Reprinted from CIRCULATION
Vol. XX, No. 6 December, 1959
Printed in U.S.A.

Relationship Between Plasma and Extracellular Fluid
Volume Depletion and the Antihypertensive
Effect of Chlorothiazide

By Iuse M. Witson, M.D., anp Epwarp D. Freis, M.D.

With the technical assistance of Mary J. Taylor

The importance of plasma and extracellular fluid volumes in the mechanism of the anti-
hypertensive effect of chlorothiazide is disputed. The present investigation indicates that
the lowering of blood pressure is accompanied by reductions in plasma and extracellular

fluid volumes and in body weight.

salt-free dextran reverses the antihypertensive effect.

Furthermore, re-expansion of plasma volume with

However, since gradual reaccumu-

lation of extracellular fluid oceurs during 1 year of continuous treatment, the late anti-
hypertensive effects of chlorothiazide cannot be explained by the volume-depletion

mechanism.

HE discovery of the antihypertensive

action of chlorothiazide!? raised cer-
tain questions as to its mechanism of action.
These questions included (1) whether the
hypotensive response was caused by salt de-
pletion or by some independent factor; (2)
if produced by salt loss, by what mechanism it
reduced blood pressure; (3) the factors that
lead to increased reactivity to other antihy-
pertensive agents; and (4) the reason for
the moderate antihypertensive action of the
drug when used alone in hypertensive pa-
tients and the absence of such activity im
normotensive subjects.* + Complete answers to
these questions cannot yet be given. This
report concerns an attempt to elucidate some
of the favtors involved.

MarTERIALS AND METHODS

Twenty male hypertensive patients with no
symptoms or signs of congestive heart failure or
edema were hospitalized and placed on a stand-
ard diet containing 1.5 Gm. of salt per day plus
a supplement of 3 Gm. of salt in tablet form.
This supplied a sodium intake of approximately
75 mEq. per day. Most of these patients were

 

From the Veterans Administration Hospital and
the Cardiovascular Researeh Laboratory, Department.
of Medicine, Georgetown University School of Medi-
cine, Washington, 1D. C.

Supported in part by U. 8. Public Health Service
grant H-720 (National Heart Institute) and by a
research grant from Merck, Sharp and Dohme.

1028

under treatment with other antihypertensive agents
(table 1), which were continued during the entire
period of study. A 4-day period for acclimatiza-
tion to the diet was instituted in order to obtair
stabilization of body weight and electrolyte exere-
tions as well as average basal blood pressure
readings. Determinations of plasma and extra-
cellular volumes, serum electrolyte concentrations,
and serum bicarbonate were carried out on the
morning of the fifth day. Each patient was then
given 500 mg. of chlorothiazide twice daily, and
the studies were repeated after a period of 3 to
8 (average 6) days.

In 11 patients carefully screened for reliability
and conscientiousness in regard to taking their
medications, studies were repeated at approxi-
nately 6-month intervals for a period of 12
months. Seven of these patients were hospitalized
and placed on the controlled salt intake for 2
to 3 days prior to repeating the determinations.
while the remaining 4 were hospitalized for 1 day
only because of their inability to take time out
from work. Thus, dietary intake of salt cannot
be regarded as well controlled in the long-term
studies. At approximately 6 months following
initiation of treatment the determinations of
plasma and extracellular volumes in 7 patients
were repeated. Chlorothiazide therapy was then
discontinued for 1 week, after which another series
of determinations were carried out. The patients
remained hospitalized throughout this latter
period.

Plasma volume was determined in the fasting
subject after one half hour of rest in the supine
position with the Evans blue dye method of Gibson
and Evans® adapted and modified for use in the
Coleman spectrophotometer,® * 3 specimens being

Circulation, Volume XX, December 1959
 

 

 

 

 

 

FLUID VOLUME DEPLETION 1029
TABLE 1.—Data Following Short-Term Therapy with Chlorothiazide,
= ~ zg
a ¥ ~ = x
4 = Ce & en 8 ce 5 wey
& ¢ 32 = a z e g E. 5 Es g oe 2 tb Sernm ie pana
é 2 ee gee art ec 36 €é¢ Se! Na creak? CO:
CE. 46 — 8 —20/15 $91 —412 — 9 ~—80 —5 +45 -3
J.Wr, 47 Reserpine 6 — 26/16 +4.4 —501 -—- 9 — 9 - 4 7 +
NLM. 60 Reserpine 6 —17/6 +5.0 —783 -—2.4 --3.3 0 -—3 0 ---6
AW. 46 Reserpine 7 — 30/14 - 5 —1382 — 3B +18 +4 --10 8
J.Br. 43 Reserpine 7 — 30/20 +3.0 —319 + 4 --2.0 -—-] 0 —.
J.Bu. 60 Reserpine 7 —15/10 +2.2 —568 —--1.9 ~-2.7 +4 0 +.1 0
TEAK. 40 Reserpine 8 20/20 +17 — 66 — 5 12 5 0 4 1
JZ. 57 Reserpine 5 ~-10/0 +10 —296 —3.9 2.2 +438 -10 5
ALN. 64 Reserpine 5 —20/5 — 4 — 55 —1-4 — 7 +1 +1 ~ 7
N.B. 50 Reserpine 7 —I15/15 +1.8 —460 —3.6 —4.0 —3 —2 2
J.We 66 Reserpine and 6 —40/18 +4.2 -—-265 —5.8 4.6 0 4 —~.2
Hydralazine
Cs, 36 Reserpine and 6 —-14/20 +3.0 —741 —-4,0 14 —4 +9 —-.2
Hydralazine
TAY. 66 Reserpine and = 6 40/25 +1.4 098 Of - 30 7 0 5
Meeamylamine
he 47 Reserpineand 4 - 20/22 +5 363 = 5 + ..2 ood 0 +.2 ~ 14
Meeamylamine
CP. 52 Reserpine and 8 —50/30 +1.1 177 -2.1 22 =.2 0 3 - 1
Meecamylamine
M.P. 46 Reserpine and 8 ~—25/10 +4.4 -—222 1.1 10 —83 +6 8
Mecamylamine
G.B. 44 Reserpineand 8 —-30/10 44.2 --403 —2.0 S15 oe 3 ~~ 7
Chlorisondamine
J.1) 47 Reserpine and 6 10/10 +1.4 — 39 —3.8 -- 2 1 1 —- 2
Chlorisondamine
MLC, 42 Reserpine and 3 - 10/25 + 4 -—-172 + 4 Ba 7 4 2
Chlorisondamine
S.E. 42 Reserpine, 7 25/15 42.0 590 460 of fe 0 6 4d
Hydralazine,
Pentolinium
Mean 6.4 —24/—15* 2.2 —1.8" —1.3 —1.1 —A82* - -5
1.84 9 4.8 2.4 1S

8. D,

 

*p value less than .001.

used at 10, 15, and 20 minutes for determining dye
dilution, The hematocrit value was determined
from the average of 6 samples drawn during the
various experimental procedures. Eighteen milli-
liters of 5 per cent sodium thiocyanate* were in-
jected usually at the end of the plasma volume
deternunations and samples were drawn at 2 and
3 hours for determination of thiocyanate space
according to the method of Crandall and Ander-
son® adapted to the Coleman spectrophotometer.
Radiosulfate space was determined with S%5-
labeled sulfate by the method of Walser.9 Sodium
und potassium were determined with a flame

“Kindly supplied by Eli Lilty and Company, Tndi-
nnapolis, Ind.

 

photometer, Serum chloride was determined by
the method of Schales and Schales!® and serwn
bicarbonate by a modification of the method of
Van Slyke.!!

All injectates were given in calibrated syringes;
blood samples were drawn with minimal stasis.
Blood pressure was determined by the ausenlta-
tory method before and after each experimental
period. The patients were weighed prior to the
procedures on a beam balance to =
250 Gm.

The cumulative negative balances of sodinm,
potassium, and chloride during the first 3 to 4
days of chlorothiazide treatment were estimated
us follows: beginning 48 hours after institution of
the diet 3 consecutive 24-hour collections of urine

  
 

accurate
1030

CHANGE IN SCN
L SPACE - LITERS

 

T
T
F k + *y t |: 4 .. wd
° rn ° 2

CHANGE IN ; |
BODY WT.-Kg |

"2° Le
oe i
t
|

. o. L4
. +

° 16

Fig. 1. Relationship between change in body weight
and in thiocyanate space following short-term treat-
ment with chlorothiazide.

were analyzed for total sodium, chloride, and
potassium, and the average was taken as the con-
trol level of excretion. The cumulative elevations
above this level during the first 3 to 4 days after
chlorothiazide were then used to estiniate eleetro-
lyte losses.

The effect of restoration of the plasma volume
was determined in 7 patients who exhibited sig-
nificant hypotensive responses to chlorothiazide.
Each patient received 500 ml. of 6 per cent dex-
tran in normal saline; and after several days the
procedure was repeated with substitution of a
similar volume of dextran in 5 per eent glucose
in water. The infusion rate was approximately
25 mil. per minute. Blood pressure and heart rate
were determined repeatedly before and through-
out the procedure and blood for hematocrit deter-
minations was drawn immediately preceding and
following: each infusion.

RESULTS

clcute Effects of Chlorothiazide

Following 3 to 8 (mean 6) days of chloro-
thiazide therapy the plasma volume showed
some reduction in all of 20 nonedematous
hypertensive patients (table 1). The loss of
plasma volume averaged 358 + 223 ml. and
varied in different patients from insignifi-
cant falls of less than 70 ml. to rather marked
losses above 700 ml. The hematocrit reflected
the plasma volume change in an approximate
way by exhibiting slight to moderate eleva-
tion. However, there was no close quantita-
tive relationship between hematocrit and
plasma volume changes.

TABLE 2.—Losses after Chlorothiazide

WILSON, FREIS

 

 

Change after

 

 

 

 

Chloro- Urinary losses, chlorothiazide
thiazide cumulative negative Thiocyanate
treatment balance (mEq.) space Wt.
Patient (days) Na cl Kk (L.) (Kg.)
S. E, 3 224 392 209 ~—4.6 —5.4
J. Bu. 4 301 489 181 —1.9 —27
N. M. 4 290 560 168 2.4 —3.3
J. Z. 3 162 135 154 3.9 —2,2
J. T. 4 216 268 4 — 7 —2,2
E. W. 3 349 510 298 —3.4 —2.7
Average 3.5 257* 392* 156.3* 2.8* 3.0*
Ss. D. 68 163 71 1.4 1.2

 

*p value less than .005.

The thiocyanate space declined by more
than 0.5 L. in all except 5 of the 20 cases
(table 1). For the group as a whole the mean
reduction was 2.1 + 1.75 L. Although the
reductions of both plasma and available fluid
spaces were significant (p < .001), there was
no consistent relationship between the degree
of plasma volume and available fluid changes.

Body weight decreased in 16 of the 20
patients (mean 1.8 + 1.84 Kg.). The change
in body weight was significant at the .001
level. There appeared to be a rough correla-
tion between the extent of change in body
weight and in thiocyanate space (fig. 1).
The blood pressure fell in all patients, the
average change being —24/15 mm. Hg. The
reduction occurred during the first 48 hours
following institution of chlorothiazide. There
was no quantitative correlation between the
plasma volume and arterial pressure changes.

The cumulative losses of body electrolytes
(cumulative negative balance) by the end of
the third to fourth day of chlorothiazide treat-
ment averaged 257 + 68 mEq. of sodium, 392
+ 163 mHq. of chloride, and 156 + 71 mEq. of
potassium in 6 patients studied (table 2).
All of these patients exhibited significant re-
ductions in body weight and thiocyanate
space. The correlation between the extent of
body weight reduction and electrolyte losses
was poor in this small series. However, the
accuracy of 24-hour urine collections cannot
FLUID VOLUME DEPLETION

1031

> TawLE 3.—Hffects of Intravenous Infusion of Six Per Cent Dextran in Saline
and of Salt-free Dextran

 

Following dextran infusion

 

 

 

 

 

After
ehloro- In 5 per cent In normal
. thiazide glucose saline
thravide Control change change change
treatment B.P. mean B.P.* Mean B.P. Hematocrit Mean B.P. Hematocrit
Patient (mos. ) (mm.H¢g) (%) (%) (%) (%) (%)
J.H.F 1.0 150/110 —16.2 -+10.0 — 3.2 + 9.0 — 5.8
J.C.F 5 180/110 —22.9 +18.4 —11.4 +20.0 --12.0
C.S.£ 3.0 150/110 —13.0 +10.5 —12.1 +10.0 —12.8
R.C.£ 1.0 180/108 —30.4 +28.1 — 91 $23.7 —12.6
W.E.FT 1.0 140/100 —11.3 +14.0 —13.6 +13.2 — 9.8
N.BE 1.0 145/95 —12.5 +19.0 -— 8.8 +17.0 — 6.0
JZ.F 2.0 142/96 —10.9 + 8.4 —- 5.0 —_— —
Mean 1.4 155/104 —16.7§ +15.4§ — 9.0£ +f15.4¢ — 9.8t
8.D. 8.8 6.9 3.8 5.8 3.3

 

*«<Mean’?’ blood pressure equals (systolic + diastolic) /2.

fOn no other antihypertensive drugs.

fAlso reeeiving reserpine 0.25 mg. per day. C.S. also taking hydralazine 25 mg. td,

8p value less than .005.

tp value less than .001.

be regarded as completely reliable, even in
hospitalized patients. The chauges in the
serum concentrations of sodium and chloride
were insignificant, but there was a small but
significant decrease in potassium (table 1).
The serum bicarbonate determined in 4 pa-
tients showed essentially no change.

Acute Restoration of Plasma Volume

The effect of re-expansion of plasma
volume on blood pressure was determined in
7 patients who had been under continuous
treatment with chlorothiazide for periods
ranging from 2 weeks to 3 months (mean
1.4 months). These. patients had exhibited
reductions of ‘‘mean’’ (systolic + diastolic) /2
blood pressure averaging 17.0 + 9.0 mm. Hg
(table 3). The administration of 6 per cent
dextran in isotonic saline in 6 of these pa-
tients resulted in. an immediate elevation of
blood pressure averaging 15.0 + 6.0 mm. Hg.
The mean percentage fall in the hematocrit
was 9.8 + 3.3 during the-dextran infusion.

In order to determine whether sodium ion
was important in the reversal of the anti-
hypertensive effect. the infusions were re-
peated after several days with 6 per cent

dextran in do per cent glucose in water. In 7
patients a significant elevation of blood pres-
sure was observed ranging between 8 to 28
mm. Hg (mean + 15.0 + 7.0 mm. Hg,
(p < .005) The percentage fall of hemato-
erit averaged 9.0 + 3.8.

Effects of Long-Term Treatment with Chloro-

thrazide

Whereas the reductions in plasma volume,
thiocyanate space, and body weight were
significant during short-term therapy with
chlorothiazide, this was not the ease follow-
ing long-term treatment. In 11 patients when
these paraineters were measured at the end
of 6 months, only the reduction in_ thio-
cyanate space remained significant (table 4).
At the end of 1 year the change in thiocyanate
space was no longer signifieant at the .05
level. Despite this trend, however, the blood
pressure changes, although not as marked at
the end of 1 year as in the early stages of
therapy, still showed significant reductions.

In 9 patients after 1 to 3 months (average
1.8 months) the extracellular fluid volume
was estimated by means of 2 indicators (table
9), thiocyanate and radiosulfate simultane-
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1032 WILSON, FREIS
TasBLe 4.—Changes Initially after Six and Twelve Months
Plasma volume Blood pressure Thiocyanate space Body weight
(ml.) (mm. Hg) (L.) (Ke.)

Patient 1 wk. 6mo. 12 mo. 1 wk. 6 mo. 12 mo. 1 wk. 6 mo. 12 mo. 1 wk. 6mo. 12 mo.
J.We. —265 — 22 +4165 —40/—18 —30/—15 0/8 —58 —3.9 — 7 —4.6 —l1 —2.8
J.Wr. 501 — 42 —212 —26/—16 —22/—20 — 7/0 — 9 —22 — 2 — 9 —6.0 —6.1
CS. —741 —201 —225 —14/—20 --20/--20 --30/—30 —4.0 —3.2 —1L2 —14 —80 -—4.9
N.M, —783 + 74 4132 —17/—6 —20/—19 35/14 —2.4 --24 +1.2 —3.3 4+ .2 —1.0
T.H. —598 +110 —168 --40/—25 -—30/—20 —20/—10 —24 — A ~~ 7 —3.0 +7.7  +6.7
S.E.  —-596 + 58 —19 -—25/--15 —20/—20 —#0/—2®2 —4.6 —4.7 +49 —5.4 —6.3 +5.7
J.C. —3638 323 —402 —20/—22 -—30/—20 30/—5 — 5 — 5 — 1 + 2 + 9 +49
J.Br. 319 —248 534 30/20 —28/—15 -—10/0 + 1 —2.1 —2.1 +2.0 +44 41.9
J.Bu. —568 —164 -——336 —15/—10 -—-15/—20 -—10/—15 —19 — 8 — 6 —2.7 + 8 —14
CP. —177 — 51 — 51 —50/—30 -—30/—20 -—-18/—20 —2.1 —tL1 —15 —2.2 —4.5 —-7.0
H.K. — 66 — 75 — 56 —20/—20 —20/—20 —20/—15 — 5 —~-— 9 — 7 —12 — 1 4+.
Mean —452* — 80 -—-155f —-97/18* —24/19* —18/13* —2.38t —20f — 6 2.0 —10 — 4
8s. D. 232 128 212 11.6/6.8 5.5/2 10,5/2.4 1.8 1.5 9 2.1 4.6 47

*p value less than .001.

Tp value Jess than .05,

tp value less than .01.

TABLE 5.—Comparative Changes in Extracellular Fluid Volume before and after Chlorothiazide

As Measured by Two Different Indicators

Duration Thiocyanate space S*-labeled sulfate space Thiocyanate ss

treatment L. {L.) ehange change
Patient (mos.)} Control After drug Change Control After drug Change (%) (%)
J.W. 2 25.4 23.2 —2.2 19.9 15.7 —4,2 11.5 —21
NM. 3 19.8 17.4 —2.4 14.1 13.6 —0.5 —I2 —4
A.J. 2 24.0 oL7 23 17.7 17.3 — 4 -—10 a
TH. 1 25.6 25.0 -— .6 19.6 19.6 -——0 —2 0
S.E. 1 24,5 21.1 —3.4 18,38 16.0 —2.3 —14 13
J.Bu. 25 19,0 18.2 —1.8 14.4 13.2 —1.2 —li —12
J.Br. 7 224 18.9 3.5 18.6 15.8 —2.8 —16 —15
GP. L 18.9 16.8 —-2.6 15.8 V4e1 —1.7 —l4 ll
ALK. 1 Vl 16.8 — 3B 13.2 12.4 — 8 —2 mon
Mean 1.83 21.8 19.9 —2,1* 16.8 15.3 —1.5 —-10,3* ~-9.3T
8.D. 11 1.3 5 7.3

 

*p value less than .001.
Tp value less than .05.

ously, as an additional check on the validity
of the change demonstrated with thiocyanate
alone. Considerably more variation was ob-
served with radiosulfate as the indicator.
However, the average change for the group
as a whole was similar with either thiocyanate
or S*-labeled sulfate and the reduction in
estimated extracellular space was still sig-
nificant with either method.

Chiorothiazide was withdrawn for a l-week

period in 8 patients after 3 to 7 (average
5.4) months of continuous treatment in order
to evaluate the changes that might occur fol-
lowing discontinuation of long-term therapy
(table 6). There was a prompt elevation of
plasma volume averaging 241 + 100 ml., a
rise in blood pressure (mean 23/15 + 8/9
mm. Hg), and a gain in body weight aver-
aging 16 + 1.3 Kg. All these changes were
significant at the .01 level or less. The in-
FLUID VOLUME DEPLETION

1033

TaBLe 6.— Changes Following Discontinuation of Chlorothiazide

 

 

Plasma volume
(

Blood pressure

Thiocyanate space Body weight
(L.)

 

 

 

 

Duration ml.) . {mm. Hg) (Ke.)

treatment After Repeat Gain or After Repeat Gain or After Repeat Gain or After Repeat Gain or
Patient (mos,) chlor. control loss chlor. control loss chlor. contro! loss chlor. control loss
J.We. 4 3720 3827 +107 170/100 200/120 +-30/20 23, 26.9 +2.1 85.3 85.3 0
J.Wr. 7 3717 4106 +389 151/103 180/130 +29/27 26.4 25.4 —1.0 103.3 104 + 7
CS. 7 3537 3871 +234 128/87 150/110 +22/93 19.6 20.7 +1.1 73.0 74.3 +1.3
T.H. 7 8943 4152 +209 158/100 160/110 + 7/10 24.7 25.6 + .9 92 94,3 +2.3
S.E. 7 3183 3525 +342 160/110 180/125 +20/15 24.2 25.6 414 715 72.7 +1.2
JC. 5 3340 3663 -+323 170/100 200/110 +30/10 22.8 23.0 + .2 95.4 99.5 +4.1
J.Br, 3 3200 3350 +150 140/100 165/100 +25/0 18.9 224 43.5 80.3 83.0 +2.7
HK. 3 2931 38106 +175 130/90 150/110 +20/20 16.4 17.2 4 8 70 70.7 + .7
Average 5.4 3446 3700 +241* 150/99 173/144 +23*/157 21.5 23.4 +1.1 83.8 85.2 +1.6f
8.D. 100.4 7.8/8.8 1.4 1.3

 

*y value less than .001.
Tp value less than .05.

crease in thiocyanate space averaged 1.1 +
1.4 L. The latter change was of questionable
significance (p < .06).

As compared to pretreatment values it was
of interest that following discontinuation of
the drug plasma volume increased 120 + 18
ml]. and thiocyanate space rose to 0.5 + 0.3 Lh.
above the original, untreated level. These
slight overshoots. were not accompanied by
parallel behavior of the blood pressure, the
latter remaining -4/6 + 7/7 mm. Hg below
the pretreatment level.

The changes in serum electrolyte concen-
trations and in serum bicarbonate were de-
termined after 1 year of chlorothiazide
treatment in 11 patients (table 7). Serum
sodium concentrations were essentially un-
changed. Serum chloride levels decreased
slightly in most patients but increased slightly
in 2, the average change being 4.4 + 6.8 mq.
There was a significant decline in serum po-
tassium concentrations. The mean reduction
was 1.1 + 0.6 mEq. (p .001). The serum bi-
carbonate remained unchanged.

Discussion

In the present observations the reduction
of blood pressure paralleled the saluresis and
did not precede or follow it. The saluretic
effect was essentially complete in 48 hours.
The question arises as to the source of the

sodium depletion, specifically as to whether
it represented reductions in intracellular
stores or came from the extracellular fluid.
The maintenance of normal serum sodium
levels indicated that the drug did not ex-
tract sodium from the extracellular water.
However, the extent of the depletion of ex-
tracellular fluid volume indicated that the
majority if not all of the sodium removed
could be accounted for by an exeretion of
isotonic, extracellular fluid. This is in keep-
ing with the observations on other saluretic
agents such as mereurials, which produce pri-
marily a reduction of extracellular fluid
volume.?*

Sinee the plasma volume is in equilibrium
with the interstitial fluids, it also shared in
the general reduction of extracellular fluid
space. However, the relationship between
change in extracellular and plasma volumes
is only approximate.’? Because of the failure
to find any evidence for significant cellular
losses of sodium, the question arose whether
the antihypertensive effect was related simply
to the decrease in total circulating blood
volume and to a reduction in tissue pressure
secondary to the extracellular fluid loss.
Such an interpretation is supported by the
observation of Crosley and his associates’
that right heart pressures and cardiac output
are reduced by chlorothiazide. Preliminary
1034

TaBLE 7.—Changes Following One Year of
Treatment with Chlorothiazide

 

 

Change in mEq.

 

 

 

 

Patient Na Cl K CO:

J.We. —3 — 4 — 8 —1.

J.Wr. --1 —12 — —1.6
CS. 0 — 9 — 6 —2.0
NLM. —l1 —9 — 6 +4.8
TAIT. +2 —l1 —l1.1 +3.1
S.E, 0 —§ —1L5 +3.2
J.C, —4 — 1 — 4 +2.0
J.Bu. —5 + 9 —1.2 —3.0
J.Br. —1 — 3 — 4 —1.0
CP. —4 + £5 —2.1 —4.0
H.K. 0 — 5 —1.0 +1.0
Mean —1.5 — 4,4* — 95 + .2
S.D. 2.5 6.8 6 2.9

 

*p value less than .05.
Tp value less than .001.

results in this laboratory tend to confirm
these observations. Dustan and her co-work-
ers also found a reduction in cardiae out-
put following chlorothiazide.

If the antihypertensive effects of chloro-
thiazide were due to a change in intracellular
sodium concentration producing a deerease
either in arteriolar tone or a dehydration
of ‘‘waterlogged’’ arterioles,!® the hemody-
namie effeets should be reflected in a redue-
tion of total peripheral resistance rather
than a fall in cardiac filling pressures and
output. The latter findings, however, are
readily explained by the decreases in plasma
and extracellular flnid volumes. It is of in-
terest that the normotensive individual com-
pensates so that the basal blood pressure is
not lowered, whereas this compensation seems
to be deficient in hypertensive patients. 4
Whether such failure of homeostasis is due
to a decreased responsiveness of the barore-
ceptors or to other factors cannot be deter-
mined on the basis of present evidence.

The importance of the plasma volume
change in the mechanism of the antihyper-
tensive effect was indicated by the fact that
restoration of plasma volume either with or
without any replenishment of sodium re-
stored the blood pressure to essentially pre-

WILSON, FREIS

treatment levels. The lack of a quantitative
relationship between the decrease in plasma
volume and the fall of blood pressure can
be explained on the basis of variations in
the activity of compensatory mechanisms in-
cluding the baroreceptor reflexes, and in-
trinsic vascular distensibility in different
individuals. This variability was further
enhanced by the fact that some of the patients
were taking ganglion-blocking drugs and
others were not.

These conclusions on the importance of
oligemia in the mechanism of the autihyper-
tensive action of chlorothiazide confirm our
initial findings? and are similar to those ex-
pressed by Dustan, Tapia, and associates. 17
They propose that the oligemia enhances
vasomotor ‘‘tone’’? which in turn makes the
patient more responsive to ganglion-blocking
agents. Hlowever, the observation that the
hypertensive patient often shows some redue-
tion of blood pressure with chlorothiazide alone
whereas the normotensive subject does not® 4
suggests that the compensatory mechanisms
for plasma volume depletion often are in-
adequate in hypertension. Also, the decreased
responsiveness to norepinephrine following
chlorothiazide™ !® in normotensive subjects
is not readily explained on the basis of in-
ereased vasomotor tone alone, and suggests
that reactivity is dependent to some extent
on the degree of filling of the vascular system.

The failure to observe a significant redue-
tion of plasma volume after 6 to 12 months
of treatment reflects either tolerance to the
saluretic effects of the drug or the establish-
ment of compensatory mechanisms for res-
toration of homeostasis. Compensatory mech-
anisms become active after the first 48 hours
of treatment when the output of sodium
comes back into balance with the intake.
However, in the first month of treatment the
depletion achieved during the initial salure-
sis is maintained and only gradually, there-
after, is the plasma volume deficit made up.

Other investigators have postulated that
chlorothiazide may have antihypertensive
effects additional to its saluretic action.* This
FLUID VOLUME DEPLETION

is based on the observation that the blood
pressure remains below pretreatment levels
after the body weight has been restored. It
should be pointed out, however, that disap-
pearance of acute drug effects without return
to pretreatment levels of blood pressure is
not unique for chlorothiazide. For example,
the acute hemodynamic effects of hydralazine,
characterized by tachycardia and palpitation
reflecting an increase in cardiac output, gen-
erally disappear with long-term treatment.
Similarly, the manifestations of ganglionic
blockade such as impaired visual accommoda-
tion, dry mouth, and postural hypotension
often diminish with long-continued therapy
even though basal arterial pressure remains
below the pretreatment level. The observations
of Perry and Schroeder? indicate that vigor-
ous treatment to obtain a continuous and pro-
longed reduction of arterial pressure, often
modifies the severity of the hypertension, so
that less intensive or no further treatment
is required. Thus, after long-term therapy
many factors, such as tolerance, compensa-
tory reactions, and modification of the basal
level of blood pressure, come into play to
obscure the initial relationship between drug
action and antihypertensive effect. It does
not seem possible to draw valid conclusions
concerning the antihypertensive activity of
a drug at this late stage, especially if the
effects of drug withdrawal are not deter-
mined.
SUMMARY AND CONCLUSIONS

Plasma and extracellular finid volumes,
serum electrolyte concentrations, arterial
pressure, body weight, and electrolyte excre-
tions were determined in hypertensive pa-
tients treated with chlorothiazide.

Plasma and extracellular fluid volumes
decreased promptly during the early phases
of treatment. This was accompanied by a re-
duction of a terial pressure and body weight.
Sodium losses could be accounted for on the
basis of extracellular fluid volume depletion.
Restoration of plasma volume either with or
withont partial replenishment of sodium
reversed the antihypertensive effect of chloro-

1035

thiazide. Withdrawal of chlorothiazide after
3 to 7 mouths of treatment was followed by
an elevation of plasma volume usually to
levels slightly above the control. Blood pres-
sure rose to levels slightly below the control.

Significant reductions of plasma volume
and body weight were not found after 6
mouths and of extracellular fluid volume after
12 months of therapy even though the blood
pressure remained redneed. However, the
latter may not be a valid eriterion of drug
activity following long-term modification of
the blood pressure level. The only significant
change in serum electrolyte concentration
was a reduction in serum potassium. Serum
hicarbonate levels were not altered.

On the basis of this and other evidence
discussed it is suggested that the decrease in
plasma volume is an important factor pro-
ducing the initial antihypertensive effect. Re-
duction in tissue pressure secondary to extra-
cellular fluid volume depletion also could
contribute to this response.

SUMMARIO IN INTERLINGUA

Le volumines del plasma e del liquido
extracellular, le concentrationes del electro-
lytos in le sero, le tension arterial, le peso
eorporee, e le excretion de electrolytos esseva
determinate in patientes hypertensive sub-
traectamento con chlorothiazido.

Tie volumines de plasma e de liquido extra-
cellular descendeva promptemente durante le
phases initial del tractamento. Isto esseva
accompaniate per un reduction del tension
arterial e del peso corporee. Le perditas de
natrium esseva explicabile super le base del
depletion del volumine de hquido extracellu-
lar. Le restauration del plasma, tanto con
como etiam sin le restitution partial de nat-
rium reverteva le effecto antihypertensive de
chlorothiazide. Le privation de chlorothiazido
post 3 a 7 menses de tractamento esseva se-
quite per un elevation del volumine del plas-
ma, usualmente usque a nivellos levemente
supra le nivellos de controlo. Le tension del
sanguine montava usque a nivellos levemente
infra le nivellos de controlo,
1036

Significative reductiones del volumine del
plasma e del peso corporee non esseva cvon-
statate post 6 menses de therapia, e simile-
mente nulle significative reductiones del vol-
umine del liquido extracellular esseva con-
statate post 12 menses de therapia ben que
le tension de sanguine remaneva reducite.
Tamen, iste ultime facto es possibilemente
invalide come criterio del activitate del droga
post un prolongate modification del nivello
del tension de sanguine. Le sol significative
alteration in le concentration de eleetrolytos
in le sero esseva un reduction del contento de
kalium. I-e nivellos de bicarbonato in le sero
non esseva alterate.

Super le base de iste e altere observationes
diseutite in le presente reporto, le these es
formulate que le reduction del volumine del
plasma es un importante factor in le produc-
tion del effecto antihypertensive initial. Le
reduction del pression del histos que seque
le depletion del volumine de liquido extracel-
lular es etiam possibilemente un contributor
a ile respousa.

REFERENCES

1. Freis, EK. D., Ann Witson, 7. M.: Potentiat-
ing effect of chlorothiazide (Diuril) in con-
bination with antihypertensive agents. Med.
Ann, District of Columbia 26: 468, 1957.

2. Hotuanper, W., Anp Winxins, R. W-.:
Chlorothiazide. A new type of drug for
treatinent of hypertension. Boston U. Med.
Quart. 8: 69, 1957.

3. Frets, E, D., Wanko, A., Witson, I. M,,
AND Panrrisu, A. E.: Chlorothiazide in hy-
pertensive and normotensive patients. Ann.
New York Acad. Se. 71: 450, 1958.

4+. Winkins, R. W., Ho_uanper, W., ann CHo-

RBANIAN, A. V.: Chlorothiazide in hyper-
tension. Studies on its mode of action, Ann.
New York Aead. Se. 71: 465, 1958.

). Girson, J. G, Irn, axp Evans, W. A, Jr:
Clinical studies on the blood volume. J.
Clin. Invest. 16: 317, 1937.

6. Kuerson, C. P., anp Enert, R. V.: Study of
shock in battle casualties: Measurement of
blood volume changes in responses to thera-
py. Ann. Surg. 122: 745, 1945.

Freis, E. D., anp Smituwick, R. H.: Effects
of lumbodorsal splanchnicectomy on the
blood volume and “thiocyanate space” of
patients with essential hypertension. Am. J.
M. Se. 214: 363, 1947.

10.

ll.

It.

15.

16.

19.

20.

WILSON, FREIS

Cranpaun, L. A. Jk, AND ANDERSON, M. X.:
Kstimation of state of hydration by amount
of water available for solution of thioeya-
nate. Am. J. Digest. Dis. & Nutrition 1:
126, 1934.

Wauser, M., SELLIN, D. W., AND GROLLMAN,
A.: An evaluation of radiosulfate for the
determination of the volume of extracellular
fluid in man and dogs. J. Clin. Invest. 32:
299, 1953.

Scuaves, O., AND Scitanes, 8. S.: Simple aud
aceurate method of determination of ehlor-
ide in biological fluids. J. Biol. Chem. 140:
879, 1941.

Van StyKe, D. D.: Studies of acidosis; de-
termination of bicarbonate coneentration of
blood and plasma. J. Biol. Chem. 52: 495,
1922.

Learp, 8. E., anp Freis, EK. D.: Changes in
volume of the plasnia, interstitial and in-
tracellular fluid spaees during hydration
and dehydration in normal and edematous
subjects. Am. J. Med. 7: 697, 1949.

Laverr, M. Fl: Physiological considerations
of edema. J, Mount Sinai Hosp. 19: 571,
1952.

Crosury, A. P., Jr. Cosreuio, C., FREE-
MAN, J. D., Warire, D. H., Jr, anp Rowe,
G. C.: The aeute effects of carbonic an-
hydrase inhibitors on systemie hemody-
names. J. Clin. Invest. 37: 887, 1958.

Dustan, H. P., Cumminas, G. R., Corcoran,
A. G., ann Pace, J. H.: A mechanism of
chlorothiazide enhanced effectiveness of
antihypertensive ganglioplegie dings. Cir-
culation 19: 360, 1959.

Topraxn, L., AND Repiear, P. D.: Effects of
hypertension on arterial wall electrolytes
during desoxyeorticosterone administration.
Cireulation Researeh 4: 671, 1956.

Tapia, F. A., Dusran, H. P., ScrnecKLoru,
R. A., Corcoran, A. C., axp Pace, I. H.:
Enhaneed effectiveness of ganglionic block-
ing agents. Laneet 2: 831, 1957.

Merritn, J. P., Guinanp-Baupo, A., Anp
Gtorpano, C.: Effeets of chlorothiazide on
hovepinephrine response im human hyper-
tension. Clinical Research 6: 230, 1958.

Wanko, A., AND Frets, E. D.: Arterial vascu-
lar responsiveness following chlorothiazide
or mereurial diuresis in normotensive sub-
jects. Cireulation 10: 793, 1958.

Perry, M. H., Jr., anp Scuroeper, H. A.:
Studies on control of hypertension. VI.
Some evidence of reversal of process during
hexamethonium and hydralazine therapy.
Cireulation 8: 528, 1956.