Does this patient have a condition treatable with diuretics?

What are diuretics?

Diuretics are commonly used drugs that act on different sites of the nephron to limit the reabsorption of sodium, resulting in a decrease in extracellular volume. Diuretics are used in the treatment of hypertension and in various edematous states. In addition, specific diuretics are often used for various non-hypertensive, non-edematous indications, as will be discussed below.

Irrespective of their mechanism of action, diuretics induce negative sodium balance until a new steady state is achieved, wherein sodium intake equals output to prevent further volume depletion. All diuretics other than aldosterone antagonists need to be secreted into the tubular lumen to exert their effects – an important consideration in dosing, especially in the setting of kidney disease.

What are the clinical uses of diuretics?


Continue Reading

One of the most common indications for diuretics, particularly thiazides, is essential hypertension. As per the report of the Eighth Joint National Committee (JNC 8): 2014 Evidence-Based Guideline for the Management of High Blood Pressure in Adults; thiazide type diuretics, not loop diuretics, should be considered initially for hypertension management, and are recommended as add-on therapy if not used first line.

There are many clinical trials supporting this recommendation, the largest of which was the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), published in 2002. ALLHAT showed that chlorthalidone, amlodipine and lisinopril were equally efficacious in preventing the primary endpoint of a combination of fatal coronary heart disease or nonfatal myocardial infarction. Chlorthalidone was beneficial in preventing some of the secondary endpoints. Specifically, there was less heart failure in the chlorthalidone group compared with the amlodipine and lisinopril groups, and fewer strokes compared to the lisinorpril group.

There have been no trials comparing different thiazides against one another with respect to “hard” outcomes such as mortality or cardiovascular events; however, there is data showing that chlorthalidone more effectively lowered blood pressure than hydrochlorothiazide. Further, chlorthalidone has been effective in all trials to date, whereas the results are mixed with hydrochlorothiazide, suggesting that chlorthalidone may be more efficacious and is the preferred agent. This may be an effect of the longer half-life of chlorthalidone, which permits once daily dosing. Studies have shown that twice daily hydrochlorothiazide is needed to achieve a similar degree of blood pressure lowering as that obtained with daily chlorthalidone.

Thiazides produce their antihypertensive effect by reducing total body sodium stores, and therefore extracellular volume until a new steady state is achieved. This process induces increased reabsorption of sodium in the distal tubules due to activation of the renin angiotensin system (RAS). For this reason, thiazides work well if combined with a medication that blocks the RAS such as an angiotensin converting enzyme inhibitor or an angiotensin receptor blocker.

As mentioned above, thiazides lose their efficacy in patients with renal insufficiency, particularly if it is advanced. In these patients, it may be necessary to substitute a loop diuretic.

Spironolactone has emerged as a useful therapy in treatment resistant hypertension (TRH – elevated blood pressure despite optimal dose of 3 antihypertensive agents of different classes, including a diuretic). The largest of several trials to demonstrate benefit of this agent was the PATHWAY-2 trial. Patients with treatment resistant hypertension were received spironolactone, doxazosin and bisoprolol in a 12-week crossover design. Spironolactone reduced blood pressure significantly more (by 8.70 mm Hg) than doxasozin (4.03 mg) and bisoprolol (4.48 mm Hg). Whether eplerenone, a weak mineralocorticoid antagonist with less side effects, will be as beneficial in TRH requires more study.

In addition to thiazides, other diuretics may be used in specific hypertension syndromes. Spironolactone and epleronone are the agents of choice in patients with hyperaldosteronism. Epleronone has a better side effect profile, as there is less gynecomastia and breast tenderness. The potassium-sparing diuretics amiloride and triamterene are used preferentially to treat hypertension associated with Liddle’s syndrome, a rare genetic disease caused by a gain of function mutation of the amiloride sensitive sodium channels (ENaC) in the collecting tubule.

Edematous states

The most common edematous states encountered clinically are congestive heart failure, cirrhosis and nephrotic syndrome. These will be considered individually. Although not discussed in detail, dietary sodium restriction is imperative for each of the disorders discussed below.

Heart failure

Diuretics are a mainstay in heart failure management due to their role in treating both pulmonary and peripheral edema. It should be noted however, that unlike treatment with beta blockers and angiotensin converting enzyme inhibitors, there are no data showing that diuretics improve heart failure mortality or prevent progression (with the exception of aldosterone antagonists – see below).

In using diuretics for heart failure, one must consider not only the class of diuretic, but also individual diuretics within the class, the dose, route of administration and frequency of dosing. Given that loop diuretics are the most potent of all the diuretic classes, they are typically the first agent used. Irrespective of agent used, proper attention to dosage and frequency of dosing is critical for maximal efficacy.

Loop diuretics need to reach tubular lumen to exert their effects. Many heart failure patients also have concomitant renal insufficiency rendering secretion of the drug into the lumen less effective, necessitating a higher dose. Although commonly used, once daily furosemide and bumetanide may be associated with worsening sodium balance. There is an initial natriuresis when these agents are given, but as the diuretic effect wanes, there is a period of enhanced renal sodium reabsorption such that the 24-hour sodium balance is at best no better, or possibly even worse than if no diuretic had been given.

Clinically, furosemide is the agent most often used, but there is some, albeit not conclusive, evidence that torsemide may be a better choice. Some studies have shown lower hospitalization rates and improved quality of life in patients treated with torsemide rather than furosemide, but properly designed placebo controlled trials are lacking. A feasible explanation for potential differences in efficacy is the difference in bioavailability between these agents.

Diuretic management for heart failure is aimed at relieving the vascular congestion that accompanies volume overload.

Outpatient treatment of heart failure typically starts with oral therapy but patients with refractory edema may require hospitalization and intravenous therapy. Loop diuretics may also induce venodilatation such that there is a decrease in cardiac filling pressures and symptomatic relief before the ensuing diuresis.

Thiazides are second line diuretics in heart failure. They are often added when the response to loop diuretics is unsatisfactory. Thiazides will block sodium reabsorption distal to the site of action of loop diuretics, leading to an increase in sodium excretion. Although rarely done, carbonic anhydrase inhibitors may be used in a similar manner.

The aldosterone antagonists spironolactone and epleronone deserve special mention in treatment of heart failure. As demonstrated in the RALES, EPHESUS and EMPHASIS-HF trials, they are the only diuretics shown to improve heart failure mortality. In addition to their independent indication for heart failure, aldosterone antagonists may help ameliorate potassium wasting induced by loop and thiazide diuretics. Amiloride and triamterene may also be used for potassium sparing effect but do not have the added mortality benefit of the aldosterone antagonists.


Ascites is one of the most common complication of cirrhosis. Sodium and water retention lead to ascitic fluid accumulation and diuretics are used to achieve a negative sodium balance.

In cirrhosis, much of the sodium retention is via the collecting tubule, likely related to hyperaldosteronism. For this reason, the aldosterone antagonists spironolactone and epleronone are the first line therapy for cirrhotic patient with ascites. Loop diuretics may be used as adjunctive therapy.

The recent clinical practice guideline from The European Association for the Study of the Liver suggests that the first occurrence of ascites be treated with an aldosterone antagonist alone, whereas recurrent episodes be treated with the addition of a loop diuretic. Patients with massive ascites typically require large volume paracentesis rather than diuretics.

Nephrotic syndrome

Mechanisms for edema in nephrotic syndrome include decreased oncotic pressure due to severe hypoalbuminemia with associated arterial underfilling, and more importantly enhanced renal sodium retention. Patients may either be volume contracted or volume expanded.

Although there may be some degree of arterial underfilling, and intravascular volume contraction, most patients with nephrotic syndrome tolerate diuretic therapy well and, along with sodium restriction, this is the mainstay of therapy. Loop diuretics are most often employed, although thiazides may be added if needed.

As diuretics are highly protein bound, their effectiveness may be diminished in nephrotic syndrome. Although unproven, it has been hypothesized that in severely hypoalbuminemic patients (Albumin <2 gm/L) diuretics infused along with salt poor albumin may increase drug delivery to the tubular lumen and enhance diuresis.

Other uses of diuretics


As most diuretics cause potassium wasting, these agents may be used to treat hyperkalemia. Intravenous administration of loop diuretics in a patient with adequate urine output may be preferred to Kayexylate in hyperkalemic patients due to an association with bowel perforation in the latter.

Hypercalciuria and nephrolithiasis

Due to their ability to decrease urinary calcium excretion, thiazide diuretics are often used in hypercalciuric patients with nephrolithiasis. Amiloride may be used as well, as it increases distal calcium reabsorption and increases potassium, which mitigates hypocitrauria.


Loop diuretics promote calcium excretion and therefore may be used in patients with hypercalcemia. Patients should be adequately volume repleted prior to starting a loop diuretic, as volume depletion may exacerbate the hypercalcemia.

Altitude sickness

Acute mountain sickness is caused by a rapid ascent to high altitude areas leading to severe tissue hypoxia. The exact mechanism of action still remains unclear. Inhibition of carbonic anhydrase in the proximal convoluted tubule promotes bicarbonaturia and metabolic acidosis. This is essentially a “speeding up” of the bicarbonaturia, which would normally occur in response to hypoxia induced hyperventilation and subsequent respiratory alkalosis. Induction of metabolic acidosis may also remove the hypocapnic inhibition of the full hyperventilatory response to hypoxia.

Nephrogenic diabetes insipidus

Thiazides have a paradoxical antidiuretic effect in nephrogenic diabetes insipidus (NDI). The volume depletion induced by thiazide use causes more salt and water to be reabsorbed in the proximal tubule, and a decrease in the amount delivered to more distal nephron segments. Less flow to the collecting duct (which in NDI is rendered impermeable to water due to lack of response to ADH) causes a decrease in urine output. Care must be taken in patients on lithium (a common cause of NDI), as thiazides may cause increase proximal lithium reabsorption, resulting in toxic levels.

Amiloride may also be used in lithium-induced NDI. Amiloride blocks ENaC, which is also lithium’s conduit of entry into the cell.

Syndrome of inappropriate antidiuretic hormone secretion (SIADH)

As will be discussed below, the so called aquaretics are used to effect a free water diuresis, increasing the serum sodium in patients with SIADH. An alternative strategy is the use of loop diuretics which increase free water excretion via the production of hypotonic urine. One caveat is that the sodium loss induced in euvolemic SIADH patients may precipitate hypovolemia. If needed, excess salt can be given in the diet, via oral supplements or intravenous administration (via a solution hypertonic to the urine so as not to worsen the hyponatremia) to mitigate this problem.

What tests to perform?


How should patients with diseases of volume regulation who receive diuretic therapy be managed?

How is sodium handled by the kidneys?

Diuretics are distinguished by their site of action. We will first present a brief overview of renal sodium reabsorption as this is key to understanding the mechanism and potential side effects of various diuretics.

The primary job of the kidneys is the reabsorption of the vast majority (>99%) of filtered sodium so as to defend extracellular volume. This reabsorption occurs in all nephron segments and most of the sodium transport mechanism is susceptible to inhibition by one of the classes of diuretics. Paracellular absorption of sodium does occur, but is not amenable to blockade by diuretics. and will not be considered further.

The driving force behind sodium reabsorption in the nephron is the basolateral sodium-potassium exchanger, present in all nephron segments, which transports sodium out of and potassium into the cell in a 3:2 ratio. The end result is a net negative intracellular compartment with a low sodium concentration. This provides a favorable electrochemical gradient for sodium movement from the tubular lumen into the cell.

The bulk of sodium (about 60-70%) is reabsorbed in the proximal tubule. Sodium transport in the proximal tubule is via two mechanisms: cotransport with another solute, including glucose, amino acids and phosphate, and by a sodium-hydrogen antiporter, which contributes to acid-base hemostasis as well.

The loop of Henle is responsible for approximately 25% of renal sodium reabsorption via the sodium-potassium-2-chloride transporter. The sodium-chloride cotransporters of the distal convoluted tubules reabsorb 5-10% of the filtered load, and final sodium balance is determined by the principal cells of the collecting duct, which are responsible for the final 2-3%. In the cortical collecting duct, the electroneutral sodium channel reabsorbs sodium, at the same time driving potassium and hydrogen excretion to maintain charge balance.

What are the practical issues in diuretic use?

Dosing and route of administration

Dosing of diuretics depends on the clinical indication, the pharmakokinetics of the particular diuretic in question, and the individual circumstances of the patient, such as presence of kidney dysfunction. Typical doses are summarized in Table I.

Table I.

Typical doses for diuretics.

Several clinical scenarios provide particular challenges in dosing. Kidney dysfunction limits the amount of drug reaching the tubular lumen. This may be an issue for all diuretics except for aldosterone antagonists. As mentioned, thiazides (except metolazone) lose their efficacy as GFR decreases to <40 ml/min. Loop diuretics may need a larger dose and more frequent dosing to be effective in kidney diseases. In cirrhosis and congestive heart failure there is diminished nephron response to loop diuretics, such that a strategy of more frequent dosing rather than increasing the dose may be more effective.

For outpatients, diuretics are typically administered orally. With more severe disease, however, practitioners may prefer to use intravenous dosing. Although not supported by clinical trials, the hypothesis is that bowel edema limits absorption of oral diuretics, rendering the intravenous route more effective. For loop diuretics, once the intravenous route is chosen, the dose may be given as a bolus or via continuous infusion.

A recent randomized clinical trial reported no difference in outcomes between those two strategies in the treatment of decompensated heart failure. Furosemide is the most common agent given as a continuous infusion. Typical doses range between 1 mg per hour and 15 mg per hour. Loop diuretics should not be “pushed,” as rapid infusion rates may increase the risk of ototoxicty.

The DOSE-AHF trial reported no difference in outcomes between those two strategies in the treatment of decompensated heart failure. The CARRESS-HF trial showed that stepped pharmacologic therapy was superior to mechanical ultrafiltration with respect to kidney function and adverse events at 96 hours, with similar effects on weight loss. In this study, diuretics were used to achieve a urine output of 3-5 liters per day. The stepped pharmacologic therapy is shown in Table II. Furosemide is the most common agent given as a continuous infusion. Typical doses range between 1 mg per hour and 15 mg per hour. Loop diuretics should not be “pushed”, as rapid infusion rates may increase the risk of ototoxicty.

Table II.

Stepped pharmacologic therapy for diuretic use in decompensated heart failure as per the CARRESS-HF trial.

Monitoring patients on diuretics

Patients on diuretics should be monitored for both efficacy and complications. Efficacy can be monitored by the following:

Daily weight: The American College of Cardiology/ American Heart Association Heart Failure Guidelines recommends monitoring weights in patients being treated for heart failure. Weight gain, or too rapid weight loss would both be of concern. Ideal would be to establish dry weight for each patient, and use that as a goal of therapy.

Urinary sodium excretion: Monitoring urinary sodium excretion may be helpful when diuretic therapy is not as effective as desired. A low urinary sodium indicates a lack of diuretic effect, whereas levels >100 mEq/day may suggest dietary noncompliance. A spot urinary Na/K ratio > 2.5 was shown to predict elevated 24-hour sodium excretion in edematous cirrhotic patients.

Diuretic resistance

Potential mechanisms of diuretic resistance

  • Insufficient dose

  • Decreased effectiveness from bowel wall edema

  • Decreased secretion into proximal convoluted tubule from renal insufficiency

  • “Braking effect”: refers to short term tolerance after the first dose from activation of renal angiotensin aldosterone system

  • Decreased protein binding

  • Hypertrophy of distal nephron

Treatment for diuretic resistance

  • Strict sodium and fluid restriction

  • Increasing the dose or change the frequency of dosing

  • Changing form oral to IV

  • Combining diuretics (Loop and thiazide)

  • Consider continuous infusion (although a recent clinical trial found no difference)

  • Mechanical ultrafiltration

What are the ongoing controversies on diuretic use?

Diuretics in acute kidney injury

Acute kidney injury (AKI) is common in hospitalized patients, particularly in the intensive care unit. Observational data show that oliguric AKI is associated with a higher mortality than non-oliguric AKI. Diuretics are often used to “convert” oliguric AKI to non-oliguric, but, studies have not demonstrated a benefit to this approach. Observational studies have associated diuretic use with higher mortality, but causation has not been shown.

The routine use of diuretics in AKI should be discouraged, and limited to the treatment of hyperkalemia, and volume overload causing respiratory compromise. There is also renewed interest and research in the use of natriuretic peptides, particularly atrial natriuretic peptide (ANP) in AKI. Several small, single center studies have shown that ANP may prevent AKI in the settings of cardiac surgery and contrast administration, but thus far data do not support its routine use.

Mannitol for increased intracranial pressure

Mannitol creates a high osmotic gradient in the renal tubules, especially in the proximal tubules and loop of Henle, causing water excretion that leads to decreased intracranial pressure. It has also been shown to increase cerebral perfusion pressure. Serum osmolarity and sodium should be monitored closely, as it can cause hypernatremia. Blood pressure should also be monitored as mannitol can cause hypotension which decreases cerebral perfusion pressure.

Caution is needed in patients with kidney dysfunction, as mannitol may be retained and lead to hyponatremia by movement of water out of cells. If used, the dose of mannitol is 1 g/kg, and then repeated every six hours as needed. Serum chemistries must be monitored closely, and the drug discontinued if the serum sodium is >150 mEq/dl or the serum osmolarity >320 mOsm/liter.

What happens to patients with diseases of volume regulation receiving diuretic therapy?

Mechanisms of diuretic action

Knowing the site of action where the various diuretics act can help predict both their efficacy and their acid-base and electrolyte side effects.

Proximal convoluted tubules

Carbonic anhydrase inhibitors

Carbonic anhydrase inhibitors such as acetazolamide act in the proximal tubule, leading to the loss of both sodium chloride and sodium bicarbonate. Normally sodium is exchanged for hydrogen in this segment. The secreted hydrogen then combines with filtered bicarbonate.

Carbonic anhydrase catalyzes the formation of carbonic acid (H2CO3), which then dissociates into water and carbon dioxide. The CO2 and water can diffuse across the cell membrane. The water dissociates into H+ and OH-. The H+ is secreted into the lumen with sodium as described above, whereas the CO2 combines with OH- to form HCO3- in a reaction catalyzed by intracellular carbonic anhydrase. The newly synthesized HCO3- is then returned to the systemic circulation via a basolateral sodium-bicarbonate transporter.

The net result is that inhibiting carbonic anhydrase limits the production of luminal CO2, resulting in less CO2 diffusing into the cell, less HCO3 being synthesized, and less H+ available for secretion. Without the secretion of hydrogen ion, neither sodium reabsorption via the Na/H antiporter or bicarbonate reclamation occurs. Thus, carbonic anhydrase inhibitors lead to renal wasting of sodium (and chloride) as well as bicarbonate.

Given that about 60-70% of the total filtered sodium load is reabsorbed in the proximal tubule, it is rational to assume that diuretics working in this segment, such as carbonic anhydrase inhibitors, would be the most potent diuretics available. This is not the case, for several reasons. First, sodium escaping reabsorption in the proximal tubule can be efficiently reabsorbed in the remainder of the nephron, particularly the loop of Henle. Second, the ensuing metabolic acidosis due to bicarbonate loss leads to less bicarbonate being filtered, and a less important role for the carbonic anhydrase mediated sodium reabsorption.

Diuretics inhibiting the other pathways of sodium transport in the proximal tubule (i.e., that which is coupled to phosphorus, glucose and amino acids) do not exist in clinical practice.

Carbonic anhydrase inhibitors are not commonly used. They are rarely used in edematous states, reserved mostly as an adjunct therapy for patients with severe edema, refractory to a combination of other diuretics, particularly those who have developed metabolic alkalosis. As discussed above, there are other, non-edematous conditions that call for the preferential use of carbonic anhydrase inhibitors.

Osmotic diuretics

Osmotic diuretics are substances which exert an osmotic force in the tubular lumen, particularly in the proximal tubule and loop of Henle, resulting in increased tubular fluid osmolarity and interference with water and sodium reabsorption. Osmotic diuretics tend to produce more water loss than sodium. Mannitol is a non-absorbable polyol which is freely filtered at the glomerulus and creates a high osmotic gradient in the renal tubules. Glucosuria in poorly controlled diabetics has a similar effect.

Osmotic diuretics are not used in edematous disorders, as initial administration leads to hyperosmolarity in the vascular space, with resultant movement of water from the intracellular to the extracellular space, and transient worsening of volume overload. Osmotic diuretics have been used in several non-edematous states, and this will be discussed below.

Loop diuretics

As their name implies, loop diuretics lead to sodium wasting in the loop of Henle. Specifically, they inhibit the sodium-potassium-2-chloride transporter in the ascending loop. This group includes furosemide, torsemide, bumetanide and ethacrynic acid. Ethacrynic acid is unique because it is non-sulfa containing and may be used in patients with sulfa allergies. Loop diuretics are the most potent diuretic available, and are often the first choice for edematous disorders. They can completely block sodium reabsorption in this segment, and lead to excretion of 20-25% of the filtered sodium load.

Loop diuretics differ with respect to bioavailability and half-life, which translates into varying doses and/or dosing frequencies. Furosemide and bumetanide, with half-lives of less than 2 hours, require either twice or thrice daily dosing, whereas torsemide lasts twice as long and may be dosed either daily or twice daily. In terms of bioavailability, oral bumetanide and torsemide have better absorption (>80%) compared to furosemide. On average 50% of administered furosemide is absorbed, however, this number varies widely (from 10-100%) between individuals.

Administration with food decreases the bioavailability of both furosemide and bumetanide but not that of torsemide. These compounds are protein bound so their effectiveness decreases with nephrotic syndrome. As normally half of the administered furosemide is excreted into the urine, with the rest metabolized in the kidneys, renal insufficiency may prolong duration of action.

Bumetanide and torsemide may be hepatically metabolized, and do not have a prolonged effect in the setting of renal insufficiency. As mentioned above, loop diuretics also have some venodilating properties, which may explain the rapid symptom relief provided in congestive heart failure, which may precede diuresis.

Loop diuretics are useful in several other clinical scenarios, as discussed above.

Thiazide diuretics

Thiazide diuretics act by blocking the sodium-chloride cotransporter in the distal convoluted tubule. Representative diuretics in this class include hydrochlorothiazide, chlorthalidone, metolazone, and indapamide. As this segment is responsible for 5-10% of sodium reabsorption, these agents are weaker diuretics compared to those acting in the loop of Henle. They are typically not used as a monotherapy in edematous disorders, but rather as add on therapy when a loop diuretic is not as effective as hoped.

The primary use of thiazide diuretics is in the treatment of hypertension, where they are considered a first line agent. Thiazides need to be excreted into the tubular lumen to exert their effect. The efficiency of this secretion diminishes with the advancing kidney dysfunction.

Most thiazides lose their effect at a glomerular filtration rate (GFR) of <40 ml/min, although metolazone is efficacious down to a GFR of about 20 ml/min. Half-lives in this group tend to be longer than those of loop diuretics, and most thiazides are dosed daily, although the different durations of action may have clinical ramifications (see above-hypertension). The other non-hypertensive, non-edematous clinical scenarios in which thiazides are useful were discussed above.

Potassium-sparing diuretics

Potassium sparing diuretics act on ENaC on the collecting tubules, either directly or via inhibition of aldosterone itself. This group includes amiloride and triamterene (ENaC inhibitors), and spironolactone and epleronone (aldosterone antagonists). Trimethoprim, particularly in high doses, can also function as a potassium sparing diuretic due to ENaC blockade.

As only 1-2% of sodium is reabsorbed in the collecting tubule, these agents are rather poor diuretics. An exception is in cirrhosis as the major mechanism of sodium retention in these patients is hyperaldosteronism. Spironolactone and epleronone are highly efficacious in this setting. Cirrhosis notwithstanding, the most common use potassium sparing diuretics is in conjunction with other diuretics to mitigate hypokalemia.

The pharmacokinetics of potassium sparing diuretics allow for once daily dosing. The aldosterone antagonists are the only diuretics that do not need to reach the tubular lumen to exert their effect.


Aquaretics are not diuretics in the true sense, in that rather than leading to salt and water excretion, they selectively inhibit water reabsorption. These agents are inhibitors of argenine vasopression (AVP). In the kidneys, AVP acts on vasopressin receptors in the collecting tubule (V2 receptors) leading to insertion of aquaporin channels in their apical membrane, thus promoting water reabsorption.

This group includes conivaptan, tolvaptan and satavaptan. Tolvaptan and satavaptan are available in oral form, and they act only on V2 receptors, whereas conivaptan has blocks both V1 (those which cause vasoconstriction) and V2 receptors. These agents are useful in treating hyponatremia in syndrome of inappropriate antidiuretic hormone (SIADH) and CHF, although clinical trials have not shown any survival advantage.

The Federal Drug Administration has instituted several warnings regarding tolvaptan. First, due to the of hyponatremia, this medication should be started as an inpatient with close monitoring. Second, clinical trials have demonstrated higher degrees of liver toxicity, as evidenced by serum alanine aminotransferase (ALT) elevation and clinically significant increases in serum total bilirubin. For this reason, duration of therapy should not exceed 30 days.

B-natriuretic peptide

Nesiritide is recombinant form of beta natriuretic peptide produce by the ventricle. It binds to smooth muscle cells increasing cyclic GMP, which causes smooth muscle relaxation leading to vasodilation and natriuresis. It is indicated only for hospitalized acute decompensated heart failure patients with dyspnea at rest as it is only available in intravenous form. Multiple trials, including the largest trial ASCEND-HF, have not shown any benefit in terms rehospitalization and mortality. It is not a first line agent nor a substitute for diuretics.

Side effects of diuretics

Diuretic use may be associated with various side effects. As they promote renal sodium loss, all diuretics may lead to volume depletion and a decrease in glomerular filtration rate. Other side effects depend on the diuretic in question. A convenient classification scheme is to divide these into electrolyte/acid-base abnormalities and other side effects.

Electrolyte and acid-base side effects

Potassium disorders

Hypokalemia is a consequence of many diuretics. Renal handling of potassium is complex, but with respect to diuretic use, most important is the tubular secretion occurring in the cortical collecting duct. In this segment, one K+ is secreted into the lumen for each Na+ that is reabsorbed by ENaC. As the absorption of Na+ via ENaC is flow dependent, so is the K+ secretion. Thus, carbonic anhydrase inhibitors, osmotic diuretics, loop diuretics and thiazides, all of which increase distal Na+ delivery, will enhance K+ secretion, predisposing to hypokalemia.

In addition, the aforementioned volume depletion will cause up regulation of the renal angiotensin aldosterone system, resulting in an increased aldosterone concentration. Aldosterone stimulates Na+ reabsorption via ENaC, leading to further K+ secretion.

The degree of hypokalemia tends to be dose dependent, and may have serious consequences such as ventricular arrhythmias. Thus, diuretic induced hypokalemia should always be treated, either with K+ supplementation or by concurrent use of potassium sparing diuretics.

Not surprisingly, hyperkalemia may complicate use of a potassium sparing diuretic. This is particularly problematic in patients with renal insufficiency and those on other medications that interfere with renin angiotensin system. The potential for hyperkalemia with these agents is illustrated by the effect of widespread use of spironolactone in heart failure after the RALES trial. There was a 4-times higher incidence of hyperkalemia related hospitalizations and greater than 6-fold increase in mortality among these patients. Any potassium supplements should be discontinued prior to beginning these medications, and serum chemistries should be monitored frequently.

Sodium disorders

Hyponatremia is a common side effect of thiazide diuretics. Hyponatremia may occur either due to defect in free water excretion, or by urinary loss of sodium and potassium in excess of water. The kidneys are normally quite efficient in water excretion, maintaining the serum osmolarity in a narrow range. The bulk of water handling is done via proximal tubule, where water is reabsorbed, iso-osmotically, along with sodium.

The complex solute and water handling in the loop of Henle results in the generation of an osmotic gradient both in the tubules, and the medullary interstitium. Tubular fluid passing from the loop of Henle into the water impermeable distal convoluted tubule undergoes further dilution, as Na-Cl is reabsorbed in this segment, but water is not. This results in a maximally dilute tubular fluid at the end of the distal tubule.

The final urine osmolarity is determined by the cortical collecting duct. In times of water deprivation, plasma hyperosmolarity stimulates the production of antidiuretic hormone.

ADH serves to increase aquaporin insertion into the apical membrane of the tubular cell, such that water can move down the osmotic gradient between the dilute tubular lumen, and more concentrated interstitium. In times of water surfeit, ADH is maximally suppressed, resulting in the excretion of a dilute urine.

As thiazides inhibit sodium chloride reabsorption in the distal convoluted tubule, they prevent the formation of a maximally dilute tubular fluid. Further, thiazide induced hypovolemia results in ADH secretion, and increased water reabsorption along the collecting duct as described above. The final result is the retention of water, despite progressive systemic hypoosmolarity.

While all other diuretics may cause hypovolemia and ADH secretion, none of them aside from thiazide affect the dilution of urine by the distal convoluted tubule. In addition to sparing urinary dilution, loop diuretics, by inhibiting Na+ and Cl- reabsorption in the loop of Henle, attenuate the osmotic gradient created by the countercurrent mechanism, blunting the movement of water out of the tubular lumen despite the presence of ADH. In clinical practice, thiazides are over 10 times as likely to cause hyponatremia as compared to other diuretics.

Due to their effect on the countercurrent mechanism, and the resultant decrease in water reabsorption in the collecting duct, loop diuretics tend to produce a urine hypoosmotic to plasma, and may lead to hypernatremia, particularly during times of reduced water intake. This be treated by replacement of free water, while still preserving the desired natriuretic effect.

By inhibiting the vasopressin receptors, and thus the action of ADH, aquaretics may lead to hypernatremia if not properly monitored.

Magnesium and calcium disorders

Both loop and thiazide diuretics may lead to magnesium depletion. In the loop of Henle, magnesium absorption is paracellular and coupled with sodium reabsorption. Although seemingly electroneutral, the sodium-potassium-2-chloride transporter back leaks some of the reabsorbed potassium and leads to lumen positivity, promoting magnesium reabsorption. The mechanism of thiazide induced hypomagnesemia is unclear. Hypomagnesemia should be suspected when diuretic induced hypokalemia is resistant to potassium supplentation. Potassium sparing diuretics may also mitigate the loss of magnesium in the urine.

Loop and thiazides diuretics have opposite effects on renal calcium handling. In the loop, calcium reabsorption is coupled to that of sodium, similar to magnesium. The resultant urinary calcium loss typically does not lead to hypocalcemia due to efficient hormonal regulation of serum calcium. Thiazides, conversely, lead to a decrease in renal calcium excretion, as sodium and calcium reabsorption are uncoupled in the distal convoluted tubule. If hypercalcemia occurs at all, it is typically mild, due to concomitant decrease in the secretion of parathyroid hormone.

Acid-base disorders

Carbonic anhydrase inhibitors and potassium sparing diuretics both induce a non-anion gap metabolic acidosis, although by different mechanisms. As previously described, carbonic anhydrase inhibitors lead to urinary bicarbonate loss, with a resultant acidosis. In the collecting duct, inhibition of sodium reabsorption by ENaC attenuates the lumen negative voltage needed not only for potassium excretion, but for hydrogen excretion via the intercalated cells. Hyperkalemia may exacerbate the acidosis due to exchange of extracellular K+ for intracellular H+.

Loop and thiazide diuretics are associated with metabolic alkalosis. This is likely due to volume contraction, as well as loss of hydrogen in the collecting tubule due to hypovolemia induced secondary hyperaldosteronism.

Other side effects of diuretics

Some of the more common complications associated with diuretics include:

Hyperuricemia: This is typically seen with both loop and thiazide diuretics. The increase in serum uric acid is usually mild, but gout may develop necessitating drug discontinuation.

Hyperglycemia: A modest increase in blood sugar may be seen with thiazides, and, to a lesser degree, loop diuretics. The clinical significance of this is not known. It has been suggested that hypokalemia, rather than the diuretic itself, may be the culprit.

Hyperlipidemia: Thiazides have been associated with a small increase in total cholesterol, LDL, and triglycerides. Some data also implicate loop diuretics in the development of hyperlipidemia as well. As above, the clinical significance of this small increase is not known.

Ototoxicity: Ototoxicity is a dose dependent complication of loop diuretics, and is typically seen with rapid, intravenous infusion. Ototoxicity is much more common with ethacrynic acid compared to other loop diuretics. Consequently, this agent is not commonly used.

Nephrocalcinosis and nephrolithiasis: Triamterene is rarely found as a component of kidney stones. Carbonic anhydrase inhibitors may predispose to calcium phosphate stones due to decreased solubility of calcium phosphate on alkaline urine.

Gynecomastia and breast tenderness: This is seen with spironolactone, due to its estrogenic effect. This does not seem to be an issue with epleronone.

How to utilize team care?


Are there clinical practice guidelines to inform decision making?


Other considerations


What is the evidence?

“Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to Pravastatin vs usual care”. JAMA. vol. 288. December 18, 2002. pp. 2998-3007. (The ALLHAT trial was a large scale, double blind, randomized controlled trial comparing the efficacy of a thiazide diuretic, an angiotensin converting enzyme inhibitor [ACE-I], a calcium channel blocker [CCB] and an alpha adrenergic antagonist in treating hypertension, and preventing cardiovascular events. The thiazide [chlorthalidone] was as effective as ACE-I’s and CCB’s in preventing fatal coronary events and fatal and non-fatal myocardial infarction, whereas the alpha blocker arm was stopped early due to increased risk. The conclusion from this trial was that no other agents were superior to thiazides [specifically chlorthalidone] in treating hypertension, and preventing hypertension related cardiovascular events. The cost effectiveness of chlorthalidone make this a good first line agent in the treatment of essential hypertension.)

James, PA, Oparil, S, Carter, BL, Cushman, WC. “2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA”. 2014 Feb. vol. 311. 5. pp. 507-20. (The JNC 8 report provides an update on hypertension management. The important messages of the report with respect to diuretics is that thiazides remain a viable option for the initial management of hypertension, or as add on therapy if further agents are needed.)

Greenberg, A. “Diuretic complications”. . vol. 319. 2000. pp. 10

Sarafidis, PA, Georgianos, PI, Lasaridis, AN. “Diuretics in clinical practice. Part I: mechanisms of action, pharmacological effects and clinical indications of diuretic compounds”. . vol. 9. 2010. pp. 243-257.

Sarafidis, PA, Georgianos, PI, Lasaridis, AN. “Diuretics in clinical practice. Part II: electrolyte and acid-base disorders complicating diuretic therapy”. . vol. 9. 2010. pp. 259-273. (These two review articles by the same authors are an excellent summary about the difference classes of diuretics, their mechanisms of action, pharmokinetics, pharmacodynamics, clinical indications and complications. This review succinctly describes both the metabolic and non-metabolic complications of all classes of diuretics via a literature review and review of large-scale antihypertensive trials.)

Pitt, B, Zannad, F, Remme, WJ, Cody, R. “The effect of spironolactone on morbidity and mortality in patients with severe heart failure”. . vol. 341. 1999. pp. 709-717. (The Randomized Aldosterone Evaluation Study [RALES] tested the effect of adding spironolactone [an aldosterone antagonist] to loop diuretics and ACE inhibitors in patients with severe congestive heart failure. Both mortality and hospitalization were 30% lower in the spironolactone group. Gynecomastia occurred in 10% of spironolactone treated patients [compared with 1% of placebo treated].)

Pitt, B, Williams, G, Remme, W, Martinez, F. “The EPHESUS Trial: Eplerenone in patients with heart failure due to systolic dysfunction complicating acute myocardial infarction”. . vol. 15. 2001. pp. 79-87. (This randomized controlled trial tested the hypothesis demonstrated that addition of eplerenone [an aldosterone antagonist] to optimal medical therapy in patients with a myocardial infarction and left ventricular dysfunction and heart failure led to decrease morbidity and mortality. The trial also showed a predilection to hyperkalemia in the eplerenone group.)

Zannad, F, McMurray, JJ, Krum, H, van Veldhuisen, DJ. “Eplerenone in patients with systolic heart failure and mild symptoms”. . vol. 364. 2011. pp. 11-21. (This randomized controlled trial showed that in mild heart failure, the addition of eplerenone to optimal medical management resulted in a 37% decrease in cardiovascular mortality and hospitalization, and a 24% decrease in all-cause mortality, as well as decreases in all-cause and heart failure hospitalization. There was also more hyperkalemia in the eplerenone treated group. Taken together, the RALES, EPHESUS and EMPHASIS-HF trials establish the use of aldosterone antagonists in moderate to severe heart failure, heart failure after a myocardial infarction, and mild heart failure respectively.)

Felker, GM, Lee, KL, Bull, DA, Redfield, MM. “NHLBI Heart Failure Clinical Research Network. N Engl J Med”. 2011 Mar. vol. 364. 3. pp. 797-805. (This study, the DOSE-AHF study, found that there was no difference in the primary end point of global assessment of symptoms or kidney function when diuretics were given as a bolus compared with a continuous infusion strategy.)

Bart, BA, Goldsmith, SR, Lee, KL, Givertz, MM. “Ultrafiltration in decompensated heart failure with cardiorenal syndrome. Heart Failure Clinical Research Network. N Engl J Med”. 2012 Dec. vol. 367. 13. pp. 2296-304. (This randomized controlled trial, known as the CARRESS-HF study showed that a stepped pharmacologic strategy leads to superior renal preservation, less adverse events and a similar weight loss compared to mechanical ultrafiltration. The stepped approach to diuretics in heart failure has been widely adopted in clinical practice.)

Williams, B, MacDonald, TM, Morant, S, Webb, DJ. “British Hypertension Society's PATHWAY Studies Group: Spironolactone versus placebo, bisoprolol, and doxazosin to determine the optimal treatment for drug-resistant hypertension (PATHWAY-2): a randomised, double-blind, crossover trial”. Lancet. vol. 386. 2015 Nov 21. pp. 2059-68. (The PATHWAY-2 study compared spironolactone with doxazosin and bisporol in treatment resistant hypertension. Spironolactone was associated with a larger blood pressure reduction than either of the other agent. This trial, along with other, smaller studies, suggests preferential use of spironolactone in this group of patients.)