THE CARDIOVASCULAR EFFECTS OFACUTELY INDUCED HYPOTHERMIA

O. Prec, … , S. Rodbard, L. N. Katz

J Clin Invest. 1949;28(2):293-300. https://doi.org/10.1172/JCI102071.

Research Article

Find the latest version:

http://jci.me/102071/pdf

Pdf

THE CARDIOVASCULAREFFECTSOF ACUTELYINDUCEDHYPOTHERMIA1, 2

By 0. PREC,3 R. ROSENMAN,K BRAUN,4 S. RODBARD,AND L. N. KATZ

(From the Cardiovascular Department,5 Medical Research Institute, Michael ReeseHospital, Chicago)

(Received for publication July 19, 1948)

Under the conditions of relatively constantbody temperature which exist in warm bloodedanimals including man, the physiological variationsoccurring in circulatory functions such as bloodpressure, cardiac output, and circulatory rate aresmall. This relative constancy has led to the de-velopment of the concept of circulatory homeo-stasis. In recent years there has been a growingappreciation that the warm blooded animal maysurvive at greatly reduced body temperature dur-ing which marked circulatory changes may occur.The interdependence of the several cardiovascularfunctions has indicated the need for a correlationof the circulatory adjustments which occur dur-ing such progressive hypothermia. Only in thismanner would it be possible to interpret the sig-nificance of the changes and their respective rolesin the integration of the animal under the stressimposed by the hypothermia. Such studies mayalso have value in the development of rationaltherapy for individuals who have undergone pro-longed exposure to cold.

METHOD

Fifteen heparinized dogs were anesthetized with intra-venously administered sodium pentobarbital. Three ani-mals were used as controls and 12 were cooled. Nine ofthese latter were maintained in a lighter and three in adeeper plane of anesthesia. The results were comparedwith those of six other dogs, in a parallel study (1),subjected to hyperthermia.

Respiratory rate and oxygen consumption were re-corded with a basal metabolism apparatus attached to atracheal cannula. Kymographic tracings of arterialblood pressure were obtained from the femoral artery

1 Read before the III Interamerican CardiologicalCongress, June 13-17, 1948, Chicago, Ill.

2 Aided in part by a grant from the Department of theArmy to the Michael Reese Hospital (Dr. L. N. Katz,responsible investigator).

8 Rockefeller Fellow from Prague, Czechoslovakia.4Hadassah Fellow from Jerusalem, Israel.5This department is supported in part by the Michael

Reese Research Foundation.

with a mercury manometer. Pulse rates were obtainedfrom electrocardiograms taken simultaneously with theother measurements.

Cardiac output was calculated from the oxygen con-sumption and the arteriovenous oxygen difference accord-ing to Fick's formula (2). Two radio-opaque cardiaccatheters were introduced through the external jugularveins and, with the aid of fluoroscopy, the distal endswere placed in the right auricle and in the pulmonaryartery, respectively. The proximal ends of the catheterswere connected with saline manometers for the recordingof mean pressures. The location of the catheters andtheir zero pressures were verified at post-mortem ex-amination. To ensure maximal mixing of the bloodreturning to the heart, samples of venous blood weretaken from the pulmonary artery. Arterial blood wasobtained from the cannulated femoral artery. In thecourse of each experiment, 100 to 150 cc. of blood werewithdrawn and replaced with isotonic saline solution.The Van Slyke-Neill technic (3) was used in theanalysis of blood Qa and CO2. Hematocrits were deter-mined by the Wintrobe method (4), arterial blood beingemployed.

Control measurements were made when the blood pres-sure and respiration of the animal became stabilized.The animals were then covered with chipped-ice packs.Body temperatures were recorded from thermometersplaced deeply in the rectosigmoid area and protected fromthe ice packs. The mean cooling period was about twohours during which the body temperatures fell to 290 C.The animal was then dried, covered with thin sheets, andexposed to radiant heat. However, the body tempera-tures continued to fall to 270 C, before beginning to rise.The mean rewarming period lasted five to six hours, atthe end of which body temperatures usually had returnedto 37° C. Ten or 11 determinations were obtained ineach experiment, measurements being taken after each 2degree change of body temperature.

Evaluation of the methodMeasurements of right auricular and pulmonary arte-

rial pressures represent only the approximate mean values.The changes in pressure were considered to be of greatersignificance. During periods of extreme bradypnea re-sulting from severe respiratory center depression, greatvariations in concentration of blood gases occur duringeach respiratory cycle, and this produces errors in thedirect Fick method. This inaccuracy is further in-creased by the fact that the usual means of collectingblood samples makes it impossible to obtain exactly

293

0. PREC, R. ROSENMAN,K. BRAUN, S. RODBARD, AND L. N. KATZ

TEr ATR - 'IC

37 -36F

OXYGENCONYSVMPTIQNOG/6 SOD WEIGHT

RESPIRTY RATE40._30 _ .. _ _ _

20 ..-_

PULSE RATE210r

ISot -.- -_GARDPSOUTPUTCC. BODYWT.

0 I 2 3 4TIME -HOURS

FIG. 1. COMPOSITEGRAPHOF DATA OBTAINED FROMTHREE CONTROL DOGS, ANESTHETIZED BUT NOT SUB-JECTED TO COOLING, DURING FOUR-HOURPERIOD

Values for oxygen consumption and cardiac output aregiven in cc./kg./min. Respiratory and heart rates arein frequency per minute. The results in each individualdog are indicated by a different type of line.

simultaneously withdrawn arterial and mixed venousblood. These samples were drawn within 10 seconds ofeach other. This error is shown by the progressiveaccumulation of carbon dioxide in both arterial andvenous bloods during apneic periods. Thus arterial bloodtaken at the end of apneic periods may contain morecarbon dioxide than is found in venous blood withdrawnat the beginning of the respiratory cycle. For this rea-son, data obtained during marked bradypnea are not usedin the conclusions of this study, although these areincluded in the tables.

Measurements similar to those described above wereobtained over four-hour periods in a control group ofthree dogs, anesthetized but not subjected to coolingprocedures (Figure 1).

The site from which true mixed venous blood can beobtained has been a controversial point, and has beendiscussed by many investigators in connection with theapplication of the direct Fick method in man. In thelatter instance it has been demonstrated that there areonly slight differences between blood taken from theright ventricle or pulmonary artery and that from theright auricle provided that the catheter tip is placed near

the tricuspid valve orifice. A different situation existsin dogs, our experience having shown that the oxygencontent of blood taken from the right ventricle or pul-monary artery may differ markedly from that of theright auricle. We therefore adopted catheterization ofthe pulmonary artery as a more exact technic. Evengreater errors may be introduced when venous blood istaken from the inferior vena cava due to the presenceof more highly oxygenated renal venous blood which ispoorly mixed with that from other viscera.

DISCUSSION OF RESULTS

Oxygen consumptionThe initial values for oxygen consumption lay

within a relatively small range, 4.0 to 6.6 cc./kg./min. In the control series no significant changesin oxygen consumption occurred during the four-hour period of observation (Figure 1). In eightcooled dogs there was a progressive decrease inoxygen consumption as body temperature fell.In two less deeply anesthetized animals the oxygenconsumption increased from 4.7 to 10.8 and from4.0 to 4.7 cc./kg./min. during shivering, and thensuddenly decreased. In two other dogs the oxy-gen consumption did not vary significantly untilbody temperature fell to 31'C. Very little varia-tion was noted in the pattern of change afterthe temperature fell below 310C. The lowest oxy-gen consumption was 1.4 cc./kg./min. Duringrewarming periods there was a progressive in-crease in oxygen consumption in all instances.Data on three deeply anesthetized animals aregiven in Figure 2.

The direct relationship which was found betweenoxygen consumption and body temperature issimilar to the results of other investigators in dogs(5). Variations in oxygen consumption up to200%o which we observed in the four less deeplyanesthetized dogs mentioned above were associatedwith marked shivering and increase in pulse rate.There is general agreement that shivering causesan increase in oxygen consumption (6). Visibleshivering occurred at temperatures as low as 290C in our experiments and did not prevent furtherdrop in body temperature. Figure 3 represents anexperiment in which marked shivering did notprevent further fall in body temperature. It shouldbe noted that the rate of temperature change dur-ing both the cooling and rewarming periods wasnot altered by marked shivering. The administra-tion of 100% oxygen did not inhibit or delay the

294

ACUTELY INDUCED HYPOTHERMIA-CARDIOVASCULAREFFECTS

OxYGE CONSUMPTIONCC. /KG. BODYWT-6

--RESPIRATORY2 RATEI 1_40/

/

___

PLERATE- --.

200

ISO[

1601401201.

100so

60

STROKEVOLUMEIN cc,

14r12101

6 a>

MMOOIN G*- REWARMING-~a-a-- A L.J..L

37 36 33 31 29 2BOUWTEMPERATU -'C

FIG. 2. A COMPOSITEGRAPHOF AVERAGEDATA OB-TAINED FROM THREE DEEPLY ANESTHETIZED ANIMALSDURING COOLING AND REWARMING

It should be noted that the cardiac output failed toincrease during the rewarming period despite the ac-celeration of the heart rate. Values for oxygen consump-tion and cardiac output are given in cc./kg./min., pulserate is in beats per minute.

onset of shivering although such effects have beenreported by others (7).

Respiratory rateIn the control series there was very little change

in the respiratory rate (Figure 1). During bothcooling and rewarming a close correlation betweenthe respiratory rate and body temperature wasfound. This direct relationship was seen in thethree deeply anesthetized (Figure 2) and in mostof the less deeply anesthetized dogs. In two in-stances transient increases in respiratory ratefrom 28 to 40/min. and from 6 to 15/min. oc-curred during the early cooling period. The ini-tial rates varied from 6 to 28/min. and at the

lowest temperature varied from 0.5 to 7/min. ex-cept in one dog. In this animal the lowest ratewas 22, this dog maintaining a high rate during theentire experiment. In almost every case, the low-est respiratory rate occurred at the lowest bodytemperature. Following rewarming the range ofvariation in the rates was greater than that priorto cooling. Similar results have been obtainedby others in animals (4, 8).

Pulse rate

The direct relationship found between changesin body temperature and pulse rate is shown inFigure 2. During the induction of cooling, in-creases in pulse rate occurred in the two animals

OXYGENCAMSUMPTIONCC/ KG.OFBODYWT.-

60L <<"a+IEICOOUNG- R~EWARMIN9-_>*

37 35 33 31 29 27 29 31 33 35 37BODYTEMPERATURE-T0

FIG. 3. A COMPOSITEGRAPHSHOWINGTHE EFFECT OFSHIVERING IN ONE DOG

It should be noted that the curves obtained in this ani-mal do not differ significantly from those obtained fromdogs in which shivering was absent (Figure 2). Valuesof oxygen consumption and cardiac output are given incc./kg./min.

295

0. PREC, R. ROSENMAN,K. BRAUN, S. RODBARD,AND L. N. KATZ

in which the rates were measured immediately af-ter the ice was applied, although there was visibleshivering in only one dog. In both there weretransient increases in pulse and respiratory ratesand blood pressure. After these initial changesan almost linear relationship was found, thechanges in pulse rate closely paralleling thechanges in body temperature. The lowest pulserate was 59/min. occurring at 270 C. In the con-trol series very slight changes were noted (Fig-ure 1).

Similar results have been described by others(6). The direct effect of cold on the sinus pace-maker appears to be responsible for the variationsin heart rate. The absence of changes in the con-trol series lends support to the assumption that thechanges are not due to the anesthetic.

ElectrocardiogramElectrocardiograms (Lead II) were taken simul-

taneously with other measurements in 11 dogs.The progressive sinus bradycardia occurring dur-ing cooling has already been discussed. Therewere no marked abnormalities of the P waves butthe P-R intervals increased progressively as theheart rate slowed. A gradual widening of theventricular complexes occurred during cooling,being maximal at the lowest temperatures in allbut one dog. The average duration of the QRScomplex was 0.062 seconds prior to cooling andthis was doubled at 27° C. At the lowest tempera-tures an intraventricular block with notched Swave was noted in all but one animal. Figure 4illustrates the relationship between the body tem-perature and QRS duration. Electrical systole,calculated from Bazett's formula (9), was pro-gressively increased from a normal value for K of0.33 to 0.43 at 290C (Figure 4). An elevation ofthe S-T segment was observed in nine dogs and adepression in one dog. In two dogs the T wavereversed its direction and in four there was a de-creased amplitude. No change in T wave wasnoted in four dogs. Ventricular premature beatswere seen in three dogs and in two instances, ven-tricular fibrillation developed.

The initial effect of cooling is on the heart rateand on the duration of electrical systole. Changein contour and duration of ventricular complexesand S-T-T changes occur later. The effect ofcooling on repolarization in the ventricles is seen

before the effect on depolarization. All thesechanges were entirely reversible during the re-warming phase. The observed abnormalities areprobably a direct effect of cold on the metabolicprocesses in the myocardium, since similar changesoccur with direct application of cold (10). Amore severe hypothermia than we used has beenreported to induce auricular fibrillation and A-Vblock (11), but we did not observe these effects.

Blood pressure

Sodium pentobarbital anesthesia induces a fallin blood pressure, apparently via a direct effect onthe vasomotor center and this is often associatedwith a spontaneous fall in body temperature (12).In confirmation of this we noted in the controlseries a fall in temperature of as much as 20 C, as-sociated with a decrease in mean blood pressureduring the first two hours of anesthesia; after this,the pressure was maintained at the lower level intwo dogs and returned to control values in thethird (Figure 1).

A slight but transient rise in blood pressureusually occurred immediately after the applica-tion of the ice packs. Shivering usually elicitedfurther increases as great as 25%, but in two dogs,despite shivering, mean blood pressures fell 6%and 17%o respectively. The fall in blood pres-sure which occurred during the early phase was

DURATION OFORS -SECONDS

0.13

012

0.11 /

040 /009 /008 /007

2:6005

K VALUE046

0.44

24

04

0.038

036

16' 034

0.30

0 <1 COOUNG- REWARMING-*37 35 33 31 29 27 29 31 33 35 37

BODYTEMPERATURE- .GFIG. 4. RELATIONSHIP OF BODY TEMPERATURETO

ELECTROCARDIoGRAPHICCHANGESDiscussed in text.

296

ACUTELY INDUCED HYPOTHERMIA-CARDIOVASCULAREFFECTS

TABLE I

Cardiac output (cc./kg./min.) changes during cooling and rearming

Cooling temperature 0 C. Warming temperature 0 C.Dogno.

380 370 350 330 310 300 290 28° 270 290 310 330 350 370 380

C2 150 70 59 56 36 34 46 49C7 175 43 47 39 30 30 29 46CHi 144 100 83 66 63 35t 73 29 29 35 31C1 82 98* 111* 76* 129* 22 21 30 18tC3 72 96* 121* 53 46 77* 43 44 144tC5 63 143* 92 104t 80: 95* 68 64 57 66 66C8 112 65 86 88 40 101:t 78 67 82C9 75 39 173 42tCIO 141 155 113* 168* 79 1001: 63 72 86 80C12 200 77 f

t Dog died shortly afterward.

not great in most instances and not markedly dif-ferent from the changes observed in the controlseries. However, during prolonged cooling, therewas a tendency to a progressive fall in blood pres-

sure and at the lowest body temperature in thedogs that survived, mean pressures had fallento as low as 54%o of the original level (Figure 2).During rewarming there was a consistent rise inblood pressure, the latter returning almost to ini-tial values when the body temperature returned to370 C.

This laboratory has previously called attentionto a relationship between body temperature andblood pressure which may be demonstrated in sev-

eral species of animals. In these, lowering of thetemperature is accompanied by a fall in blood pres-

sure (13, 14). Certain data in the literature on

the effect of cooling on the blood pressure of man

(15), and animals (16) are in accord with thisconcept. However, other data (17) show that theblood pressure may be maintained at control lev-els in cooled anesthetized rats until the body tem-perature falls below 290 C. Proskauer et al. (18)found that in rats an initial decrease was followedafter 30 minutes of cooling by a rise in blood pres-

sure. These responses may be a part of the homeo-static mechanisms which maintain the body tem-perature at normal levels, and in the face of a

falling temperature maintain the blood pressure

for a period.

Cardiac output

Control values for cardiac output in these andother dogs that we have studied averaged 112 cc./kg./min. and varied from 60 to 175 (Table I).

I Extreme bradypnea. Estimation inaccurate.

In the deeply anesthetized dogs there was a markeddecrease in output during cooling and a small in-crease during rewarming (Figure 2). The lowestoutputs occurred, not at the lowest body tempera-tures, but during the early rewarming periods, andthese ranged between 35%o and 56%o of the con-

trol values. There were greater variations in car-

diac output in the six less deeply anesthetized dogs.In 10 of the 12 dogs, there was a definite increasein output as great as 126%o during the first periodof cooling. These changes were associated withshivering, increased oxygen consumption and pulserate, or a combination of these factors. For ex-

ample, dog C1, which shivered violently duringearly cooling, showed an increase of oxygen con-

sumption from 64 to 146 cc./min., of respiratoryrate from 28 to 40 and of pulse rate from 176 to196. The sudden rise in output in dog C9 at 310 Cwas related to an acute fall in arterial oxygen con-

tent which persisted until sudden vasomotor col-lapse led to death. This sudden increase in outputmay have been due to the acute anoxia but the eti-ology of the sudden drop in arterial oxygen con-

tent could not be accounted for. Complete re-

covery from cooling was not achieved as can beseen from the fact that the final values for cardiacoutput during rewarming were only about 50%of the original values. Increases in output duringrewarming of the less deeply anesthetized dogswere more rapid and of a higher order and almostreturned to the original values.

In two dogs of the control group there was a fallin cardiac output coincident with a spontaneousfall in body temperature up to 20 C (Figure 1).The reduction in cardiac output may have been due

* Shivering.

297

0. PREC, R. ROSENMAN,K. BRAUN, S. RODBARD, AND L. N. KATZ

partly to an anesthetic action and partly to the re-duction of body temperature. No change in eitherbody temperature or cardiac output occurred inthe third control dog.

Right auricular and pulmonary arterial pressures

The mean right auricular pressure was meas-ured in 10 dogs during cooling and in three con-trols, and was found to lie in the range of -+- 1 mm.Hg in the control state (compared to the atmos-pheric pressure). In the control group variationsoccurred, usually in a negative direction, the largestbeing - 2 mm. Hg. In the animals which werecooled the changes were as great as - 5 mm. Hg.

The mean pulmonary arterial pressure wasmeasured in eight cooled and three control dogs.The initial values lay in a range of 7 to 12 mm.Hg except in one dog in which it was 15 mm. Hg.In both control and cooled dogs there were rapidand substantial variations in pressures as muchas 6 mm. Hg. The pressure changes were notconsistent and bore no relationship to pulse rate,blood pressure, stroke volume or cardiac output.

With the development of the direct Fick methodin man, much attention has been paid to the rightauricular pressure and its relation to cardiac out-put. McMichael and his group (19) have prom-ulgated the idea that within physiological limitsin the compensated heart, the cardiac outputchanges concordantly with right auricular pres-sure but this concept has recently been criticized(20, 21). In our experiments there was no rela-tionship between mean right auricular pressureand stroke volume, pulmonary arterial pressure,blood pressure or work of the right heart in eitherthe cooled or the control series. Changes in themean right auricular and mean pulmonary arterialpressures usually occurred in the same directionbut were not proportional and on occasion the di-rection of change was discordant.

HemoconcentrationHematocrits were taken periodically in 10 dogs.

It should be noted that about 100 to 150 cc. ofblood were withdrawn and replaced with isotonicsaline solution during each experiment. Therewas an average increase in the hematocrit valueof 10% at the lowest temperatures with a tendencyto hemodilution during rewarming. Barbour (22)has demonstrated a reflex which originates in

chilled skin and is mediated via the hypothalamus,to produce a shift in the body fluids from the circu-lating blood to the interstitial spaces in hypo-thermic states.

Causes of deathFive animals died in the course of these experi-

ments. Cardiac failure has been suggested as acause of death in hypothermia in man (23), indogs (5) and in rats (17). However, this view isnot substantiated by our observations. One deathwas due to ventricular fibrillation which occurredearly in the cooling period; ventricular prematurebeats were frequent from the onset of the ex-periment and in this case an irritating effect of theintracardiac catheter could not be excluded. Inthree dogs there was a sudden fall of blood pres-sure prior to the cessation of respiration. Thisoccurred during rewarming at body temperaturesof 350 C in one dog, at 340 C, in another, and inthe third dog it occurred at 27.5° C. Markedbradypnea and respiratory irregularity developedsoon after the blood pressure fell and respirationceased shortly thereafter. In these dogs the pulserates were 146, 120 and 90/min. respectively, andthe cardiac outputs were at high levels during theperiod immediately preceding the fall in blood pres-sure. Wemust conclude that in such cases therewas a sudden vasomotor paralysis. In the fifthdog there was a primary central respiratory ar-rest. A sudden anoxic rise in blood pressure oc-curred and this was followed by a progressive fallin pressure. In this case the vasomotor centerwas able to respond to anoxia in a normal manner.Artificial respiration was instituted and the heartcontinued to beat for about two hours afterrespiratory arrest. Therefore, in our cases deathwas due to failure in the central nervous systemeither of the respiratory or of the vasomotor cen-ters, as stated in some other reports (5, 24). Inmore prolonged and more intense hypothermiaemployed by others, other causes may be involved.

GENERALCOMMENTS

The present studies, correlated with those ofother investigators, make possible a more com-plete analysis of the hemodynamic changes occur-ring during hypothermic states. Figure 2 repre-sents a composite graph in which the relationshipbetween the hemodynamic factors which we have

298

ACUTELY INDUCED HYPOTHERMIA-CARDIOVASCULAREFFECTS

studied and changes in body temperature can

readily be seen. The general effect of cold on allbody tissues is to reduce progressively all cellularmetabolic processes in accordance with van'tHoff's Law. Early in the cooling period of a

warm blooded animal such as the dog, reflexes are

stimulated which tend to decrease the loss of heatand to increase the rate of endogenous heat pro-

duction, thus tending to maintain the body tem-perature at normal levels. Some of these reflexesresult in a reduction in peripheral blood flow,shivering and redistribution of body fluids. Thecardiovascular response during this early periodis manifested by tachycardia, elevated blood pres-

sure and increased cardiac output, as was observedearly in the cooling of the lightly anesthetizedanimals. Anesthesia tends to inhibit these protec-tive reflexes to some extent but shivering can beseen in most anesthetized animals during cooling.

As cooling becomes more intense and more pro-

longed there develops a progressive bradycardia,bradypnea, decreased oxygen consumption, de-creased cardiac output and hypotension. Themetabolic demands of the tissues for oxygen andsubstrate are so markedly decreased that true an-

oxia must be minimal despite a reduced avail-ability of oxygen. This accounts for the failureof the administration of 100% oxygen to be ofbenefit or to prolong life during hypothermia. Atvery low body temperatures a progressive respira-tory center depression occurs, due to the effectof cooling, and in our experiments to the effect ofanesthesia. This depression may markedly reducethe oxygen supply and lead to the death of the an-

imal in anoxia. During this phase of respiratoryfailure, oxygen and artificial respiration shouldbe of value.

The depressant effect of cold on all body tis-sues holds true for the myocardium itself. Thisis reflected by the development of progressive sinusbradycardia, abnormalities in depolarization andlater in repolarization, prolongation of electricalsystole and the development of intraventricularblock, as well as prolongation of auriculo-ventric-ular conduction time. The direct effect of coldon the heart is well shown by the development inthe isolated heart preparation of changes similarto those described above. We observed no evi-dence of an inability of the heart to deal satisfac-

torily with the venous return of blood from theperiphery.

The marked decrease of cardiac output whichoccurred during cooling is primarily the resultof changes in the peripheral circulation. There isa progressive decrease in the circulating bloodvolume with the development of hemoconcentra-tion, due to a shift of body fluids from the vascu-lar bed into the interstitial spaces. These factorstend to reduce the venous return to the heart andto decrease the cardiac output. There are cer-tain compensatory mechanisms, however, as isshown by the maintenance of the blood pressureat relatively high levels in the face of a markedlyreduced cardiac output. These facts may thus beconsidered as evidence that marked peripheralvasoconstriction occurs, partially compensating forthe decreased blood volume and cardiac output.When these compensatory mechanisms fail to meetthe stress produced by hypothermia, vasomotorcollapse occurs and leads to death.

All the changes which we observed were en-tirely reversible upon rewarming, and the re-covery of function was in general related to the de-gree of rewarming accomplished. Thus the heartand respiratory rates, the blood pressure and thecardiac output tended to return to control valuesas the animal approached its normal body tempera-ture. This would suggest that the induction ofhypothermia in a warm blooded animal such as thedog may not be an altogether unphysiologicalprocedure, in the sense that the compensatingadjustment patterns, which appear during the cool-ing and the rewarming periods, are ordinarily ca-pable of coping with the situation, within limits.

SUMMARY

1. Circulatory changes and cardiac output werestudied in 12 dogs during cooling induced by ex-posure to chipped ice packs and during rewarm-ing, and in three control dogs.

2. During the early cooling period, thermo-genic reflexes which cause shivering are set offwhich act to produce an increase in oxygen con-sumption, respiratory and heart rates, blood pres-sure and cardiac output. These reflexes may beinhibited to some extent by deep anesthesia.

3. As cooling continues, and the combined de-pressing effects of hypothermia and anesthesia act,hemoconcentration and a progressive fall in oxy-

299

0. PREC, R. ROSENMAN,K. BRAUN, S. RODBARD, AND L. N. KATZ

gen consumption, in respiratory and heart rates,and in blood pressure occur. The cardiac outputalso falls. Marked prolongation of electrical sys-tole, with the occurrence of intraventricular andA-V block, and changes in the S-T-T contourare seen. The changes in these values are re-lated to the degree of hypothermia obtained.

4. During rewarding, these effects are reversedwith a return of the blood pressure, respiratoryrate, and pulse rate to normal, and with a tend-ency to a return to normal in the degree of hemo-concentration and cardiac output. Electrocardio-graphic changes were also reversed.

5. When it occurred, death in hypothermia wasattributable to failure of the vasomotor or respira-tory centers, rather than to failure of the heart.

6. No consistent correlated changes in rightauricular pressure, on the one hand, and pulmon-ary arterial pressures, cardiac output, or the workof the right fieart, on the other, were observed dur-ing either the cooling or rewarming periods.

BIBLIOGRAPHY

1. Prec, O., Rosenman, R., Braun, K., Harris, R.,Rodbard, S., and Katz, L. N., The circulatory re-sponses to hyperthermia induced by radiant heat.J. Clin. Invest., 1949, 28, 301.

2. Fick, A., Ueber die Messung des Blutquantums inder Herzventrikeln. Sitzungsb. der phys. med.Gesellsch. zu Wurzbung, 1870, p. 16.

3. Van Slyke, D. D., and Neill, J. M., The determina-tion of gases in blood and other solutions by vac-uum extraction and manometric measurement. J.Biol. Chem., 1924, 61, 523.

4. Wintrobe, M. M., Clinical Hematology. Lea &Febiger, Philadelphia, 1946, page 242.

5. Woodruff, L. M., Survival of hypothermia by dog.Anesthesiology, 1941, 2, 410.

6. Dill, D. B., and Forbes, W. H., Respiratory andmetabolic effects of hypothermia. Am. J. Physiol.,1941, 132, 685.

7. Kottke, F. J., Phalen, J. S., and Visscher, M. B.,An effect of breathing high oxygen mixtures onshivering in man. Federation Proc., 1944, 3, 26.

8. Hamilton, J. B., Effect of hypothermic states uponreflex and central nervous system activity. YaleJ. Biol. & Med., 1937, 9, 327.

9. Bazett, H. C., An analysis of time-relations in theelectrocardiogram. Heart, 1920, 7, 353.

10. Nahum, L. H., Hoff, H. E., and Kaufman, W., In-fluence of temperature on electrogram and mono-phasic action potential of mammalian heart. Proc.Soc. Exper. Biol. & Med., 1941, 46, 395.

11. Hamilton, J. B., Dresbach, M., and Hamilton, R. S.,Cardiac changes during progressive lbypothermia.Am. J. Physiol., 1937, 118, 71.

12. Beaton, L. E., Leininger, C. R., and McKinley, W. A.(& others), Neurogenic hyperthermia and itstreatment with soluble pentobarbital in monkey.Arch. Neurol. & Psychiat., 1943, 49, 518.

13. Rodbard, S., and Tolpin, M., Relationship betweenbody temperature and blood pressure in thechicken. Am. J. Physiol., 1947, 151, 509.

14. Rodbard, S., and Feldman, D., Relationship betweenbody temperature and blood pressure. Proc. Soc.Exper. Biol. & Med., 1946, 63, 43.

15. Wayburn, E., Immersion hypothermia. Arch. Int.Med., 1947, 79, 77.

16. Hook, W. E., and Stormont, R. T., Effect of low-ered body temperature on heart rate, blood pres-sure and the electrocardiogram. Am. J. Physiol.,1941, 133, 334.

17. Crismon, J. M., Effect of hypothermia on the heartrate, arterial pressure and electrocardiogram of therat. Arch. Int. Med., 1944, 74, 235.

18. Proskauer, G. G., Neumann, C., and Graef, I., Themeasurement of the blood pressure in rats withspecial reference to the effect of changes in tem-perature. Am. J. Physiol., 1945, 143, 290.

19. McMichael, J., and Sharpey-Schafer, E. P., Theaction of intravenous digoxin in man. Quart. J.Med., 1944, 13, 123.

20. Brannon, E. S., Merrill, A. J., Warren, J. V., andStead, E. A., Jr., The cardiac output in patientswith chronic anemia as measured by the techniqueof right atrial catheterization. J. Clin. Invest.,1945, 24, 332.

21. Warren, J. V., Brannon, E. S., Stead, E. A., Jr., andMerrill, A. J., The effect of venesection and pool-ing of blood in the extremities on the atrial pres-sure and cardiac output in normal subjects withobservations on acute circulatory collapse in threeinstances. J. Clin. Invest., 1945, 24, 337.

22. Barbour, H. G., McKay, E. A., and Griffith, W. P.,Water shifts in deep hypothermia. Am. J. Physiol.,1943, 140, 9.

23. Talbott, J. H., Medical progress; physiologic andtherapeutic effects of hypothermia. New EnglandJ. Med., 1941, 224, 281.

24. Ariel, I., Bishop, F. W., and Warren, S. L., Studieson effect of hypothermia. Cancer Research, 1943,3, 448.

300