Evaluation of the adult patient with hypokalemia

INTRODUCTION — Hypokalemia is a common clinical problem, the cause of which can usually be determined from the history (as with diuretic use, vomiting, or diarrhea). In some cases, however, the diagnosis is not readily apparent.

The diagnostic approach to the patient with hypokalemia will be reviewed here. There are two major components to the diagnostic evaluation: assessment of urinary potassium excretion to distinguish renal potassium losses (eg, diuretic therapy, primary aldosteronism) from other causes of hypokalemia, and assessment of acid-base status, since some causes of hypokalemia are associated with metabolic alkalosis or metabolic acidosis.

The causes, clinical manifestations, and treatment of hypokalemia are discussed separately:

●(See "Causes of hypokalemia in adults".)

●(See "Clinical manifestations and treatment of hypokalemia in adults".)

REGULATION OF POTASSIUM EXCRETION — The majority of the potassium filtered at the glomerulus is reabsorbed in the proximal tubule and loop of Henle, whereas most of the potassium that is excreted is derived from tubular secretion by the principal cells in the connecting tubule and cortical collecting tubule (figure 1) [1,2]. Aldosterone plays a central role in this process. An increase in plasma potassium stimulates the secretion of aldosterone, which then appropriately increases potassium secretion to return the plasma potassium to normal.

The renal tubular response to potassium depletion is twofold:

●Decreased potassium secretion by the principal cells – The effect of hypokalemia and potassium depletion on potassium secretion by the principal cells is mediated at least in part by the renin-angiotensin aldosterone system [3-5]. Hypokalemia suppresses aldosterone secretion [6], which results in reduced sodium reabsorption by principal cells. Decreased sodium reabsorption by these cells diminishes the electrogenic, lumen-negative stimulus for potassium secretion via apical potassium channels (figure 1). In addition, potassium depletion increases renin release and systemic angiotensin II formation, which in turn downregulates apical secretory potassium channels, thereby directly inhibiting potassium secretion [5]. Hypokalemia also activates the Na-Cl cotransporter (NCC) in distal convoluted tubule cells, which increases sodium reabsorption by the distal tubule and reduces delivery of sodium to principal cells, thereby blunting potassium secretion [7-9].

●Increased active potassium reabsorption by type A intercalated cells – Hypokalemia activates H-K-ATPase pumps in the apical membrane of type A intercalated cells, which are adjacent to the principal cells in the cortical collecting tubules (figure 2) [2,10,11]. These pumps reabsorb potassium and secrete hydrogen.

EVALUATION — The cause of hypokalemia is usually apparent from the history. As examples, patients may report vomiting, diarrhea, or the use of loop or thiazide diuretics. (See "Causes of hypokalemia in adults".)

However, the cause is occasionally uncertain. In such cases, there are two major components to the diagnostic evaluation:

●Assessment of urinary potassium excretion to distinguish renal potassium losses (eg, diuretic therapy, primary aldosteronism) from other causes of hypokalemia (eg, gastrointestinal losses, transcellular potassium shifts). (See 'Assessment of urinary potassium excretion' below.)

●Assessment of acid-base status, since some causes of hypokalemia are associated with metabolic alkalosis or metabolic acidosis. (See 'Assessment of acid-base status' below.)

Assessment of urinary potassium excretion is best accomplished by measuring potassium excretion in a 24-hour urine collection. Excretion of more than 30 mEq of potassium per day indicates inappropriate renal potassium loss. Measurement of the potassium and creatinine concentrations in a spot urine is an alternative if collection of a 24-hour urine is impractical. A spot urine potassium-to-creatinine ratio greater than 13 mEq/g creatinine (1.5 mEq/mmol) usually indicates inappropriate renal potassium loss. Causes of urinary potassium wasting are reviewed elsewhere. (See "Causes of hypokalemia in adults", section on 'Increased urinary losses'.)

After determining whether renal potassium wasting is present, assessment of acid–base status can further narrow the differential diagnosis. (See 'Assessment of acid-base status' below.)

It is critical to distinguish hypokalemia caused by redistribution, given the risk of rebound hyperkalemia that these patients may have after potassium repletion [12]. As an example, rebound hyperkalemia (serum potassium greater than 5 mEq/L) developed in 19 of the 30 patients with periodic paralysis after administration of a mean of 63 mEq of potassium chloride [13]. This effect is due both to reversal of the shift of extracellular potassium into the cells and to the administered potassium. (See "Hypokalemic periodic paralysis" and "Thyrotoxic periodic paralysis".)

Assessment of urinary potassium excretion — The best method for assessing renal potassium excretion is a 24-hour urine collection. However, the potassium concentration or, preferably, potassium-to-creatinine ratio on a spot urine are alternatives.

24-hour urine potassium — A 24-hour urine collection is the most accurate method to measure urinary potassium excretion. A normal individual can, in the presence of potassium depletion that is not due to urinary losses, lower urinary potassium excretion below 25 to 30 mEq per day on a 24-hour urine collection [14]. Higher values suggest at least a contribution from urinary potassium wasting.

In patients with severe hypokalemia, measurement of 24-hour potassium excretion is impractical because potassium repletion is urgently needed. In such patients, a spot urine potassium may be helpful as long as the urine sodium is greater than 30 to 40 mEq/L and the urine osmolality is greater than the plasma osmolality (suggesting that the patient is not polyuric). If the urine sodium is lower than 30 mEq/L and/or the urine osmolality is lower than the plasma osmolality, then the spot urine potassium concentration may be misleading. If so, comparable information can be obtained by calculating the urine potassium-to-creatinine ratio on a single urine specimen.

Spot urine potassium concentration — Random measurement of the urine potassium concentration can also be used. The minimum urine potassium concentration that can be achieved with hypokalemia is 5 to 15 mEq/L [14]. Some have suggested that extrarenal losses are present if the urine potassium concentration is less than 15 mEq/L [15], while substantially higher values suggest at least a component of potassium wasting.

However, random measurements may be misleading since the urine potassium concentration is determined by both the amount of potassium in the urine and the urine volume. As examples:

●A urine potassium concentration of less than 15 mEq/L can be seen in patients with renal potassium wasting in two settings: if the cause of potassium wasting has resolved (eg, the effect of the diuretic has worn off), or if the patient is polyuric, which may be due to impaired urinary concentrating ability and, possibly, increased thirst induced by the hypokalemia. (See "Hypokalemia-induced kidney dysfunction", section on 'Impaired urinary concentrating ability'.)

●A urine potassium concentration of 40 mEq/L, which seems to favor urinary potassium wasting, can also represent appropriate potassium conservation of 20 mEq per day if the urine volume is only 500 mL due to a low rate of water intake.

●In patients who are volume depleted, sodium and water delivery to the distal potassium secretory site may be substantially reduced. In this setting, the urine potassium concentration may be relatively high due to secondary hyperaldosteronism, but the urine volume and absolute amount of potassium excreted are relatively low [3]. Urinary sodium excretion should be above 30 to 40 mEq per day to avoid this problem.

Given these limitations, measurement of the urine potassium-to-creatinine ratio is preferred because it is not influenced by the urine volume. (See 'Urine potassium-to-creatinine ratio' below.)

Another approach to distinguishing between extrarenal and renal causes of potassium loss is to assess the response to the administration of potassium. Patients with previous extrarenal losses (eg, diarrhea) that have ceased will respond to potassium replacement with an increase in serum potassium that will eventually return to normal. A similar response will occur with diuretic-induced hypokalemia if the diuretic has been discontinued. By contrast, potassium replacement in patients with ongoing renal potassium losses will modestly raise the serum potassium, which will be followed by an increase in urinary potassium losses to match intake, and a lesser degree of hypokalemia will persist. In addition, the serum potassium will fall back to the baseline level if potassium therapy is discontinued. (See "Clinical manifestations and treatment of hypokalemia in adults", section on 'Ongoing losses and the steady state'.)

Urine potassium-to-creatinine ratio — Since creatinine is excreted at a near-constant rate, the urine potassium-to-creatinine ratio corrects for variations in urine volume. The urine potassium-to-creatinine ratio on a spot urine correlates with 24-hour urine potassium excretion [16-18].

The urine potassium-to-creatinine ratio is usually less than 13 mEq/g creatinine (1.5 mEq/mmol creatinine) when hypokalemia is caused by transcellular potassium shifts, gastrointestinal losses, or prior use of diuretics (ie, after the effect of the diuretic has dissipated) [13,19]. Higher values are seen with active renal potassium wasting (including diuretics, when the effect on the kidney is still present).

The utility of the urine potassium-to-creatinine ratio was evaluated in a study of 43 patients with severe hypokalemia (range, 1.5 to 2.6 mEq/L) associated with paralysis [13]. The urine potassium-to-creatinine ratio reliably distinguished between the 30 patients with hypokalemic periodic paralysis (whose hypokalemia was caused by an internal shift of potassium from the extracellular fluid into the cells) and the 13 patients with hypokalemia due mostly to renal potassium wasting (10 had renal tubular acidosis or Gitelman syndrome). The urine potassium-to-creatinine ratio was significantly lower in the patients with periodic paralysis (11 versus 36 mEq/g creatinine [1.3 versus 4.1 mEq/mmol creatinine]). The cutoff value was approximately 22 mEq/g creatinine (2.5 mEq/mmol). In addition, patients with periodic paralysis had normal acid-base balance, while patients without periodic paralysis often had metabolic alkalosis or metabolic acidosis. (See 'Assessment of acid-base status' below.)

One theoretical limitation of the urinary potassium-to-creatinine ratio is patients with low creatinine excretion, for example, due to very low muscle mass; the urinary transtubular potassium gradient (TTKG) may be a preferable test in such patients. (See 'Transtubular potassium gradient' below.)

The urine potassium-to-creatinine ratio may be elevated in many patients with gastrointestinal causes of hypokalemia (eg, bulimia or laxative abuse) [19]. These causes of hypokalemia are associated with metabolic alkalosis rather than metabolic acidosis, and the associated metabolic alkalosis and rise in plasma bicarbonate concentration increase the filtered bicarbonate load above its reabsorptive threshold. As a result, more sodium and bicarbonate are delivered to the distal potassium secretory site. In combination with a hypovolemia-induced increase in aldosterone release, the net effect is increased potassium secretion and potentially large urinary potassium losses. In such cases, the ratio of urinary sodium-to-chloride concentrations can be helpful. Patients with renal tubular disorders such as Gitelman syndrome and Bartter syndrome tend to have coupled urinary sodium and chloride losses (ie, ratio of 1:1). By contrast, patients with bulimia tend to have high, "uncoupled" urinary sodium-to-chloride ratios (approximately 5:1), whereas patients with laxative abuse exhibit low urinary sodium-to-chloride ratios (approximately 0.4) [19].  

Transtubular potassium gradient — Another widely used index of urinary potassium excretion is the "transtubular potassium gradient" (TTKG) [20], which is defined as follows:

In the absence of renal potassium wasting, the expected value of the TTKG is less than 3 to 4 in the presence of hypokalemia. Values higher than this suggest renal potassium wasting.

Some groups have suggested that the TTKG is unreliable since it fails to incorporate postulated effects of distal tubular urea reabsorption in potassium excretion; however, neither urea transporter knockout mice nor rats treated with a urea transporter inhibitor demonstrate abnormalities in potassium homeostasis [21]. Regardless of the theoretical limitations, the TTKG continues to be used as an index of urinary potassium wasting in clinical studies and in practice [13,19]. The author typically utilizes both the TTKG and the urine potassium-to-creatinine ratio in the bedside evaluation of patients with hypokalemia.

ASSESSMENT OF ACID-BASE STATUS — Once urinary potassium excretion is measured, the following diagnostic possibilities should be considered in the patient with hypokalemia of uncertain origin [1].

Assessment of acid-base status should include measurement of the serum bicarbonate, albumin, chloride, and in some cases, urine pH, urine electrolytes, and venous or arterial blood gas analysis.

●Metabolic acidosis with a low rate of urinary potassium excretion is, in an asymptomatic patient, suggestive of lower gastrointestinal losses due to laxative abuse or a villous adenoma [22]. (See "Acid-base and electrolyte abnormalities with diarrhea" and "Factitious diarrhea: Clinical manifestations, diagnosis, and management".)

●Metabolic acidosis with urinary potassium wasting is most often due to diabetic ketoacidosis or to type 1 (distal) or type 2 (proximal) renal tubular acidosis (RTA). (See "Overview and pathophysiology of renal tubular acidosis and the effect on potassium balance" and "Etiology and diagnosis of distal (type 1) and proximal (type 2) renal tubular acidosis".)

●Metabolic alkalosis with a low rate of urinary potassium excretion may be due to surreptitious vomiting (often in bulimic patients in an attempt to lose weight) or diuretic use if the urine collection is obtained after the diuretic effect has worn off. In addition, some patients with laxative abuse present with metabolic alkalosis, rather than the expected metabolic acidosis [23,24]. (See "Pathogenesis of metabolic alkalosis".)

●Metabolic alkalosis with urinary potassium wasting:

•In the setting of normal blood pressure, this is most often due to diuretic use, vomiting, or Gitelman or Bartter syndrome [25,26]. In this setting, measurement of the urine chloride concentration is often helpful [25,26], being normal (equal to intake) in Gitelman or Bartter syndrome, high or low with diuretics depending upon whether the diuretic is still acting, and low in vomiting at a time when urinary sodium and potassium excretion may be relatively high due to the need to maintain electroneutrality as some of the excess bicarbonate is being excreted. The last possibility can be determined at the bedside from the urine pH, which should be above 7 if significant bicarbonaturia is present. (See "Inherited hypokalemic salt-losing tubulopathies: Pathophysiology and overview of clinical manifestations".)

•In the setting of hypertension, this is suggestive of surreptitious diuretic therapy in a patient with underlying hypertension, renovascular disease, or one of the causes of primary mineralocorticoid excess. (See "Diagnosis of primary aldosteronism".)

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 5^th to 6^th 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 10^th to 12^th 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.)

●Basics topic (see "Patient education: Hypokalemia (The Basics)")

SUMMARY

●In most patients with hypokalemia, the cause is apparent from the history (eg, vomiting, diarrhea, diuretic therapy). However, the cause is not apparent from the history in some patients. In such cases, there are two major components to the diagnostic evaluation (see 'Evaluation' above):

•Assessment of urinary potassium excretion to distinguish renal potassium losses (eg, diuretic therapy, primary aldosteronism) from other causes of hypokalemia (eg, gastrointestinal losses, transcellular potassium shifts). (See 'Assessment of urinary potassium excretion' above.)

•Assessment of acid-base status, since some causes of hypokalemia are associated with metabolic alkalosis or metabolic acidosis. (See 'Assessment of acid-base status' above.)

●The best method for assessing renal potassium excretion is a 24-hour urine collection. However, the potassium concentration or, preferably, potassium-to-creatinine ratio on a spot urine are alternatives (see 'Assessment of urinary potassium excretion' above):

•A normal individual can, in the presence of potassium depletion that is not due to urinary losses, lower urinary potassium excretion below 25 to 30 mEq per day on a 24-hour urine collection. Higher values suggest at least a contribution from urinary potassium wasting. (See '24-hour urine potassium' above.)

•Since creatinine is excreted at a near-constant rate, the urine potassium-to-creatinine ratio corrects for variations in urine volume. The urine potassium-to-creatinine ratio is usually less than 13 mEq/g creatinine (1.5 mEq/mmol creatinine) when hypokalemia is caused by transcellular potassium shifts, gastrointestinal losses, previous use of diuretics, or poor dietary intake. Higher values are seen with renal potassium wasting. (See 'Urine potassium-to-creatinine ratio' above.)

●Acid-base status should be evaluated (see 'Assessment of acid-base status' above):

•Metabolic acidosis with a low rate of potassium excretion suggests lower gastrointestinal losses.

•Metabolic acidosis with renal potassium wasting may be seen with diabetic ketoacidosis and type 1 (distal) or type 2 (proximal) renal tubular acidosis (RTA).

•Metabolic alkalosis with a low rate of renal potassium excretion suggests surreptitious vomiting, diuretic use (when the urine collection is obtained after the diuretic effect has worn off), or, rarely, laxative abuse.

•Metabolic alkalosis with renal potassium wasting and a normal blood pressure is most often due to diuretic use, vomiting, or Gitelman or Bartter syndrome.

•Metabolic alkalosis with potassium wasting and hypertension occurs with primary mineralocorticoid excess or a disorder that mimics a mineralocorticoid excess state, renovascular disease, or the surreptitious ingestion of diuretics by a patient with underlying hypertension.

●Measurement of urine sodium and chloride can also provide important diagnostic clues (see 'Urine potassium-to-creatinine ratio' above):

•Patients with hypokalemia due to bulimia can have urine sodium concentrations five times the urine chloride concentration, whereas patients with hypokalemia due to laxative abuse may have urine sodium concentrations that are substantially lower than the urine chloride concentration.

•Patients with renal loss of potassium (Gitelman and Bartter syndrome, distal RTA, active diuretic therapy) tend to have urine sodium and chloride concentrations that are equivalent.