| |||||||||||||||||||||||
Hyperkalemia develops in 20 to 53 percent of patients with AIDS while they are receiving high doses of trimethoprim in combination with sulfamethoxazole or dapsone for the treatment of Pneumocystis carinii pneumonia8. We recently encountered a patient with AIDS and P. carinii pneumonia in whom hyperkalemia abruptly developed during therapy with trimethoprim and sulfamethoxazole. The temporal association of hyperkalemia and an inappropriately low urinary potassium concentration with the administration of these drugs suggested that they may interfere with renal potassium excretion. Trimethoprim is a heterocyclic weak base that is structurally related to the potassium-sparing diuretic agents amiloride and triamterene. These diuretic agents block the reabsorption of sodium in the distal nephron and secondarily impair renal potassium secretion9,10,11. We used a cell line (A6) derived from the nephron of the toad12,13,14 to examine whether trimethoprim inhibits sodium transport. The results demonstrate that trimethoprim is a sodium-channel inhibitor and therefore functions as a potassium-sparing diuretic agent.
Case Report
A 53-year-old man was admitted to the Veterans Affairs Medical Center with a productive cough and dyspnea. Physical examination revealed a temperature of 37.2 °C, a respiratory rate of 26 per minute, a blood pressure of 90/60 mm Hg measured with the patient supine, and a pulse rate of 108 beats per minute. The blood pressure fell to 80/50 mm Hg and the pulse rate increased to 112 beats per minute when the patient stood up. The patient also had oropharyngeal thrush and rales at the bases of both lungs. A chest film revealed infiltrates in the right upper and lower lobes. The serum sodium concentration was 134 mmol per liter, potassium 4.6 mmol per liter, chloride 97 mmol per liter, bicarbonate 21 mmol per liter, blood urea nitrogen 29 mg per deciliter (10.4 mmol per liter), and serum creatinine 2.4 mg per deciliter (212 µmol per liter). The pH of arterial blood was 7.59, the partial pressure of carbon dioxide 22 mm Hg, and the partial pressure of oxygen 54 mm Hg. Blood and sputum cultures were negative. The patient was given trimethoprim (1.9 g per day), sulfamethoxazole (9.6 g per day), and prednisone (80 mg per day), with rapid clinical improvement. P. carinii was subsequently identified in sputum, and his serum tested positive for human immunodeficiency virus antibodies.
During the first five days of therapy with trimethoprim-sulfamethoxazole the serum potassium concentration rose slightly and then stabilized at 5.3 to 5.5 mmol per liter. The urinary potassium and sodium concentrations were 5 mmol per liter and 52 mmol per liter, respectively. On the ninth day of trimethoprim-sulfamethoxazole therapy, the serum potassium concentration was 7.9 mmol per liter, blood urea nitrogen 47 mg per deciliter (16.8 mmol per liter), and serum creatinine 3.1 mg per deciliter (274 µmol per liter). The blood pressure measured with the patient supine was 112/78 mm Hg, and the pulse rate was 100 beats per minute; these measurements were 84/70 mm Hg and 120 beats per minute, respectively, after he stood up. The urinary potassium and sodium concentrations were 14 mmol per liter and 56 mmol per liter, respectively. The calculated transtubular potassium-concentration gradient was 1.9, indicating impaired renal potassium secretion15,16,17,18. Isotonic saline was administered intravenously, and sodium polystyrene sulfonate orally, with rapid correction of the hypovolemia, azotemia, and hyperkalemia.
The patient's treatment was changed from prednisone to dexamethasone (1.5 mg daily). After the intravenous administration of 250 µg of cosyntropin for eight hours daily for three days, the patient's serum cortisol concentration was 38 µg per deciliter (1049 nmol per liter), confirming normal cortisol production. The administration of up to 0.3 mg per day of fludrocortisone acetate failed to raise the urinary potassium concentration above 17 mmol per liter, to increase the transtubular potassium-concentration gradient above 2.8, or to raise the urinary sodium concentration to 80 mmol per liter or more.
The patient's serum potassium concentration increased to a peak of 5.7 mmol per liter during the subsequent week, and the urinary potassium concentrations ranged from 7 to 20 mmol per liter. After trimethoprim and sulfamethoxazole were discontinued, the serum potassium concentration stabilized at 4.8 mmol per liter, and the urinary potassium concentration increased to 38 to 45 mmol per liter. The plasma renin activity and plasma aldosterone concentrations were normal in response to furosemide-induced volume depletion. The plasma renin activity with the patient supine was 5.0 ng per milliliter per hour (1.4 ng per liter • second; normal range, 1.5 to 4.5 ng per milliliter per hour [0.41 to 1.25 ng per liter • second]); this value increased to 9.1 ng per milliliter per hour (2.5 ng per liter • second; normal range, 6.0 to 11.0 ng per milliliter per hour [1.7 to 3.1 ng per liter • second]) in the upright position. The plasma aldosterone concentration was 18 ng per deciliter (499 pmol per liter; normal range, 5 to 20 ng per deciliter [139 to 555 pmol per liter]) with the patient supine and 36 ng per deciliter (999 pmol per liter; normal range, 18 to 34 ng per deciliter [499 to 943 pmol per liter]) in the upright position. Ten days after his discharge, the patient's serum potassium concentration was 4.5 mmol per liter, serum creatinine 1.6 mg per deciliter (141 µmol per liter), and urinary potassium 56 mmol per liter; the transtubular potassium-concentration gradient was 8.9.
Methods
Materials
Trimethoprim, sulfamethoxazole, and N-acetylsulfamethoxazole were gifts from Roche. Trimethoprim was also purchased from Sigma Chemical. Amiloride was a gift from Merck.
Electrophysiologic Measurements
A subclone (2F3) of the A6 cell line derived from Xenopus laevis kidney was used between passages 91 and 96. Cells were seeded on collagen-coated polycarbonate filters (Costar) at a density of 0.5 to 1.0 x 106 cells per square centimeter and were maintained at 28 °C in an incubator with 5 percent carbon dioxide with amphibian medium supplemented with 5 percent fetal-calf serum for 7 to 21 days, as described elsewhere19. The medium was supplemented with aldosterone (10-6 M) for more than 18 hours before the electrophysiologic measurements, in order to increase the rate of amiloride-sensitive sodium transport. The cell monolayers were transferred to a modified Ussing chamber and bathed in a modified Ringer's solution containing the following ingredients per liter of solution: sodium chloride, 100.0 mmol; potassium chloride, 4.0 mmol; sodium bicarbonate, 2.5 mmol; dibasic potassium phosphate, 1.0 mmol; calcium chloride, 1.0 mmol; glucose, 11.0 mmol; and HEPES buffer, 10 mmol; the pH was 7.4. Electrical measurements were performed with a DVC-1000 voltage clamp (World Precision Instruments). The short-circuit current was allowed to stabilize before the addition of drug or vehicle. All test compounds were dissolved in dimethyl sulfoxide and added to the luminal compartment. The amiloride-sensitive component of the short-circuit current was determined by adding amiloride (10-5 M) to the luminal solution at the end of each experiment20. All results are expressed as means ±SE.
Transtubular Potassium-Concentration Gradient
The transtubular potassium-concentration gradient is a measurement of the capacity of the kidney to increase the concentration of potassium in the urine through potassium secretion by the distal convoluted tubule and the cortical collecting tubule. The gradient is calculated by an equation in which the numerator is the urinary potassium concentration divided by the quotient of the urinary osmolality divided by the plasma osmolality and in which the denominator is the serum potassium concentration.
The numerator of the equation corrects for the effect of increasing urinary potassium concentration through osmolal concentration of the urine in the medullary collecting duct15,16,17,18. The transtubular potassium-concentration gradient is an indirect measure of the capacity of the kidney to secrete potassium, and it should be higher than 6 in subjects receiving physiologic doses of mineralocorticoids16,17. The measurement of the gradient assumes minimal reabsorption of potassium in the collecting ducts and adequate delivery of sodium to the distal tubules. It cannot be used to estimate the kidney's capacity to secrete potassium if the urine is hypotonic with respect to plasma.
Results
The amiloride-sensitive component of short-circuit current is a measure of net sodium transport across the A6-cell monolayer12,13. The addition of trimethoprim to the luminal compartment inhibited short-circuit current in a dose-dependent manner (Figure 1). The effect of trimethoprim was rapid, and a concentration of 10-3 M inhibited the amiloride-sensitive short-circuit current by a mean (±SE) of 83 ±3 percent (Figure 2A). The concentration of trimethoprim required to inhibit 50 percent of the amiloride-sensitive short-circuit current (IC50) was 1.2 x 10-4 M. Amiloride, by comparison, inhibits sodium transport in A6 cells with an IC50 of 3.5 x 10-7 M20. The addition to the luminal solution of sulfamethoxazole (10-3 M, in three experiments) or its principal metabolite, N-acetylsulfamethoxazole21 (2.5 x 10-4 M, in two experiments), did not inhibit short-circuit current (data not shown). N-acetylsulfamethoxazole was insoluble at a concentration of 5 x 10-4 M.
|
|
Discussion
Sodium reabsorption by cells in the distal nephron and by A6 cells is mediated by an amiloride-sensitive sodium channel in the luminal membrane and a sodium-potassium-ATPase in the basolateral (serosal) membrane12,13,19. The amiloride-sensitive component of the short-circuit current in A6 cells is a simple and reliable measure of net transepithelial sodium transport12,13. Our results suggest that trimethoprim reversibly inhibits sodium transport across A6 cells by blocking sodium channels. The reversal of trimethoprim-induced inhibition of short-circuit current by amphotericin B argues against an inhibitory effect of trimethoprim on the sodium-potassium-ATPase or on cellular metabolism. Trimethoprim also blocks sodium channels in frog-skin epithelium23. These findings are consistent with the notion that trimethoprim, amiloride, and triamterene, which are structurally related heterocyclic weak bases, act in a similar manner. It is likely that, as in the case of amiloride, only protonated trimethoprim blocks sodium channels12,20. For trimethoprim, the negative log of the dissociation constant (pKa) is 7.224. In our experiments, only 41 percent of trimethoprim was in the cationic form at a pH of 7.4, and the concentration of protonated trimethoprim that blocked 50 percent of the amiloride-sensitive short-circuit current in A6 cells was 5 x 10-5 M. Acidification of the urine results in a larger fraction of trimethoprim that is protonated.
Patients receiving 320 mg of oral trimethoprim per day in two doses have mean urinary concentrations (approximately 2 x 10-4 M) similar to the IC for the inhibition of sodium transport in A6 cells25. Patients with AIDS and P. carinii pneumonia receive 1 to 2 g of intravenous trimethoprim per day in four divided doses8. Urinary concentrations of trimethoprim above the IC reported in this study are probably reached in the distal nephrons of these patients. Recurrent hyperkalemia with impaired kaliuresis developed in our patient while he received large doses of trimethoprim. His urinary potassium concentration was low, his transtubular potassium-concentration gradient was less than 6,15,16,17,18 and neither responded to large doses of fludrocortisone. The discontinuation of trimethoprim resulted in a rapid correction of hyperkalemia and in increases in the urinary potassium concentration and the transtubular potassium-concentration gradient, suggesting that the potassium-sparing diuretic property of trimethoprim contributed to the hyperkalemia.
The antikaliuretic properties of amiloride, and presumably those of trimethoprim, result from the inhibition of luminal sodium channels9,10,22. Potassium secretion in the distal nephron is mediated by a sodium-potassium-ATPase in the basolateral membrane and by potassium channels in the luminal membrane. The inhibition of the entry of sodium hyperpolarizes the luminal membrane and secondarily blocks potassium secretion9,10.
Although hyperkalemia is a common complication of treatment with potassium-sparing diuretic agents,26,27,28 it has not been associated with the usual doses of trimethoprim. The large doses of trimethoprim used in the treatment of P. carinii pneumonia may explain the high incidence of hyperkalemia in patients with AIDS receiving this drug8. The only patient without AIDS in whom hyperkalemia has been observed in association with trimethoprim also received high doses of that drug29. Other factors may contribute to the high incidence of hyperkalemia in patients with AIDS. Impaired renal function is common in AIDS1,2 and probably played a part in the development of hyperkalemia in our patient. Although hyperkalemia due to isolated hyporeninemic hypoaldosteronism has been reported in four patients with AIDS, they were also receiving trimethoprim3. Studies of adrenal function in patients with AIDS have revealed remarkably normal function of the renin-angiotensin-aldosterone axis, as compared with the often diminished secretory reserve of cortisol4,5.
Trimethoprim has the potential to induce hyperkalemia by interfering with urinary potassium excretion. Serum potassium concentrations should be monitored in patients with P. carinii pneumonia and AIDS who receive high doses of trimethoprim, as well as in patients with renal insufficiency who receive this drug.
Supported by a grant (DK-07006) from the National Institutes of Health and by grants from the Department of Veterans Affairs and the Southeastern Pennsylvania Chapter of the American Heart Association. Dr. Choi is the recipient of a postdoctoral fellowship award from the National Kidney Foundation. Dr. Kleyman is an Established Investigator of the American Heart Association.
We are indebted to Drs. R. Grossman and J. Getsy for their critical reviews of the manuscript, and to Ms. Susan Adamson for her assistance in the preparation of the manuscript.
Source Information
From the Philadelphia Veterans Affairs Medical Center (M.J.C., P.C.F., D.D., H.S., T.R.K.); the Renal and Electrolyte Division, Departments of Medicine (M.J.C., P.C.F., A.P., B.C.-G., D.D., H.S., T.R.K.) and Physiology (T.R.K.), University of Pennsylvania; and the Department of Medicine, Medical College of Pennsylvania (P.C.F.) -- all in Philadelphia.
Address reprint requests to Dr. Kleyman at Medical Research (151), Veterans Affairs Medical Center, Philadelphia, PA 19104.
References
| |||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
