Bumetanide

Congestive heart failure (CHF) affects 2 to 3 million Americans with approximately 400,000 new cases diagnosed each year.[1] The prevalence increases with advancing age such that CHF is the most common diagnosis-related group for hospitalized patients 65 years of age or older. As the heart failure progresses, it becomes refractory to conventional oral medications and diuretic resistance frequently occurs. Gerlag and van Meijel[2] demonstrated that patients with severe refractory CHF may require IV doses of furosemide as high as 4,000 mg daily. The continuous infusion of a loop diuretic has been shown to provide a more efficient delivery of diuretic to the nephron, eliminate the diuretic-free interval during which compensatory sodium retention occurs, and decrease the development of tolerance.[3] The efficacy of continuous infusions of both bumetanide and furosemide has been demonstrated in patients with renal insufficiency and CHF.[2-4] It has also been suggested that the lower peak drug concentrations associated with continuous IV infusion may allow the administration of high daily doses while minimizing the risk of ototoxicity and other adverse effects.[5] In patients with refractory CHF, the safety of furosemide has been clemonstrated using continuous infusion rates as high us 160 to 168 mg/h.[24]

A variety of musculoskeletal symptoms have been reported with bumetanide, including pain, cramping, and weakness. These adverse effects are usually mild and self-limiting with a reported incidence of less than 2%[6]. There have been reports, however, of severe disabling musculoskeletal symptoms following oral or intermittent IV injections of bumetanide.[7] In a randomized, crossover trial, Rudy et al[3] reported that myalgias in three of eight patients receiving intermittent bolus therapy did not recur when the same patients received the equivalent dose as a continuous infusion. In contrast, we report a series of eight patients with heart failure with a total of 11 drug exposures that resulted in severe musculoskeletal symptoms during continuous IV infusion of bumetanide.

MATERIALS AND METHODS

After encountering a patient with severe musculoskeletal symptoms during a continuous IV infusion of bumetanide, we monitored all patients with severe refractory heart failure who received bumetanide infusions during the period from March 1994 to December 1995. A log was maintained for all patients who met the following four criteria: (1) administration of a continuous IV infusion of bumetanide for severe CHF; (2) development of severe musculoskeletal symptoms while receiving the infusion; (3) resolution of symptoms following discontinuation of the infusion; and (4) redevelopment of musculoskeletal symptoms if rechallenged with the bumetanide infusion. Medical records of these patients were reviewed and the following data were abstracted- demographics, primary diagnoses, comorbidities, concurrent medications, laboratory results, drug infusion data, symptoms, time course for symptoms, effects of rechallenge, alternative treatment, clinical course, and outcomes. These data were evaluated to characterize the adverse reaction and identify potential risk factors.

RESULTS

During the monitoring period, 34 patients received bumetanide continuous infusions for the treatment of refractory edema secondary to severe CHF. Eight of these patients (23.5%) experienced a total of 11 episodes of severe musculoskeletal symptoms during the infusions. Patients were both black and white, male and female, ranging in age from 36 to 59 years. The patient profiles are presented in Table 1. All were in severe heart failure with excess fluid volumes estimated at 8 to 15 L. The etiology of the CHF varied. The immediate causes for the exacerbation of congestive failure were often difficult to determine but were most often due to noncompliance with medications, fluid restriction, and sodium restriction or progression of the disease. The patients had multiple medical problems and a wide spectrum of comorbidities. All patients were receiving a variety of oral medications, including digoxin, loop diuretics, angiotensin-converting enzyme inhibitors, and various other medications for ancillary medical problems.

[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]

In all 11 episodes, the patients described the musculoskeletal symptoms as severe, resulting in significant pain, loss of mobility, and inability to perform ordinary functions. In every case, the reaction necessitated discontinuation of the infusion. All of the patients were subsequently diuresed with an equivalent or higher dose of the loop diuretic, furosemide, without further symptoms. Table 2 summarizes the diuretic dosage regimens and time course of the musculoskeletal symptoms. Pertinent laboratory findings during the adverse reaction are summarized in Table 3. The laboratory values did not change significantly prior to, during, or after the bumetanide infusions and no consistent abnormalities were observed. The following case report is representative of the adverse musculoskeletal symptoms induced by bumetanide in eight patients with severe CHF.

(*) 1a & 1b = patient 1 rechallenges; 2a = patient 2 rechallenge; po = oral; IVB = IV bolus.

(*) 1a and 1b=patient 1 rechallenges; 2a=patient 2 rechallenge. Laboratory values were obtained during the adverse reaction; no significant differences were noted in values immediately before or after the reaction. Normal range is listed in parentheses.

CASE REPORT

A 47-year-old black woman (case 1) was admitted to the hospital with a 2- to 3-week history of swelling in her legs, increasing abdominal girth, shortness of breath, and a 13.5-kg weight gain. The patient reported noncompliance with her diuretic regimen of furosemide, 400 mg daily. Her medical history included a thymoma treated with radiation in 1968, radiation pneumonitis, pulmonary hypertension, restrictive lung disease, right-sided CHF, and iron deficiency anemia.

The patient was diagnosed as having an exacerbation of CHF secondary to pulmonary fibrosis and pulmonary hypertension and was started on a regimen of a bumetanide drip at 1 mg/h. After 24 h, the rate was increased to 1.5 mg/h, at which time she began to complain of mild stiffness in her joints. Three days later, the bumetanide infusion was increased to 2.0 mg/h. Within 24 h of this dosage change, she reported severe muscle aches and cramping throughout her entire body. She was unable to get out of bed due to the severity of the pain and was unable to sleep. Acetaminophen provided only minimal relief During the 7 days of bumetanide infusion, she had diuresed a total of 9 kg.

Concurrent medications during the time when the patient was experiencing musculoskeletal symptoms included captopril, acetazolamide, potassium chloride, magnesium oxide, ferrous sulfate, estradiol patch, medroxyprogesterone, ammonium chloride, albuterol, and ipratropium. Laboratory values were as follows: serum glucose, 92 mg/dL; sodium, 145 mEq/L; potassium, 3.5 mEq/L; chloride, 98 mEq/L; total carbon dioxide, 36 mEq/L; BUN, 24 mg/dL; creatinine, 1.1 mg/dL; uric acid, 8.7 mg/dL; calcium, 9.1 mg/dL; phosphate, 4.1 mg/dL; magnesium, 2.1 mEq/L; albumin, 4.1 g/dL; total bilirubin, 1.3 mg/dL; aspartate aminotransferase (AST), 41 IU/L; alkaline phosphatase, 137 IU/L; creatine kinase, 66 IU/L; leukocytes, 6.3x[10.sup.9]/L; hematocrit, 14.7%; and platelet count, 159x[10.sup.9]/L. The bumetanide infusion was discontinued and the musculoskeletal symptoms resolved completely within 24 h. The patient was placed on a regimen of oral furosemide, 400 mg alternating with 480 mg every other day, and diuresed an additional 3 kg without complaint before hospital discharge.

Approximately 6 weeks later, the patient was readmitted to the hospital for increasing shortness of breath and edema. Her problem list during this hospitalization included right-sided CHF, pneumococcal pneumonia, and Clostridium difficile colitis. Her furosemide therapy was discontinued and a bumetanide drip was initiated at 2 mg/h. Concurrent medications in addition to those listed previously were heparin, piperacillin/tazobactam, and diphenhydramine. Less than 24 h later, she complained of severe muscle cramps and pain similar in character to that which she had experienced with the first episode. She had diuresed approximately 4 kg with the bumetanide infusion.

Laboratory values were as follows: serum glucose, 82 mg/dL; sodium, 143 mEq/L; potassium, 3.4 mEq/L; chloride, 97 mEq/L; [CO.sub.2], 37 mEq/L; BUN, 21 mg/dL; creatinine, 1.1 mg/dL; uric acid, 9.8 mg/dL; calcium, 9.2 mg/dL; phosphate, 4.6 mg/dL; maguesium, 1.7 mEq/L; albumin, 4.1 g/dL; total bilirubin, 0.9 mg/dL; AST, 50 IU/L; alkaline phosphatase, 104 IU/L; leukocytes, 8.3x[10.sup.9]/L; hematocrit, 12.6%; and platelet count, 219x 109/L. The bumetanide infusion was stopped and the symptoms resolved within 48 h. The patient remained hospitalized for an additional 2 weeks during which time she was given mechanical ventilatory assistance due to worsening pneumonia and respiratory failure. During this time, she received daily IV doses of furosemide, 240 ma, and was diuresed an additional 5 kg. She was discharged from the hospital on a regimen of oral furosemide without further complication.

Approximately 5 months later, the patient was readmitted to the hospital for an exacerbation of CHF. Her oral furosemide dose of 360 mg was discontinued and she was started on a bumetanide infusion at a rate of 1 mg/h. No other changes were made in her long-term medications. The following day, the infusion rate was increased to 2 mg/h at which time she began to complain of severe cramping in her upper and lower extremities and requested that the bumetanide infusion be stopped. She had diuresed approximately 5 kg with the infusion. All of her symptoms resolved completely within 24 to 48 h of discontinuing the infusion. She was started on a furosemide infusion at an equivalent dose of 80 mg/h. Over the next 3 days, she diuresed an additional 8 kg without further complication.

In summary, the severe musculoskeletal reactions associated with the bumetanide infusion. were observed in this patient during both the initial infusion and two subsequent rechallenges. Furthermore, conversion to an equivalent dose of a furosemide infusion resulted in a diuresis of greater magnitude that was not associated with musculoskeletal complaints.

DISCUSSION

Early published reports of musculoskeletal symptoms with bumetanide were anecdotal with no information on temporal relationships, potential mechanisms, or associated risk factors. To date and to our knowledge, the most extensive documentation of this adverse reaction was provided by Johnson et al.[7] These authors describe in detail nine patients who developed severe musculoskeletal symptoms following the administration of bumetanide. The reactions were seen most often in patients on the nephrology and renal transplant services. All nine patients had moderate to severe renal failure with a creatinine clearance ranging from [is less than]10 to 42 mL/min. Eight of the patients had hypoalbuminemia ranging from 1.7 to 3.0 g/dL. The patients were receiving a variety of other medications, including immunosuppressants; however, no drug was common to all nine except bumetanide.

The adverse reactions reported by Johnson et al[7] occurred following both oral and IV bolus administration of bumetanide in doses of 2 to 8 mg. A wide variety of symptoms were described, including muscle tenderness, cramping, tightness, pain, weakness, and marked loss of mobility. The severity of the symptoms appeared to correlate with dose and in most cases required discontinuation of treatment with the medication. In most patients, the onset of symptoms occurred approximately 2 to 4 h after dosing and the symptoms were not associated with electrolyte abnormalities. In all cases, the adverse effects resolved within 12 to 48 h of discontinuing the bumetanide therapy. Rechallenge in two of the patients resulted in an increased severity of the reaction. Two patients were changed to furosemide and one to ethacrynic acid with no further problems.

The authors were unable to explain the mechanism for these observed reactions; however, they suggested that renal failure and hypoalbuminemia are important risk factors. Since bumetanide is a highly protein-bound drug and is eliminated through the kidneys, reduced concentrations of albumin and decreased renal clearance would result in higher concentrations of the free active drug.

Although our cases share some characteristics with the series reported by Johnson et al,[7] there are at least four significant differences. First, all of our patients were being treated for CHF. Second, this is, to the best of our knowledge, the first report of severe musculoskeletal symptoms during the continuous infusion of a loop diuretic. Third, most of our patients did not have significant renal failure. During six of the episodes, the serum creatinine concentration was within the normal range of 0.5 to 1.2 mg/dL. In the remaining five instances, the serum creatinine concentration was elevated ranging from 1.3 to 3.3 mg/dL. Most of these patients probably had prerenal azotemia as determined by the BUN:creatinine ratio. Fourth, hypoalbuminemia was not a factor in most of our patients. In six instances, the serum albumin level was within the normal range of 3.8 to 5.5 g/dL. The albumin level was [is less than or equal to] 3.0 g/dL in only two patients.

In other respects, the adverse reactions we observed were similar to those reported by Johnson et al.[7] The intensity of the reaction was severe, resulting in extreme weakness, pain, and loss of mobility. The pain was not relieved by analgesics, quinine, or heating pads. Patients were unable to sleep due to the muscle cramping and severity of the pain, and several patients required assistance to stand or walk during the reaction. The pain was widely distributed throughout the body involving joints and muscles of the upper and lower extremities, the back, and sternum. There was no erythema, swelling, warmth, or other signs of inflammation over the tender areas. One patient reported pain and stiffness in his jaw with ensuing difficulty opening his mouth to talk. Another patient reported that her legs felt so weak, she thought she was paralyzed. This patient had a history of peripheral neuropathy and degenerative joint disease. However, she was emphatic that this pain was different from any she had experienced previously and it was not relieved by her usual dose of carbamazopine. Furthermore, her symptoms began to subside within 4 h of discontinuing the bumetanide infusion and completely resolved within 24 h.

The reactions that we observed appeared to be dose-related. In seven instances, the pain occurred after increasing the infusion rate to 2.0 mg/h. In the remaining four episodes, musculoskeletal symptoms began at rates of 1.0 mg/h and increased in severity as the infusion rate was increased to 1.5 to 2.0 mg/h. The times to onset and resolution of symptoms were remarkably consistent. The onset was approximately 12 to 24 h after increasing the rate of infusion to 2 mg/h in all but three patients in whom the onsets were approximately 8, 8 to 12, and 48 h, respectively. With the continuous infusion, the onset of symptoms appeared to be more gradual as compared to the abrupt onset reported by Johnson et al[7] following bolus IV administration. This may reflect the fact that bolus administration produces a rapid increase in serum drug concentration as compared to the more gradual increase produced by the infusion; however, this was not measured. Similarly, the resolution of symptoms was gradual with some patients reporting a decrease in the symptoms within 4 to 8 h of discontinuing the infusion and complete resolution within 24 to 48 h. In the two patients who were rechallenged, the onset was approximately the same with a slightly longer time for resolution. We were not able to quantitatively determine any difference in the severity of the symptoms with the second reaction. The occurrence of musculoskeletal symptoms did not correlate with the degree of diuresis. In several cases, both daily and total diuresis of an equal or greater magnitude were achieved with alternate loop diuretic therapy without complication.

The adverse reactions did not correlate with any specific laboratory abnormality. No significant differences were observed in laboratory values obtained before, during, or after the reactions. During most episodes, the serum potassium level was normal. In four instances, the potassium level was 3.4 to 3.5 mEq/L, which is slightly below the lower normal limit of 3.7 mEq/L. None of the patients were hypomagnesemic during the reactions. Taking into account a correction for the two patients with low albumin levels, the serum calcium level was normal for all patients. During eight of the 11 episodes, the serum uric acid level was elevated. Renal function was normal in five of these patients. Although the increased uric acid level may have been caused in part by the bumetanide, it was not a consistent finding in all of the patients with symptoms. Results of liver function tests (AST, bilirubin, and alkaline phosphatase) were elevated in some but not all patients. These changes were believed to be due to hepatic disease or congestion secondary to CHF. The creatine kinase concentration was measured in four patients during the adverse event and was within the normal range. The erythrocyte sedimentation rate was slightly increased in three patients during the time of the reaction; however, there were numerous factors that could have accounted for this finding.

Although all of our patients were receiving multiple medications at the time of the reaction, no significant drug interactions were identified that could consistently account for the musculoskeletal symptoms. There was no cross-reaction with another loop diuretic, furosemide, since all patients were successfully changed to oral or IV furosemide. In five instances, the furosemide was given as a continuous infusion at an equivalent or higher dose with no further symptoms. Another potential alternative agent for these patients would have been the loop diuretic, torsemide. However, we have no personal experience with this drug and to our knowledge, there is no published literature on its use as a continuous infusion.

We were unable to determine the underlying cause for the observed reaction or identify specific risk factors. Although renal failure and hypoalbuminemia may contribute, as suggested by Johnson et al,[7] these factors do not account for all patients. Based on our patients and those previously described in the literature, it appears that the reaction is dose related and occurs irrespective of the route of bumetanide administration. We do not know if reducing the dose during the reaction would resolve the symptoms, since none of our patients were willing to continue with the infusion. In conclusion, the continuous infusion of high-dose bumetanide may be associated with severe disabling musculoskeletal symptoms in some patients. Patients who develop this reaction may be successfully treated with an alternative loop diuretic such as furosemide.

REFERENCES

[1] Parmley WW. Pathophysiology and current therapy of congestive heart failure. J Am Coll Cardiol 1989; 13:771-85

[2] Gerlag PG, van Meijel JJ. High dose furosemide in the treatment of refractory congestive heart failure. Arch Intern Med 1988; 148:286-91

[3] Rudy DW, Voelker JR, Greene PK, et al. Loop diuretics for chronic renal insufficiency: a continuous infusion is more efficacious than bolus therapy. Ann Intern Med 1991; 115: 360-66

[4] Van Meyel JJM, Smits P, Dormans T, et al. Continuous infusion of furosemide in the treatment of patients with congestive heart failure and diuretic resistance. J Intern Med 1994; 235:329-34

[5] Yelton SL, Gaylor MA, Murray KM. The role of continuous infusion loop diuretics. Ann Pharmacother 1995; 29:1010-14

[6] Tuzel IH. Comparison of adverse reactions to bumetanide and furosemide. J Clin Pharmacol 1981; 21:615-19

[7] Johnson CA, Grant KL, Madalon MG. Severe musculoskeletal syndrome in patients with renal failure and hypoalbuminemia receiving bumetanide. Clin Pharm 1987; 6:735-41

(*) From the Department of Pharmacy Practice (Dr. Howard) and the Division of Cardiovascular Medicine, Department of Internal Medieine (Dr. Dunn), University of Kansas Medical Center Kansas City.

Manuscript received June 11, 1996; revision accepted August 29. Reprint requests: Dr Marvin I. Dunn, KU Medical Center Division of Cardiovascular Medicine-Room 1001E, 3901 Rainbow Blvd, Kansas City, KS 66160-7378

COPYRIGHT 1997 American College of Chest Physicians
COPYRIGHT 2004 Gale Group