Comparison of two different loading doses of digoxin in severe renal impairment

Summary The correct loading dose of digoxin in patients with advanced renal failure is still a matter of discussion. The effect has been studied of loading doses of digoxin 0.625 mg or 1.25 mg given over 48 h according to randomized crossover design to healthy volunteers and to two different groups of patients with renal impairment and the same mean endogenous creatinine clearance of about 15 ml/min. The subsequent maintenance dose for 4 days was digoxin 0.25 mg in the volunteers and 0.125 mg in both groups of patients. The minimum plasma digoxin concentrations before each dose was measured by radioimmunoassay and the plasma levels in the different groups have been compared. In the healthy volunteers no significant difference was found during the study, despite wide variation in the plasma digoxin concentration. In contrast, in patients with renal failure, the group with the higher loading dose showed significantly higher plasma concentrations 24, 36 and 48 h after drug administration, reaching the highest mean value of 2.2 ng/ml at 48 h. However, after 120 h of maintenance therapy a mean digoxin concentration of 1.3 mg/ml was found in both groups. Thus, despite different loading doses identical plasma concentrations were reached during administration of the same maintenance therapy. The higher plasma digoxin concentration obtained during administration of a higher loading dose might be the cause of arrythmias in individual patients.     

The effect of captopril on pharmacokinetics of digoxin in patients with mild congestive heart failure.

The effect of captopril on steady-state pharmacokinetics of digoxin was studied in 12 patients with mild congestive heart failure (CHF; New York Heart Association functional class 1 or 2). Serum and urine digoxin concentrations were determined before and after a repeated administration of captopril in the patients on chronic digoxin therapy. The patients were taking digoxin, 0.25-0.375 mg/day, once daily, and were concurrently administered captopril, 37.5 mg/day, three times daily, for seven days. Peak serum concentration of digoxin (SCD) before and after captopril was 2.1 +/- 0.2, mean +/- SEM, and 2.0 +/- 0.1 ng/ml; the time to peak was 1.1 +/- 0.2 and 1.8 +/- 0.3 h; the terminal half-life (t1/2 alpha) was 10.9 +/- 1.0 and 8.7 +/- 0.9 h, and the area under the concentration-time curve to 24 h was 26.9 +/- 2.4 and 27.6 +/- 2.0 ng.h/ml. There was no significant difference between patients without and with captopril in SCD and its pharmacokinetic parameters. Renal digoxin clearance and creatinine clearance also showed no significant difference. After an administration of captopril, angiotensin-converting-enzyme (ACE) activity was well suppressed. These results suggest that captopril does not increase SCD in patients with CHF, and effectively suppresses ACE activity. Thus, modification in the dosage regimen of digoxin may be unnecessary in the case of coadministration with captopril.     

Passage of digoxin into cerebrospinal fluid in man

The passage of digoxin into the cerebrospinal fluid (CSF) was studied in 8 infants on maintenance therapy with digoxin, 11 adult patients on long-term digoxin therapy, and 15 patients, previously non-digitalized, who were given 0.5 mg digoxin orally 1 hr to 12 hrs prior to lumbar puncture. Digoxin in the serum and CSF was determined by radioimmunoassay. In the infants a mean serum concentration of 1.5 ng/ml (range 0.7-2.3 ng/ml) was found, and a simultaneous mean CSF concentration of 0.5 ng/ml (range 0.3-1.1 ng/ml). In the adults on long-term therapy, the corresponding figures were 1.1 ng/ml (range 0.5-2.2 ng/ml) and 0.3 ng/ml (range 0-0.6 ng/ml). Among the 15 patients given a single oral dose of digoxin, detectable CSF concentrations (0.2-0.3 ng/ml) were found in five, 1-12 hrs after the administration of the drug. In three paediatric patients with hydrocephalus (3 months-5 years) digoxin therapy was started as an attempt to decrease CSF production. In these patients, the production of CSF was reduced by 17, 25 and 30%, respectively.     

Evaluation of a sex-based difference in the pharmacokinetics of digoxin

STUDY OBJECTIVE: To determine whether a sex-based difference in digoxin pharmacokinetics exists in patients receiving long-term digoxin therapy for chronic heart failure or atrial fibrillation. DESIGN: Single-center, retrospective review of medical records. SETTING: University-based teaching hospital and outpatient clinic. PATIENTS: Sixty-seven adults (32 men, 35 women) with chronic heart failure or atrial fibrillation who were receiving digoxin therapy. MEASUREMENTS AND MAIN RESULTS: Serum digoxin concentrations and daily digoxin doses were obtained from patients' medical records. Daily doses were adjusted for patients' actual and ideal body weight and body mass index (BMI). The ratio between the serum digoxin concentration and each of the adjusted daily doses of digoxin was compared between men and women. The mean +/- SD serum digoxin concentration was 0.85 +/- 0.51 ng/ml for men compared with 1.02 +/- 0.51 ng/ml for women. Mean +/- SD unadjusted doses of digoxin were 0.180 +/- 0.063 and 0.164 +/- 0.059 mg/day for men and women, respectively; the difference was not statistically significant. Ratios of serum digoxin concentration to daily digoxin doses did not differ by sex when doses were estimated with actual or ideal weight. Only the ratio of the digoxin concentration to the BMI-adjusted dose was significantly different between men and women (0.14 +/- 0.09 and 0.19 +/- 0.11, respectively, p<0.05). CONCLUSION: Sex-based differences in digoxin pharmacokinetics were absent when actual or ideal body weight was used. However, the ratio of serum digoxin concentration to daily digoxin dose adjusted for BMI differed by sex. Because digoxin is distributed to lean body mass, use of the BMI could have overadjusted body weight, leading to inaccurate pharmacokinetic assumptions and calculations. The pharmacokinetics of digoxin do not appear to differ by sex.     

The effects of captopril on serum digoxin levels in patients with severe congestive heart failure

The effects of captopril on serum digoxin concentrations were studied in 8 patients with severe (NYHA Class IV) congestive heart failure. Serum digoxin concentrations were determined before and after the administration of captopril for 1 week in patients on chronic digoxin therapy. Each patient who was taking 0.25 mg of digoxin PO q.d., was administered 12.5 mg of captopril PO t.i.d. for 7 days. The peak serum concentration of digoxin (Cmax) before and after (on Days 0 and 7) captopril administration was 1.7+/-0.2 ng/ml and 2.7+/-0.2 ng/ml, the time to peak (tmax) was 2.4+/-0.5 h and 1.3+/-0.2 h, and the area under the 24-hour digoxin concentration-time curve (AUC0-24h) was 30.0+/-1.5 ng x h/ml and 41.7+/-3.4 ng x h/ml, respectively. While captopril caused a significant increase in peak serum concentration and the area under the digoxin concentration-time curve, it decreased the time to digoxin peak (p = 0.01, p = 0.04, p = 0.01, respectively). No patient developed evidence of digoxin toxicity. Concomitant administration of captopril with digoxin increases serum digoxin concentration in patients with severe congestive heart failure.     

Interaction between digoxin and propafenone

The pharmacokinetic and pharmacodynamic interactions between digoxin and propafenone were investigated in 10 hospitalized patients with heart disease and cardiac arrhythmias. During steady state (0.25 mg/day) the glycoside was combined with 600 mg of propafenone daily for 1 week. The mean +/- SD serum digoxin concentration (SDC) was 0.97 +/- 0.29 ng/ml before and 1.54 +/- 0.65 ng/ml (p less than 0.003) during propafenone administration. Propafenone induced a mean decrease in 31.1 and 31.7% in total and renal digoxin clearances, respectively. The increase in SDCs was accompanied by a decrease in heart rate (HR) and shortening of QTC (QT interval corrected for HR). In patients receiving digoxin and propafenone simultaneously, the SDCs should be monitored and the digoxin dose reduced if there is evidence of toxicity.     

The influence of diltiazem hydrochloride on trough serum digoxin concentrations

A significant drug interaction between verapamil and digoxin, resulting in elevated serum digoxin concentrations, has been well documented in the medical literature. However, a similar interaction between digoxin and the calcium channel blockers nifedipine and diltiazem has not been conclusively established. This study investigated the influence of diltiazem hydrochloride on trough serum concentrations of concurrently administered digoxin in eight healthy volunteers. During the control phase of the study, volunteers were administered digoxin 0.25 mg/d for 13 days, and subsequently judged to be at steady state by serial determinations of digoxin serum concentrations. Twenty-four hour urine collections were done for creatinine clearance and urinary digoxin clearance determinations. Phase II of the study involved the addition of diltiazem hydrochloride 30 mg qid to the on-going, daily regimen of digoxin. After 14 days of concomitant therapy, steady-state trough digoxin concentrations were again determined, as well as creatinine clearances and urinary digoxin clearances. This investigation demonstrates that concomitant administration of diltiazem hydrochloride with digoxin results in significantly elevated steady-state trough digoxin concentrations (0.32 +/- 0.07 ng/ml increasing to 0.48 +/- 0.06 ng/ml, p less than 0.01). Urinary digoxin clearance decreased from 223.5 +/- 35.7 ml/min to 153.4 +/- 17.5 ml/min (p less than 0.05). Creatinine clearances were unaltered. A review of the current literature on this topic is included.     

Failure of CAVH to remove digoxin-Fab complex in piglets

Digoxin toxicity may be associated with renal failure and an inability to excrete the digoxin-Fab (antibody fragment) complex used in detoxification. We are unaware of any previous reports regarding the removal of digoxin-Fab fragment complex by continuous arteriovenous hemofiltration. Continuous arteriovenous hemofiltration allows ultrafiltration of molecules less than 50,000 daltons. Because the digoxin-Fab fragment complex has a molecular weight of 45 - 50,000 daltons, we evaluated the efficiency of continuous arteriovenous hemofiltration in removing the digoxin-Fab fragment complex. Three piglets were given 100 mcg/kg digoxin IM, in divided doses. Animals were anesthetized and continuous arteriovenous hemofiltration was begun using a Diafilter 10 cartridge. A mean ultrafiltration rate of 3.6 +/- 0.4 ml/min was obtained. The system equilibrated for 30 minutes and initial serum and ultrafiltrate digoxin levels were obtained. Mean serum values were total 6.20 +/- 1.74 ng/ml and free 3.72 +/- 0.88 ng/ml. Digibind 40 mg IV was given and then samples of serum and ultrafiltrate were obtained at 30, 60 and 90 minutes for digoxin levels. Mean values were as follows: 30 min serum total 54.23 +/- 26.13 ng/ml and free 0.08 +/- 0.08 ng/ml; 60 min serum total 61.24 +/- 27.31 ng/ml and free 0.07 +/- 0.04 ng/ml; and 90 min serum total 67.63 +/- 26.78 ng/ml and free 0.10 +/- 0.10 ng/ml. Ultrafiltrate levels throughout the experiment were negligible (less than or equal to 0.04 ng/ml). Continuous arteriovenous hemofiltration appears to be ineffective in removing the digoxin-Fab fragment complex.     

The pharmacokinetics of piroxicam in elderly persons with and without renal impairment.

1. Piroxicam pharmacokinetics were assessed in three groups of subjects: (1) young healthy volunteers, (2) healthy elderly subjects (mean +/- s.d. creatinine clearance 88 +/- 13 ml min-1), and (3) elderly patients with renal insufficiency (creatinine clearance 60 +/- 10 ml min-1) following the administration of piroxicam 20 mg as a single dose and after chronic dosing of 20 mg once daily for 4 weeks. 2. Piroxicam and 5'-hydroxypiroxicam concentrations were measured by h.p.l.c. in serum and urine samples collected for 96 h after the single dose and for 144 h after chronic dosing. Unbound concentrations of piroxicam were determined by ultrafiltration. 3. Elimination half-lives, steady state concentrations of piroxicam and 5'-hydroxypiroxicam, clearances of total and unbound piroxicam, volumes of distribution normalized for body weight, and urinary recovery of 5'-hydroxypiroxicam were not influenced by age or renal function. Volumes of distribution after the single dose were significantly lower in women compared with men (mean +/- s.d. 10.0 +/- 2.9 l vs 12.9 +/- 5.0 l; 95% confidence interval of the difference 0.1 to 5.6). 4. Percent unbound piroxicam values were 1.46 +/- 0.3% after the single dose and 1.45 +/- 0.2% at steady state. There were significant reductions in clearance and clearance of unbound piroxicam between single and chronic doses. The half-lives of 5'-hydroxypiroxicam (80.9 +/- 44 h) were significantly longer than those of piroxicam (54.9 +/- 26 h) after chronic dosing.     

Determination of cefalexin pharmacokinetics and dosage adjustments in relation to renal function.

Following oral administration of a single 500-mg dose of cefalexin, serum and urine levels of the antibiotic were determined comparatively in ten normal subjects, ten patients with renal impairment, and ten patients with chronic nephritis on maintenance hemodialysis. In normal subjects, mean serum peak levels (12.0 +/- 0.8 mcg/ml) were observed 2 hours after drug administration. Absorption half-time (Ta1/2) averaged 0.82 hour and mean serum half-life (T 1/2) was 1.03 hours. Urinary recovery of cefalexin over a 6-hour period amounted to 64 per cent of the ingested dose. The renal clearance of the drug was 214 ml/min. In patients with renal impairment and in patients on maintenance hemodialysis, total elimination rate constant (Ke) was markedly lower (CrCl=0; Ke=Km=0.0766), whereas serum half-life (T 1/2) was significantly increased, reaching theoretically 8.47 hours in patinets with creatinine clearance of 0 ml/min. A correlation was established between Ke values and the creatinine clearances of the patients under study (Ke=0.0766 + 0.0060 CrCl). Initial loading doses, maintenance doses, and intervals adjusted to creatinine clearances were calculated from these data; accurate dosage schedules well adjusted to the renal status of each individual patient were derived from the calculated values.     

Comparison of pharmacokinetic and safety profiles of amantadine 50- and 100-mg daily doses in elderly nursing home residents.

Pharmacokinetic and adverse event profiles of 50- and 100-mg amantadine doses administered daily for up to 21 days for influenza prophylaxis were compared in 82 elderly nursing home residents (mean age 85 yrs, 68% female). We sought to determine if a standard daily dose of 50 mg would achieve mean steady-state trough serum concentrations (CPSSt) of 300 ng/ml and be associated with a lower frequency of adverse events than 100-mg doses. Statistically significant relationships were found between CPSSt and dosage (in mg/kg/day) and serum creatinine. Adverse events were more common with the 100-mg dose (24% vs 14%); 94% occurred in women. Amantadine CPSSt and apparent clearance were not significantly different between sexes. Thirty-nine residents (89%) receiving 50 mg daily achieved CPSSt below 300 ng/ml compared to 42% receiving 100 mg. Standard daily amantadine doses of 50 mg may not achieve adequate CPSSt in elderly nursing home residents, but 100 mg may lead to excessive CPSSt and adverse events, especially in the presence of renal impairment.     

Loading dose of digoxin in renal failure.

1 Twenty-two dialysis dependent patients received an intravenous loading dose of digoxin of 10 microgram/kg and the mean +/- s.d. serum digoxin concentration 24 h later was 1.5 ng/ml +/- 0.4 (range 0.8--2.1 ng/ml). No evidence of toxicity was observed. 2 The average apparent volume of distribution of digoxin in dialysis dependent patients was found to be reduced by about one third compared with that reported for individuals with normal renal function, but there was great variation. 3 An appropriate intravenous loading dose of digoxin in most patients with advanced renal failure is 10 microgram/kg.     

Steady-state pharmacokinetics of rufloxacin in elderly patients with lower respiratory tract infections.

The pharmacokinetics of rufloxacin, after repeated doses, was evaluated in 12 elderly patients with lower respiratory tract infections. Patients were given a single loading dose of 400 mg on the first day of treatment and single daily maintenance doses of 200 mg for the next 6-9 days. Serum concentrations of the drug were determined by high-performance liquid chromatography (HPLC) at regular intervals during treatment and fitted to a one-compartment open model for repeated doses. The maximum serum concentration after the first dose was 6.46 +/- 1.06 (mean +/- SEM) micrograms/ml and was reached in 4.3 +/- 0.8 h after the first administration. The elimination half-life was 28.7 +/- 4.1 h. The area under the serum levels-time curve from 0 to 24 h was 103 +/- 14 micrograms/h/ml after the first dose. On the last day of observation it increased to 155 +/- 28 micrograms/h/ml, with a mean extent of accumulation of 2.3 +/- 0.3 times. The elimination half-life was comparable to those in other studies in healthy young subjects, while plasma levels were about 80% higher. These results suggest that in elderly patients elevated drug concentrations may be reached in the serum. Although no untoward reactions related either to the drug concentration in serum or the dose have been noted with rufloxacin, this patient population should nevertheless be monitored carefully for adverse effects.     

Effect of tiaprofenic acid on serum digoxin concentration

The effect of concomitant tiaprofenic acid (Surgam) administration (200 mg t.i.d.) on serum digoxin concentration (SDC) was evaluated in 12 healthy volunteers on digoxin maintenance treatment. During a 10-day coadministration period with tiaprofenic acid no significant increase in SDC was observed (0.97 +/- 0.24 vs. 1.12 +/- 0.21 ng/ml, p less than 0.05). Mean tiaprofenic acid concentration amounted to 2.85 +/- 1.94 micrograms/ml 14 h after last drug intake. The incidence of adverse reactions was minimal with gastrointestinal upset in one person. Tiaprofenic acid had no influence on red or white blood cell count. Thus, in contrast to various other nonsteroidal antiinflammatory drugs coadministration of tiaprofenic acid (600 mg daily) has no relevant influence on serum digoxin levels.     

Effect of digoxin on the sensitivity to flickering light.

1 The sensitivity to flickering light at various light frequencies (DeLange curve) was determined in 20 controls and 45 patients receiving maintenance doses of digoxin. 2 Flicker thresholds (mean percentage of maximal light modulation +/- s.d.; F 30 Hz) were 7.6 +/- 1.7 in controls and 9.4 +/- 1.7 in patients with optimal plasma digoxin levels (0.5-1.9 ng/ml), but they rose to 15.5 +/- 1.9 at subtoxic levels (2.0-3.0 ng/ml), and to 21.8 +/- 2.6 at toxic levels (above 3.0 ng/ml). 3 Flicker sensitivity was inversely correlated with plasma digoxin levels and returned to baseline values when the administration of digoxin was interrupted. 4 The DeLange curve seems to be a valuable tool to measure the toxic effects of digitalis on the visual system.     

Pharmacokinetics of digoxin and main metabolites/derivatives in healthy humans

Three healthy, young male volunteers received doses of 0.6 and 1.2 mg of specifically labelled [3H]digoxin each by intravenous (i.v.) bolus injection and oral (p.o.) administration in accordance with a randomized four-way crossover design. Plasma, urine, and feces samples were taken over an interval of 144 h after drug administration. Total radioactivity and individual radioactivity assignable to digoxin and its metabolites were measured. After i.v. administration, the mean +/- SD recovery of total radioactivity, as percent of dose, was complete, urine 81.3 +/- 2.0% and feces 17.1 +/- 2.8%. The mean recovery of digoxin and that of its metabolites in urine was digoxin 75.6 +/- 3.0%, dihydrodigoxin 2.8 +/- 1.6%, digoxigenin bisdigitoxoside 1.6 +/- 0.1%, and additional metabolites 1.5 +/- 0.3%. Judging from the metabolite data in urine and considering the 5% impurity of the administered dose, metabolism of digoxin appeared to be insignificant after i.v. administration. The total and renal clearances of digoxin were, on average, 193 +/- 25 ml min-1 and 152 +/- 24 ml min-1. The mean steady state volume of distribution was 489 +/- 73 L and the mean residence time 41 +/- 5 h. For the metabolites dihydrodigoxin and digoxigenin bisdigitoxoside the mean residence times were on average 35 +/- 9 h and 53 +/- 11 h; the renal clearances were 79 +/- 13 ml min-1 and 100 +/- 26 ml min-1. After p.o. administration, the mean recovery of total radioactivity, as percent of the dose, was also complete, urine 65.7 +/- 1.98% and feces 31.6 +/- 7.6%. The mean recovery of digoxin and that of its metabolites, as percent of dose, in urine was digoxin 51.5 +/- 11.4%, dihydrodigoxin 4.5 +/- 3.9%, digoxigenin bisdigitoxoside 1.9 +/- 0.1%, polar metabolites 5.5 +/- 3.8%, and additional metabolites 1.3 +/- 0.6%. After p.o., as compared to i.v. administration, larger amounts of all the metabolites were formed in accordance with first pass metabolism/degradation. Maximum mean plasma concentrations of 4.3 +/- 2.5 ng ml-1 and 9.5 +/- 1.1 ng ml-1 for digoxin were observed at 40 +/- 10 min after p.o. administration of 0.6 and 1.2 mg of the drug. The mean absolute bioavailability of digoxin from an aqueous solution was 0.67 +/- 0.14. Renal clearance and mean oral residence time for digoxin were on average 176 +/- 28 ml min-1 and 37 +/- 4 h after p.o. administration.(ABSTRACT TRUNCATED AT 400 WORDS)     

Serum digoxin levels in patients of a general practice in Germany

Summary Serum digoxin levels were determined in 33 outpatients of a general practice in the countryside, on three occasions at intervals of 8 weeks. All the patients were on long term digoxin treatment, about 2 years on average. About 14 days after the first and the second visits the results of the measurements were sent to the patients, with a comment about their reliability in taking treatment according to the serum digoxin level. At the first visit half of the serum digoxin level were lower than 0.5 ng/ml; the mean serum concentration was 0.52 ng/ml. There was no correlation between serum concentration and age, dose or creatinine level; but there was with replies to the question about regularity of drug intake. The mean serum level at the second and the third visits was 0.88 ng/ml and 0.89 ng/ml, respectively. A correlation was found between the dose and the serum digoxin level. From these results it seems that compliance by the patient plays a major role in producing steady state levels of drugs.Dedicated to Professor Dr. R. Aschenbrenner on the occasion of his 70th birthday     

Effect of multiple doses of losartan on the pharmacokinetics of single doses of digoxin in healthy volunteers.

1. Losartan (DuP 753, MK-954) is a novel, potent and highly selective AT1 angiotensin II receptor antagonist. The effect of multiple oral doses of losartan on digoxin pharmacokinetics was evaluated in healthy male subjects. 2. In a double-blind and randomized fashion, subjects received 50 mg losartan or placebo once daily for 15 days in each period. At least 7 days elapsed between the two treatment periods. On days 4 and 11 of each period, subjects also received a single 0.5 mg dose of digoxin intravenously and orally respectively. 3. Eleven of 13 subjects completed the study. Side effects were mild and transient (12 out of 13 subjects reported at least one adverse experience). During the study, no laboratory abnormalities were noted. 4. Multiple oral doses of losartan (50 mg daily) did not affect the pharmacokinetic parameters of 0.5 mg of digoxin i.v. AUC(0.48h) of immunoreactive digoxin during losartan 28.8 +/- 2.9 vs 28.5 +/- 3.9 ng ml-1 h during placebo; not significant, and 96 h urinary excretion [% dose] during losartan 54.0 +/- 7.2 vs 51.9 +/- 6.5% during placebo; not significant). Geometric mean ratios (90% confidence interval) for AUC and urinary excretion were respectively, 1.03 (0.98, 1.08) and 1.09 (0.98, 1.21). 5. Multiple oral doses of losartan did not affect the pharmacokinetic parameters of oral digoxin AUC(0.48 h) during losartan 23.6 +/- 3.7 ng ml-1 h vs 22.4 +/- 2.6 ng ml-1 h during placebo; not significant, Cmax 3.5 +/- 0.7 ng ml-1 with vs 3.1 +/- 0.5 ng ml-1 without losartan; not significant and tmax 0.6 +/- 0.2 h with vs 0.9 +/- 0.7 h without losartan; not significant, and 96 h urinary excretion [% dose] during losartan 51.2 +/- 6.3 vs 46.3 +/- 2.4% during placebo; not significant). Geometric mean ratios (90% confidence interval) for AUC and urinary excretion were respectively, 1.06 (0.98, 1.14) and 1.12 (0.97, 1.28). 6. We conclude that multiple oral doses of losartan (50 mg daily) do not alter the pharmacokinetics of immunoreactive digoxin, following either intravenous or oral digoxin. Furthermore, the co-administration of digoxin with losartan is well tolerated by healthy male volunteers.     

Serum cystatin C is not a better marker of creatinine or digoxin clearance than serum creatinine

AIMS: To assess whether cystatin C, a new serum marker of renal function, is a better index of creatinine or digoxin clearance than serum creatinine in older people. METHODS: Twenty-two volunteers over the age of 65 years (mean 73 +/- 5) were recruited from a healthy elderly volunteer database. None of the volunteers was taking digoxin or other medication known to interfere with digoxin kinetics or assay. Digoxin was infused at a dose of 7-10 microg kg(-1) and blood samples were taken over the following 48 h and assayed for serum digoxin. Serum cystatin C, creatinine and creatinine clearance were measured and a calculated creatinine clearance was estimated using the Cockcroft Gault formula. Digoxin clearance was calculated using a pharmacokinetic software package. All values were log transformed to normalize their distribution. RESULTS: Of the 22 volunteers enrolled into the study, 18 completed the study. Serum cystatin C ranged between 0.72 and 1.89 mg l(-1) and serum creatinine ranged from 69.6 to 153.9 micromol l(-1). Measured creatinine clearance ranged from 38 to 123 ml min(-1) and calculated creatinine clearance from 29.5 to 88.0 ml min(-1). Digoxin clearance ranged from 51.0 to 103.5 ml min(-1). Cystatin C correlated extremely well with creatinine (r=0.93, P<0.001, 95% CI 0.82, 0.97) and with creatinine clearance (r=0.67, P=0.002, 95% CI 0.3, 0.87). Neither serum cystatin C nor serum creatinine correlated with digoxin clearance (r=0.25, P=0.31, 95% CI -0.25, 0.64 and r=0.44, P=0.068, 95% CI -0.03, 0.75, respectively). Measured creatinine clearance, however, did correlate well with digoxin clearance (r=0.55, P=0.018, 95% CI 0.11, 0.81). CONCLUSIONS: Serum cystatin C and serum creatinine show very similar correlations with creatinine and digoxin clearances. Serum cystatin C does not offer any advantages in this respect. It remains to be seen whether cystatin C offers any advantage over creatinine in elderly people in other respects.