Pharmacokinetics
of Digoxin and Digitoxin in Patients
Undergoing Hemodialysis
FREDRIC 0. FINKELSTEIN, M.D. New Haven, Connecticut JOHN A. GOFFINET, M.D. ERNEST0
D. HENDLER, M.D.
West Haven, Connecticut JOHN LINDENBAUM, M.D. New York, New York
From the Department of Medicine, West Haven Veterans Administration Hospital, West Haven, Connecticut 065 16; the Yale University Medical School, New Haven, Connecticut 06510; and the Department of Medicine, Harlem Hospital Center, College of Physicians and Surgeons, Columbia University, New York, New York 10037. This study was supported in part by grants from the Connecticut Heart Association (Grant 519-(1972)), the Health Research Council of the City of New York (U-2230) and the Burroughs-Wellcome Co. Requests for reprints shoufd be addressed to Dr. Fredric 0. Finkelstein, Department of Internal Medicine, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut 06510. Manuscript accepted June 4, 1974.
The pharmacokinetics of digoxin and digitoxin in patients undergoing long-term hemodiaiysis were examined to determine which is the preferred cardiac giycoside in this patient population. Absorption curves from 0 to 24 hours afler an oral dose of digitoxin were similar in dialyzed patients and in control patients. Serum giycoside concentrations after an oral dose of digoxin were higher in dialyzed patients than in control patients, significantly so from 2 to 24 hours, reflecting the absence of the predominantly renal route of excretion of digoxin. When nine dialyzed patients were placed on a maintenance dose of digoxin, 0.125 mg 5 days a week, serum levels plateaued at 30 days at a mean concentration (& SE) of 0.84 f 0.05 ng/ml. Maintenance therapy with 0.1 mg digitoxin 5 days a week resuited in stabilization of serum levels within 30 days at a mean concentration of 19 f 1 ng/ml. Variability in the serum glycoside concentrations was determined after stabilization of levels during 2 to 19 week follow-up periods with each drug. Variability in serum levels was somewhat increased during maintenance therapy with digitoxin. On the basis of the pharmacokinetic data obtained in this study, no clear cut preference for one giycoside over the other could be established. Congestive heart failure secondary to associated myocardial disease, hypertension and fluid overload frequently develops in patients with renal failure. Consequently, therapy with a cardiac glycoside, usually digoxin or digitoxin, is often prescribed. However, which of these two glycosides and what doses of the drugs are preferable in patients with chronic renal failure have not been established. Some investigators recommend digoxin [l] whereas others prefer digitoxin [2,3]. Appropriate dosage schedules of these drugs have been suggested, but the serum glycoside levels achieved in patients with renal failure following these schedules have been incompletely documented. Moreover, no studies are available to indicate how well digoxin and digitoxin are absorbed in patients undergoing dialysis. We have undertaken to answer these questions by examining three aspects of the pharmacokinetics of digoxin and digitoxin therapy in patients undergoing long-term dialysis. These include (1) measurement of serum glycoside concentrations after the oral administration of a single dose of the drugs in patients with renal failure and in patients with normal renal function, (2) determination of
April 1975
The American Journal of Medicine
Volume 58
525
CARDIAC GLYCOSIDES IN HEMODIALYSIS PATIENTS-FINKELSTEIN
I
2
3
6
8
24
ET AL.
I
HOURS
METHODS AND MATERIALS The patients studied were stable subjects with chronic renal failure, who underwent dialysis for 10 hours biweekly. All patients had creatinine clearances less than 3 cc/ min, and none was in frank congestive heart failure during the study. The control patients in whom the absorption curves were studied were hospitalized patients with orthopedic or dermatologic conditions; they had no chemical or laboratory evidence of renal, cardiac, hepatic, thyroid or gastrointestinal disease. Mean body weights (68 kg f 6 SD), age (50 years f 7) and serum total proteins (7.4 g/ 100 ml f 0.2) of the control patients did not differ significantly from those of the study patients (see Table IV). All the digoxin tablets (scored 0.25 mg tablets) employed in this study were kindly supplied by BurroughsWellcome Co., and all were from the same lot (351C). All digitoxin tablets (scored 0.1 mg tablets) were kindly supplied by Eli Lilly & Co. and were from a single lot (6NF 79A). The digoxin tablets, as assayed in vitro for uniformity of content, averaged 103.8 f 0.9 per cent (mean f SE) of the stated 0.25 mg dosage; the mean dissolution rate of this lot of tablets was 83 per cent at 1 hour (R. Cresswell, personal communication). Similar data were not available for the digitoxin tablets. Drug absorption was studied by measuring serum glycoside concentrations at 0, 1, 1 l/2, 2, 3, 6, 8 and 24 hours after the oral administration of 0.25 mg of either digoxin or digitoxin. Tablets were taken after an overnight fast, and patients were maintained in the fasting state for the initial 4 hours after ingestion of the tablets. No other medications were ingested during this fasting period. In the patients with renal failure absorption studies were performed on the day preceding their dialyses. None of the patients had received any cardiac glycoside in the previous 3 months. In the maintenance part of the study, nine patients undergoing long-term dialysis were initially given a standard dosage of digoxin, 0.125 mglday, 5 days a week. April 1975
The American Journal of Medicine
3 HOURS
the serum levels achieved and the time needed to reach these levels when standard maintenance dosages are given to patients undergoing hemodialysis, and (3) analysis of the variability in the serum levels of the drugs once the serum levels have apparently stabilized.
526
2
Volume 58
6
8
24
Figure 1. tion curves tients (left) undergoing (right) after 0.25 mg.
Digoxin absorpin control paand in patients hemodialysis an oral dose of
Other medications taken by the patients during the study period included folic acid, multivitamins, aluminum hydroxide gel and ferrous sulfate. At the onset of the study, three patients had not previously received any cardiac glycoside, four patients were taking the same dosage of regular pharmacy digoxin (Burroughs-Wellcome), and two patients were receiving regular pharmacy digitoxin. Patients were instructed to ingest the tablet each morning upon arising, and to omit it on the day of dialysis. Blood for determining serum glycoside concentrations was obtained at the onset of dialysis, 24 hours after ingestion of the most recent dose. After 3 to 4 months of digoxin therapy, 8 patients were changed to digitoxin. At this time, all serum digitoxin determinations were not significantly different from 0. In addition, one new patient with renal failure was begun on maintenance digitoxin therapy. Digitoxin, 0.1 mg, was given daily except on dialysis days. All patients were seen by one of us during each dialysis, checked for possible digitalis toxicity, and reminded to take the pills as instructed. Informed, written consent was obtained from each patient. Serum digoxin concentrations were measured by the radioimmunoassay method of Hayes et al. [4] (using a tritiated t3H) marker, New England Nuclear Corp.), except that a dilution of antidigoxin serum was used that produced approximately 45 per cent binding of the 3H digoxin in the absence of the unlabelled glycoside. Serum digitoxin levels were determined by the radioimmunoassay technic of Oliver et al. [5], using the Schwarz-Mann Kit with the 12? marker. To avoid chemiluminescence as a potential source of error in the digoxin assay [6], the final mixtures of reagents and scintillation fluid were allowed to equilibrate for 8 to 24 hours before counting, and at least two complete sets of counts were obtained to insure that successive counting determinations did not vary significantly. The coefficients of variation (standard deviation/mean X 100) of the serum assay methods were calculated by averaging 16 different determinations of the same serum specimen on 8 different days over a 4 month period. Coefficients of variation for both assays were less than 7 per cent. Since digitoxin levels using standard methods for the single dose absorption curves were very low, four times the usual amount of serum was added for increased accuracy, and the levels were divided by four. Control curves for the digitoxin absorption determinations were obtained
CARDIAC GLYCOSIDES IN HEMODIALYSIS PATIENTS-FINKELSTEIN
TABLE I
TABLE II
Mean Serum Digoxin Levels After a Single Oral Dose of Digoxin
Mean Serum Digitoxin Levels After a Single Oral Dose of Digitoxin Serum Digitoxin =k SE (ng/ml)
Serum Digoxin :k SE (ng/ml)
Hours 0 1 1 ‘I? 2 3 6 8 24
Control Patients (8) ________ 0.67 0.57 0.64 0.51 0.23 0.13 0.09
0 XII III -L jl T i *
0.18 0.10 0.08 0.04 0.04 0.02 0.03
NOTE: NS = not significant. cate the number of patients
Dialyzed Patients (7)
p Values
Hours
0 + It i i i i f
NS NS NS co.02
of Digoxin and Digitoxin in Patients
Undergoing Hemodialysis
FREDRIC 0. FINKELSTEIN, M.D. New Haven, Connecticut JOHN A. GOFFINET, M.D. ERNEST0
D. HENDLER, M.D.
West Haven, Connecticut JOHN LINDENBAUM, M.D. New York, New York
From the Department of Medicine, West Haven Veterans Administration Hospital, West Haven, Connecticut 065 16; the Yale University Medical School, New Haven, Connecticut 06510; and the Department of Medicine, Harlem Hospital Center, College of Physicians and Surgeons, Columbia University, New York, New York 10037. This study was supported in part by grants from the Connecticut Heart Association (Grant 519-(1972)), the Health Research Council of the City of New York (U-2230) and the Burroughs-Wellcome Co. Requests for reprints shoufd be addressed to Dr. Fredric 0. Finkelstein, Department of Internal Medicine, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut 06510. Manuscript accepted June 4, 1974.
The pharmacokinetics of digoxin and digitoxin in patients undergoing long-term hemodiaiysis were examined to determine which is the preferred cardiac giycoside in this patient population. Absorption curves from 0 to 24 hours afler an oral dose of digitoxin were similar in dialyzed patients and in control patients. Serum giycoside concentrations after an oral dose of digoxin were higher in dialyzed patients than in control patients, significantly so from 2 to 24 hours, reflecting the absence of the predominantly renal route of excretion of digoxin. When nine dialyzed patients were placed on a maintenance dose of digoxin, 0.125 mg 5 days a week, serum levels plateaued at 30 days at a mean concentration (& SE) of 0.84 f 0.05 ng/ml. Maintenance therapy with 0.1 mg digitoxin 5 days a week resuited in stabilization of serum levels within 30 days at a mean concentration of 19 f 1 ng/ml. Variability in the serum glycoside concentrations was determined after stabilization of levels during 2 to 19 week follow-up periods with each drug. Variability in serum levels was somewhat increased during maintenance therapy with digitoxin. On the basis of the pharmacokinetic data obtained in this study, no clear cut preference for one giycoside over the other could be established. Congestive heart failure secondary to associated myocardial disease, hypertension and fluid overload frequently develops in patients with renal failure. Consequently, therapy with a cardiac glycoside, usually digoxin or digitoxin, is often prescribed. However, which of these two glycosides and what doses of the drugs are preferable in patients with chronic renal failure have not been established. Some investigators recommend digoxin [l] whereas others prefer digitoxin [2,3]. Appropriate dosage schedules of these drugs have been suggested, but the serum glycoside levels achieved in patients with renal failure following these schedules have been incompletely documented. Moreover, no studies are available to indicate how well digoxin and digitoxin are absorbed in patients undergoing dialysis. We have undertaken to answer these questions by examining three aspects of the pharmacokinetics of digoxin and digitoxin therapy in patients undergoing long-term dialysis. These include (1) measurement of serum glycoside concentrations after the oral administration of a single dose of the drugs in patients with renal failure and in patients with normal renal function, (2) determination of
April 1975
The American Journal of Medicine
Volume 58
525
CARDIAC GLYCOSIDES IN HEMODIALYSIS PATIENTS-FINKELSTEIN
I
2
3
6
8
24
ET AL.
I
HOURS
METHODS AND MATERIALS The patients studied were stable subjects with chronic renal failure, who underwent dialysis for 10 hours biweekly. All patients had creatinine clearances less than 3 cc/ min, and none was in frank congestive heart failure during the study. The control patients in whom the absorption curves were studied were hospitalized patients with orthopedic or dermatologic conditions; they had no chemical or laboratory evidence of renal, cardiac, hepatic, thyroid or gastrointestinal disease. Mean body weights (68 kg f 6 SD), age (50 years f 7) and serum total proteins (7.4 g/ 100 ml f 0.2) of the control patients did not differ significantly from those of the study patients (see Table IV). All the digoxin tablets (scored 0.25 mg tablets) employed in this study were kindly supplied by BurroughsWellcome Co., and all were from the same lot (351C). All digitoxin tablets (scored 0.1 mg tablets) were kindly supplied by Eli Lilly & Co. and were from a single lot (6NF 79A). The digoxin tablets, as assayed in vitro for uniformity of content, averaged 103.8 f 0.9 per cent (mean f SE) of the stated 0.25 mg dosage; the mean dissolution rate of this lot of tablets was 83 per cent at 1 hour (R. Cresswell, personal communication). Similar data were not available for the digitoxin tablets. Drug absorption was studied by measuring serum glycoside concentrations at 0, 1, 1 l/2, 2, 3, 6, 8 and 24 hours after the oral administration of 0.25 mg of either digoxin or digitoxin. Tablets were taken after an overnight fast, and patients were maintained in the fasting state for the initial 4 hours after ingestion of the tablets. No other medications were ingested during this fasting period. In the patients with renal failure absorption studies were performed on the day preceding their dialyses. None of the patients had received any cardiac glycoside in the previous 3 months. In the maintenance part of the study, nine patients undergoing long-term dialysis were initially given a standard dosage of digoxin, 0.125 mglday, 5 days a week. April 1975
The American Journal of Medicine
3 HOURS
the serum levels achieved and the time needed to reach these levels when standard maintenance dosages are given to patients undergoing hemodialysis, and (3) analysis of the variability in the serum levels of the drugs once the serum levels have apparently stabilized.
526
2
Volume 58
6
8
24
Figure 1. tion curves tients (left) undergoing (right) after 0.25 mg.
Digoxin absorpin control paand in patients hemodialysis an oral dose of
Other medications taken by the patients during the study period included folic acid, multivitamins, aluminum hydroxide gel and ferrous sulfate. At the onset of the study, three patients had not previously received any cardiac glycoside, four patients were taking the same dosage of regular pharmacy digoxin (Burroughs-Wellcome), and two patients were receiving regular pharmacy digitoxin. Patients were instructed to ingest the tablet each morning upon arising, and to omit it on the day of dialysis. Blood for determining serum glycoside concentrations was obtained at the onset of dialysis, 24 hours after ingestion of the most recent dose. After 3 to 4 months of digoxin therapy, 8 patients were changed to digitoxin. At this time, all serum digitoxin determinations were not significantly different from 0. In addition, one new patient with renal failure was begun on maintenance digitoxin therapy. Digitoxin, 0.1 mg, was given daily except on dialysis days. All patients were seen by one of us during each dialysis, checked for possible digitalis toxicity, and reminded to take the pills as instructed. Informed, written consent was obtained from each patient. Serum digoxin concentrations were measured by the radioimmunoassay method of Hayes et al. [4] (using a tritiated t3H) marker, New England Nuclear Corp.), except that a dilution of antidigoxin serum was used that produced approximately 45 per cent binding of the 3H digoxin in the absence of the unlabelled glycoside. Serum digitoxin levels were determined by the radioimmunoassay technic of Oliver et al. [5], using the Schwarz-Mann Kit with the 12? marker. To avoid chemiluminescence as a potential source of error in the digoxin assay [6], the final mixtures of reagents and scintillation fluid were allowed to equilibrate for 8 to 24 hours before counting, and at least two complete sets of counts were obtained to insure that successive counting determinations did not vary significantly. The coefficients of variation (standard deviation/mean X 100) of the serum assay methods were calculated by averaging 16 different determinations of the same serum specimen on 8 different days over a 4 month period. Coefficients of variation for both assays were less than 7 per cent. Since digitoxin levels using standard methods for the single dose absorption curves were very low, four times the usual amount of serum was added for increased accuracy, and the levels were divided by four. Control curves for the digitoxin absorption determinations were obtained
CARDIAC GLYCOSIDES IN HEMODIALYSIS PATIENTS-FINKELSTEIN
TABLE I
TABLE II
Mean Serum Digoxin Levels After a Single Oral Dose of Digoxin
Mean Serum Digitoxin Levels After a Single Oral Dose of Digitoxin Serum Digitoxin =k SE (ng/ml)
Serum Digoxin :k SE (ng/ml)
Hours 0 1 1 ‘I? 2 3 6 8 24
Control Patients (8) ________ 0.67 0.57 0.64 0.51 0.23 0.13 0.09
0 XII III -L jl T i *
0.18 0.10 0.08 0.04 0.04 0.02 0.03
NOTE: NS = not significant. cate the number of patients
Dialyzed Patients (7)
p Values
Hours
0 + It i i i i f
NS NS NS co.02
