- Email: [email protected]
Dose-response study of intravenous torsemide in congestive heart failure
Dose-response study of intravenous torsemide in congestive heart failure In a double-blind dose-response study, 49 patients with New York Heart Association functional class III or IV heart failure were randomized to receive a single intravenous dose of 5, 10, or 20 mg torsemide or 40 mg furosemide. Torsemide produced dose-related decreases in body weight and increases in sodium and chloride excretion and urine volume. Wiih the 20 mg dose of torsemide and the 40 mg dose of furosemide, body weight decreased significantly relative to baseline, and total and fractional 24-hour urinary excretion of sodium, chloride, and potassium and urine volume increased significantly. The 10 mg torsemide dose also produced a significant increase in urine volume. The results indicate that intravenous torsemide is effective for the acute treatment of sodium and fluid retention resulting from moderate to severe congestive heart failure. (AM HEART J 1994;128:352-7.)
Robert J. Hariman, MD,av b Siobhan Bremner, RN, MPH,al b Eric K. Louie, MD,b William J. Rogers, MD,c John B. Kostis, MD,d Michael A. Nocero, MD,e and John P. Jones, MDf Chicago, Ill., Birmingham, Ala., New Brunswick, N. J., Orlando, Flu., and Hampton, Vu.
Loop diuretics are regarded as essential in medical therapy for congestive heart fai1ure.l Torsemide (known in Europe as torasemide) is a new loopdiuretic agent of the pyridine sulfonylurea class. It has been studied in both the intravenous and oral forms for the treatment of edema and sodium and fluid retention caused by congestive heart failure, renal insufficiency, or cirrhosis. The oral form has been studied for the treatment of hypertension. In patients with congestive heart failure, diuresis begins within 10 minutes after intravenous administration of torsemide and reaches a peak within the 1st hour.2 Torsemide is well absorbed after oral administration; bioavailability is approximately 80 % in healthy subjects.3 Liver metabolism is the major route of elimination of torsemide. 4*5 Therefore, renal insufficiency has relatively little effect on torsemide’s half-life.6> 7 Previous clinical studies have demonstrated that torsemide has diuretic activity8 and therapeutic’-l2 and hemodynamic13-16 effects similar to those of fuFrom the *University of Illinois, Chicago, and currently affiliated with bLoyola University Medical Center, Chicago; the %Jniversity of Alabama, Birmingham; the dUniversity of Medicine and Dentistry of New JerseyRobert Wood Johnson Medical School, New Brunswick; %entral Florida Cardiology Group, Orlando; and fTidewater Heart Institute, Inc., Hampton. Supported by a grant from Boehringer Mannheim Pharmaceuticals. Received for publication Aug. 20, 1993; accepted Nov. 19, 1993. Reprint requests: Robert J. Hariman, The Edward Hines VA Hospital, Hines: IL 60141. Copyright Q 1994,by Mosby-Year Book, Inc. 000%8703/94/$3.00 + 0 4/l/66687
352
rosemide in congestive heart failure. We investigated the pharmacodynamics, efficacy, and dose-response relation of single intravenous doses of torsemide in patients with congestive heart failure of New York Heart Association (NYHA) functional class III or IV who had peripheral edema. METHODS This investigation was a multicenter (five centers), randomized, double-blind, parallel-group study of a range of torsemide doses, with 40 mg intravenous furosemide as the comparison treatment. The five centers were the University of Illinois at Chicago; the University of Alabama, Birmingham; the University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, New Brunswick; Central Florida Cardiology Group, Orlando; and Tidewater Heart Institute, Hampton, Va. After each center’s institutional review board had approved the protocol, enrollment began. Forty-nine patients were admitted to the hospital for the 3-day (baseline, treatment, and posttreatment days) study. Enrollment criteria. After giving informed consent, patients 2 21 years of age were invited into the study if they met all of the following criteria: (1) congestive heart failure meeting the definition of NYHA class III or IV; (2) low ejection fraction (5 35%), measured by multiple-gated radionuclide ventriculogram; (3) at least moderate leg edema or ‘Z 3% weight gain during an optional qualification period; and (4) hepatojugular rel?ux, jugular venous distention, or basilar rales on physical examination. Patients with evidence of acute pulmonary edema, unstable angina, recent myocardial infarction, significant ventricular arrhythmias (> 7 to 10 premature ventricular contrac-
Volume Amehwl
128, Number 2 Heart Journal
Hariman
Table 1. Baseline demographic characteristics and body weight Torsemide (mg) 5 (n=12)
Gender 11(92%) Men Women 1 (S?L)
Furosemide 40 mg (n = 11)
lo fn = 12)
(n = 14)
9(75?) 3 (25W
12(86%) 2(14%‘0)
6(55%b) 5 (45%)
7 (58%) 4 (33%) 1(9$J)
8(57%) 6(43%) 0
6 (55%) 5(45%) 0
68 44-82
55 27-76
62 38-78
20
Race
White 9 (754,) Black 3(25”,) Other 0 Age(yd Mean 68 Range 52-82 Body weight(kg) Mean 86.7 Range 68.8-106.5
76.9 58.2-98.7
82.0 56.2-115.0
81.8 52.1-132.3
tions per minute) or atria1 arrhythmias or atria1fibrillation with ventricular responseof > 115 beats/min), digoxin toxicity, electrolyte imbalance,kidney or liver dysfunction, chronic obstructive pulmonary disease,urinary tract obstruction, mental illness, uncontrolled diabetes mellitus, hypersensitivity to sulfonamides, childbearing potential, morbid obesity, alcoholism, or drug addiction were excluded from the study. Concomitant medications were allowed, with the exception of diuretics, potassium-sparing drugs, aspirin, and nonsteroidal antiinflammatory drugs. Patients receiving potassiumsupplementswhen they entered the study were allowed to continue receiving them. For the treatment of congestive heart failure, captopril, enalapril, or lisinopril wasadministered to all participants starting at least 7 days before the study. Study plan. After study candidatesagreedto participate and signedthe informed consent,they underwent baseline qualification tests that included a complete history and physical examination, chest roentgenography (posteroanterior and lateral views), a clinical laboratory examination (complete blood count with differential count and measurement of serum electrolytes, magnesium,cholesterol, triglycerides, total protein, and alkaline phosphatase; liver and renal function tests; and urinalysis), and a 12-lead electrocardiogram. Twenty-four hours before the baseline phaseof the study, patients beganreceiving a diet containing 75 mEq sodium per day; this diet was continued throughout the study. On the first morning of the baseline phase, the investigator recorded vital signs and body weight and began 24-hour urine collection for baseline measurement of volume and electrolyte concentrations. Serum electrolytes and creatinine were measuredat 8:OO AM and 8:OOPM. Body weights weremeasuredagain at 2:00 PM and 8:OOPM. On the morning of the 2nd day of the study (treatment day), baselinevital signsand body weight were measured, and blood and urine were collected for electrolyte determination. Patients then were randomly assignedto receive a
et al.
353
singledoseof either intravenous torsemide (5,10, or 26mg) or intravenous furosemide (40 mg). Biinding of the investigator, the staff members,and the patient wasmaintained by each study center’s pharmacy. Patients then were administered the study medication diluted in 4 ml of normal saline, intravenously for a period of :j minutes. Vital signswere measuredand blood samplesdrawn for determination of serumelectrolytes and creatinine at 2,4,6,8, 12, and 24 hours after intravenous administration. A 24-hour urine collection was initiated and the collected samplethen partitioned into aliquots (0 t.o 2-2 to 4,4 to 8, 8 to 12, and 12 to 24 hours); the volume of eachaliquot,was measured,and sampleswere obtained for the measurement of electrolytes and creatinine. Body weight wasmeasured at 6,12, and 24 hours after intravenous administration of the study medication. After the 24-hour treatment period, patients underwent a complete physical examination; examination by electrocardiography and chest roentgenography; and a repeat of the clinical laboratory examination. Variables measured. Changesin body weight from the baselineperiod (immediately before dosing)to the end of treatment period (24 hours after dosing) were compared. Each time a patient wasweighed, we usedthe samescale and the patients wore the sameclothing, usually a hospital gown. Mean changesin fractional urinary excretions were calculated asthe ratio of the clearanceof the electrolyte to creatinine clearance and were expressed as percentage point differences. For example, an increase in fractional excretion from 1% to 3% was recorded as an increaseof 24;. Total urinary excretion of sodium, chloride, and potassiumand total urine volume were determined for the 24-hour period after drug administration. The baselinefor all analysesof electrolyte excretions was the 24-hour control value. Statistical methods. The mean change in body weight and the meanfractional and total urinary sodium,chloride, and potassium excretions within each treatment group were evaluated by two-way analysis of variance that included the effects of treatment, the study center, and their interaction. In addition to the within-group analysis, comparisonswere madeamongthe four treatment groups. Treatment group meansestimated by the analysis of variancegave equal weight to eachstudy center. The onsetand duration of action for the four treatment groups were assessedfrom graphs of meantotal electrolyte excretion versusthe midpoint of each collection interval. Whenever a patient wasunableto void during a given interval, the output during the subsequentinterval was averagedover t,he two intervals. RESULTS Forty-nine patients at five medical centers enrolled in and completed the study. Demographic characteristics and body weights are summarized in Table I. Congestive heart failure characteristics, signs and symptoms, and creatinine clearance data are summarized in Table II. The demographic and clinical characteristics and the distribution between NYHA classesIII and IV were similar among the four
354
August 1994
Hariman et al.
Table II. Baseline
American
congestive
(mg) Furosemide 40 mg (n = 11)
(n =5 1.2)
LVEF Mean Range NYHA functionalclass III IV Jugular venous distention
23.2 5-35
Edema None 1+ 2+ 3+ 4+ Hepatojugular reflux Digoxin use Potassium supplementation Creatinine clearance (ml/min) mean range ejection
fraction;
21.2 8-35
9 (75%) 3 (25%) 5 (42%) 11(92%)
Rales
Left ventricular
Journal
heart failure characteristics, signsand symptoms, and creatinine clearance Torsemide
LVEF,
Heart
9 (75%) 3 (25%) 6 (60%)
1(8%)
0 0
: (25%) 8 (67%) 0 3 (25%) 9 (75%) 4 (33%)
8 (67%) 3 (25%)
New York Heart
19.5 11-35
9 (64%) 5 (36%)
6 (55%) 4 (45%) 8 (73%) 6 (55%)
10 (71%)
8 (57%)
11(92%)
0 0 7 (50%)
3 (25%)
5 (36%) 2 (14%) 7 (50%) 13 (93%) 5 (36%)
75 39-121
46 23-76
1(8%) 1(9%‘0) 1.2 (100%)
56 7-171 NYHA,
24.4 10-34
0 0
6 (55%) 3 (27%) 2 (18%) 7 (64%) 11(100%70) 3 (27%) 45 5-111
Association.
Table Ill. Changefrom baseline* in body weight, electrolyte excretion, and urine volume (mean + SEM [nt]) Torsemide
5 Body weight (kg) 0.04 + 0.23 (12) 24-Hour fractional electrolyte excretion (%) Sodium 0.17 -r- 0.21 (12) Chloride 0.34 ” 0.33 (12) Potassium 2.19 rf: 2.35 (12) 24-Hour total electrolyte excretion (mEq) Sodium 29.8 -t 20.6 (12) Chloride 39.2 rt 23.3 (12) Potassium 12.3 + 6.1 (12) 24-Hour urine volume (ml) 262 t 208 (12)
(mg)
10
20
-0.14 t 0.23 (12)
-0.94 t 0.22$(13)
0.25 + 0.21 (12) 0.60 -+ 0.33 (12) -1.35 -+ 2.35 (12) 29.5 44.3 1.3 703
+ + f +
20.6 (12) 23.3 (12) 6.1 (12) 208s (12)
1.12 I!z 0.21$ (12) 1.97 + 0.33$ (12) 5.09 * 2.35(( (12) 136.9 166.9 20.2 1380
+ 20.6$(12) + 23.3$ (12) IL 6.15 (12) + 208$(12)
Furosemide 40 mg
-0.74 + 0.259 (11) 1.11 + 0.24$ (10)
2.09 * 0.45f (9) 5.78 + 2.62(((10) 88.8 108.6 16.1 858
-t * + f
23.0$ (10) 31.55 (9) 6.811(10) 2323 (10)
*Baseline body weight was measured immediately before administration of study medication. Baseline electrolyte excretions and urine volume were determined from the 24-hour collection period immediately before administration of study medication. tData from one patient in the 20 mg toraemide group was excluded from analysis of body weight and electrolyte excretions because he inadvertently was given furosemide 22 hours after the torsemide dose. Values for n in the 20 mg torsemide group and the furosemide group reflect missing urine data for a few patients. $Significant change from baseline @ < 0.001). @ignificant change from baseline @ < 0.01). I)Significant change from baseline @ < 0.05).
treatment groups. Although
creatinine clearance values varied widely, the mean values indicated that moderate renal impairment was common. Effectiveness. The effect of torsemide on fluid mobilization increased with increasing dose. Patients given 10 mg torsemide showed a small, statistically insignificant decrease from baseline in mean body weight. There were statistically significant decreases
in mean body weight in patients given 20mg torsemide (p < 0.001) or 40 mg furosemide (p < 0.01; Table III). The mean baseline values of body weight (measured immediately before dosing) differed by 9.8 kg between the heaviest and lightest treatment groups (Table I). When added as a covariate in the analysis, this baseline body weight difference did not have an appreciable effect on relative treatment differences.
Volume 128, Number 2 American Heart Journal
Harimnn
et al.
355
Chloride
Sodium
60 o-
0
TorsZ7ide
6
Dose
Dose
(mg)
(mgl
Volume
Potassium
3,000
2607
-
P
T
2,600200
6
Furo~mide
20
2,oOa
rf
-
1,6W-
-t
E l,Ooo-
600-
c
0 6
TArtide
Dose *
o-
6
20
Torsezde
FurosGmide
20
C.05
Furoszmide
Dose
(mg) vs.
Baseline
q
Baseline
cla
(mg) Drug
Effect
1. Mean 24-hour total excretion of electrolytes and urine volume after diuretic treatment in patients with congestiveheart failure NYHA functional classIII or IV. Baseline, Excretion during 24-hour baseline period immediately before administration of study drug; drug effect, increasein excretion during 24 hours after administration. Values obtained by two-way analysis of variance. Bars, standard error.
Fig.
Torsemide produced dose-related increases in fractional and total sodium and chloride excretion and in urine volume (Table III) but had no clear doserelated effect on fractional or total potassium excretion. The changes from baseline in 24-hour fractional and total sodium, chloride, and potassium excretions were statistically significant in both the 20 mg torsemide and the 40 mg furosemide treatment groups, whereas the changes in 24-hour urine volume were statistically significant in the groups receiving 10 or 20 mg torsemide or 40 mg furosemide. Total electrolyte excretion and urine volume during the 24 hours after treatment are depicted in Fig. 1. The effect of 20 mg torsemide on body weight and excretion of sodium and chloride was significantly greater than the effects of 5 or 10 mg torsemide. Although increases in fractional electrolyte excretion were similar for 20 mg torsemide and 40 mg furosemide, the decrease in body weight and the increases in total electrolyte excretion and urine volume were some-
what greater with the 20 mg dose of torsemide. These differences were not statistically significant. Duration of action. Intravenous torsemide and furosemide were similar in onset and duration of action (approximately 6 to 8 hours; Fig. 2). Most diuresis occurred within the first 4 hours after administration. Tolerability. Torsemide and furosemide were equally well tolerated. Minor adverse events, most of which probably were related to the patients’ disease rather than to diuretic treatment, included dizziness, mild arrhythmias, dyspnea, abdominal pain, and pleural effusion. In a 75-year-old woman treated with 10 mg torsemide, accelerated idioventricular heart rhythm developed. This rhythm had been observed in the patient before the study. Sinus rhythm was restored after treatment with lidocaine and quinidine gluconate. In none of the patients was serum potassium concentration < 3.5 mEq/L after treatment. Minor, transient changes in other clinical laboratory values were observed in this single-dose study.
356 Harimun et al.
American
August 1904 Heart Journal
30 -
Q
0
2
4
8
Midpoint GToR
12
of Collection 5 OTOR
lo
7~0~
24
Interval 20 +FUR
(hours) 40
Fig. 2. Mean excretion rate of sodium vs time after administration of diuretic agent in patients with congestive heart failure NYHA functional class III or IV. FUR, Furosemide; TOR, torsemide.
DISCUSSION
In patients with NYHA class III or IV (moderate to severe) heart failure, single intravenous doses of torsemide produced dose-related decreases in body weight and increases in urinary sodium and chloride excretion and in urine volume. The 24-hour effects on body weight and on sodium and chloride excretion were statistically significant with the 20 mg dose, whereas the increases in 24-hour urine volume were significant with both the 10 and 20 mg doses. Fractional and total potassium excretion did not increase in a dose-related manner. In the present study, the equilibration period consisted of a 24-hour lead-in phase followed by a 24-hour baseline phase. A longer equilibration period was not used because of the patients’ need for diuretic therapy for treatment of their edema. Thus true sodium balance may not have been achieved, a factor that may have contributed to variability in the data. The dose-response relation of loop-diuretic agents has been characterized by Brater.17 Because the site of action of loop diuretics is the luminal side of the
kidney tubule, the pharmacodynamic response is correlated with the urinary excretion rate of the drug. As the urinary excretion rate of loop diuretics increases, the pharmacodynamic effect, increased sodium excretion, also occurs. Brater et all8 showed that, like furosemide, torsemide has a threshold rate of excretion that must be achieved before sodium excretion increases above its basal rate. Above the threshold, increasing the torsemide excretion rate (by increasing the dose) will increase sodium excretion until a plateau is reached, at which additional increases in torsemide dose and excretion rate will no longer produce increases in sodium excretion. Although in the present study urinary excretion rates of torsemide were not measured, the results of analysis of 24-hour sodium excretion suggest that a 5 mg torsemide dose may be at the pharmacodynamic threshold in patients with moderate to severe congestive heart failure, whereas a 20 mg dose clearly is effective. Higher doses of torsemide would be expected to produce greater sodium excretion. The dose at which a plateau occurs remains to be defined. In healthy subjects, the pharmacodynamic effects
Volume 128, Number 2 American Neml Journal
of torsemide have been shown to increase linearly with dose over the range of 2.5 to 20 mg.3118 In the present study, the minimal response at 5 mg and the relatively small response at 10 mg suggest that these patients with moderate to severe congestive heart failure were less responsive to torsemide than were the healthy subjects. A decrease in responsiveness (a shift to the right in the pharmacodynamic response curve) in patients with congestive heart failure compared with responsiveness in healthy subjects also have been reported for furosemide.lg This decrease in responsiveness could be related to increased sodium retention (a feature of congestive heart failure) and to slower delivery of the drug to the kidney tubule because of renal impairment, which is evident from the low creatinine clearance values of the patients in this study. Despite decreased renal elimination, torsemide does not accumulate in these patients with reduced renal function because the drug is eliminated largely by hepatic metabolism.4y 5 In this study, the 20 mg dose of torsemide was somewhat more effective than 40 mg furosemide in decreasing body weight and increasing total sodium excretion. This finding is consistent with the results of two other, small studies of intravenous torsemide and furosemide in patients with heart failure.l2,20 In conclusion, this study demonstrated that i&ravenous torsemide is effective and well tolerated in the treatment of sodium and fluid retention from moderate to severe congestive heart failure. REFERENCES
1. Smith TW, Braunwald E, Kelley RA. The management of heart failure. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine. 3rd ed. Philadelphia: WB Saunders, 198&485-543. 2. Isbary J, Achhammer I, Wetzels E. The influence of 20 mg torasemide i.v. and 20 mg furosemide i.v. on hemodynamics and diureais in patients with high grade left heart failure. Prog Pharmacol Clin Pharmacol 1990;8:137-46. 3. Barr WH, Smith HL, Kames HT, Sica D, Vetticaden SJ, Prsad VK, Kramer WG, Scott DI, Lindberg SE. Torasemide doseproportionality of pharmacokinetica and pharmacodynamics. Prog Pharmacol Clin Pharmacol 1990;8:29-37. 4. Von Moellendorff E, Neugebauer G. Pharmacokinetics of oral
Harimn
5. 6. 7. 8.
9. 10. 11
12.
13.
14.
15.
16. 17. 18. 19. 20.
et al.
357
torasemide in patients with congestive heart failure. Prog Pharmacol Clin Pharmacol 1990,8:73-9. Neugebauer G, Besenfelder E, Von Moellendorff E. Pharmacokinetics and metabolism of torasemide in man. Arzneimittelforschung 1988;38:164-6. Knauf H, Spahn H, Mutschler E. The loop diuretic torasemide in chronic renal failure: pharmacokinetics and pharmacodynamics. Drugs 1991;41(sippl 3):23-34. Rudv D. Gehr T. Sica D. Brater D. Matzke G. Hevman G. Kramer ’W, Linciberg S. ‘Pharmacokinetics of ‘torsemide in chronic renal insufficiency. Clin Pharmacol Ther 1991; 49149. Herchuelz A, Deger F, Douchamps J, Ducame H, Broekhuysen J. Comparative pharmacodynamics of torasemide and furosemide in patients with oedema. Arzneimittelforschung 1988;38:180-3. Achhammer I, Haecker W, Glocke M. Efficacy and safety of torasemide in patients with chronic heart failure. Anneimittelforschung 1988;38:184-7. Duesing R, Piesche L. Second line therapy of congestive heart failure with torasemide. Prog Pharmacol Clin Pharmacol 1990;8:105-20. Stauch M, Stiehl L. Controlled, double-blind clinical trial on the efficacy and tolerance of torasemide in comparison with furosemide in patients with congestive heart failure: a multicenter studv. Proa Pharmacol Clin Pharmacol 1990:8:121-6. Scheen AJ,-Vanciombreucq JC, Delarge J, Luyckx AS. Diuretic activity of torasemide and furosemide in chronic heart failure: a comparative double blind cross-over study. Eur J Clin Pharmacol 1986;31:35-42. Fiehring H, Achhammer I. Influence of 10 mg torasemide i.v. and 20 mg furosemide i.v. on intracardiac pressures in patients with heart failure at rest and during exercise. Prog Pharmacol Clin Pharmacol 1990;8:97-104. Isbary J, Achhammer I, Wetzels E. The influence of 20 mg torasemide i.v. and 20 mg furosemide i.v. on hemodynamics and diuresis in patients with high grade left heart failure. Prog Pharmacol Clin Pharmacol 1990;8:137-46. Langbehn AI?, Achhammer I, Bolke T. Acute hemodynamic effects of 20 mg torasemide and 20 mg furosemide given intravenously to patients with congestive heart, failure. Prog Pharmacol Clin Pharmacol 1990;8:147-55. Podszus T, Piesche L. Effect of torasemide on pulmonary and cardiac haemodynamics after oral treatment if chronic beart failure. ProR Pharmacol Clin Pharmacol 1990:8:157-66. Brater DC.-Clinical pharmacology of loop d&t&-Drugs 1991;41:14-22. Brater CD, Leinfelder BS, Anderson SA. Clinical pharmacology of torasemide, a new loop diuretic. Clin Pharmacol Ther 1987;42:187-92. Brater DC, Chennavasin P, Seiwell R. Furosemide in patients with heart failure: shift in dose-response curves. Clin Pharmacol Ther 1980;28:182-6. Stroobandt R, Dodion L, Kesteloot H. Clinical efficacy of torasemide, a new diuretic agent, in patients with acute heart failure: a double blind comparison with furosemide. Arch Int Pharmacodyn Ther 1982;260:151-8.
Robert J. Hariman, MD,av b Siobhan Bremner, RN, MPH,al b Eric K. Louie, MD,b William J. Rogers, MD,c John B. Kostis, MD,d Michael A. Nocero, MD,e and John P. Jones, MDf Chicago, Ill., Birmingham, Ala., New Brunswick, N. J., Orlando, Flu., and Hampton, Vu.
Loop diuretics are regarded as essential in medical therapy for congestive heart fai1ure.l Torsemide (known in Europe as torasemide) is a new loopdiuretic agent of the pyridine sulfonylurea class. It has been studied in both the intravenous and oral forms for the treatment of edema and sodium and fluid retention caused by congestive heart failure, renal insufficiency, or cirrhosis. The oral form has been studied for the treatment of hypertension. In patients with congestive heart failure, diuresis begins within 10 minutes after intravenous administration of torsemide and reaches a peak within the 1st hour.2 Torsemide is well absorbed after oral administration; bioavailability is approximately 80 % in healthy subjects.3 Liver metabolism is the major route of elimination of torsemide. 4*5 Therefore, renal insufficiency has relatively little effect on torsemide’s half-life.6> 7 Previous clinical studies have demonstrated that torsemide has diuretic activity8 and therapeutic’-l2 and hemodynamic13-16 effects similar to those of fuFrom the *University of Illinois, Chicago, and currently affiliated with bLoyola University Medical Center, Chicago; the %Jniversity of Alabama, Birmingham; the dUniversity of Medicine and Dentistry of New JerseyRobert Wood Johnson Medical School, New Brunswick; %entral Florida Cardiology Group, Orlando; and fTidewater Heart Institute, Inc., Hampton. Supported by a grant from Boehringer Mannheim Pharmaceuticals. Received for publication Aug. 20, 1993; accepted Nov. 19, 1993. Reprint requests: Robert J. Hariman, The Edward Hines VA Hospital, Hines: IL 60141. Copyright Q 1994,by Mosby-Year Book, Inc. 000%8703/94/$3.00 + 0 4/l/66687
352
rosemide in congestive heart failure. We investigated the pharmacodynamics, efficacy, and dose-response relation of single intravenous doses of torsemide in patients with congestive heart failure of New York Heart Association (NYHA) functional class III or IV who had peripheral edema. METHODS This investigation was a multicenter (five centers), randomized, double-blind, parallel-group study of a range of torsemide doses, with 40 mg intravenous furosemide as the comparison treatment. The five centers were the University of Illinois at Chicago; the University of Alabama, Birmingham; the University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, New Brunswick; Central Florida Cardiology Group, Orlando; and Tidewater Heart Institute, Hampton, Va. After each center’s institutional review board had approved the protocol, enrollment began. Forty-nine patients were admitted to the hospital for the 3-day (baseline, treatment, and posttreatment days) study. Enrollment criteria. After giving informed consent, patients 2 21 years of age were invited into the study if they met all of the following criteria: (1) congestive heart failure meeting the definition of NYHA class III or IV; (2) low ejection fraction (5 35%), measured by multiple-gated radionuclide ventriculogram; (3) at least moderate leg edema or ‘Z 3% weight gain during an optional qualification period; and (4) hepatojugular rel?ux, jugular venous distention, or basilar rales on physical examination. Patients with evidence of acute pulmonary edema, unstable angina, recent myocardial infarction, significant ventricular arrhythmias (> 7 to 10 premature ventricular contrac-
Volume Amehwl
128, Number 2 Heart Journal
Hariman
Table 1. Baseline demographic characteristics and body weight Torsemide (mg) 5 (n=12)
Gender 11(92%) Men Women 1 (S?L)
Furosemide 40 mg (n = 11)
lo fn = 12)
(n = 14)
9(75?) 3 (25W
12(86%) 2(14%‘0)
6(55%b) 5 (45%)
7 (58%) 4 (33%) 1(9$J)
8(57%) 6(43%) 0
6 (55%) 5(45%) 0
68 44-82
55 27-76
62 38-78
20
Race
White 9 (754,) Black 3(25”,) Other 0 Age(yd Mean 68 Range 52-82 Body weight(kg) Mean 86.7 Range 68.8-106.5
76.9 58.2-98.7
82.0 56.2-115.0
81.8 52.1-132.3
tions per minute) or atria1 arrhythmias or atria1fibrillation with ventricular responseof > 115 beats/min), digoxin toxicity, electrolyte imbalance,kidney or liver dysfunction, chronic obstructive pulmonary disease,urinary tract obstruction, mental illness, uncontrolled diabetes mellitus, hypersensitivity to sulfonamides, childbearing potential, morbid obesity, alcoholism, or drug addiction were excluded from the study. Concomitant medications were allowed, with the exception of diuretics, potassium-sparing drugs, aspirin, and nonsteroidal antiinflammatory drugs. Patients receiving potassiumsupplementswhen they entered the study were allowed to continue receiving them. For the treatment of congestive heart failure, captopril, enalapril, or lisinopril wasadministered to all participants starting at least 7 days before the study. Study plan. After study candidatesagreedto participate and signedthe informed consent,they underwent baseline qualification tests that included a complete history and physical examination, chest roentgenography (posteroanterior and lateral views), a clinical laboratory examination (complete blood count with differential count and measurement of serum electrolytes, magnesium,cholesterol, triglycerides, total protein, and alkaline phosphatase; liver and renal function tests; and urinalysis), and a 12-lead electrocardiogram. Twenty-four hours before the baseline phaseof the study, patients beganreceiving a diet containing 75 mEq sodium per day; this diet was continued throughout the study. On the first morning of the baseline phase, the investigator recorded vital signs and body weight and began 24-hour urine collection for baseline measurement of volume and electrolyte concentrations. Serum electrolytes and creatinine were measuredat 8:OO AM and 8:OOPM. Body weights weremeasuredagain at 2:00 PM and 8:OOPM. On the morning of the 2nd day of the study (treatment day), baselinevital signsand body weight were measured, and blood and urine were collected for electrolyte determination. Patients then were randomly assignedto receive a
et al.
353
singledoseof either intravenous torsemide (5,10, or 26mg) or intravenous furosemide (40 mg). Biinding of the investigator, the staff members,and the patient wasmaintained by each study center’s pharmacy. Patients then were administered the study medication diluted in 4 ml of normal saline, intravenously for a period of :j minutes. Vital signswere measuredand blood samplesdrawn for determination of serumelectrolytes and creatinine at 2,4,6,8, 12, and 24 hours after intravenous administration. A 24-hour urine collection was initiated and the collected samplethen partitioned into aliquots (0 t.o 2-2 to 4,4 to 8, 8 to 12, and 12 to 24 hours); the volume of eachaliquot,was measured,and sampleswere obtained for the measurement of electrolytes and creatinine. Body weight wasmeasured at 6,12, and 24 hours after intravenous administration of the study medication. After the 24-hour treatment period, patients underwent a complete physical examination; examination by electrocardiography and chest roentgenography; and a repeat of the clinical laboratory examination. Variables measured. Changesin body weight from the baselineperiod (immediately before dosing)to the end of treatment period (24 hours after dosing) were compared. Each time a patient wasweighed, we usedthe samescale and the patients wore the sameclothing, usually a hospital gown. Mean changesin fractional urinary excretions were calculated asthe ratio of the clearanceof the electrolyte to creatinine clearance and were expressed as percentage point differences. For example, an increase in fractional excretion from 1% to 3% was recorded as an increaseof 24;. Total urinary excretion of sodium, chloride, and potassiumand total urine volume were determined for the 24-hour period after drug administration. The baselinefor all analysesof electrolyte excretions was the 24-hour control value. Statistical methods. The mean change in body weight and the meanfractional and total urinary sodium,chloride, and potassium excretions within each treatment group were evaluated by two-way analysis of variance that included the effects of treatment, the study center, and their interaction. In addition to the within-group analysis, comparisonswere madeamongthe four treatment groups. Treatment group meansestimated by the analysis of variancegave equal weight to eachstudy center. The onsetand duration of action for the four treatment groups were assessedfrom graphs of meantotal electrolyte excretion versusthe midpoint of each collection interval. Whenever a patient wasunableto void during a given interval, the output during the subsequentinterval was averagedover t,he two intervals. RESULTS Forty-nine patients at five medical centers enrolled in and completed the study. Demographic characteristics and body weights are summarized in Table I. Congestive heart failure characteristics, signs and symptoms, and creatinine clearance data are summarized in Table II. The demographic and clinical characteristics and the distribution between NYHA classesIII and IV were similar among the four
354
August 1994
Hariman et al.
Table II. Baseline
American
congestive
(mg) Furosemide 40 mg (n = 11)
(n =5 1.2)
LVEF Mean Range NYHA functionalclass III IV Jugular venous distention
23.2 5-35
Edema None 1+ 2+ 3+ 4+ Hepatojugular reflux Digoxin use Potassium supplementation Creatinine clearance (ml/min) mean range ejection
fraction;
21.2 8-35
9 (75%) 3 (25%) 5 (42%) 11(92%)
Rales
Left ventricular
Journal
heart failure characteristics, signsand symptoms, and creatinine clearance Torsemide
LVEF,
Heart
9 (75%) 3 (25%) 6 (60%)
1(8%)
0 0
: (25%) 8 (67%) 0 3 (25%) 9 (75%) 4 (33%)
8 (67%) 3 (25%)
New York Heart
19.5 11-35
9 (64%) 5 (36%)
6 (55%) 4 (45%) 8 (73%) 6 (55%)
10 (71%)
8 (57%)
11(92%)
0 0 7 (50%)
3 (25%)
5 (36%) 2 (14%) 7 (50%) 13 (93%) 5 (36%)
75 39-121
46 23-76
1(8%) 1(9%‘0) 1.2 (100%)
56 7-171 NYHA,
24.4 10-34
0 0
6 (55%) 3 (27%) 2 (18%) 7 (64%) 11(100%70) 3 (27%) 45 5-111
Association.
Table Ill. Changefrom baseline* in body weight, electrolyte excretion, and urine volume (mean + SEM [nt]) Torsemide
5 Body weight (kg) 0.04 + 0.23 (12) 24-Hour fractional electrolyte excretion (%) Sodium 0.17 -r- 0.21 (12) Chloride 0.34 ” 0.33 (12) Potassium 2.19 rf: 2.35 (12) 24-Hour total electrolyte excretion (mEq) Sodium 29.8 -t 20.6 (12) Chloride 39.2 rt 23.3 (12) Potassium 12.3 + 6.1 (12) 24-Hour urine volume (ml) 262 t 208 (12)
(mg)
10
20
-0.14 t 0.23 (12)
-0.94 t 0.22$(13)
0.25 + 0.21 (12) 0.60 -+ 0.33 (12) -1.35 -+ 2.35 (12) 29.5 44.3 1.3 703
+ + f +
20.6 (12) 23.3 (12) 6.1 (12) 208s (12)
1.12 I!z 0.21$ (12) 1.97 + 0.33$ (12) 5.09 * 2.35(( (12) 136.9 166.9 20.2 1380
+ 20.6$(12) + 23.3$ (12) IL 6.15 (12) + 208$(12)
Furosemide 40 mg
-0.74 + 0.259 (11) 1.11 + 0.24$ (10)
2.09 * 0.45f (9) 5.78 + 2.62(((10) 88.8 108.6 16.1 858
-t * + f
23.0$ (10) 31.55 (9) 6.811(10) 2323 (10)
*Baseline body weight was measured immediately before administration of study medication. Baseline electrolyte excretions and urine volume were determined from the 24-hour collection period immediately before administration of study medication. tData from one patient in the 20 mg toraemide group was excluded from analysis of body weight and electrolyte excretions because he inadvertently was given furosemide 22 hours after the torsemide dose. Values for n in the 20 mg torsemide group and the furosemide group reflect missing urine data for a few patients. $Significant change from baseline @ < 0.001). @ignificant change from baseline @ < 0.01). I)Significant change from baseline @ < 0.05).
treatment groups. Although
creatinine clearance values varied widely, the mean values indicated that moderate renal impairment was common. Effectiveness. The effect of torsemide on fluid mobilization increased with increasing dose. Patients given 10 mg torsemide showed a small, statistically insignificant decrease from baseline in mean body weight. There were statistically significant decreases
in mean body weight in patients given 20mg torsemide (p < 0.001) or 40 mg furosemide (p < 0.01; Table III). The mean baseline values of body weight (measured immediately before dosing) differed by 9.8 kg between the heaviest and lightest treatment groups (Table I). When added as a covariate in the analysis, this baseline body weight difference did not have an appreciable effect on relative treatment differences.
Volume 128, Number 2 American Heart Journal
Harimnn
et al.
355
Chloride
Sodium
60 o-
0
TorsZ7ide
6
Dose
Dose
(mg)
(mgl
Volume
Potassium
3,000
2607
-
P
T
2,600200
6
Furo~mide
20
2,oOa
rf
-
1,6W-
-t
E l,Ooo-
600-
c
0 6
TArtide
Dose *
o-
6
20
Torsezde
FurosGmide
20
C.05
Furoszmide
Dose
(mg) vs.
Baseline
q
Baseline
cla
(mg) Drug
Effect
1. Mean 24-hour total excretion of electrolytes and urine volume after diuretic treatment in patients with congestiveheart failure NYHA functional classIII or IV. Baseline, Excretion during 24-hour baseline period immediately before administration of study drug; drug effect, increasein excretion during 24 hours after administration. Values obtained by two-way analysis of variance. Bars, standard error.
Fig.
Torsemide produced dose-related increases in fractional and total sodium and chloride excretion and in urine volume (Table III) but had no clear doserelated effect on fractional or total potassium excretion. The changes from baseline in 24-hour fractional and total sodium, chloride, and potassium excretions were statistically significant in both the 20 mg torsemide and the 40 mg furosemide treatment groups, whereas the changes in 24-hour urine volume were statistically significant in the groups receiving 10 or 20 mg torsemide or 40 mg furosemide. Total electrolyte excretion and urine volume during the 24 hours after treatment are depicted in Fig. 1. The effect of 20 mg torsemide on body weight and excretion of sodium and chloride was significantly greater than the effects of 5 or 10 mg torsemide. Although increases in fractional electrolyte excretion were similar for 20 mg torsemide and 40 mg furosemide, the decrease in body weight and the increases in total electrolyte excretion and urine volume were some-
what greater with the 20 mg dose of torsemide. These differences were not statistically significant. Duration of action. Intravenous torsemide and furosemide were similar in onset and duration of action (approximately 6 to 8 hours; Fig. 2). Most diuresis occurred within the first 4 hours after administration. Tolerability. Torsemide and furosemide were equally well tolerated. Minor adverse events, most of which probably were related to the patients’ disease rather than to diuretic treatment, included dizziness, mild arrhythmias, dyspnea, abdominal pain, and pleural effusion. In a 75-year-old woman treated with 10 mg torsemide, accelerated idioventricular heart rhythm developed. This rhythm had been observed in the patient before the study. Sinus rhythm was restored after treatment with lidocaine and quinidine gluconate. In none of the patients was serum potassium concentration < 3.5 mEq/L after treatment. Minor, transient changes in other clinical laboratory values were observed in this single-dose study.
356 Harimun et al.
American
August 1904 Heart Journal
30 -
Q
0
2
4
8
Midpoint GToR
12
of Collection 5 OTOR
lo
7~0~
24
Interval 20 +FUR
(hours) 40
Fig. 2. Mean excretion rate of sodium vs time after administration of diuretic agent in patients with congestive heart failure NYHA functional class III or IV. FUR, Furosemide; TOR, torsemide.
DISCUSSION
In patients with NYHA class III or IV (moderate to severe) heart failure, single intravenous doses of torsemide produced dose-related decreases in body weight and increases in urinary sodium and chloride excretion and in urine volume. The 24-hour effects on body weight and on sodium and chloride excretion were statistically significant with the 20 mg dose, whereas the increases in 24-hour urine volume were significant with both the 10 and 20 mg doses. Fractional and total potassium excretion did not increase in a dose-related manner. In the present study, the equilibration period consisted of a 24-hour lead-in phase followed by a 24-hour baseline phase. A longer equilibration period was not used because of the patients’ need for diuretic therapy for treatment of their edema. Thus true sodium balance may not have been achieved, a factor that may have contributed to variability in the data. The dose-response relation of loop-diuretic agents has been characterized by Brater.17 Because the site of action of loop diuretics is the luminal side of the
kidney tubule, the pharmacodynamic response is correlated with the urinary excretion rate of the drug. As the urinary excretion rate of loop diuretics increases, the pharmacodynamic effect, increased sodium excretion, also occurs. Brater et all8 showed that, like furosemide, torsemide has a threshold rate of excretion that must be achieved before sodium excretion increases above its basal rate. Above the threshold, increasing the torsemide excretion rate (by increasing the dose) will increase sodium excretion until a plateau is reached, at which additional increases in torsemide dose and excretion rate will no longer produce increases in sodium excretion. Although in the present study urinary excretion rates of torsemide were not measured, the results of analysis of 24-hour sodium excretion suggest that a 5 mg torsemide dose may be at the pharmacodynamic threshold in patients with moderate to severe congestive heart failure, whereas a 20 mg dose clearly is effective. Higher doses of torsemide would be expected to produce greater sodium excretion. The dose at which a plateau occurs remains to be defined. In healthy subjects, the pharmacodynamic effects
Volume 128, Number 2 American Neml Journal
of torsemide have been shown to increase linearly with dose over the range of 2.5 to 20 mg.3118 In the present study, the minimal response at 5 mg and the relatively small response at 10 mg suggest that these patients with moderate to severe congestive heart failure were less responsive to torsemide than were the healthy subjects. A decrease in responsiveness (a shift to the right in the pharmacodynamic response curve) in patients with congestive heart failure compared with responsiveness in healthy subjects also have been reported for furosemide.lg This decrease in responsiveness could be related to increased sodium retention (a feature of congestive heart failure) and to slower delivery of the drug to the kidney tubule because of renal impairment, which is evident from the low creatinine clearance values of the patients in this study. Despite decreased renal elimination, torsemide does not accumulate in these patients with reduced renal function because the drug is eliminated largely by hepatic metabolism.4y 5 In this study, the 20 mg dose of torsemide was somewhat more effective than 40 mg furosemide in decreasing body weight and increasing total sodium excretion. This finding is consistent with the results of two other, small studies of intravenous torsemide and furosemide in patients with heart failure.l2,20 In conclusion, this study demonstrated that i&ravenous torsemide is effective and well tolerated in the treatment of sodium and fluid retention from moderate to severe congestive heart failure. REFERENCES
1. Smith TW, Braunwald E, Kelley RA. The management of heart failure. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine. 3rd ed. Philadelphia: WB Saunders, 198&485-543. 2. Isbary J, Achhammer I, Wetzels E. The influence of 20 mg torasemide i.v. and 20 mg furosemide i.v. on hemodynamics and diureais in patients with high grade left heart failure. Prog Pharmacol Clin Pharmacol 1990;8:137-46. 3. Barr WH, Smith HL, Kames HT, Sica D, Vetticaden SJ, Prsad VK, Kramer WG, Scott DI, Lindberg SE. Torasemide doseproportionality of pharmacokinetica and pharmacodynamics. Prog Pharmacol Clin Pharmacol 1990;8:29-37. 4. Von Moellendorff E, Neugebauer G. Pharmacokinetics of oral
Harimn
5. 6. 7. 8.
9. 10. 11
12.
13.
14.
15.
16. 17. 18. 19. 20.
et al.
357
torasemide in patients with congestive heart failure. Prog Pharmacol Clin Pharmacol 1990,8:73-9. Neugebauer G, Besenfelder E, Von Moellendorff E. Pharmacokinetics and metabolism of torasemide in man. Arzneimittelforschung 1988;38:164-6. Knauf H, Spahn H, Mutschler E. The loop diuretic torasemide in chronic renal failure: pharmacokinetics and pharmacodynamics. Drugs 1991;41(sippl 3):23-34. Rudv D. Gehr T. Sica D. Brater D. Matzke G. Hevman G. Kramer ’W, Linciberg S. ‘Pharmacokinetics of ‘torsemide in chronic renal insufficiency. Clin Pharmacol Ther 1991; 49149. Herchuelz A, Deger F, Douchamps J, Ducame H, Broekhuysen J. Comparative pharmacodynamics of torasemide and furosemide in patients with oedema. Arzneimittelforschung 1988;38:180-3. Achhammer I, Haecker W, Glocke M. Efficacy and safety of torasemide in patients with chronic heart failure. Anneimittelforschung 1988;38:184-7. Duesing R, Piesche L. Second line therapy of congestive heart failure with torasemide. Prog Pharmacol Clin Pharmacol 1990;8:105-20. Stauch M, Stiehl L. Controlled, double-blind clinical trial on the efficacy and tolerance of torasemide in comparison with furosemide in patients with congestive heart failure: a multicenter studv. Proa Pharmacol Clin Pharmacol 1990:8:121-6. Scheen AJ,-Vanciombreucq JC, Delarge J, Luyckx AS. Diuretic activity of torasemide and furosemide in chronic heart failure: a comparative double blind cross-over study. Eur J Clin Pharmacol 1986;31:35-42. Fiehring H, Achhammer I. Influence of 10 mg torasemide i.v. and 20 mg furosemide i.v. on intracardiac pressures in patients with heart failure at rest and during exercise. Prog Pharmacol Clin Pharmacol 1990;8:97-104. Isbary J, Achhammer I, Wetzels E. The influence of 20 mg torasemide i.v. and 20 mg furosemide i.v. on hemodynamics and diuresis in patients with high grade left heart failure. Prog Pharmacol Clin Pharmacol 1990;8:137-46. Langbehn AI?, Achhammer I, Bolke T. Acute hemodynamic effects of 20 mg torasemide and 20 mg furosemide given intravenously to patients with congestive heart, failure. Prog Pharmacol Clin Pharmacol 1990;8:147-55. Podszus T, Piesche L. Effect of torasemide on pulmonary and cardiac haemodynamics after oral treatment if chronic beart failure. ProR Pharmacol Clin Pharmacol 1990:8:157-66. Brater DC.-Clinical pharmacology of loop d&t&-Drugs 1991;41:14-22. Brater CD, Leinfelder BS, Anderson SA. Clinical pharmacology of torasemide, a new loop diuretic. Clin Pharmacol Ther 1987;42:187-92. Brater DC, Chennavasin P, Seiwell R. Furosemide in patients with heart failure: shift in dose-response curves. Clin Pharmacol Ther 1980;28:182-6. Stroobandt R, Dodion L, Kesteloot H. Clinical efficacy of torasemide, a new diuretic agent, in patients with acute heart failure: a double blind comparison with furosemide. Arch Int Pharmacodyn Ther 1982;260:151-8.