Download Thirst: Understanding Its Mechanisms and Clinical Significance and more Slides Medicine in PDF only on Docsity!
THIRST AS A SYMPTOM
By JOSEPH H. HOLMES, M.D.
DENVER
Most clinicians pay little^ attention^ to^ thirst^ as a^ symptom
unless it becomes so pronounced that it^ deserves^ the^ term^ "poly-
dipsia." Perhaps this is largely due to the fact^ that^ we have no
clear understanding of the physiological mechanisms involved, nor
is there any good objective measurement of the^ sensation^ of
thirst. Subjectively many times the term has become confused
with a longing for a specific liquid. Perhaps as Dr. Adolph has
frequently said to me,^ several mechanisms^ may be^ responsible^ for
the sensation of thirst and attempts to simplify the problem^ to
a single mechanism will not be successful. Perhaps the man who
wrote me a penny post card is right. He said: "I could save you
lots of time and useless research. It seems silly to spend money
to learn of information so simple it^ cannot^ be^ comprehended."
Current thinking on the causation of thirst is^ confined^ to
two general concepts, the first that the sensation of^ thirst^ is^ localized
in the mouth and subjectively is associated with dryness of that
region, which in turn is a function of the salivary flow. Cannon strongly supported^ this^ concept^ and^ when^ stranded in^ a^ London
hotel after World War^ I,^ deprived himself of^ water^ and^ demon-
strated a^ decrease^ insalivary^ secretion^ in^ association^ with^ thirst.'
Adolph's observation made^ in^ the Arizona^ Desert^ showed^ a^ direct
correlation between water deficit and^ salivary flow.2^ The^ second
concept is based on the homeostatic pattern that thirst^ represents
an expression of the need of the body cells^ for^ fluid.^ Examination of various fluid patterns has shown that the^ one^ most^ consistently
associated with^ thirst^ is^ that^ of^ intracellular^ dehydration.^ It^ might
well be that the^ two^ concepts could^ be^ combined^ as a^ single^ concept if it were shown that^ a^ reduction^ in^ salivary^ flow could be^ produced
by intracellular^ dehydration.
From the Department of Medicine, University of^ Colorado^ School^ of
Medicine, Denver.
THIRST AS A SYMPTOM
The following clinical situations have been selected^ for dis-
cussion and represent situations in which the symptom of thirst
can be of use diagnostically:
1. Dehydration
- Endocrine disorders such as diabetes (^) insipidus,
hyperparathyroidism, and hyperthyroidism
3. Edematous phase of cardiac failure
4. Hemorrhage and shock
5. Disturbances of salivary secretion
6. Psychiatric or emotional disturbances
Thirst is a well recognized symptom of simple dehydration.
As previously mentioned, the salivary flow and dryness of the
mouth appear to correlate well with the extent of dehydration.
However, there are certain pitfalls^ in^ assuming^ that^ when thirst
is absent, the patient has no need for fluids.^ For^ example,^ sodium
depletion alone as after intraperitoneal glucose solutions4 or sodium depletion together with water in concentrations equal to that of
the body fluids as in sweating or drainage from a Miller Abbott
tube may elicit no thirst.5'6 Yet, the reduction in plasma and
extracellular volumes may be sufficient to cause a drop in blood
pressure and a^ rapid^ pulse.^ Frequently, thirst^ is^ assuaged^ before
the body fluids are restored to normal. For^ example,^ with^ de- hydration on the desert, man tends to drink^ back only^ 75%o of his water deficit.7 In contrast, injection of hypertonic solutions of salt and glucose will^ induce thirst^ even^ though^ there^ has^ been no change in^ the^ total^ amount^ of body^ fluids,^ but^ merely^ a^ shift of water from the intracellular^ to^ the extracellular^ compartments. That such injections produce a^ drop in^ salivary flow is^ evident from Figure 1.9 In^ treating dehydration^ it^ should^ be^ remembered that in patients with excessive sodium^ loss^ thirst^ may be minimal and yet the need for fluids be^ great. A^ good clinical^ example of this is the case of acute food poisoning with^ severe^ diarrhea^ where the patient frequently does not complain of^ severe^ thirst till the second or third day. In such patients, fluid^ therapy administered
THIRST AS A SYMPTOM
TABLE 1
Shows the fluid changes observed in a case of diabetes insipidus (E.D.)
after administration of pitressin (20 pressor units Q. 6 h.) and after forcing of fluids with no pitressin (Holmes and Gregersen'0).
Forced Untreated Pitressin Fluids Fluid intake (L) -16.3 3.1 17. Urine (L) ------------------------- 15.6 2.6 15. Weight (kg.) -60.8^ 62.0 62. Saliva (^) (cc) -0.6 4.0 (^) 4. Plasma (cc) -2440 3070 2800 NaScn (L) -- ------------------ 13.8 14.8 16. Serum Na (m eq./L) -155 134 133 Urine NaCl (gm) (24 hrs.) -6.3 6.5 6.
I would like to digress in order to point out a useful diag- nostic aid to differentiate the true polyuric from the excessive water drinker on a psychiatric basis. The increase in (^) salivary flow after pitressin has^ been^ found in^ all^ fourteen cases of diabetes insipidus which we have studied. Typical examples are shown in
TABLE 2
Shows the increase in salivary flow produced in patients with diabetes
insipidus by administration of pitressin or forcing of fluids.
Salivary flow (cc/5 min.)
Force
Patient Untreated Pitressin Fluids
ED -0.6^ 4.0^ 4.
TP -0.3 2.0 2.
SD -0.2 2.6 1.
L -0.2 2.8 1.
BA -0.2 1.6 -
Table 2. In contrast in the normal person, pitressin produces a
decrease in salivary flow. (Table 3). Further, a dose as small as .07 pressor units will increase the (^) salivary flow in (^) diabetes insipidus, although it has no effect on the salivary flow of the normal (^) person. In 2 cases (^) representing excessive water drinkers on a psychiatric basis, the salivary flow after (^) pitressin was either
JOSEPH H. HOLMES
unchanged or decreased. In the^ normally hydrated^ person,^ ex-
cessive water intakes (7-9 L) for a 12 hour period have^ no^ effect
TABLE 3
Shows the drop of salivary flow in two normal men^ following the^ sub- cutaneous injection of^3 and^8 -^ pressor units^ of^ pitressin.
Salivary rate (cc/5 min)
Minutes after Pitressin Subject Control 10 20 30
B 2.0 1.5 ---^ 0.
B 3.3 1.9^ 1. S 2.1 2.1^ 1.
S 2.4 1.7^ 1.
on the salivary flow. The salivary test is done by^ blocking the
nose with a nose clip to insure mouth breathing and^ then^ having
the (^) patient collect all saliva during a 5 minute period in a graduated centrifuge tube.^ This^ test^ has also^ been^ of^ use^ in^ following^ the duration of pitressin effect^ in cases^ of^ diabetes^ insipidus^ where urine collections are not^ convenient.
Fluid measurements similar to^ those for diabetes^ insipidus have not been made in cases of hyperparathyroidism or^ hyper-
thyroidism. However,^ preliminary^ observations and work^ with
animals given thyroid extract, would^ suggest^ a^ similar^ mechanism to that observed in^ diabetes^ insipidus; namely,^ a^ polydipsia^ sec- ondary to a pronounced polyuria, and^ a^ thirst^ no^ different from that of simple dehydration.1'
The cardiac patient who^ is^ accumulating edema^ fluid^ frequently complains of thirst. In^ fact, several^ patients, because of^ their
thirst have^ apprised us^ of^ the^ accumulating^ edema before it^ became
evident clinically. Figure^2 shows^ that^ the^ salivary^ flow^ may^ be reduced with^ cardiac^ decompensation^ and increase^ as^ the^ patient
becomes compensated. It^ is^ interesting^ to^ speculate^ on^ the^ possible
role of^ thirst^ in^ edema formation.^ The^ cause^ of thirst in^ cardiac-
JOSEPH H. HOLMES
secretion. In the untreated disease, salivary flow is reduced and
returns within the normal range with prostigmine therapy. In 2
cases we were unable to demonstrate any concomitant changes
in plasma and extracellular (NaSCN) volumes.
Thirst, or at least^ dryness^ of the mouth, has long^ been^ asso-
ciated with emotional disturbances, such^ as fear^ and^ anxiety.
A college student before an^ important examination^ is a good ex-
ample. This knowledge was used in the Asiatic countries as a
means of picking out the guilty person through his inability to
swallow a bowl of dry rice.
That the gastrointestinal tract may play an important role
in thirst is shown by^ animal experiment.^ Fluids^ given by^ stomach
40 minutes prior to inducing thirst by injection of^ hypertonic^ salt
solution will inhibit drinking. The same fluids given intravenously do not alter the drinking response.15 This inhibition is not abolished
by section of the vagus or sympathetic nerves to the stomach. A
distended balloon in the^ stomach will^ also^ inhibit drinking.^ This effect can be^ abolished by local^ application of^ cocaine^ to^ the^ stomach mucosa.16 Curiously enough, in^ man, water^ given^40 minutes^ prior to injection of hypertonic salt solution, prevents the^ decrease^ in salivary flow usually observed after injection of salt.9 This^ re- lationship between salivary flow and the stomach may be of further medical^ interest, since^ recent^ reports show^ an^ increased salivary flow in^ cases of peptic ulcer.
Our habit of questioning all^ patients about thirst has often been of value in diagnosis or in^ planning the^ course^ of^ therapy. Incidentally, it always leads to a^ good story, frequently pointing to another illness which was not revealed by routine^ questioning. I have discovered that everyone has had some particular experience with thirst, which he^ loves^ to^ recount.
SUM1MARY
This has been a very brief presentation of^ a^ number^ of clinical situations in which thirst may be important to^ the^ diagnosis or
THIRST AS A SYMPTOM 101
therapy. These are dehydration, cardiac edema, polyurias of en-
docrine origin, hemorrhage, diseases altering salivary secretion,
and emotional disturbances. The rate of salivary secretion has
been presented as one method of differentiating between^ true^ thirst
and excessive water drinking. It is important to keep in mind
that fluid imbalance may well represent a defect of fluid intake
rather than a defect of fluid elimination.
REFERENCES
1. CANNON, W. B.: The Physiological Basis of Thirst, Proc. Roy Soc.
- ADOLPIH, E. F., (^) ani( Associates: Physiology of Mlani in the Desert, Inter- science Press, New York, 1947.
3. (GREGERSEN, M. I.: Macleod's Physiology in Mloderni Medicinie, Chap. 79,
The C. V. Mosby Co., St. Louis, 1941.
4. GILMIAN, A.: The Relation Between Blood Osmotic Pressure, Fluid
Distribution anid Voluntary Water Intake, Am. J. Physiol., 120: 323,
25. DILL, D. B.: Life, Heat and Altitude, Chal). IN' Cambridge, Harvard,
6. 'NADAL, J. W., PEDERSEN, S., and MADDO(K, WV. G.: A Comparisoni
Between Dehydration from Salt Loss and from Water Deprivation, J. Clin. Invest., 20: 691, 1941.
.7. PITTS, G. C., JOHNSON, R. E., and CONSOLAZIo. F. C.: Work in Heat as
Affected by Intake of 'Vater, Salt and Glucose. Amz. J. Physiol. 142:
8 HoLIN,ES, J. H., anid GREGFRSENX, M. I.: Observations oni Drinking Induced
by Hypertonic Solutions. Am. J. Physiol., 162: 326, 1950.
9. HOLAIES, J. H., and GREGERSEN, M. I.: Relaticon of the Salivary Flow to
the Thirst Produced in Man by Intravenious Injectioni of Hypertonic
Salt Solution. Am. J. Physiol., 151: 252, 1947.
10. HOLMES, J. H., and GREGERSEN, M. I.: Origin of Thirst in Diabetes
Insipidus, Am. J. Med., 4: 503, 1948.
11. HOLMES, J. H., and BEHAN, M.: Thirst with Thyroid Feeding, Fed.
Proc. 10: 66, 1951.
THIRST AS A SYMPTOMI
interestinlg shifts. \Ve could make no observationi that they depenlded on (legrees of (^) dehydrationi, pulse rate, temperature. or (^) other points in (^) the experi- ment.
DR<. MAURICE C. PJNcoFFS (^) (Baltimnore) I (^) amil very mluclh intereste(d in
wvhat Dr. Bean recounted and Dr. Holmes' very interestinig paper. I am
snre that this disproportional factor of thirst is not a guide of need at a
certaini point. It is very important, especially, in tropical climate. One of
the lost opportunities for (^) study in the Southwest Pacific area was the (^) rela- tioIn between this (^) factor and the very high occurrence of urinary calculi.
That has beeni notedl in many other parts of the tropical area to be at an
extraordinarily h.igh level. It was noted by many that those who came (^) up
from the more temperate climate of Australia had an inordilnate thirst for
a (^) week, and often (^) they mentioned they drank no more wrater than they did
at home, in spite of the fact that sweating was a prominient factor in everv-
onle's life all (lav lon-g.
IPerhaps this very higlh inlcidenice of urinary calcuili trouble bore soille
relationi to (^) that effect in adjustmenits.
One other poinlt that perhaps Dr. Holmes might not agree on is this:
In or(linary clinical practice, one of the (^) most striking manifestations of thirst occurs in a patient with an acute gastric dilatation. I have seen a number
of those in xvhom tlhirst was so urgent that they Nvere wanting; to have water
every few minuttes unitil one had emptied the stomiiach of two or three liters
of retainied fluidl Nvhe tlhirst subsided.
DR. JAME.S^ A. GREENE (Houston, Texas): Did yotu analyze this sputuml
for solids and electrolytes? The reason I wondered about that is that I have
seenl two patienits wvho lhad very thick saliva for several mlolnths; the volume
was alnmost normal, and( yet the saliva was very thick ancd stringy. Thirst
svas iot presenit in those individuals.
DR. 1OI.MES (Closinig): I am goinig to answer Dr. Greene's questionl
first. \Ve have lots of electrolyte data on (^) saliva, but still lack an under-
standinig of the mechaniisms involved. In general sodium and( chloride con-
centrationis increase wvith increases in rate of salivary flow while potassiumll
and plhosplhate concenitrations remain relatively constant. However, in the
dehydrated surgical patient .we have observed increase(l salivary concentra-
tionls of sodiumli and clhloride in the presence of a decreased rate of flow. I
(lo not have the specific answer to Dr. Greene's question. 'Many patienlts
with thirst state that they have a dlefinite taste sensationi in addition to the
mouth beinig dry anid( the saliva sticky. It is difficult to get any good
descriptioni of this taste. They call it metallic in character, but they canllot
(lescribe it anyimore (lefiniitely.
JOSEPH H. HOLMES
WVith regard to Dr. Pinicoffs' interestitng comments oin thirst ili Cases
of acute dilatation of the stomach. WVe have had no (^) experienice witl thirst
in this particular situation. In the rabbit, an animal which cannot vomit,
wlheni one ties off the pylorus, it will continue to drink under certain cir-
cumiistanices. I)r. Bean, I have nio aniswer (^) to your question except in an inidirect (^) way. \Vith the dogs in (^) particular and also with man, I think it is true that (^) drink- ing l)ecomes (^) conditionied to the individual. In the dog, we have nioticed (^) that
each aniimal has a differenit drinikinig pattern. Some drink far in excess of
whlat they nieed to dilute (^) body fluids back to normal. Others drinlk (^) less
thani is needed, and depend on the kidney to readjtust the fluid patterns to
normiial by excreting a hypertoniic urinie, yet each animal conisistently follows
his owni drinking pattern. If after drinking is completed the ilngested fluid
is (^) remove(d by opening a gastric fistula anid the (^) animals are allowed to drink
againi, the total initakes after this procedure are approximately the same.
Perhaps initial drinlking has two components, namely the fluidc needed (^) to restore hydration of the cells and the fluid needed to satisfy the local sensa- tion of (^) dryniess in the imouth and pharynx. Some (^) animals may require more fluid than others to satisfy the local sensation. (^) However, onice the local
sensationi is satisfied by the act of driniking then the total intakes when the
aniimal is allowed to drink againi are approximately equal from animal to
aniimal. I forgot to (^) menition one thinig which I know will interest Dr. (^) Jones.
\VXhen water is giveni by stomach, 40 minutes prior to the salt injection, it
inihibits the drinikinig responise. However if the samiie amount of fluid is given
initravenously 40 minutes prior to the salt injection there is no inhibitionl of
the drinking even (^) though there is greater dilution of the body (^) fluids. This
emiphlasizes again^ the^ importance of the stomach in the satisfactionl of thirst.
It is of interest to note that dogs tend to drink all the water right (^) away, whereas man (^) tends to drink it in stages over a period (^) of several hiours.
