European Heart Journal (1992) 13 (Supplement G), 10-14
Clinical pharmacology of loop diuretics in health and disease D. C. BRATER
Department of Medicine, Indiana University School of Medicine, U.S.A. KEY WORDS: Diuretics, comparative pharmacokinetics, comparative pharmacodynamics, renal failure, liver failure.
Downloaded from by guest on July 12, 2015
There are differences in metabolism and excretion of the loop diuretics which extrapolate to differences in pharmcokinetic behaviour in different disease states. For example, furosemide is eliminated in equal portions by renal and non-renal routes; the non-renal route involves primarily glucuronidation. Both renal and non-renal pathways are impaired during renal insufficiency, such that the elimination half-life is prolonged considerably in this disease state. In contrast, there seems to be little change in disposition in patients with liver disease. Bumetanide and torasemide have non-renal elimination pathways via the hepatic cytochrome system. In patients with renal insufficiency these non-renal pathways of elimination are left intact so that there is little prolongation of half-life in such patients. In contrast, in patients with liver disease or with congestive hepatopathy, there is impairment in non-renal elimination so that relatively more drug appears in the urine. With all loop diuretics, response is governed by the amount of drug appearing in the urine. By assessing the relationship between urinary excretion rates of the diuretic and sodium excretion rate, a maximal response which amounts to a fractional excretion of sodium of approximately 20% can be defined. Thus it is possible to define a maximal dose as the amount of drug necessary to attain a fractional excretion of sodium of 20%. Studies in different disease states using escalating doses can thereby use this relationship to define a dose above which little is to be gained in terms of therapeutic efficacy. Analysing the relationship between amount of diuretic in the urine and response also allows assessment of different mechanisms by which resistance to diuretics occurs in different disease states. For example, in patients with renal insufficiency, the primary mechanismfor resistance is diminished delivery ofdrug to the site ofaction. As such, administration of large doses ofdrug to deliver 'normal' amounts ofdiuretic into the urine result in the same response as occurs in subjects with normal renal function. However, this response is in terms of the fractional excretion of sodium meaning that the total amount of sodium eliminated in the urine is still diminished as a result of decreasedfilteredsodium. In contrast, in patients with liver disease or with congestive heart failure even when normal amounts of diuretic reach the urinary site of action there is diminished response which may be accountedfor by either increased proximal or increased distal tubular reabsorption of sodium. In such syndromes, frequent administration of drug is needed to attain the desired cumulative response and, in addition, combination drug therapy can be helpful by inhibiting sodium reabsorption at several distinct sites of the nephron. This latter strategy can be particularly helpful in subjects who have received long-term loop diuretic therapy who likely have hypertrophy of distal nephrons which are thereby able to reabsorb more sodium and diminish overall response. These distal nephron sites can be inhibited by thiazide diuretics accounting for the synergistic response that often occurs with a combination of loop and thiazide diureticsIntroduction characterized in healthy subjects to use as a comparison Recent studies have allowed elucidation of disease- t 0 patients. Pharmacokinetic parameters are generally induced effects on pharmacokinetics and pharmaco- similar (Table 1), excepting the somewhat lower clearance dynamics of loop diuretics. Such data have permitted a a n d concomitantly longer half-life of torasemide"-3". greater understanding of mechanisms for the diminished S " " ^ Perusing these pharmacokinetic parameters, howresponse to diuretics that occurs in most oedematous dis- e v e r ' ' g n o r e s differences among these drugs that become orders. Mechanisms differ in different clinical conditions, ™POrtant in disease states. For example the elimination and their definition has allowed derivation of rational ^alf-hfe reported for furosem.de is that after intravenous strategies for treatment. The goal of this review is to d o s i n S" A f t e r o r a I d o s i n 8 t h e t e r m i n a l h a l f " l l f e o f f u r o s " present an overview of the clinical pharmacology in health e u m i d e «J. 0 "* 6 ! l * a n i n t r a v j! n °"f d o s ' n « a n d " P " * * " * and disease of furosemide, bumetanide, and torasemide, t h J h a l f -' l f e ?ff absorption^. Thus, the time course of the three loop diuretics that have been studied in most absorption of furosemide is slower than that of bumet^ {L anide and torasemide. Moreover, in the oedematous disorders of congestive heart failure (CHF) and cirrhosis, the absorption of furosemide is further slowed'5"71. This Pharmacokinetics of loop diuretics in health and disease c ^ e s the terminal half-life to be yet longer and could The pharmacokinetics and pharmacodynamics of lead to the misinterpretation that elimination (rather than furosemide, bumetanide, and torasemide have been absorption) is slowed. This phenomenon does not seem to occur with bumetanide and torasemide. C
^^^^tll^%r^Ts^H7^X
IN 46202, U.S.A. 0I95-668X/92/0GOO10+05 $08.00/0
Pharmacokinetic parameters also do not stress differences in metabolism among these loop diuretics. The © 1992 The European Society of Cardiology
Clinical pharmacology of loop diuretics 11
Table 1 Pharmacokinetics offurosemide, bumetanide, and torasemide
Bioavailability (%) Volume of distribution (1. kg"1) Clearance (ml. min"'. kg-') Half-life (h) Fraction of intravenous dose excreted unchanged (%)
Table 2
Furosemide
Bumetanide
Torasemide
50 016 2-2 10 60
80 017 26 1-2 65
80 0-16 08 3-0 20
Effects of liver disease on the pharmacokinetics of bumetanide and torasemide
Bioavailability (%) Volume of distribution (1. k g ' ) Clearance (ml. min"'. kg"1) Half-life (h) Fraction of intravenous dose excreted unchanged (%)
Bumetanide
Torasemide
95 017 0-6 2-3 70
95 0-34 0-5 8-4 27
diminished non-renal clearance with preservation of renal clearance; half-life concomitantly doubles1"1. The combination of these effects causes a small increase in the amount of diuretic appearing in the urine, since there is a higher plasma concentration of drug (owing to decreased total clearance) available for renal excretion. The duration of effect of bumetanide in cirrhotic patients would be expected to be prolonged concomitant with the increased half-life. The effect of hepatic disease on torasemide pharmacokinetics differs somewhat from that with bumetanide. Non-renal clearance is similarly decreased, but renal clearance increases such that substantially more drug appears in the urine. Half-life remains about the same as in healthy subjects so that the duration of effect is not prolonged. These pharmacokinetic characteristics of furosemide, bumetanide, and torasemide allow predictions as to effects of disease on diuretic response. Pharmacodynamics of loop diuretics in health and disease
All loop diuretics are highly protein bound (about 98% to albumin)11'31 and have been shown in both animals and man to block renal sodium reabsorption from the lumenal rather than the peritubular side of the nephron"21. The high degree of protein binding precludes access of meaningful amounts of a loop diuretic to the lumen via filtration at the glomerulus (even in hypoalbuminemic states); the major mode of entry into the urine is thus by active secretion via the organic acid secretory pump of the proximal tubule. After secretion, diuretic travels with the tubular fluid to the thick ascending limb of the loop of Henle, where it inhibits the Na + -K + -2C1" reabsorptive pump. This nephron site normally reabsorbs about 20% of filtered sodium. Interestingly, maximal doses of loop
Downloaded from by guest on July 12, 2015
non-renal elimination of furosemide occurs for the most part via glucuronidation, a component of which may be renal""31. Diminished renal function not only decreases the renal clearance of furosemide, but also decreases non-renal clearance (by unknown mechanisms). Consequently, total clearance offurosemide decreases substantially, and elimination half-life increases to as long as 4 to 6 h in patients with severe renal insufficiency"-2'891. In contrast, the non-renal clearance of bumetanide and torasemide appears to occur via the mixed function oxygenases of the liver, a pathway that is not influenced by renal disease18'1. Consequently, as one would predict, non-renal clearance of both of these drugs is not affected by renal disease and, in fact, this route of elimination compensates for the decreased renal clearance sufficiently that the elimination half-life of bumetanide and torasemide is not prolonged in patients with renal insufficiency1891. Similarly, duration of action is not prolonged, and accumulation with chronic dosing is less likely to occur compared to furosemide. These different routes of metabolism also impact upon the influence of hepatic disease on the pharmacokinetics of loop diuretics. With furosemide, neither hepatic cirrhosis nor congestive hepatopathy from CHF affects pharmacokinetics171. This may be because non-renal clearance offurosemide occurs at extra-hepatic sites (e.g. renal glucuronidation). Though not carefully studied in man, the observation that hepatectomy of dogs has no effect on non-renal clearance offurosemide supports this notion"01. With bumetanide and torasemide, bioavailability increases in patients with cirrhosis from about 80% to 95-100%(l", perhaps due to diminished first-pass elimination. This magnitude of change is unlikely to be clinically important. Cirrhosis, however, has important effects on the pharmacokinetics of bumetanide and torasemide (Table 2). Total clearance of bumetanide decreases due to
12 D. C. Brater
Downloaded from by guest on July 12, 2015
Administering a dose of loop diuretic sufficient to compensate for the changes in pharmacokinetics that occur in Bumetanide Furosemide the particular clinical condition may still not elicit the expected response if there are disease-induced changes in pharmacodynamics; namely, a change in the relationsship between the amount of drug at the active site and response. Recent data have shown that remaining 1-0 W KK) 50 100 nephrons in patients with renal insufficiency respond Diuretic excretion rate (/ig.min"1) normally to amounts of diuretic reaching them1813'. If a Figure 1 Schematic representation of therelationshipbetween the sufficient dose of the diuretic is administered, maximal urinary excretion rate of a loop diuretic and response showing the response actually exceeds a FE Nl of 20%, likely due to relative potencies of furosemide, bumetanide, and torasemide. the increased reabsorptive load of remnant nephrons in patients with renal disease1'3'. Fundamentally, however, the pharmacodynamics of response to loop diuretics in diuretics cause excretion of 20% of filtered sodium patients with renal insufficiency is normal. (FE Nl = 20%) in healthy volunteers, indicating that loop Even though remaining nephrons respond satisfacdiuretics are able to completely block solute reabsorption torily when sufficient doses of loop diuretics are adminisin this nephron segment. tered, overall response is greatly diminished compared to Since loop diuretics exert their effects at the lumenal subjects with normal renal function18'. In such patients, surface of the nephron, it is amounts of diuretic in the however, it is filtered sodium that limits response rather urine that best correlate to response, as opposed to than the dynamics of interaction between the diuretic and concentrations in plasma'121. The relationship between tubular solute reabsorption. For example, a maximal urinary diuretic and natriuretic response is characterized FE excretion of 20% in a subject with normal renal Na by a sigmoid-shaped curve that fits the so-called Sigmoid function results in a peak sodium excretion rate of about E ^ model (Fig. 1). Each loop diuretic has the same 3 mEq . min"'. A similar response in terms of FE in a Na maximal response ( E J and similar slopes (y), but they patient with severe renal insufficiency results in a peak differ in terms of potency ( E D ^ (Table 3), accounting for sodium excretion rate of only about 0-3 mEq. min" 1 the differing doses necessary to achieve the same overall purely on the basis of diminished filtered sodium. Thus, response1''. while remnant nephrons may respond 'normally' in In assessing the effects of disease on response to loop patients with renal insufficiency, the overall natriuresis diuretics, one must examine not only the effect of the clini- that ensues is limited by the decreased filtered load of cal condition on the pharmacokinetics of the diuretic, but solute. also the effect on pharmacodynamics. Two prototypic Thus, in patients with renal insufficiency, the pharmadiseases, renal insufficiency and hepatic cirrhosis, will be cokinetics of loop diuretics are altered, and these changes discussed and related to the pharmacodynamics that have have been characterized sufficiently to allow accurate estibeen delineated in healthy subjects. mation of the doses needed to attain adequate amounts of diuretic at the tubular site of action. The pharmacodynamics of loop diuretics are normal in this condition, Renal disease though the overall natriuretic response is limited by As would be predicted, patients with renal insufficiency diminished filtered sodium. have diminished renal clearance of loop diuretics due to decreased numbers of functioning nephrons and additionally to blockade of secretion of diuretics, into the Hepatic disease lumen at the proximal tubule by accumulated endogenous Whereas renal insufficiency typifies a condition with organic acids of azotemia1'"3'. Thus, relative to dose, a altered pharmacokinetics of loop diuretics, hepatic cirsmaller amount of diuretic reaches the lumen, and it is not rhosis is a disease with altered pharmacodynamics. In surprising that large doses are needed to attain effective patients with cirrhosis but preserved renal function, the amounts at the site of action, particularly in patients with renal clearance of loop diuretics is normal, the exception severe renal insufficiency18'. being the increased renal clearance observed with toras4 191 With all loop diuretics, the relationship between the emide'' " . As a consequence, 'normal' amounts of level of renal function and renal clearance of the diuretic diuretic reach the tubular site of action, and no change has been defined'1"3'. For example, in patients with severe in dosing is needed in such patients unless concomitant renal insufficiency (creatinine clearance approximately decreases in renal function occur. 15 ml. min"'), renal clearance of furosemide is about oneWhen the relationship between amounts of diuretic in fifth that of a subject with normal renal function'8'. From urine and natriuretic response is assessed, however, the such data, one can readily estimate the dose escalation curve is shifted downward and to the right (Fig. 2).1201. necessary to attain 'normal' amounts of diuretic in the Maximal response is suppressed so that such patients will urine; thus, in a patient with severe renal insufficiency, five not attain the same magnitude of natriuresis as a healthy times the customary dose of furosemide would be needed subject. Thus, cumulative sodium excretion relative to a compared to a patient with normal creatinine clearance. dose is less in a patient with cirrhosis than in a healthy
Clinical pharmacology of loop diuretics 13
Table 3
Pharmacodynamic parameters of furosemide, bumelanide, and lorasemide Furosemide
Bumetanide
Torasemide
16 1-2 70
17 1-8 10
16 20 14
Maximal response (FE t o in %) Slope Excretion rate causing half-maximal response (jig. min~')
,
20 -
odiurr
£
/ Normal
(A
1 f
/
5
excretion
"o
o Fracl
JOI.
J'
/ Cirrhosis
/
Log urinary diuretic excretion rate
Figure 2 Schematic illustrating the change in pharmacodynamics characteristic of patients with cirrhosis.
4
r
Conclusion
The pharmacokinetics of loop diuretics in different disease states have, with the exception of nephrotic syndrome, been well clarified, allowing accurate estimation of doses needed to attain sufficient amounts at the tubular site of action to elicit a response. Disease states in which altered pharmacodynamics of loop diuretics occurs have been identified so that clinicians understand that even maximally effective doses of loop diuretics in such patients will have subnormal responses. Coupling knowledge of pharmacokinetics and pharmacodynamics allows understanding of mechanisms of diuretic action and development of rational treatment strategies that can be tailored to individual patients.
Heart failure
Two additional oedematous disorders, CHF and nephrotic syndrome, represent combinations of pharmacokinetic and pharmacodynamic changes in loop diuretics. In congestive heart failure, renal function is usually diminished to at least a small degree, and this causes changes in delivery of diuretic into the urine15-20'. By escalating the dose relative to changes in renal function, sufficient amounts of diuretic gain access to the site of action; however, response is subnormal'211. Similar to observations in patients with cirrhosis, the mechanism of the changed pharmacodynamics is unknown. Nephrotic syndrome
In patients with nephrotic syndrome and normal glomerular filtration rate, renal clearance of furosemide is
References [1] Brater DC. Diuretics. In: Williams RL, Brater DC, Mordenti J, eds. Rational Therapeutics: A Clinical Pharmacologic Guide for the Health Professional. New York: Marcel Dekker, 1989: 269-315. [2] Beermann B, Groschinsky-Grind M. Clinical pharmacokinetics of diuretics. Clin Pharmacokinet 1980; 5: 221-45. [3] Hammarlund-Udenaes M, Benet LZ. Furosemide pharmacokinetics and pharmacodynarmcs in health and disease—An update. J Pharmacokinet Biopharm 1989; 17: 1-46. [4] Hammarlund MM, Paalzow LK, Odlind B. Pharmacokinetics of furosemide in man after intravenous and oral administration. Application of moment analysis. Eur J Clin Pharmacol 1984; 26: 197-207. [5] Brater DC, Day B, Burdette A, Anderson S. Bumetanide and furosemide in heart failure. Kidney Int 1984; 26: 183-9. [6] Vasko MR, Cartwright DB, Knochel JP, Nixon JB, Brater DC. Furosemide absorption altered in decompensated congestive heart failure. Ann Intern Med 1985; 102:314-8.
Downloaded from by guest on July 12, 2015
•
counterpart, even though the dose is sufficient to attain maximal response in that patient. The mechanism for altered pharmacodynamics is unknown. Increased reabsorption of solute in the proximal nephron could theoretically cause such an effect, as could increased distal reabsorption of solute. These possibilities should be amenable to testing by assessing whether co-administration of diuretics blocking proximal and/or distal sites restores response toward normal. It is also possible that disease-induced abnormality at the loop of Henle itself is causal. In summary, hepatic cirrhosis is a disease wherein pharmacokinetics of loop diuretics (at least in terms of delivery of drug to the site of action) are normal, and altered response is due to changed pharmacodynamics, the mechanism of which is unknown.
normal'22-23'. However, it is now clear, at least in animal studies, that the loop diuretic reaching the tubular lumen binds to the albumin that has leaked into the lumen through the glomerulus124-25'. This binding renders the diuretic inactive; activity can be restored in experimental animals by displacing the diuretic from albumin'261. Whether such a strategy might be employed clinically is unknown. In addition to this unique kinetic effect of intratubular albumin, it appears that there is also a pharmacodynamic abnormality wherein response assessed relative to unbound diuretic in the urine is subnormal in nephrotic syndrome'271. Again, the mechanisms of this change are unknown. Similarly, in man it is not clear how large a dose of diuretic must be administered to attain normal amounts of unbound diuretic at the site of action. Further studies are needed to clarify this issue.
14 D. C. Brater
[18] Verbeeck RK, Patwardhan RV, Villeneuve JP, Wilkinson GR, Branch RA. Furosemide disposition in cirrhosis. Clin Pharmacol Ther 1982; 31:719-25. [19] Keller E, Hoppe-Seyler G, Mumm R, Schollmeyer P. Influence of hepatic cirrhosis and end-stage renal disease on pharmacokinetics and pharmacodynamics of furosemide. Eur J Clin Pharmacol 1981; 20: 27-33. [20] Brater DC, Seiwell R, Anderson S, Burdette A, Dehmer GJ, Chennavasin P. Absorption and disposition of furosemide in congestive heart failure. Kidney Int 1982; 22: 171-6. [21] Brater DC, Chennavasin P, Seiwell R. Furosemide in patients with heart failure. Shift of the dose-response relationship. Clin Pharmacol Ther 1980; 28: 182-6. [22] Rane A, Villeneuve JP, Stone WJ, Nies AS, Wilkinson GR, Branch RA. Plasma binding and disposition of furosemide in the nephrotic syndrome and in uremia. Clin Pharmacol Ther 1978; 24: 199-207. [23] Keller E, Hoppe-Seyler G, Schollmeyer P. Disposition and diuretic effect of furosemide in the nephrotic syndrome. Clin Pharmacol Ther 1982; 32:442-9. [24] Green TP, Mirkin BL. Resistance of proteinuric rats to furosemide: Urinary drug protein binding as a determinant of drug effect. Life Sci 1980; 26:623-30. [25] Kirchner KA, Voelker JR, Brater DC. Intratubular albumin blunts the response to furosemide — a mechanism for diuretic resistance in the nephrotic syndrome. J Pharmacol Exp Ther 1990; 252: 1097-101. [26] Kirchner KA, Voelker JR, Brater DC. Binding inhibitors restore furosemide potency in tubule fluid containing albumin. Kidney Int 1991; 40: 418-24. [27] Smith DE, Hyneck ML, Berardi RR, Port FK. Urinary protein binding, kinetics, and dynamics of furosemide in nephrotic patients. J Pharm Sci 1985; 74: 603-7.
Downloaded from by guest on July 12, 2015
[7] Fredrick MJ, Pound DC, Hall SD, Brater DC. Furosemide absorption in patients with cirrhosis. Clin Pharmacol Ther 1991; 49: 241-7. [8] Voelker JR, Brown-Cartwright D, Anderson S, el al. Comparison of loop diuretics in patients with chronic renal insufficiency: Mechanism of difference in response. Kidney Int 1987; 32: 572-8. [9] Brater DC. Clinical pharmacology of loop diuretics. Drugs 1991;41(Suppl3): 14-22. [10] Verbeeck RK, Gerkens JF, Wilkinson GR, Branch RA. Disposition of furosemide in functionally hepatectomized dogs. J Pharmacol ExpTher 1981; 216:479-83. [11] Marcantonio LA, Auld WHR, Murdoch WR, Purohit R, Skellern GG, Howes CA. The pharmacokinetics and pharmacodynamics of the diuretic bumetanide in hepatic and renal disease. Br J Clin Pharmacol 1983; 15: 245-52. [12] Chennavasin P, Seiwell R, Brater DC, Liang WMM. Pharmacodynamic analysis of the furosemide-probenecid interaction in man. Kidney Int 1979; 16: 187-95. [13] Brater DC, Anderson SA, Brown-Cartwright D. Response to furosemide in chronic renal insufficiency: Rationale for limited doses. Clin Pharmacol Ther 1986; 40: 134-9. [14] Villeneuve JP, Verbeeck RK, Wilkinson GR, Branch RA. Furosemide kinetics and dynamics in patients with cirrhosis. Clin Pharmacol Ther 1986; 40: 14-20. [15] Gonzalez G, Arancibia A, Rivas MI, Caro P, Antezana C. Pharmacokinetics of furosemide in patients with hepatic cirrhosis. Eur J Clin Pharmacol 1982; 22: 315-20. [16] Traeger A, Hantze R, Penzlin M, et al. Pharmacokinetics and pharmacodynamic effects of furosemide in patients with liver cirrhosis. Int J Clin Pharmacol Ther Toxicol 1985; 23: 129-33. [17] Fuller R, Hoppel C, Ingalls ST. Furosemide kinetics in patients with hepatic cirrhosis with ascites. Clin Pharmacol Ther 1981; 30:461-7
Clinical pharmacology of loop diuretics in health and disease D. C. BRATER
Department of Medicine, Indiana University School of Medicine, U.S.A. KEY WORDS: Diuretics, comparative pharmacokinetics, comparative pharmacodynamics, renal failure, liver failure.
Downloaded from by guest on July 12, 2015
There are differences in metabolism and excretion of the loop diuretics which extrapolate to differences in pharmcokinetic behaviour in different disease states. For example, furosemide is eliminated in equal portions by renal and non-renal routes; the non-renal route involves primarily glucuronidation. Both renal and non-renal pathways are impaired during renal insufficiency, such that the elimination half-life is prolonged considerably in this disease state. In contrast, there seems to be little change in disposition in patients with liver disease. Bumetanide and torasemide have non-renal elimination pathways via the hepatic cytochrome system. In patients with renal insufficiency these non-renal pathways of elimination are left intact so that there is little prolongation of half-life in such patients. In contrast, in patients with liver disease or with congestive hepatopathy, there is impairment in non-renal elimination so that relatively more drug appears in the urine. With all loop diuretics, response is governed by the amount of drug appearing in the urine. By assessing the relationship between urinary excretion rates of the diuretic and sodium excretion rate, a maximal response which amounts to a fractional excretion of sodium of approximately 20% can be defined. Thus it is possible to define a maximal dose as the amount of drug necessary to attain a fractional excretion of sodium of 20%. Studies in different disease states using escalating doses can thereby use this relationship to define a dose above which little is to be gained in terms of therapeutic efficacy. Analysing the relationship between amount of diuretic in the urine and response also allows assessment of different mechanisms by which resistance to diuretics occurs in different disease states. For example, in patients with renal insufficiency, the primary mechanismfor resistance is diminished delivery ofdrug to the site ofaction. As such, administration of large doses ofdrug to deliver 'normal' amounts ofdiuretic into the urine result in the same response as occurs in subjects with normal renal function. However, this response is in terms of the fractional excretion of sodium meaning that the total amount of sodium eliminated in the urine is still diminished as a result of decreasedfilteredsodium. In contrast, in patients with liver disease or with congestive heart failure even when normal amounts of diuretic reach the urinary site of action there is diminished response which may be accountedfor by either increased proximal or increased distal tubular reabsorption of sodium. In such syndromes, frequent administration of drug is needed to attain the desired cumulative response and, in addition, combination drug therapy can be helpful by inhibiting sodium reabsorption at several distinct sites of the nephron. This latter strategy can be particularly helpful in subjects who have received long-term loop diuretic therapy who likely have hypertrophy of distal nephrons which are thereby able to reabsorb more sodium and diminish overall response. These distal nephron sites can be inhibited by thiazide diuretics accounting for the synergistic response that often occurs with a combination of loop and thiazide diureticsIntroduction characterized in healthy subjects to use as a comparison Recent studies have allowed elucidation of disease- t 0 patients. Pharmacokinetic parameters are generally induced effects on pharmacokinetics and pharmaco- similar (Table 1), excepting the somewhat lower clearance dynamics of loop diuretics. Such data have permitted a a n d concomitantly longer half-life of torasemide"-3". greater understanding of mechanisms for the diminished S " " ^ Perusing these pharmacokinetic parameters, howresponse to diuretics that occurs in most oedematous dis- e v e r ' ' g n o r e s differences among these drugs that become orders. Mechanisms differ in different clinical conditions, ™POrtant in disease states. For example the elimination and their definition has allowed derivation of rational ^alf-hfe reported for furosem.de is that after intravenous strategies for treatment. The goal of this review is to d o s i n S" A f t e r o r a I d o s i n 8 t h e t e r m i n a l h a l f " l l f e o f f u r o s " present an overview of the clinical pharmacology in health e u m i d e «J. 0 "* 6 ! l * a n i n t r a v j! n °"f d o s ' n « a n d " P " * * " * and disease of furosemide, bumetanide, and torasemide, t h J h a l f -' l f e ?ff absorption^. Thus, the time course of the three loop diuretics that have been studied in most absorption of furosemide is slower than that of bumet^ {L anide and torasemide. Moreover, in the oedematous disorders of congestive heart failure (CHF) and cirrhosis, the absorption of furosemide is further slowed'5"71. This Pharmacokinetics of loop diuretics in health and disease c ^ e s the terminal half-life to be yet longer and could The pharmacokinetics and pharmacodynamics of lead to the misinterpretation that elimination (rather than furosemide, bumetanide, and torasemide have been absorption) is slowed. This phenomenon does not seem to occur with bumetanide and torasemide. C
^^^^tll^%r^Ts^H7^X
IN 46202, U.S.A. 0I95-668X/92/0GOO10+05 $08.00/0
Pharmacokinetic parameters also do not stress differences in metabolism among these loop diuretics. The © 1992 The European Society of Cardiology
Clinical pharmacology of loop diuretics 11
Table 1 Pharmacokinetics offurosemide, bumetanide, and torasemide
Bioavailability (%) Volume of distribution (1. kg"1) Clearance (ml. min"'. kg-') Half-life (h) Fraction of intravenous dose excreted unchanged (%)
Table 2
Furosemide
Bumetanide
Torasemide
50 016 2-2 10 60
80 017 26 1-2 65
80 0-16 08 3-0 20
Effects of liver disease on the pharmacokinetics of bumetanide and torasemide
Bioavailability (%) Volume of distribution (1. k g ' ) Clearance (ml. min"'. kg"1) Half-life (h) Fraction of intravenous dose excreted unchanged (%)
Bumetanide
Torasemide
95 017 0-6 2-3 70
95 0-34 0-5 8-4 27
diminished non-renal clearance with preservation of renal clearance; half-life concomitantly doubles1"1. The combination of these effects causes a small increase in the amount of diuretic appearing in the urine, since there is a higher plasma concentration of drug (owing to decreased total clearance) available for renal excretion. The duration of effect of bumetanide in cirrhotic patients would be expected to be prolonged concomitant with the increased half-life. The effect of hepatic disease on torasemide pharmacokinetics differs somewhat from that with bumetanide. Non-renal clearance is similarly decreased, but renal clearance increases such that substantially more drug appears in the urine. Half-life remains about the same as in healthy subjects so that the duration of effect is not prolonged. These pharmacokinetic characteristics of furosemide, bumetanide, and torasemide allow predictions as to effects of disease on diuretic response. Pharmacodynamics of loop diuretics in health and disease
All loop diuretics are highly protein bound (about 98% to albumin)11'31 and have been shown in both animals and man to block renal sodium reabsorption from the lumenal rather than the peritubular side of the nephron"21. The high degree of protein binding precludes access of meaningful amounts of a loop diuretic to the lumen via filtration at the glomerulus (even in hypoalbuminemic states); the major mode of entry into the urine is thus by active secretion via the organic acid secretory pump of the proximal tubule. After secretion, diuretic travels with the tubular fluid to the thick ascending limb of the loop of Henle, where it inhibits the Na + -K + -2C1" reabsorptive pump. This nephron site normally reabsorbs about 20% of filtered sodium. Interestingly, maximal doses of loop
Downloaded from by guest on July 12, 2015
non-renal elimination of furosemide occurs for the most part via glucuronidation, a component of which may be renal""31. Diminished renal function not only decreases the renal clearance of furosemide, but also decreases non-renal clearance (by unknown mechanisms). Consequently, total clearance offurosemide decreases substantially, and elimination half-life increases to as long as 4 to 6 h in patients with severe renal insufficiency"-2'891. In contrast, the non-renal clearance of bumetanide and torasemide appears to occur via the mixed function oxygenases of the liver, a pathway that is not influenced by renal disease18'1. Consequently, as one would predict, non-renal clearance of both of these drugs is not affected by renal disease and, in fact, this route of elimination compensates for the decreased renal clearance sufficiently that the elimination half-life of bumetanide and torasemide is not prolonged in patients with renal insufficiency1891. Similarly, duration of action is not prolonged, and accumulation with chronic dosing is less likely to occur compared to furosemide. These different routes of metabolism also impact upon the influence of hepatic disease on the pharmacokinetics of loop diuretics. With furosemide, neither hepatic cirrhosis nor congestive hepatopathy from CHF affects pharmacokinetics171. This may be because non-renal clearance offurosemide occurs at extra-hepatic sites (e.g. renal glucuronidation). Though not carefully studied in man, the observation that hepatectomy of dogs has no effect on non-renal clearance offurosemide supports this notion"01. With bumetanide and torasemide, bioavailability increases in patients with cirrhosis from about 80% to 95-100%(l", perhaps due to diminished first-pass elimination. This magnitude of change is unlikely to be clinically important. Cirrhosis, however, has important effects on the pharmacokinetics of bumetanide and torasemide (Table 2). Total clearance of bumetanide decreases due to
12 D. C. Brater
Downloaded from by guest on July 12, 2015
Administering a dose of loop diuretic sufficient to compensate for the changes in pharmacokinetics that occur in Bumetanide Furosemide the particular clinical condition may still not elicit the expected response if there are disease-induced changes in pharmacodynamics; namely, a change in the relationsship between the amount of drug at the active site and response. Recent data have shown that remaining 1-0 W KK) 50 100 nephrons in patients with renal insufficiency respond Diuretic excretion rate (/ig.min"1) normally to amounts of diuretic reaching them1813'. If a Figure 1 Schematic representation of therelationshipbetween the sufficient dose of the diuretic is administered, maximal urinary excretion rate of a loop diuretic and response showing the response actually exceeds a FE Nl of 20%, likely due to relative potencies of furosemide, bumetanide, and torasemide. the increased reabsorptive load of remnant nephrons in patients with renal disease1'3'. Fundamentally, however, the pharmacodynamics of response to loop diuretics in diuretics cause excretion of 20% of filtered sodium patients with renal insufficiency is normal. (FE Nl = 20%) in healthy volunteers, indicating that loop Even though remaining nephrons respond satisfacdiuretics are able to completely block solute reabsorption torily when sufficient doses of loop diuretics are adminisin this nephron segment. tered, overall response is greatly diminished compared to Since loop diuretics exert their effects at the lumenal subjects with normal renal function18'. In such patients, surface of the nephron, it is amounts of diuretic in the however, it is filtered sodium that limits response rather urine that best correlate to response, as opposed to than the dynamics of interaction between the diuretic and concentrations in plasma'121. The relationship between tubular solute reabsorption. For example, a maximal urinary diuretic and natriuretic response is characterized FE excretion of 20% in a subject with normal renal Na by a sigmoid-shaped curve that fits the so-called Sigmoid function results in a peak sodium excretion rate of about E ^ model (Fig. 1). Each loop diuretic has the same 3 mEq . min"'. A similar response in terms of FE in a Na maximal response ( E J and similar slopes (y), but they patient with severe renal insufficiency results in a peak differ in terms of potency ( E D ^ (Table 3), accounting for sodium excretion rate of only about 0-3 mEq. min" 1 the differing doses necessary to achieve the same overall purely on the basis of diminished filtered sodium. Thus, response1''. while remnant nephrons may respond 'normally' in In assessing the effects of disease on response to loop patients with renal insufficiency, the overall natriuresis diuretics, one must examine not only the effect of the clini- that ensues is limited by the decreased filtered load of cal condition on the pharmacokinetics of the diuretic, but solute. also the effect on pharmacodynamics. Two prototypic Thus, in patients with renal insufficiency, the pharmadiseases, renal insufficiency and hepatic cirrhosis, will be cokinetics of loop diuretics are altered, and these changes discussed and related to the pharmacodynamics that have have been characterized sufficiently to allow accurate estibeen delineated in healthy subjects. mation of the doses needed to attain adequate amounts of diuretic at the tubular site of action. The pharmacodynamics of loop diuretics are normal in this condition, Renal disease though the overall natriuretic response is limited by As would be predicted, patients with renal insufficiency diminished filtered sodium. have diminished renal clearance of loop diuretics due to decreased numbers of functioning nephrons and additionally to blockade of secretion of diuretics, into the Hepatic disease lumen at the proximal tubule by accumulated endogenous Whereas renal insufficiency typifies a condition with organic acids of azotemia1'"3'. Thus, relative to dose, a altered pharmacokinetics of loop diuretics, hepatic cirsmaller amount of diuretic reaches the lumen, and it is not rhosis is a disease with altered pharmacodynamics. In surprising that large doses are needed to attain effective patients with cirrhosis but preserved renal function, the amounts at the site of action, particularly in patients with renal clearance of loop diuretics is normal, the exception severe renal insufficiency18'. being the increased renal clearance observed with toras4 191 With all loop diuretics, the relationship between the emide'' " . As a consequence, 'normal' amounts of level of renal function and renal clearance of the diuretic diuretic reach the tubular site of action, and no change has been defined'1"3'. For example, in patients with severe in dosing is needed in such patients unless concomitant renal insufficiency (creatinine clearance approximately decreases in renal function occur. 15 ml. min"'), renal clearance of furosemide is about oneWhen the relationship between amounts of diuretic in fifth that of a subject with normal renal function'8'. From urine and natriuretic response is assessed, however, the such data, one can readily estimate the dose escalation curve is shifted downward and to the right (Fig. 2).1201. necessary to attain 'normal' amounts of diuretic in the Maximal response is suppressed so that such patients will urine; thus, in a patient with severe renal insufficiency, five not attain the same magnitude of natriuresis as a healthy times the customary dose of furosemide would be needed subject. Thus, cumulative sodium excretion relative to a compared to a patient with normal creatinine clearance. dose is less in a patient with cirrhosis than in a healthy
Clinical pharmacology of loop diuretics 13
Table 3
Pharmacodynamic parameters of furosemide, bumelanide, and lorasemide Furosemide
Bumetanide
Torasemide
16 1-2 70
17 1-8 10
16 20 14
Maximal response (FE t o in %) Slope Excretion rate causing half-maximal response (jig. min~')
,
20 -
odiurr
£
/ Normal
(A
1 f
/
5
excretion
"o
o Fracl
JOI.
J'
/ Cirrhosis
/
Log urinary diuretic excretion rate
Figure 2 Schematic illustrating the change in pharmacodynamics characteristic of patients with cirrhosis.
4
r
Conclusion
The pharmacokinetics of loop diuretics in different disease states have, with the exception of nephrotic syndrome, been well clarified, allowing accurate estimation of doses needed to attain sufficient amounts at the tubular site of action to elicit a response. Disease states in which altered pharmacodynamics of loop diuretics occurs have been identified so that clinicians understand that even maximally effective doses of loop diuretics in such patients will have subnormal responses. Coupling knowledge of pharmacokinetics and pharmacodynamics allows understanding of mechanisms of diuretic action and development of rational treatment strategies that can be tailored to individual patients.
Heart failure
Two additional oedematous disorders, CHF and nephrotic syndrome, represent combinations of pharmacokinetic and pharmacodynamic changes in loop diuretics. In congestive heart failure, renal function is usually diminished to at least a small degree, and this causes changes in delivery of diuretic into the urine15-20'. By escalating the dose relative to changes in renal function, sufficient amounts of diuretic gain access to the site of action; however, response is subnormal'211. Similar to observations in patients with cirrhosis, the mechanism of the changed pharmacodynamics is unknown. Nephrotic syndrome
In patients with nephrotic syndrome and normal glomerular filtration rate, renal clearance of furosemide is
References [1] Brater DC. Diuretics. In: Williams RL, Brater DC, Mordenti J, eds. Rational Therapeutics: A Clinical Pharmacologic Guide for the Health Professional. New York: Marcel Dekker, 1989: 269-315. [2] Beermann B, Groschinsky-Grind M. Clinical pharmacokinetics of diuretics. Clin Pharmacokinet 1980; 5: 221-45. [3] Hammarlund-Udenaes M, Benet LZ. Furosemide pharmacokinetics and pharmacodynarmcs in health and disease—An update. J Pharmacokinet Biopharm 1989; 17: 1-46. [4] Hammarlund MM, Paalzow LK, Odlind B. Pharmacokinetics of furosemide in man after intravenous and oral administration. Application of moment analysis. Eur J Clin Pharmacol 1984; 26: 197-207. [5] Brater DC, Day B, Burdette A, Anderson S. Bumetanide and furosemide in heart failure. Kidney Int 1984; 26: 183-9. [6] Vasko MR, Cartwright DB, Knochel JP, Nixon JB, Brater DC. Furosemide absorption altered in decompensated congestive heart failure. Ann Intern Med 1985; 102:314-8.
Downloaded from by guest on July 12, 2015
•
counterpart, even though the dose is sufficient to attain maximal response in that patient. The mechanism for altered pharmacodynamics is unknown. Increased reabsorption of solute in the proximal nephron could theoretically cause such an effect, as could increased distal reabsorption of solute. These possibilities should be amenable to testing by assessing whether co-administration of diuretics blocking proximal and/or distal sites restores response toward normal. It is also possible that disease-induced abnormality at the loop of Henle itself is causal. In summary, hepatic cirrhosis is a disease wherein pharmacokinetics of loop diuretics (at least in terms of delivery of drug to the site of action) are normal, and altered response is due to changed pharmacodynamics, the mechanism of which is unknown.
normal'22-23'. However, it is now clear, at least in animal studies, that the loop diuretic reaching the tubular lumen binds to the albumin that has leaked into the lumen through the glomerulus124-25'. This binding renders the diuretic inactive; activity can be restored in experimental animals by displacing the diuretic from albumin'261. Whether such a strategy might be employed clinically is unknown. In addition to this unique kinetic effect of intratubular albumin, it appears that there is also a pharmacodynamic abnormality wherein response assessed relative to unbound diuretic in the urine is subnormal in nephrotic syndrome'271. Again, the mechanisms of this change are unknown. Similarly, in man it is not clear how large a dose of diuretic must be administered to attain normal amounts of unbound diuretic at the site of action. Further studies are needed to clarify this issue.
14 D. C. Brater
[18] Verbeeck RK, Patwardhan RV, Villeneuve JP, Wilkinson GR, Branch RA. Furosemide disposition in cirrhosis. Clin Pharmacol Ther 1982; 31:719-25. [19] Keller E, Hoppe-Seyler G, Mumm R, Schollmeyer P. Influence of hepatic cirrhosis and end-stage renal disease on pharmacokinetics and pharmacodynamics of furosemide. Eur J Clin Pharmacol 1981; 20: 27-33. [20] Brater DC, Seiwell R, Anderson S, Burdette A, Dehmer GJ, Chennavasin P. Absorption and disposition of furosemide in congestive heart failure. Kidney Int 1982; 22: 171-6. [21] Brater DC, Chennavasin P, Seiwell R. Furosemide in patients with heart failure. Shift of the dose-response relationship. Clin Pharmacol Ther 1980; 28: 182-6. [22] Rane A, Villeneuve JP, Stone WJ, Nies AS, Wilkinson GR, Branch RA. Plasma binding and disposition of furosemide in the nephrotic syndrome and in uremia. Clin Pharmacol Ther 1978; 24: 199-207. [23] Keller E, Hoppe-Seyler G, Schollmeyer P. Disposition and diuretic effect of furosemide in the nephrotic syndrome. Clin Pharmacol Ther 1982; 32:442-9. [24] Green TP, Mirkin BL. Resistance of proteinuric rats to furosemide: Urinary drug protein binding as a determinant of drug effect. Life Sci 1980; 26:623-30. [25] Kirchner KA, Voelker JR, Brater DC. Intratubular albumin blunts the response to furosemide — a mechanism for diuretic resistance in the nephrotic syndrome. J Pharmacol Exp Ther 1990; 252: 1097-101. [26] Kirchner KA, Voelker JR, Brater DC. Binding inhibitors restore furosemide potency in tubule fluid containing albumin. Kidney Int 1991; 40: 418-24. [27] Smith DE, Hyneck ML, Berardi RR, Port FK. Urinary protein binding, kinetics, and dynamics of furosemide in nephrotic patients. J Pharm Sci 1985; 74: 603-7.
Downloaded from by guest on July 12, 2015
[7] Fredrick MJ, Pound DC, Hall SD, Brater DC. Furosemide absorption in patients with cirrhosis. Clin Pharmacol Ther 1991; 49: 241-7. [8] Voelker JR, Brown-Cartwright D, Anderson S, el al. Comparison of loop diuretics in patients with chronic renal insufficiency: Mechanism of difference in response. Kidney Int 1987; 32: 572-8. [9] Brater DC. Clinical pharmacology of loop diuretics. Drugs 1991;41(Suppl3): 14-22. [10] Verbeeck RK, Gerkens JF, Wilkinson GR, Branch RA. Disposition of furosemide in functionally hepatectomized dogs. J Pharmacol ExpTher 1981; 216:479-83. [11] Marcantonio LA, Auld WHR, Murdoch WR, Purohit R, Skellern GG, Howes CA. The pharmacokinetics and pharmacodynamics of the diuretic bumetanide in hepatic and renal disease. Br J Clin Pharmacol 1983; 15: 245-52. [12] Chennavasin P, Seiwell R, Brater DC, Liang WMM. Pharmacodynamic analysis of the furosemide-probenecid interaction in man. Kidney Int 1979; 16: 187-95. [13] Brater DC, Anderson SA, Brown-Cartwright D. Response to furosemide in chronic renal insufficiency: Rationale for limited doses. Clin Pharmacol Ther 1986; 40: 134-9. [14] Villeneuve JP, Verbeeck RK, Wilkinson GR, Branch RA. Furosemide kinetics and dynamics in patients with cirrhosis. Clin Pharmacol Ther 1986; 40: 14-20. [15] Gonzalez G, Arancibia A, Rivas MI, Caro P, Antezana C. Pharmacokinetics of furosemide in patients with hepatic cirrhosis. Eur J Clin Pharmacol 1982; 22: 315-20. [16] Traeger A, Hantze R, Penzlin M, et al. Pharmacokinetics and pharmacodynamic effects of furosemide in patients with liver cirrhosis. Int J Clin Pharmacol Ther Toxicol 1985; 23: 129-33. [17] Fuller R, Hoppel C, Ingalls ST. Furosemide kinetics in patients with hepatic cirrhosis with ascites. Clin Pharmacol Ther 1981; 30:461-7
