SYSTEMIC HYPERTENSION

Early Changes in Plasma and Urinary Potassium in
Diuretic-Treated Patients with Systemic Hypertension

VASILIOS PAPADEMETRIOU, MD, MICHAEL PRICE, MD, ELIZABETH JOHNSON, RN,
        MARLENE SMITH, RN, and EDWARD D. FREIS, MD

Two groups of patients with uncomplicated systemic
hypertension were studied. Group 1 included 11
patients who had overt hypokalemia with diuretic
drug treatment, and group 2 included 11 patients
who remained normokalemic. After baseline studies
without treatment were performed, both groups
received hydrochlorothiazide, 50 mg twice daily.
Plasma potassium (PK) was significantly reduced
within the first day of treatment and stabilized by day
7 in both groups. The average decrease in PK was
1.0 f 0.1 mEq/liter (p <O.Ol) in the first group and
0.6 f 0.2 mEq/liter (p <O.Ol) in the second group.
Cumulative losses of K were approximately 200
mEq in the hypokalemic group and were minimal in

Diuretic-induced hypokalemia has been a matter of
major concern, leading often to the routine use of po-
tassium chloride (KCl) supplements and K-sparing
diuretic therapy. For example, in 1981 more than 7.5
million prescriptions for KC1 and 19 million prescrip-
tions for K-sparing diuretic drugs were prescribed for
patients with hypertension, at a cost exceeding $250
milli0n.i

Hypokalemia, defined as plasma potassium (PK)
13.5 mEqiIiter, occurs in approximately 20% of patients
with essential hypertension who receive diuretic ther-
apy.2b However, severe hypokalemia (PK <3.0 mEq/
liter) occurs in only a small percentage.4y5 Whether the
hypokalemia reflects a true deficit of total body K is
controversial. Some investigators6 found significant
decreases in the total body K with diuretic treatment,
whereas others found minimal reductions with long-
term diuretic treatment.T-lo Total body K remained
unchanged when KC1 supplements were added to di-
uretic therapy.sJi Although numerous studies have

From the Veterans Administration and Georgetown University Medical
Centers, Washington, D.C. Manuscript received April 26, 1984; revised
manuscript received July 6, 1984, accepted July 20, 1984.
Address for reprints: Vasilios Papademetriou, MD, Hypertension
Research, Veterans Administration Medical Center, 50 Irving Street,
N.W., Washington, D.C. 20422.

the normokalemic group as assessed by 24-hour
urinary collections. Patients in the hypokalemic
group also had a greater reduction in body weight
and blood pressure. Supplementation with KCI, 96
mEq/day, or triamterene, 200 mg/day, in 9 hypo-
kalemic patients resulted in an increase of PK to
approximately 3.5 mEq/liter leveling off by day 7,
and a cumulative K retention of approximately 200
mEq. Thus, overt thiazide-induced hypokalemia was
associated with small and biologically unimportant
losses of K from body stores. With replacement
therapy the estimated amount of retained K was also
small.

(Am J Cardiol 1964;54:1015-1019)

shown that diuretic therapy results in a mild decrease
of PK7-i1 and an initial increase in urinary K excre-
tion,12J3 studies assessing the duration and severity of
kaliuresis in patients with overt hypokalemia are
scarce.

This study investigated 2 groups of hypertensive
patients. The first group included patients known to
have overt hypokalemia (PK 13.2 mEq/liter) when
receiving diuretic therapy. The second group main-
tained normal PK (23.5 mEq/liter) when receiving di-
uretic therapy. The following questions were studied:
(1) Are there any differences between these 2 groups in
blood pressure (BP) response, fluid loss or electrolyte
excretion in the urine? (2) How rapidly does hypoka-
lemia occur and how much K is lost in the urine? (3) In
the hypokalemic patients, when large doses of KC1
supplements or K-sparing diuretic drugs are added,
how much K is retained and how soon is hypokalemia
corrected?

Methods

Patients: Twenty-two patients with mild to moderate hy-
pertension uncomplicated by heart or renal disease were
studied. Eleven of these patients had overt hypokalemia while
receiving therapy, whereas the other 11 remained normo-
kalemic. All patients were black. The average age was 54 f 3

1015


1016   DIURETIC-INDUCED POTASSIUM CHANGES

TABLE I  Control Data Before Diuretic Treatment

Variables

Normokalemic  Hypokalemic
  Patients      Patients

Age W
Body weight (Kg)
Blood pressure
Systolic (mm Hg)
Diastolic (mm Hg)
Heart rate
Plasma K (mEq/liter)
Plasma Na (mEq/liter)
Plasma Cl (mEq/liter)
Plasma creatinine (mg/dl)
Urine volume (ml/24 hr)
Urine K (mEq/24 hr)
Urine Na (mEqI24 hr)
Urine Cl (mEq/24 hr)
Urine creatinine (a/24 hr)

Creatinine clear&e (mlimin)

54 f 3      56 f 2
a5 f 5       90 f 4

150f6      152 f 7
97 f 6       99 f 4
74 f 4       77 f 4
4.4 f 0.1     3.9 f 0.1 o
143f 1      141 f 1
108 f 2      105 f 9

1.2 f 0.1     1.2 f 0.1
1,606 f 140   1,405 f 108
49 f 8       58 f 7
183 f 28     196 f 24
198 f 30     203 f 32
1.8 f 0.1     1.9 f 0.1
103f5      112 f 6

Values are mean f standard error of the mean.
+ Average baseline PK values were significantly different, p <O.Ol;
for all other values, the difference was not significant.

years in the normokalemic group and 56 f 3 years in the hy-
pokalemic group (mean f standard error of the mean).
Twenty patients were treated with a diuretic drug alone, 1
received metoprolol and the other received methyldopa in
addition to a diuretic drug. In the latter 2 patients, the dosage
of methyldopa and metoprolol remained unchanged
throughout the study.

Study protocol: The study was conducted in 4 phases:
baseline, diuretic, diuretic plus KC1 and diuretic plus triam-
terene. Only the hypokalemic patients proceeded to the third
and fourth phase. At each visit the following data were col-
lected: sitting BP, heart rate, body weight, plasma and 24-hour
urine, creatinine, K, sodium and chloride. At each visit the
patient brought with him the urine that had been collected
during the preceding 24 hours.

Screening phase: Diuretic treatment was discontinued for
at least 2 weeks and the patients were given KCl, 20 mEq twice
daily, for the first 5 days in order to restore any existing

INCREASE IN URINARY POTASSIUM

l+k-#
1 , y --`+ ___---- +
13   7    14        28     13   7    14        28

DECREASE IN PLASMA POTASSIUM

TIME (DAYS,

FIGURE 1. Changes in plasma and 24hour urinary potassium levels
during the diuretic treatment alone. Left, changes in the group of patients
who remained normokalemic; rlght, changes that occurred in the hy-
pokalemic group. Dotted lines represent mean f standard error of the
mean in the 7 patients who completed 4 weeks wtth hydrochlorothiazide
therapy. Solid lines represent mean f standard error of the mean of
11 patients. + p <0.05; o * p <O.Ol.

K deficits. Baseline data were collected at the end of this
period.

Diuretic phase: After the baseline studies, treatment with
hydrochlorothiazide (HCTZ), 50 mg twice daily, was begun
and the patients were seen on days 1,3,7,14 and 28 of diuretic
treatment. After completing the 4-week period with diuretic
treatment, patients whose PK was reduced to 13.2 mEq/liter
proceeded to the next phase. If a patient had hypokalemic
symptoms or a PK of 12.6, therapy with KC1 was begun ear-
lier without completing the 4-week period with diuretic drugs
alone.

Diuretic plus KC1 phase: Nine patients from the hypo-
kalemic group proceeded to the diuretic plus KC1 phase.
In addition to the diuretic drug, these patients received 96
mEq /day of supplementary KC1 in wax matrix tablets ("slow
K"). They were examined and blood and urine were collected
on days 1,3,7,14 and 28 of this phase.

Diuretic plus triamterene phase: After the diuretic plus
KC1 phase, KC1 was discontinued. Patients then received
diuretic therapy alone for a period of time sufficient to allow
hypokalemia to recur (15 f 3 days). After this second baseline
period the patients were given triamterene, 100 mg twice daily,
and were seen again on days 1,3,7,14 and 28 of this phase.

Results were analyzed using the Student t test for paired
observations. The study protocol was approved by the Re-
search and Development Committee of the Veterans Ad-
ministration Medical Center at Washington, D.C., and all
patients gave informed written consent.

Results

At baseline, the average BP was 150/97 mm Hg in the
normokalemic group and 152/99 mm Hg in the hypo-
kalemic group (Table I). Plasma and urinary electro-
lytes and plasma creatinine levels were within normal
limits in both groups. However, the average baseline PK
was significantly lower in the group that became hy-
pokalemic while receiving diuretic therapy (3.9 f 0.1 vs
4.4 f 0.1, p <O.Ol).

The mean 24-hour urinary creatinine excretion and
the average urinary volume indicated adequate urine

DIURETIC + KCL

INCREASE IN PLASMA K      2

100    INCREASE IN URINARY K

DIURETIC t TRIAMTERENE
  INCREASE IN PLASMA K

       DAVSOFTREATMENT           DAYS OF TREATMENT

FIGURE 2. Changes in plasma and 24-hour urinary potassium in 9 hy-
pokalemic patients who received replacement therapy. o p <0.05; o o p
<O.Ol; o ** p <O.OOl.


November 1, 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 54

1017

collections. The average creatinine clearance during this
control period was 103 f 5 ml/min in the normokalemic
group and 112 f 7 ml/min in the hypokalemic group.

Diuretic phase only: Figure 1 shows the changes in
plasma and 24-hour urinary K during the diuretic phase
in both groups of patients. In the normokalemic group,
PK decreased by 0.6 f 0.1 mEq/liter (p <O.Ol) on day
7 of diuretic treatment and stabilized thereafter. In the
hypokalemic group, PK decreased 0.4 f 0.1 mEq/liter
(p <0.05) on day 1 of diuretic treatment and by day 3
PK was 1.0 f 0.2 mEq/liter lower than the baseline
value (p <O.OOl). Two of these patients exhibited a PK
of 2.3 and 2.6 mEq/liter, respectively, on day 3 and
therapy with KC1 was begun. On day 7 of diuretic
treatment, 1 more patient was found with a PK of 2.5
mEq/liter while on day 14 another patient had a PK
level of 2.4 mEq/liter. These 2 patients also began
therapy with KC1 before completing the full 4-week
period with diuretic drugs alone. Thus, only 7 of the 11
patients completed the full 4 weeks of treatment (Fig.
1). After day 7, PK levels remained relatively stable.

There was no increase in the 24-hour urinary excre-
tion of K in the group of patients who remained nor-
mokalemic after HCTZ treatment. In the hypokalemic
group, the urinary excretion of K increased by 26 f 8
mEq/24 hours (p <0.05) during the first day of treat-
ment and by 41 f 8 mEq/24 hours (p <O.OOl) on the
third day. On day 7, the 24-hour urinary K excretion was
higher, but was not significantly different from baseline
values. It was estimated that the total net loss of K from
body stores probably did not exceed 200 mEq.

Addition of KCI: The left panel of Figure 2 shows the
changes of plasma and 24-hour urinary K when KCl, 96
mEq/day in divided doses, was added to the regimen of
9 hypokalemic patients. The average PK in patients
receiving diuretic treatment alone, before KC1 was in-
itiated, was 2.8 f 0.1 mEq/liter. The average PK in-
creased by 0.4 f 0.2 mEq/liter on day 1 and by 0.7 f 0.2
mEq/liter on day 3 of KC1 treatment. PK reached its
highest level of 3.8 f 0.2 mEq/liter on day 7, leveling off
thereafter. All patients were compliant, as judged by pill
counts, which indicated they had taken an average of
80 mEq/day. The difference between the 80-mEq intake
and the increase in urinary excretion of K represented
retained K. This indicated (Fig. 2) that approximately
50 mEq of K were retained on day 1 of KC1 treatment,
about 30 mEq/day from days 2 to 3 and approximately
20 mEq/day from days 4 to 7. Thereafter, essentially all
of the administered K was excreted in the urine. Thus,
the net total quantity of K retained during this period
probably did not exceed 200 mEq.

Addition of triamterene: After the second phase of
the study, KC1 was discontinued. Patients continued
taking HCTZ alone, which resulted in return of overt
hypokalemia (15 f 3 days). The average PK with di-
uretic therapy alone, before triamterene was 2.7 f 0.1
mEq/liter and the average urinary K excretion was 53
mEq/24 hours. As shown on the right panel in Figure 2,
PK increased on day 1 of the triamterene treatment by
0.5 mEq/liter, from 2.7 f 0.1 to 3.2 mEq/liter (p <O.Ol).
On day 3 PK averaged 3.2 f 0.1 mEq/liter and by day
7 it stabilized at about 3.5 to 3.6 mEq/liter. The average

K retention was 30 f 4 mEq/24 hours greater than
baseline (p <O.OOl) on day 3 and 13 f 6 mEq/24 hours
on the seventh day (difference not significant). On days
14 and 28 of triamterene treatment, the urinary excre-
tion of K was almost identical with the HCTZ control.
Again, it can be estimated that the cumulative net re-
tention of K during the HCTZ plus triamterene phase
probably did not exceed 200 mEq.

Sodium excretion, body weight, blood pressure
and creatinine: Figure 3 shows the changes in 24-hour
urinary excretion of sodium, body weight, BP and
plasma creatinine during the diuretic-alone phase in
both groups and the diuretic plus triamterene phase in
the hypokalemic group. No changes in the above oc-
curred during KC1 treatment and therefore these data
are not shown. The 24-hour urinary excretion of sodium
was significantly increased in both groups only on the
first day of diuretic treatment (Fig. 3). The decrease in
body weight from baseline was 1.2 f 0.1 kg (p <O.Ol) in
the normokalemic group and 2.1 f 0.1 kg (p <O.OOl) in
the hypokalemic group, and stabilized on day 3 of di-
uretic treatment. The average BP was significantly re-
duced in both groups. There was a small but statistically
significant increase in plasma creatinine levels in both

FIGURE 3. Mean (& standard error of the mean) changes in urinary
sodium, body weight blood pressure and plasma creatinine. Left, di-
uretic-alone phase in the normokalemic group; middle, diuretic-alone
phase in the hypokalemic group; right, diuretic plus triamterene phase
in 9 hypokalemic patients. A = change in plasma creatinine for all 9
patients; B = change in plasma creatinine in the 8 patients excluding
the patient with the greatly elevated plasma creatinine; o p <0.05, o o p
<O.Ol, o ** p <o.ooi.


1018    DIURETICINDUCED POTASSIUM CHANCES

TABLE II  Mean Reduction8 in Plasma Potassium Body       TABLE Ill  Number of Patient8 with Normalized Plasma
     Weight and Blood Pressure on Day 7 of             Potassium (23.5 mEq/liter) on Each Day of
     Hydrochlorothlazide Treatment                      Treatment

Plasma potassium

Normokalemic Hypokalemic p
Patients      Patients    Value

0.6 f 0.1     1.0 f 0.1    <0.05

Days

                      0   1   3   7   14   28

Potassium chloride
Triamterene       8 ;   5   6
                            2   4  9  z

(mEq/liter)
Body weight (kg)       1.1 f 0.3     2.2 f 0.4    <0.05
Systolic BP (mm Hg)    15f3      22 f 5      NS
Diastolic BP (mm Hg)      6f2      12f3   p <0.05

Values are mean f standard error of the mean.
BP = blood pressure; NS = not significant.

groups during the first 3 days of diuretic treatment, but
thereafter it returned to baseline. Table II compares the
changes in PK, body weight and BP between the 2
groups on the seventh day of diuretic treatment, be-
cause by that day most variables stabilized. The de-
crease in PK was significantly greater, as expected, in
the hypokalemic group, 1.0 f 0.1 vs 0.6 f 0.1 mEq/liter
(p <0.05). However, there was also a significantly
greater reduction in body weight and diastolic BP in the
hypokalemic group (Table II).

The addition of triamterene to HCTZ resulted in
further changes in sodium excretion, body weight, BP
and plasma creatinine level. The p values shown in
Figure 3 compare the values before and after the addi-
tion of triamterene. Sodium excretion increased by 76
f 18 mEq (p <O.Ol) during the first 24 hours of treat-
ment with triamterene. The sodium excretion levels
remained somewhat, but not significantly, higher than
that in the preceding period with HCTZ treatment
alone until day 7. Body weight was correspondingly
reduced by 1.7 f 0.2 kg (p <O.Ol) on day 3, after which
it remained stable. Although diastolic BP did not
change, systolic BP was further reduced with the ad-
dition of triamterene.

Plasma creatinine also increased. One patient had
marked elevation of plasma creatinine from 1.6 to 3.1
mg/dl on day 7, which receded to 2.7 mg/dl on day 14
and to 2.3 mg/dl on day 28 of triamterene treatment.
Curve A in Figure 3 represents the change in plasma
creatinine for all 9 patients and curve B, the change for
the 8 patients excluding the patient with the markedly
elevated plasma creatinine. The average plasma creat-
inine level was significantly elevated during treatment
with triamterene and remained so to the twenty-eighth
day of treatment. This was in contrast to HCTZ alone,
which was associated with only a transient increase.

As shown in Table III, KC1 or triamterene, even in
relatively large doses, normalized PK in only approxi-
mately half of the patients. At the end of the KC1 phase,
as at the end of the triamterene phase, only 5 of 9 pa-
tients had a normal level of PK (23.5 mEq/liter).

Discussion

One of the objectives in this study was to identify
differences between the patients who maintain normal
PK with diuretic therapy and those who become overtly
hypokalemic. The 2 groups of patients were therefore
selected on the basis of their known PK response to

diuretic drugs, but were comparable in other respects
(Table I). However, the average baseline PK level in the
patients who became hypokalemic was significant-
ly lower than that in patients who maintained their
PK levels within the normal range during thiazide
therapy.

Although the hypokalemic subjects started from a
lower baseline PK than the normokalemic group, they
also exhibited an absolute decrement in PK that was
considerably greater than that in the normokalemic
patients. The hypokalemic group also exhibited a
greater reduction in body weight and diastolic BP. Di-
uretic treatment in the hypokalemic group also was
associated with an average loss of about 200 mEq of K
in the urine, whereas no detectable increase was ob-
served in the normokalemic group.

A possible explanation for the observed differences
is that the hypokalemic patients had a greater response
to diuretic therapy as indicated by BP and body weight
changes. Greater diuresis made available more sodium
in the distal nephron to be exchanged with K, which in
turn would explain the greater increase in the 24-hour
urinary excretion of K that occurred in the hypokalemic
group.

Although 24-hour urine samples were not obtained
every day, there was a sufficient number collected at
strategic points to estimate the total net losses of body
K with thiazides. The loss even in the hypokalemic
group was small, averaging approximately 200 mEq.
Because total body K approximates 4,000 mEq,4 thia-
zide treatment resulted in an average total body K loss
of about 5%. Maronde et all3 also found that the K losses
occurred during the first 3 days of diuretic therapy and
were less than 300 mEq. Kassirer and Harrington4 re-
viewed the pertinent published reports and concluded
that the decrease in total body K due to diuretic therapy
is small, even after long-term continuous treatment with
thiazides.

The hypokalemic patients in this study were selected
on the basis of exhibiting overt hypokalemia during
thiazide treatment. The dose of 50 mg of HCTZ twice
daily is more conducive to the development of hypo-
kalemia than are smaller doses. However, despite this
dosage and the relatively large decreases in PK, re-
ductions in total body K were small. Therefore, even in
this selected group with extracellular K reduced by
more than 25%, thiazide-induced hypokalemia did not
reflect a major deficit of total body K.

These observations suggest that there is a change in
the equilibrium between extracellular and intracellular
concentrations of K, which could result in part from
alkolosis and volume loss.14,15 Both volume reduction


November 1.1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 54

1019

and alkalosis, consequences of thiazide diuretic therapy,
can reduce extracellular but not intracellular K, re-
sulting in hypokalemia. This mechanism is more ap-
parent in the normokalemic group, in which PK de-
creased by an average of 0.6 mEq/liter despite no net
increase in urinary K.

The addition of KC1 in hypokalemic patients tended
to reverse the hypokalemia and restored the small losses
in total body K. The average PK increased significantly
and the estimated gain in total body K approximated
200 mEq. Both changes occurred during the first week
of replacement therapy. After the first week, most of the
administered KC1 was excreted in the urine. However,
despite the relatively large dose of KCl-96 mEq/
day-4 of the 9 patients remained hypokalemic (PK
<3.5 mEq/liter) at the end of 4 weeks of replacement
therapy. Other investigators using smaller doses of 30
to 40 mEq/day of KC1 failed to find an increase in either
plasma or total body K.sll Despite persistent hypoka-
lemia, these patients appeared to excrete most of the
administered KC1 in the urine.

A similar pattern was observed when triamterene,
instead of KCl, was added to patients with hypokalemia.
The average PK increased to nearly normal levels dur-
ing the first week of treatment, and this was associated
with an estimated net gain of about 200 mEq of total
body K. Changes during the subsequent 3 weeks of
treatment were minor or nonexistent. As occurred with
KC1 replacement therapy, 4 patients continued to ex-
hibit hypokalemia despite daily doses of 200 mg of tri-
amterene. Such results suggest that entirely normal
levels of PK may not be attained even after the small net
losses of K are replaced, again suggesting that the hy-
pokalemia resulting from thiazide diuretic therapy is
not associated with a major deficit in body stores.

Unlike the transient increase seen with HCTZ, the
addition of triamterene in 9 hypokalemic patients was
associated with further sustained increase in plasma
creatinine. Mild but significant increase of plasma

creatinine has been reported.rsJ7 Triamterene has been
reported to cause acute reversible renal failure when
combined with indomethacin,ls and has also been
identified as being present in renal calculi.19

References

1.

2.


3.




4.


5.

6.

7.


8.


9.


10.

11.


12.

13.

14.

15.


18.


17.


19.


19.

Harrlngton Jl, lsner JM, Kassirer JP. Our national obsession with potas-
sium. Am J Med 1982;73:155-159.
Glfford RW, Mattox WI, Orvls AL, Sones DA, Rosevar JW. Effects of
thiazide diuretics on plasma volume, body electrolytes and excretion of
aldosterone in hypertension. Circulation 1961;24:1197-1205.
Veterans Administration Cooperative Study Group on Antihypertensive
Agents. Effects of treatment on morbidity in hypertension. III. Influence of
age, diastolic pressure, and prior cardiovascular disease: further analysis
of side effects. Circulation 1972:65:991-1004.
Kasslrer JP, Harrington JT. Diuretics and potassium metabolism: a reas-
sessment of the need, effectiveness and safety of potassium therapy. Kidney
Int 1977;2:505-515.
Morgan DB, Davidson C. Hypokalemia and diuretics: an analysis of publi-
cations. Br Med J 1980;1:905-908.
Edmonds CJ. Jasanl B. Total bodv ootassium in hvoertensive oatients dwino
prolonged diuretic therapy. La&t 1982;2:8-12.'
Dargle MI, Boddy K, Kennedy AC, King PC, Read PR, Ward DM. Total body
potassium in long-term frusemide therapy: is potassium supplement nec-
essary? Br Med J 1984;2:316-319.
Leemhuls MP. VanDamme KJ. Struvrenbera A. Effects of chlorthalktone
on serum and total body potassium in hyperten<ive patients. Acta fv!ed Stand
1976;200:37-45.
Wllklnson PR, Hesp R, lssler H, Raftery EB. Total body and serum potas-
sium during prolonged thiazide therapy for essential hypertension. Lancet
1975;1:759-762.
Wllkinson PR. Potassium changes during diuretic therapy. Cardiovasc Med
1978;1:181-183.
Down PF, Polak A, Rao R, Mead JA. Fate of potassium supplements in six
out patients receiving long-term diuretics for oedematous disease. Lancet
1972;2:721-724.
Frels ED, Wanko A, Wilson IM, Parrish AE. Chlcrothiazide in hypertensive
and normotensive patients. Ann NY Acad Sci 1958;71:450-55.
Maronde RF, Mllgrom M, Dickey JM. Potassium loss with thiazide therapy.
Amer Heart J 1969;78:16-21.
Talso PJ, Carballo AJ. Effects of benzothiadiazines on serum and total body
electrolytes. Ann NY Acad Sci 1980;88:822-840.
Anderson J, Godfrey BE, Hlll DM, Muruotaure AD, Sheldon J. A ccmparison
of the effects of hydrochlorothiazide and of frusemide in the treatment of
hypertensive patients. Quart J Med 1971;15:541-560.
DeCarvalho GR, Emory AC, Frohllch ED. Spironolactone and triamterene
in volumedependent essential hypertension. Clin Pharmacol Ther 1980;
27:53-56.

Walker BR, Hoppe RC, Alexander F. Effects of triamterene on the renal
clearance of caidium magnesium, phosphate and uric acid. Clin Pharrnacol
Ther 1972:13:245-250.
Favre L, hlasson P, Vallotton MB. Reversible acute renal failure from
combined triamterene and indomethacin. Ann Intern Med 1982;96:317-
,,A,.

JL".

Ettlnger 8, Oldroyd NO, Sorgel F. Triamterene nephrolithiasis. JAMA
1980;244:2443-2445.