Eur J Clin Pharmacol (1994) 46:9-13 European Journa, o, (~(~(~
5>BG ecefleg @ Springer-Verlag 1994
The saluretic effect of the thiazide diuretic bemetizide in relation to the glomerular filtration rate H. Knauf 1, W. Cawello 2, G. Schmidt 2, E. Mutschler 3
1 Medizinische Klinik I, St. B ernward-Krankenhaus, Hildesheim, Germany 2 Abteilung Biomathematik und Medizinische Forschung, Schwarz Pharma AG, Monheim, Germany 3 Pharmakologisches Institut for Naturwissenschaftler, Universit~it Frankfurt/Main, Germany
Received: 17 October 1991/Accepted in revised form: 16 October 1993
Abstract. The effect of the thiazide diuretic, bemetizide, on the excretion of Na +, K +, CI-, Ca 2 +, and Mg 2 in rela- tion to the glomerular filtration rate (GFR) was studied in 17 subjects whose creatinine clearances ranged from 133 to 5 ml. min- 1.
After a 2-day fluid and salt balanced control period, 25 mg bemetizide given orally induced natriuresis and kaliuresis which lasted for 24 h and were proportional to the GFR of the patients. The ratio of bemetizide-induced K+/Na excretion was always 0.17 irrespective of individ- ual GFR. In renal failure, bemetizide increased the frac- tional Na + excretion from 3 % to about 10 %. Kaliuresis was associated with magnesiuria, whereas bemetizide-in- duced calciuresis was insignificant. The thiazide revers- ibly lowered GFR in all subjects.
Key words: Bemetizide, Salt excretion pattern; Na +, K + , CI-, Mg 2+ , Ca 2 +, chronic renal failure, renal haemodyna- mics.
It is well documented for loop diuretics [8, 14, 16] that there is a linear relationship between the magnitude of the saluretic effect and the glomerular filtration rate (GFR) of the patients. This may be interpreted as if the efficacy of the loop diuretics were proportional to the number of remaining intact nephrons in renal failure. As the furo- semide-type loop diuretics have been shown  to inter- fere with the tubulo-glomerular feedback mechanism, these "high ceiling diuretics" induce maximal diuresis by preventing the GFR from being depressed during peak diuresis. This quality renders these drugs useful in ad- vanced renal failure as well as acute left ventricular heart failure .
Thiazides - usually termed low ceiling diuretics - are the diuretics of choice for long-term treatment in cardio- vascular disease [6, 11]. This is because of their long dura-
Correspondence to: H. Knauf, Medizinische Klinik I, St. Bernward- Krankenhaus, D-31134 Hildesheim, Germany
tion of action, which allows a once a day regimen. Several pharmacodynamic features render this class of diuretics relatively safe in long-term treatment. For example, they prevent calcium loss - in contrast to loop diuretics - which might be dangerous in osteoporosis of the elderly [5, 21]; and they maintain tubulo-glomerular balance and thus impede rapid volume depletion . Thiazides are, how- ever, relatively K + wasting [9,11]. This risk can be avoided by comedication with antikaliuretics [23, 29]. In patients with renal disease, thiazides are usually replaced by loop diuretics. It is generally believed that once the GFR de- dines below 30 ml. min- 1, only the loop diuretics continue to be effective diuretic agents [7, 31]. Such a "cut-off" at a finite GFR is difficult to explain on the basis of the adap- tive changes in sodium handling by the remaining ne- phrons .
The following single-dose experiments with the thia- zide diuretic bemetizide [10, 25] focus on the pattern of electrolyte excretion as well as on the GFR in relation to the kidney function of patients with compensated chronic renal failure.
Materials and methods
This trial, approved by the local ethics committee, was carried out on 17 subjects who had given their informed consent to the study. The demographic and clinical data are given in Table 1. The participants were kept on a standard diet, receiving a defined fluid (1.5 l/day) and electrolyte (12 g NaC1 -- 200 mmol Na + ) intake 3 days prior to and throughout the trial [see 13, 14, 16, 18].
First, each subject was shown to be in steady state with regard to urinary electrolyte excretion. This was accepted when the cumulative urinary ion excretion data plotted over a 2-day period at 12 h intervals prior to administration of the diuretic formed a straight line (Fig. 1). To evaluate the effectiveness of bemetizide, urine electrolyte excretion was examined for 2 days following a single oral dose of 25 mg bemetizide before breakfast. Urine was collected over the following time intervals: 0-3, 3-6, 6-9, 9-12, 12- 24, and 24-48 h. Na +, K , Ca 2+ , Mg 2+ and C1- were measured in these samples.
The analysis of Na +, K +, Ca 2 + and Mg 2 + was done using a Zeiss automatic electrolyte analyser FL 6. For analysis of Na + and K + the samples were diluted in 1:500 with a La203 solution, and for the
Table 1. Individual demographic and clinical data of patients
Patient Sex Age Height Weight Disease Creatinine clearance (years) (cm) (kg) (ml. min- 1)
A.H. M 60 175 89.0 CG 5 J.J. M 62 169 70.0 CG 10 H.K. M 57 156 57.7 CG 11 J.L. M 63 169 67.0 CG 12 P.R. F 69 167 55.8 CP 17 R.L. F 72 179 71.0 CP 26 K.J. M 76 178 73.0 DN 34 M.A. M 65 165 74.0 - 53 B.R. F 38 165 56.0 DN 70 S. El. F 31 169 55.0 CP 72 S. Ev. F 31 170 56.0 CP 78 M.B. M 28 185 76.0 - 87 R.W. M 29 172 75.0 - 96 A.K.B. M 29 178 67.0 - 110 K. S.W. F 33 172 70.0 - 117 S.T. M 34 172 65.0 - 123 W.G. M 34 173 67.0 - 133
CG, Chronic glomerulonephritis; CR chronic pyelonephritis; DN, diabetic nephropathy
No + excretion rote (retool~h) I00 (ng /ml ) 45 ",
0 - ,~8
J , L , , , 20- -36 -24 -12 0 12 24 g8
No excretion, cumulative (mmol)
-~8 -36 -2~ - 12 0 12 2/, ~8
time after administration (h)
Fig. 1. Time course of Na + excretion following a single oral dose of 25 mg bemetizide in a healthy subject. Top, Sodium excretion rate; bottom, cumulative sodium excretion. Tdenotes the duration of action of bemetizide, and the dotted line represents the mean Na + excretion rate extrapolated from the pretreatment 48-h period ( - 48-0). ANa is the bemetizide-induced Na excretion
b b Js t ime (h)
Fig.2. Time course of plasma concentrations of parent bemetizide following the administration of 25 mg bemetizide. Data represent medians.., CLca > 90 ml/min; O, CLcR < 90 ml. min- 1
and A Mg 2 +. The fractional sodium excretion, FENa (%), represents the amount of Na + excretion (retool/time) given in percent of fil- tered load (plasma sodium times GFR). These data are related to the GFR, measured by creatinine clearance. Regression analyses were performed by the least squares method.
Serum and urine samples were stored at - 20 C until analysed. Plasma concentrations were determined before and 2, 3, 4, 6, 8, 9,12, 24, and 48 h after administration of bemetizide. Urine was collected over the time invervals given above for analyses of electrolytes. The plasma concentrations of bemetizide were determined by HPLC. The limit of detection was 10 ng. ml- 1 .
Ca 2+ and Mg z+ analysis the samples were diluted in 1:250 with a CsC1 solution. Chloride (50-btl samples) was determined titrimetri- cally (via AgNO3) with an Eppendorf Chloridemeter 6610. The ef- fectiveness of bemetizide was quantified by determining the excre- tion of each electrolyte during the period of action (T) minus the con- trol excretion assessed over the corresponding pretreatment period (bottom of Fig. 1). This was defined as A CI-, A Na +, A K +, A Ca 2 +,
Bemetizide in volunteers with normal kidney function
F igure 1 i l lustrates a representat ive exper iment in a hea l thy control . The oral admin is t ra t ion of 25 mg bemet i - z ide caused marked natr iures is ( top of the f igure), wh ich was max imal between the 3rd and 12th h and lasted for
~11111111~ x\ \ \ \ \ \ \ \ \ "
V/ / / / / / /A~\ \ \ \ \ \ \N ~/ / / / / / /~
" M H H H ,
Fig. 3. The influence of 25 mg bemetizide on the subjects' creatinine clearances. The pretreatment values were taken as 100 %. Data from 15 subjects with complete data (before, during and after treatment): , subjects with creatinine clearance > 90 ml. min ~ (n = 8); ~, subjects with creatinine clearance < 90 ml. min-~ (n = 7)
24 h. The excretion curve exhibited two peaks: at 3 and 10 h after drug administration. Then the curve fell below the pretreatment level (the latter effect can be defined as "rebound"). The period during which natriuresis ex- ceeded the pretreatment (control) value was defined as the "duration of action", z- (h). A double peak of natri- uresis was also seen in patients with advanced renal failure (see below). This phenomenon cannot be related to paral- lel changes in plasma concentrations of the drug, as the plasma level time curves were always continuous (Fig. 2). Rather, the reversible fall in natriuresis was asso- ciated with a reversible reduction of the GFR of the pa- tients (Fig.3). The pretreatment creatinine clearances (see Table 1) were taken as 100% and were then com- pared to those determined during the maximal effect of bemetizide (3rd to 10th h) and during the 2nd day after drug administration. As shown in Fig. 3, 25 mg bemetizide reduced the GFR to about 50 % of control in normals as well as in patients with advanced renal failure. This effect was completely reversible.
Table 2, Bemetizide-induced electrolyte excretion in healthy con- trols
Peak diuresis Cumulative excretion during (mmol. h 1) duration of action r (mmol)
Na + C1- K + Ca 2 + Mg z+
zXK+/2xNa + ACaZ+/ANa + AMg2+/AK +
36.5 215 26.6 171 9.3 31 0.41 1.3 0.80 3.15
0.17 0.003 0.1
At the bottom of Fig. 1, the bemetizide-induced excre- tion of sodium has been quantified by adding the hourly excretion of sodium to obtain the cumulative sodium ex- cretion curve. The diuretic-induced excretion of electro- lytes during r (minus control excretion) was defined as A Na +, A K + , A C1-, A Ca 2+, A Mg =+ . The maximal excre- tion rate as well as the cumulative excretion of electrolytes during r in healthy controls are summarized in Table 2.
Bemetizide in chronic renal failure
The efficacy of bemetizide depended on renal function: the magnitude of drug-induced electrolyte excretion de- creased with decreasing renal function. The duration of drug action increased from 24 h in healthy controls to about 48 h in end-stage renal failure. However, the rate of electrolyte excretion during the 2nd day after bemetizide administration to patients with advanced renal failure ex- ceeded the control excretion by only 10 %. Therefore, to quantify the efficacy of bemetizide in relation to renal function, A Na +, A K +, A C1 , A Ca 2+, and A Mg =+ were referred to the 24-h period after dosing and correlated with the individual creatinine clearance (Fig. 4). For every ion in question, this relationship could roughly be de- scribed by linear functions, which met the zero point of the x-axis. This means that the saluretic effect of the thiazide
ANo*/24h=~-, &Cl-/2~h=~.._ (mmol) &Ca+~'24h (mmol) 500-
0 o 2~ ~'O 6~
do ~bo 13o 1~o
Z~K/24h (mmol) 120
20 .'o 6'0 8'o 13o 13o 1~o CLcR (ml/min)
O _" L
"e"W, - - J
~o ;o s'o 8'o ibo 1~o .o
2'0 .b 6'o do lbo 1)o 1~o CLcR (ml/min)
Fig. 4. Correlation between the bemetizide-induced net excretion of electrolytes during 24 h and the creatinine clearances of the sub- jects
FENa [%1 25
20 FEN." GFR = const
5 O0 X I) 0
0 V.xX X Y, X X,)(. X
20 40 60 80 1 O0 120 140
Fig.5. Fractional sodium excretion of the subjects as a function of creatinine clearance before (x) and after () administration of 25 mg bemetizide. The normal hyperbolic function (FENa * GFR = const) is given according to Slatopolsky et al. 
diuretic was proportional to the number of remaining in- tact nephrons. The ratio of bemetizide-induced kaliuresis related to natriuresis during the 24-h period of action was 0.17 irrespective of the GFR, i.e. for every remaining intact nephron. The same held true for the ratios A Ca2+/A Na + and zX Mg2+/A K +, which were 0.003 and 0.1, respectively (Table 2).
The fractional excretion of sodium, FENa (%), is about 0.5 in healthy controls and is doubled by halving GFR, i. e. the relation between FENa and GFR follows a hyperbolic function [1, 28]. Bemetizide increased FENa from 0.5 to 3 % in normals and to 10 % and more in end-stage renal failure (Fig. 5).
Efficacy of bemetizide in relation to GFR
The trial demonstrates that the thiazide diuretic, bemeti- zide, strongly increases the excretion of Na and C1-, and also increases K excretion proportional to Na excre- tion. Excretion of Mg 2+ is small and Ca 2+ excretion is scarcely influenced. Strikingly, bemetizide did induce a saluretic effect down to end-stage renal disease, present- ing the linear relationship for salt excretion and GFR known for loop diuretics [14, 16]. Thus, there was no "cut- off" at a GFR
respectively, of about 35 mmol . h 1 in controls and about 15 mmol .h -1 in advanced renal failure. This dose-de- pendent effect has to be regarded as a physiological regu- latory mechanism to prevent vo lume deplet ion. Since furosemide-type diuretics specif ical ly b lock the Na+:K :2C1- cotransporter , which is also present in the macula densa cells , these diuretics charac- teristical ly interfere with the tubulo-g lomeru lar feedback mechanism and thus interrupt the renal feedback loop. Consequently, diuresis fol lowing the administrat ion of loop diuretics is not self - l imited as with thiazides and may therefore cause dangerous vo lume loss fol lowing high dosing.
1. Bricker NS, Fine LG, Kaplan MA, Epstein M, Bourgoignie JJ, Licht A (1978) "Magnification phenomenon" in chronic renal disease. N Engl J Med 299:1287-1293
2. Ellison DH, Velazquez H, Wright FS (1987) Thiazide-sensitive sodium chloride cotransport in early distal tubule. Am J Physiol 253:F546-554
3. Epstein M (1988) Diuretic therapy in liver disease. In: Epstein M (ed) The kidney in liver disease, 3rd edn. Williams & Wilkins, Baltimore, pp 3-30
4. Epstein M, Lepp BA, Hoffmann DS, Levinson R (1977) Poten- tiation of furosemide by metolazone in refractory edema. Curr Ther Res 21:656-667
5. Felson DT, Sloutskis D, Anderson JJ, Anthony JM, Kiel DP (1991) Thiazide diuretics and the risk of hip fracture. Results from the Framingham Study. JAMA 265:370-373
6. Flack JM, Grimm RH (1989) Diuretics. In: Kaplan NM, Brenner BM, Laragh JH (eds) New therapeutic strategies in hyperten- sion. Raven Press, New York, pp 17-32
7. Gerber JG (1981) Antihypertensive agents and diuretics. In: An- derson R J, Schrier RW (eds) Clinical use of drugs in patients with kidney and liver disease. Saunders, London, pp 211-229
8. Greger R, Wangemann P (1987) Loop diuretics. Renal Physiol 10:174-183
9. Hollifield JW (1986) Thiazide treatment of hypertension. Effects of thiazide diuretics on serum potassium, magnesium and ven- tricular ectopy. Am J Med 80 [Suppl 4A]: 8-12
10. Hoppe H-J (1980) Untersuchung zur Dosis-Wirkungs-Bezie- hung und Wirkungsdauer von Bemetizid und der Kombination Bemetizid/Triamteren (1:2) bei Probanden. In: Krtick E Schrey A (eds) Diuretika. Springer, Berlin Heidelberg New York, pp 141-148
11. Kaplan NM (1990) Diuretics versus other agents in the therapy of primary (essential) hypertension. In: Puschett JB, Greenberg A (eds) Diuretics III: chemistry, pharmacology, and clinical ap- plications. Elsevier Science, Amsterdam, pp 40%410
12. Keller E, Hoppe-Seyler G, Mumm R, Schollmeyper P (1981) In- fluence of hepatic cirrhosis and end-stage renal disease on phar- macokinetics and pharmacodynamics of furosemide. Eur J Clin Pharmaco120:27-33
13. Knauf H, Mutschler E (1984) Pharmacodynamics and pharma- cokinetics of xipamide in patients with normal and impaired kid- ney function. Eur J Clin Pharmaco126:687-693
14. Knauf H, Mutschler E (1990) S aluretic effects of the loop diuretic torasemide in chronic renal failure. Interdependence of electro- lyte excretion. Eur J Clin Pharmaco139:337-343
15. Knauf H, Mutschler E (1991) Pharmacodynamic and kinetic con- siderations on diuretics as a basis of differential therapy. Klin Wochenschr 69:23%250
16. Knauf H, Mutschler E (1992) Constant K+/Na excretion ratio by piretanide during peak diuresis, but insignificant K t loss dur- ing 24 hour. Eur J Clin Pharmaco143: 23-27
17. Knauf H, Mutschler E (1993) Low-dose segmental blockade of the nephron rather than high-dose diuretic monotherapy. In: Puschett JB (ed) Diuretics IV. Elsevier, Amsterdam, pp 449-456
18. Knauf H, Mutschler E (1993) K + loss induced by diuretics. In: Puschett JB (ed) Diuretics IV. Elsevier, Amsterdam, pp 253-256
19. Knauf H, Mutschler E (1993) Low-dose segmental blockade of the nephron rather than high-dose diuretic monotherapy. Eur J Clin Pharmaco144 [Suppl 1]: 63~58
20. Knauf H, Reuter K, Mutschler E (1985) Limitation on the use of amiloride in early renal failure. Eur J Pharmaco128:61-66
21. LaCroix AZ, Wienpahl J, White LR, Wallace RB, Scherr PA, George LK, Cornoni-Huntley J, Ostfeld AM (1990) Thiazide diuretic agents and the incidence of hip fracture. N Engl J Med 322:286-290
22. Meyer TW, Scholey JW, Brenner BM (1991) Nephron adapta- tion to renal injury. In: Brenner BM, Rector FC (eds) The kidney. Saunders, Philadelphia, pp 1871-1908
23. Multicenter Diuretic Study Group (1981) Multiclinic compari- son of amiloride, hydrochlorothiazide, and hydrochlorothiazide plus amiloride in essential hypertension. Arch Intern Med 141: 482-486
24. Oster JR, Epstein M, Smoller S (1983) Combined therapy with thiazide-type and loop diuretic agents for resistant sodium reten- tion. Ann Intern Med 99:405-406
25. Piper C, Cawello W, Bonn R, Weber E (1983) Pharmakodyna- mik und Pharmakokinetic von Bemetizid im Vergleich zu Hy- drochlorothiazid. Arzneim-Forsch/Drug Res 33 (11) 7:988-996
26. Schlatter E, Salomonsson M, Persson AEG, Greger R (1990) Macula densa cells reabsorb NaC1 via flurosemide sensitive Na -K +-2C1--cotransport. In: Puschett JB, Greenberg A (eds) Diuretics III: chemistry, pharmacology, and clinical applications. Elsevier, Amsterdam, pp 756-758
27. Sigurd B, Olesen KH, Wennevold A (1975) The supra-additive natrinretic effect of addition of bendroflumethiazide and bu- metanide in congestive heart failure. Am Heart J 89:163-170
28. Slatopolsky E, Eikan I, Weerts C, Bricker NS (1968) Studies on the characteristics of the control system governing sodium excre- tion in uremic man. J Clin Invest 47:521-530
29. Tannenbaum P J, Crosley AP (1966) A comparison of the effects of hydrochlorothiazide and hydrochlorothiazide in combination with triamterene on electrolyte balance. Clin Pharmacol Ther 7: 777-782
30. VillareaI H, Revollo A, Exaire JE, Larondo F (1966) Effects of chlorothiazide on renal hemodynamics. Circulation 26:409-412
31. Wilcox CS (1991) Diuretics. In: Brenner BM, Rector FC (eds) The kidney. Saunders, Philadelphia, pp 2123-2147
32. Wollam GL, Tarazi RC, Bravo EL, Dusten HP (1982) Diuretic potency of combined hydrochlorothiazide and furosemide ther- apy in patients with azotemia. Am J Med 72:929-938