Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Furosemide is one of the so-called high-ceiling or loop diuretics, which can achieve a much greater peak diuresis than the thiazides. It is widely and frequently used both orally and parenterally over a wider dosage range than the thiazide diuretics, because its concentration–effect curve is steeper and because it is effective in patients with moderate renal insufficiency (creatinine clearance 5–25 ml/minute), in whom the thiazide diuretics and most related compounds are ineffective.
Furosemide is quickly and almost completely excreted by kidney unmetabolized. The usual oral dose is 20–120 mg. Much larger doses (for example 1000 mg) have been used in renal insufficiency, but are not recommended. It is very effective after intravenous injection, and doses of 500 mg and more can be used in emergencies (renal insufficiency, pulmonary edema). Most adverse effects and reactions occur with the use of high doses [ ].
Disturbances of fluid and electrolyte balance, such as hyponatremia, hypokalemia, and dehydration with circulatory disturbances (such as dizziness, postural hypotension, and syncope), have been reported. Rarely gastrointestinal symptoms are problems with high dosages and in the elderly. Pancreatitis and jaundice seem to occur more often than with the thiazide diuretics, but deterioration of glucose tolerance seems to be less common. At serum concentrations over 50 μg/ml, tinnitus, vertigo, and deafness, sometimes permanent, have been reported. Hematological disorders, particularly thrombocytopenia, and serious skin disorders occur occasionally, as do hypersensitivity reactions. Neither tumor-inducing effects nor second-generation effects have been documented.
Because furosemide, apart from its diuretic effect, has transient but pronounced vasodilatory properties (changes in both venous capacitance and peripheral arteriolar resistance have been described in anephric patients), it can cause postural hypotension and syncope, particularly if given together with other blood pressure-lowering drugs [ , ]. Ischemic complications have been reported in elderly patients. Reactions due to extracellular volume depletion accounted for 9% of all adverse reactions observed in 535 patients treated with furosemide [ ]. Neuroendocrine activation and resultant increased peripheral resistance (afterload) after furosemide reduce cardiac output and stroke volume and increase cardiac work, with the possibility of worsening myocardial and tissue ischemia. Since many patients with heart failure have underlying myocardial ischemia or infarction, initial symptomatic benefit from furosemide can be followed by detrimental effects on myocardial perfusion, with extension or completion of myocardial necrosis.
When it is used in cardiac failure, furosemide acts in two ways: besides its diuretic effect it produces an immediate fall in left ventricular filling pressure, which is independent of and precedes diuresis. If furosemide is given intravenously in stable chronic heart failure (which it normally is not), this can be an unwanted effect, causing deterioration [ ], particularly in patients with pure left ventricular failure.
Visual disturbances and drowsiness have been described, but it is not clear whether these were caused by reduced cerebral perfusion or by a direct effect of the drug itself.
At high doses, and especially if serum concentrations are over 50 μg/ml, furosemide can cause ototoxic reactions, such as tinnitus, vertigo, and even deafness, sometimes permanent [ ]. Subclinical, audiometrically determined, high-tone deafness has been reported to occur in 6.4% of furosemide-treated patients [ ]. It is generally considered advisable to use another diuretic in patients whose hearing is already impaired, and to avoid using furosemide along with other ototoxic drugs, such as the aminoglycosides.
Sensorineural hearing loss occurs in a small proportion of very premature babies who are given furosemide. Various causative mechanisms have been suggested, including bilirubin, drugs, infection, and/or hypoxic brainstem injury. In a case–control study of 15 children and 30 controls born before 33 weeks of gestation, renal insufficiency and/or aminoglycoside use in conjunction with furosemide was associated with sensorineural hearing loss [ ].
A retrospective chart review (July 2000 to January 2002) of all survivors in a neonatal intensive care unit was undertaken to evaluate the effect of furosemide on hearing loss [ ]. Of the 57 neonates who had received furosemide nine had a subsequent abnormal hearing screen, and of the 207 neonates who had not received furosemide 33 also had an abnormal hearing screen. This suggests that hearing loss in these neonates is not directly related to the use of furosemide.
Furosemide rarely causes the syndrome of inappropriate antidiuretic hormone secretion (SIADH) (although it has been found useful in treating some patients with SIADH who cannot tolerate water restriction [ ]. In furosemide-induced cases [ ], serum ADH concentrations were raised, total body sodium was normal, total body potassium greatly reduced, and intracellular water raised at the expense of extracellular fluid volume. However, such cases are rare.
Patients with congestive heart failure taking high doses of furosemide can develop thiamine deficiency, which is improved by thiamine supplementation. There is whole-blood thiamine phosphate deficiency, but no reduction in the storage form of thiamine, thiamine diphosphate. These observations suggest that thiamine supplementation may not be necessary in elderly patients taking furosemide for congestive heart failure [ ].
Erythrocyte transketolase activity suggested severe thiamine deficiency in 24 of 25 patients with heart failure who were taking at least 80 mg/day of furosemide and in four of seven patients who were taking 40 mg/day (OR = 19; 95% CI = 1.1, 601) [ ]. Thiamine status was not associated with any other clinical variables. These findings suggest that thiamine deficiency occurs in a substantial proportion of patients with heart failure who are taking furosemide; however, this is of unclear clinical significance.
Agranulocytosis, thrombocytopenia, and hemolytic anemia have been reported occasionally [ , ]. The commonest complication is thrombocytopenia [ ], although it is often mild and asymptomatic [ ].
Diuretics cause altered salivary flow rate and composition and have been associated with both subjective and objective evidence of xerostomia. In a randomized trial in 12 healthy women randomly assignment to placebo, bendroflumethiazide (2.5 mg/day for 7 days) and furosemide (40 mg/day for 7 days), xerostomia increased with furosemide in conjunction with a reduction in submandibular-sublingual salivary secretion [ ]. This was particularly so at lunchtime. This suggests that diuretics may potentiate dryness of the mouth and should be used carefully in patients with abnormal salivary flow.
With normal doses, nausea, vomiting, and diarrhea are very uncommon, accounting for less than 1% of all adverse reactions [ ]. The incidence rises with higher doses and in the presence of uremia.
Although furosemide is very hepatotoxic in experimental animals, only a few cases of jaundice have been reported [ ], and no fully documented cases have so far been published. However, in patients with cirrhosis furosemide readily precipitates hepatic encephalopathy [ ], even when low doses are used [ ].
Biliary colic has been attributed to furosemide [ ].
Increases in serum isoamylase [ ] and pancreatitis have been reported in patients taking furosemide.
Excessive diuresis and dehydration often cause a transient reduction in glomerular filtration rate and a rise in serum urea (about 8% of all adverse reactions) [ ]. The sudden diuresis can cause loin pain, particularly in elderly patients, and acute urinary retention and overflow incontinence in elderly men with prostatic hyperplasia.
Although it is often described in children, medullary nephrocalcinosis with furosemide has been rarely described in adults.
A 40-year-old woman who had taken furosemide (40–160 mg/day) for 15 years developed medullary nephrocalcinosis [ ].
Chronic tubulointerstitial nephritis has been reported with furosemide.
A 25-year-old woman developed biopsy-proven chronic tubulointerstitial nephritis with accompanying distal renal tubular acidosis in association with furosemide abuse (up to 1.2 g/day for several months) [ ].
Four children with the nephrotic syndrome developed transient hypercalciuria and intraluminal calcification in renal histopathological specimens without radiological evidence of renal calcification. These children were resistant to corticosteroids and were receiving furosemide plus albumin for the management of edema [ ]. This result stresses the pervasive effect of furosemide, and probably all loop diuretics, in increasing urinary calcium excretion, with resultant nephrocalcinosis. Whenever possible, steps should be taken to limit the hypercalciuric effect of loop diuretics. Such maneuvers could include limiting the sodium content of the diet and/or combining the loop diuretic with a thiazide diuretic.
A 900-gram girl born before term with bronchopulmonary dysplasia developed ureteral obstruction, urinoma, and acute renal insufficiency as a result of furosemide-related hypercalciuria (cumulative dose 27.5 mg) and nephrolithiasis [ ]. Percutaneous drainage of the urinoma plus conversion to hydrochlorothiazide resolved the urinoma and hydronephrosis.
Acute renal insufficiency carries a high mortality and morbidity. Diuretics may increase mortality in patients with acute renal insufficiency, but this has not been studied prospectively. In a prospective, multicenter, multinational, epidemiological study of 1743 consecutive patients, who were either treated with renal replacement therapy or who fulfilled predefined criteria for acute renal insufficiency, about 70% were taking diuretics at enrolment [ ]. Severe sepsis/septic shock (48%), major surgery (39%), low cardiac output (30%), and hypovolemia (28%) were the most common conditions associated with the development of acute renal insufficiency. Furosemide was the most common diuretic used (98%). In all three multivariate models, diuretic use was not associated with a significantly increased risk of mortality. The use of diuretics in patients with acute renal insufficiency should continue, but only according to a specific need to control volume excess.
Rashes seem to be just as common with furosemide [ ] as with other oral diuretics, but severe skin reactions (exfoliative dermatitis, erythema multiforme, acquired epidermolysis bullosa), which are rare with other diuretics, have been occasionally reported with high doses of furosemide in renal insufficiency [ ].
Cases of furosemide-induced lupus-like syndrome [ ], bullous pemphigoid [ , ], and lichenoid drug eruptions [ ] have been reported.
A review of furosemide-induced skin reactions included a description of an 88-year-old man who developed an eruption that clinically and histologically simulated Sweet’s syndrome (acute febrile neutrophilic vasculitis) after 6 weeks [ ]. Atypical features and rapid resolution suggested a drug eruption rather than true Sweet’s syndrome. However, a similar mechanism may have been implicated a hypersensitivity reaction involving immune complexes.
A 46-year-old woman with congestive heart failure was given intravenous furosemide and 3 days later developed a low-grade fever followed by tender, papular, erythematous, non-pruritic skin eruptions bilaterally on the wrists, forearms, arms, and thighs [ ]. There was associated redness in both eyes and photophobia, consistent with episcleritis and iritis. A skin biopsy showed a superficial dermal nodular neutrophilic infiltrate, associated with nuclear dust and rare eosinophils, findings consistent with Sweet syndrome. Furosemide was withdrawn and the skin lesions and eye symptoms gradually subsided.
Sweet syndrome is a disorder of poorly understood pathogenesis that has several features suggesting a hypersensitivity reaction. It has been associated with autoimmune diseases, malignancies, and drugs. This patient was different from the previous one in having symptoms and signs suggestive of iritis.
Linear IgA bullous dermatosis associated with the administration of furosemide has been reported [ ].
An 86-year-old woman, with a history of stable schizophrenia, chronic obstructive pulmonary disease, ischemic cardiomyopathy, and type 2 diabetes, was admitted with cardiac insufficiency, which was treated by introduction of enalapril. A chest infection was treated with co-amoxiclav with gradual alleviation of symptoms over 10 days. At this point, furosemide was begun, because of persistent signs of heart failure. After 3 days, erythema and bullae were noted on her palms and soles, and later on the trunk, extremities, hard palate, and buccal mucosa. Biopsy showed the characteristic features of linear IgA bullous dermatosis, with linear deposition of IgA along the basement membrane. Co-amoxiclav and furosemide were withdrawn; no new lesions were noted thereafter.
Furosemide has previously been related to other bullous dermatoses, particularly bullous pemphigoid [ ]. In this case only a temporal relation was demonstrated, as rechallenge was judged to be unethical. The other putative offending drug, co-amoxiclav, was regarded as unlikely to be causative, because it had been given many times before without noticeable skin lesions.
Furosemide has been associated with disseminated superficial porokeratosis, a heritable disorder of cornification [ ].
A man developed acute generalized exanthematous pustulosis while taking furosemide [ ]. A positive lymphocyte transformation test suggested an immunological mechanism.
Although loop diuretics increase renal calcium excretion, there have been variable results in studies of their effects on the risk of fractures. In a case–control study of the risk of fracture in 44 001 patients who had taken a loop diuretic in the preceding 5 years and 194 111 age- and sex-matched controls, “ever” use of a loop diuretic was associated with a crude 51% increased risk of any fracture (OR = 1.51; 95% CI = 1.48, 155) and a 72% increased risk of hip fracture (OR = 1.72; 95% CI = 1.64, 181) [ ]. After adjustment for potential confounders, the risk reduction was only slightly increased for any fracture (OR = 1.04) and for hip fracture (OR = 1.16). With an increased average daily dose the estimates increased in former users but fell in current users. Furosemide was associated with higher risk estimates than bumetanide. This study does not provide a definitive answer to the risk of fractures with loop diuretics but suggests that particular caution should be taken.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here