study compared the effects of bendrofluazide 10 mg and metolazone 10 mg in severe resistant HF (NYHA functional class III or IV unresponsive to IV loop diuretics for 48 h).⁵¹ Most patients reported significant improvement in their CHF functional class and also showed significant weight loss when a thiazide or a thiazide-like diuretic was added.⁵¹ Median (range) maximal weight loss noted as −5.05 (−11.3 to 1.6) kg after the addition of bendrofluazide, while −5.6 (−12.2 to 4.8) kg after the addition of metolazone.⁵¹ The area under the curves for body weight loss against time reflected no significant difference between these two diuretics.⁵¹ However, currently, there is no valid explanation available to consider metolazone as superior to any other thiazide diuretic.

Loop diuretics and thiazide diuretics can block the reabsorption of approximately 25% and 5–10% of filtered sodium, respectively.¹ Additionally, thiazide diuretics are ineffective as monotherapy in patients with advanced HF due to their weak natriuretic effect.¹ It has been demonstrated in experimental animal studies that a chronic increase in the sodium load in the DCT can lead to an increase in its sodium transport capacity.^30,31 This increase in transport capacity may be due to alterations in cellular ultrastructure of that segment. As previously discussed, in patients with resistance to loop diuretics due to chronic administration, a high amount of sodium load reaches early in the DCT and gets reabsorbed by altered cells present at that site.⁵² The thiazide diuretics usually act on the early DCT to prevent the reabsorption of sodium at that site. Therefore, combining loop diuretics with thiazide diuretics in CHF patients with diuretic resistance is an effective treatment option because it takes this synergistic pathophysiological mechanism into consideration. Metolazone, a thiazide-like diuretic acts on the DCT similarly to thiazide diuretics and thus also provides a synergistic effect when combined with loop diuretics. While considering this combination therapy, it is usually recommended to administer DCT-acting diuretics 30 min before administering loop diuretics to maximize the efficacy of this approach.^53,54 The thiazide and thiazide-like diuretics have long half-lives and therefore take longer to show their peak effects after oral administration. Hence, the rationale behind the recommendation of predosing these drugs with loop diuretics can be explained by their synergistic diuretic effect with loop diuretics and long half-lives.^53,54

Many thiazide-like diuretics have been evaluated for combination therapy with loop diuretics. All combinations have shown similar results overall and thus we can say that no thiazide-like diuretic is superior to another. In patients with advanced renal failure, metolazone has been considered superior to other thiazide-like diuretics, but other thiazide-like diuretics have also improved the response to loop diuretics in the similar type of patients. A recent small retrospective single-center cohort study while comparing the two most commonly used thiazide-like diuretics, namely oral metolazone and IV chlorothiazide as an add-on to loop diuretics, did not find any statistically significant differences in safety or efficacy.⁵⁵

Spironolactone is a mineralocorticoid antagonist and a potassium-sparing diuretic. It acts primarily by competitive binding to the aldosterone-dependent sodium-potassium exchange sites located in the DCT and collecting duct. A small study reported a successful response to the introduction of spironolactone in 13 of 16 HF patients resistant to a high-dose loop diuretic.⁵⁶ However, one patient developed reversible hyperkalemia and azotemia due to dehydration but it is worth mentioning that the dose of spironolactone used (100 mg/day) in the study was much higher than the average dose (25 mg/day) that produced a survival benefit in RALES study.⁵⁷ It is advisable to monitor hydration status and serum potassium levels when this high dose is administered. This high dose should be followed by a maintenance dose of 25 mg once all the excessive fluid gets removed.¹ The use of spironolactone for the treatment of diuretic resistance in CHF patients is generally not recommended as the evidence in favor of spironolactone is limited.¹

Conclusion

Fluid overload resistant to conventional-dose diuretic therapy is a common problem encountered in patients with CHF. From the above discussion, we conclude that the high prevalence of renal impairment in CHF patients is associated with a higher risk of diuretic resistance. In addition to renal impairment, poor diuretic absorption, drug–drug interactions, chronic administration of diuretics and high sodium intake are other potential factors behind the development of diuretic resistance in HF. Diuretic resistance can be overcome by managing it in a step by step manner, such as to exclude drug noncompliance, to increase the dose of diuretic, to change the route of administration from oral to IV, and finally to use combination diuretic therapy. This stepwise approach in treating diuretic resistance might lead to a more rapid alleviation of symptoms and potentially to a decreased length of stay in patients hospitalized for decompensated HF. Once the diuretic resistance has been treated successfully, the treatment of CHF should be optimized in order to reduce further morbidity and mortality.