Pharmacokinetics of diuretics in geriatric patients

Arch. Gerontol. Geriatr., 9 (1989) 283-290

283

Elsevier A G G 00297

Pharmacokinetics of diuretics in geriatric patients W. M t i h l b e r g Chair of Internal Medicine and Gerontology, University of Erlangen-Nfwnbergo D-8500 Niirnberg 90, F.R.G. (Received 3 June 1989; accepted 25 July 1989)

Summary Pharmacokinetics of three diuretics (furosemide, spironolactone, and triamterene) were investigated in 70 geriatric patients. Comparing the data with the corresponding values of young healthy volunteers, mean plasma concentrations of furosemide and spironolactone (and peak concentrations of triamterene) were markedly higher in the geriatric patients. Different concomitant diuretic therapy (hydrochlorothiazide vs. piretanide) seems to influence the kinetic parameters of triamterene and its active metabolite in the geriatric patients. In elderly patients reduction or correction of the dosage seems to be necessary for all three diuretics investigated in our studies. Furosemide; Spironolactone; Triamterene; Pharmacokinetics; Interaction; Geriatric patients

Introduction It seems useful to investigate the pharmacokinetics of diuretics and their changes in old age for at least two reasons: 1. Diuretics are administered to geriatric patients very frequently. 2. The danger of unwanted side-effects of diuretic therapy (dehydration, increase or decrease of electrolyte levels, interaction with other drugs) is increased, especially in old age. Age dependent changes of the liver, like the decrease of enzyme activities or a diminished biliary excretion, may result in a prolonged half-life of those diuretics which are eliminated by metabolic breakdown in the liver or via biliary excretion. Protein binding may be altered, since albumin plasma concentrations are reduced in old age.

Correspondence to: Dr. W. Miihlberg, Chair of Internal Medicine and Gerontology, University of Erlangen-Niirnberg, Heimerichstrasse 58, D-8500 Ntirnberg 90, F.R.G. a Presented in the program of a Symposium entitled 'Aging at the Molecular, Cellular and Organism Level' held on the occasion of the 10th anniversary of the Gerontological Institute of FriedrichAlexander University of Erlangen-Niirnberg, 3rd June, 1989, Ni~rnberg, F.R.G. 0167-4943/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

284 An age-dependent decrease of cardiac output may influence both the half-life of diuretics with flow-limited hepatic elimination and the half-life of those diuretics (or their metabolites) that are mainly excreted via the kidneys. The impairment of renal function with increasing age is well reflected by the decrease of creatinine clearance.

Materials and Methods

We investigated the pharmacokinetics (and their changes in geriatric patients) of three diuretics: furosemide (a loop diuretic), spironolactone, and triamterene (both being so-called 'potassium-sparing' diuretics). In the furosemide study (Miihlberg et al., 1986) twenty geriatric patients with multiple diseases were given a single intravenous dose of 40 mg furosemide. Furosemide plasma and urine concentrations were measured using a thin layer chromatography method and fitted to an open two-compartment model. The pharmacokinetics of the metabolites of spironolactone (canrenone and fluorigenic metabolites) were investigated in ten geriatric female patients with multimorbidity after a multiple daily oral administration of 100 mg spironolactone under steady-state conditions (Platt et al., 1984). The concentration determinations were carried out simultaneously with a specific HPLC method and a less specific fluorimetric method. In the triamterene kinetics study 20 geriatric patients (17 male and three female) mean age 74 (60-87) years, mean body weight 68 (50-103) kg) received a single daily oral dose of 50 mg triamterene for seven days. Ten patients received, together with triamterene (TA), a single oral dose of 50 mg hydrochlorothiazide (HCT) for seven days; under equal conditions a concomitant single oral dose of 6 mg piretanide (PIR) was given to the other ten geriatric patients. Patients were entered into the two groups in randomized order. The eleven blood samples of the first 24 hours after first application were used for the determination of the (model-independent) kinetic parameters of triamterene and its phase-II-metabolite OH-TA-ester. In the second phase of the study six patients were entered in a cross-over comparison study between the two combinations, i.e. 3 patients received piretanide (as concomitant therapy) in the first phase of the study and got HCT in the second phase. To the other three patients, combinations were given vice-versa. Conditions of administration were identical in both phases of the study. The first study period (1 week) was separated from the second by a 7 day washout period. Plasma concentrations of triamterene and its metabolite OH-TA-ester were determined according to the method described by Grebian et al. (1976).

Results and Discussion

Mean plasma concentrations of furosemide were higher (Fig. 1) and mean total (plasma) clearance of furosemide was reduced in the elderly patients (Fig. 2)

285

8-

$ 6. 5

zzGERIATRICPATIENTS[N=203

•

evoUNGVOLUNTEERS[N: 8]

t,, 3

0.51

2

3

L

5

~i

Fig. 1. Concentration vs. time curve of furosemidein geriatric patients (n = 20) compared to a group of healthy volunteers (n = 8).

compared with a control group of younger adults (Beermann et al., 1977; Andreasen et al., 1978; Keller et al., 1981). In the same way, renal clearance was markedly lower in the geriatric group. Our data suggest a special function of alpha-2-globulins in binding of furosemide. Renal function (i.e., creatinine clearance) was shown to be a n important parameter for estimating the elimination rate of furosemide (Fig. 3). Comparing the pharmacokinetic parameters of the metabolites of spironolactone (canrenone and fluorigenic metabolites) with the respective values of a control group of younger healthy female subjects, the serum concentrations in the elderly patients were found to be twice as high (Fig. 4). In addition, a statistically significant positive correlation with age was found in the patients for the kinetic parameters investigated (Table I). A positive correlation existed between erythrocyte count and the pharmacokinetic parameters of canrenone

250' 200.

!

":.-

=

•

I

150" t,l.i

..~ 100"

.r .g. P g

50-

[N-12)

iN,20)

Fig. 2. Total clearance (plasma clearance) of furosemide in the geriatric patients and the corresponding values in a group of younger volunteers(Keller et al., 1981; Beermann et al., 1977).

286

~ , 25 20 15 10

• •

Y '

o o ~

•

•

• -o • o ~ o o • •

5

2'0

~o

6'o

1"=--0.502

n=20 "'~.01

o'o

CREI01NINE~

1~o

[mt/min't

Fig. 3. Inverse relationship between the creatinine clearance and the total A U C of furosernide in the geriatric patients (r = - 0.60, n = 20, p < 0.01.)

indicating that, in addition to the known high binding of canrenone to plasma proteins, there may be a possible (restrictive) binding to erythrocytes (Table I). The ratio of the high pressure liquid chromatography (HPLC)-measured canrenone concentration (actual canrenone) to the fluorimetrically determined 'canrenone' (fraction of several fluorigenic metabolites) was also displaced in the elderly patients. This difference of about 13% is statistically significant. This indicates that the elderly patients have relatively less canrenone and a higher fraction of fluorigenic metabolites. A possible explanation of this would be a capacity-limited metabolic breakdown of spironolactone in the elderly patients. A 1~1200.0

I--

-

0 0~

- ELDERLY PEOPLE ( N = 1 0 )

~---~- YOUNG PEOPLE ( N = 10)

~ 960.0

~ 720.0

0

~ 480.0

z u ~, 240.0

0.0

J . ~

,

~

,

,

,

i

r

f

92.0

96.0

100.0

104.0

108.0

112.0

110.0

120.0

HOURS

Fig. 4. Serum concentration curves (means:l: S.E.) for fluorimetrically determined 'fluorigenic metabolites.' e, geriatric patients; A, younger subjects.

287 TABLE I Correlation between clinical and laboratory diagnostic findings and the pharmacokinetic parameters (canrenone determination by H P L C a) Pharmacokinetic parameter

Clinical or laboratory diagnostic parameter

n

rs

p (2-tailed)

AUC b

Erythrocytes Cardiac output Age of patient

10 5 10

0.869 - 0.925 0.727

< 0.005 < 0.05 < 0.05

M a x i m u m concentration in the serum

Erythrocytes Cardiac output Age of patient

10 5 10

0.766 - 0.925 0.654

< 0.02 < 0.05 < 0.05

Concentration in serum after 120 h (end of last 'steady-state' interval)

Erythrocytes Hemoglobin Hematocrit Age of patient

10 10 10 10

0.803 0.785 0.690 0.690

< < < <

0.01 0.01 0.05 0.05

a High-pressure liquid chromatography. b Area under the time-concentration curve.

Mean peak concentration of triamterene was 84.0 + 90.6 n g / m l in the elderly patients (n = 20) compared with 41.4 + 19.6 n g / m l (n = 7) in young healthy volunteers (Knauf et al., 1980). Peak concentration of triamterene (TA) and OH-TA-ester as well as area under curve AUC (0-24 h) of the OH-TA-ester were closely correlated with the urea plasma levels in the geriatric patients (Fig. 5). The most important result of this study, however, seems to be the distinct influence of different concomitant diuretic therapies (50 mg hydrochlorothiazide vs.

zoooo.

isooo, l

7

10000,

b ~sooo &

,.,

,,"

80 120 IBO UREfl (MG/DL) Fig. 5. Correlation between A U C (0-24 h p . a . ) o f t h e O H - T A - ~ t e r a n d plasma u r e a l e v e l s i n t h e g e f i a t f i c patients(r = 0.754, n = 20, p < 0.001). 40

288 TABLE II

Pharmacokinetic parameters of TA and OH-TA-ester: comparison between two independent groups (geriatric patients) with different concomitant therapy (50 mg HCT vs. 6 mg piretanide) Combination

AUC (0-12 h)

c24 h tri-

AUC (0-12 h)

c24 h OH-

of drugs

triamterene

amterene

OH-TA-ester

TA-ester

(ng. h / m l )

(ng/ml)

(ng. h / m l )

(ng/ml)

765.0+837.4

99.l_+193.6

3348.1+1637.2

233.4+__182.5

280.0+204.0 p < 0.05

27.7+33.2 n.s.

4257.3+3506.6 n.s.

268.3_+176.5 n.s.

50 mg TA + 50 mg HCT (n =10) 50 mg TA + 6 mg piretanide (n =10)

U-test, 2-tailed

n.s., non-significant; c24 h, concentration after 24 h; AUC, area under curve.

6 mg piretanide) on the kinetic parameters of triamterene (and OH-TA-ester) in our geriatric patients. This influence could be demonstrated by a comparison between two independent groups (each with 10 geriatric patients, see Table II) and by a cross-over comparison in six geriatric patients (Fig. 6 and Table III). After coadministration of hydrochlorothiazide (as compared with the coadministration of piretanide) the mean A U C and mean plasma concentration after 24 h (c24 h) of triamterene were significantly higher. This could also be shown for the mean c24 h of triamterene and the OH-TA-ester in the cross-over study. A possible explanation for this result may be found by the observation that the glomerular filtration rate (GFR) is impaired (at least for some time) after administration of thiazides as well as after administration of potassium-sparing diuretics like triamterene (Knauf and

IE "N

500

400 u-

~-

300

2OO -

100

c~

TR

+

HCT

TR

+

PIR

Fig. 6. Plasma levels of the OH-TA-ester measured 24 h after administration: cross-over comparison of different concomitant diuretics (TA + HCT = 50 mg triamterene + 50 mg hydrochlorothiazide vs. TA + PIR = 50 mg triamterene + 6 mg piretanide) in six geriatric patients ( p < 0.025, 1-tailed t-test for paired

values).

289 TABLE III Pharmacokinetic parameters of TA and OH-TA-ester: cross-over comparison between different concomitant diuretics (50 mg HCT vs. 6 mg piretanide) in six geriatric patients Combination of drugs

50 mg TA + 50 mgHCT 50 mg TA + 6 mg piretanide t-test for paired values, 1-tailed

AUC (0-24 h) triamterene (ng. h/ml)

c24 h triamterene (ng/ml)

AUC (0-24 h) OH-TA-ester (ng- h/ml)

c24 h OHTA-ester (ng/ml)

606+191

33.9+24.3

6337+ 945

248+147

269+221

8.3+14.4

4354+3232

85+ 80

p < 0.05

p < 0.05

n.s.

p < 0.025

AUC, area under curve; c24 h, concentration after 24 h; n.s., non-significant.

Mutschler, 1987). With increasing age, glomerular filtration rate, renal plasma flow, and tubular excretory capacity as well as creatinine and urea clearance are significantly reduced (Schramm et al., 1981). If creatinine clearance is less than 30 ml/min, thiazides lose their diuretic activity; loop diuretics like piretanide ('highceiling' diuretics) are still effective. As compared with hydrochlorothiazide, the coadministration of piretanide does not impair the GFR and is still effective in reduced renal function, which may explain the results observed in our study. Since a second rise in the plasma levels of triamterene was observed in 16 of the 20 geriatric patients (a second rise of OH-TA-ester levels was found in eleven patients), the danger of a possible accumulation of triamterene and OH-TA-ester cannot be excluded. Viewing the results of our kinetic studies about diuretics in old age two aspects should be stressed: 1. Consequences for clinical pharmacology: in geriatric patients reduction or correction of the dosage seems to be necessary for all of the diuretics investigated in our studies. 2. Since age-dependent changes (at the molecular, cellular and organism level) are responsible for most changes in the kinetics not only of diuretic drugs, an analysis of these altered kinetics may also be a useful contribution to basic research in experimental gerontology.

References Andreasen, F., Hansen, H.E. and Mikkelsen, E. (1978): Pharmacokinetics of furosemide in anephric patients and in normal subjects. Eur. J. Clin. Pharmacol., 13, 41-48. Beermann, B., Dalen, F. and Lindstrom, B. (1977): Elimination of furosemide in healthy subjects and in those with renal failure. Clin. Pharmacol. Ther., 22, 70-78. Grebian, B., Geissler, H.E. and Mutschler, E. (1976): (lber die Bestimmung von Triamteren, Hydroxytriamteren und Hydroxyschwefels~ureester in biologischem Material durch direkte Auswertung von Diannschichtchromatogrammen. Drug Res., 26, 2125.

290 Keller, E., Hoppe-Seyler, G., Mumm, R. and Schollmeyer, P. (1981): Influence of hepatic cirrhosis and end-stage renal diseases on pharmacokinetics of furosemide. Eur. J. Clin. Pharmacol., 20, 27-33. Knauf, H., M~Shrke, W., Mutschler, E. and ViSlger, K.D. (1980): Zur Bioverfiigbarkeit von Hydrochlorothiazid und Triamteren aus Fertigarzneimitteln. Drug Res., 30, 1001-1004. Mi~hlberg, W., Platt, D. and Neubig, E. (1986): Pharmakinetics and pharmacodynamics of furosemide in geriatric patients. Arch. Gerontol. Geriatr., 5, 249-263. Platt, D., Abshagen, U., Miihlberg, W., Horn, H.J., Schmitt-Ri~th, R. and Vollmar, J. (1984): The influence of age and multimorbidity on the pharmacokinetics and metabolism of spironolactone. Arch. Gerontol. Geriatr., 3, 147-159. Schramm, A., Jennett, M. and Gerhardt, K.H. (1981): Veranderungen der Nierenfunktion und -morphologie im Alter. Z. Gerontol., 14, 354-369.