The Circulatory Responses to Hyperthermia Induced by Radiant Heat

THE CIRCULATORY RESPONSES TO HYPERTHERMIA INDUCED BY RADIANT HEAT

O. Prec, … , S. Rodbard, L. N. Katz

J Clin Invest. 1949;28(2):301-306. https://doi.org/10.1172/JCI102072.

Research Article

Find the latest version: https://jci.me/102072/pdf THE CIRCULATORY RESPONSES TO HYPERTHERMIA INDUCED BY RADIANT HEAT' By 0. PREC,2 R. ROSENMAN, K. BRAUN,8 R. HARRIS, S. RODBARD, AND L. N. KATZ (From the Cardiovascular Department,4 Medical Research Institute, Michael Reese Hospital, Chicago) (Received for publication July 19, 1948)

An elevation of the body temperature evokes femoral artery recorded mean arterial blood pressures on many physiological changes, as a direct result of a kymograph. Mean pulmonary arterial and right auricular pressures were obtained with saline manometers the increased rate of metabolism and the greater attached to their respective radio-opaque cardiac cathe- demand for oxygen and substrate, and indirectly in ters. Cardiac output was calculated according to the warm blooded animals because of the elicitation of Fick formula, blood from the pulmonary artery being thermolytic mechanisms. An understanding of the used as mixed venous blood, and oxygen consumption being recorded with a basal metabolism apparatus at- interrelationships between these factors is of ob- tached to the cannulated trachea. Hematocrits of the vious importance in an appreciation of the dynam- arterial blood were determined according to the method of ics of hyperthermia and in the establishment of a Wintrobe. Blood O0 and CO2 analyses were performed by rationale for its treatment. Despite many impor- the Van Slyke-Neill technique. A detailed description of tant contributions to this problem, there is a pau- these methods is given elsewhere (1). Measurements similar to the above were also obtained in a series of city of information concerning the effect of hyper- three dogs heated by short wave diathermy,5 but unless thermia on the cardiac output and the right intra- specified, the results given below refer only to those cardiac pressures. We have therefore undertaken experiments in which radiant heat was employed. Simi- an analytical study of the circulation in lar measurements were made on three control anes- hyper- thetized dogs kept thermia in an effort to assay these adjustments at room temperature. more satisfactorily. Our present have experiments DISCUSSION OF RESULTS been limited to a study of the changes which occur following exposure to radiant heat since this Oxygen consumption method of producing hyperthermia seemed to be Many studies have demonstrated a close parallel the simplest, being free from incidental effects pro- between the rise in body temperature and the induced by pyrogens and by other methods. A few creased oxygen consumption (2-4) which ac- experiments were done with short wave diathermy. companies the acceleration of cellular metabolism during fever. In our experiments (Figure 1) an METHOD average increase of 57% was observed at a body Hyperthermia was induced in nine dogs. Six dogs, temperature of 420 C, the greatest increase being anesthetized with sodium pentobarbital, were exposed to 100%o in one instance. During the cooling period radiant heat from several bulbs until light the body tem- oxygen consumption peratures (about 380 C) gradually increased to 42° C decreased simultaneously during a two-hour interval. Body temperatures were with the fall in body temperature and returned to reduced during the following one- to two-hour periods control levels as the temperature of the animal by using cold wet cloths applied to the body and by again reached 380 C. In the control series only draft-producing electric fans. Temperatures were ob- slight and insignificant changes in oxygen con- tained from a thermometer placed deeply in the recto- sigmoid area. A mercury manometer attached to a sumption took place. idin part by a grant from the Department of the Respiration and content of Army to the Michael Reese Hospital (Dr. L. N. Katz, blood gases responsible investigator). The production of hyperthermia was accom- 2Rockefeller Fellow from Prague, Czechoslovakia. panied by an increase in respiratory rate and 8Hadassah Fellow from Jerusalem, Israel. 4This department is supported in part by the Michael 5 The Burdick Corporation generously supplied the Reese Research Foundation. short wave equipment. 301 302 0. PREC, R. ROSENMAN, K. BRAUN, R. HARRIS, S. RODBARD, AND L. N. KATZ

A-V DIFF. 2r volume (Figure 1) similar to that which has been reported in all types of naturally occurring fever o4. 01 A-V DIFf '91 RESR (2). During recooling the respiratory rate and 70 PtLSE volume decreased proportionately to the tempera- i/ 60 22 ture fall. It is of interest to note, however, that Is 50 - 210 greater than normal rates and volumes of respira- -40 - 200 . v)~ tion persisted even after the conclusion of the cool- ' RES 30- 190 ing period and the return of body temperature to

20- 180 control levels. This effect may have been due to aP. the greater rate of destruction of the anesthetic ' PULE 1-970 agent during the febrile period and a consequent 1301 heightened activity of the respiratory center. In 10 139 120I the control series no significant changes were -1- noted. 9120 9| iB.P. . -2 Differences in the respiratory response were 'RAp _ -3 i00 noted, depending on the type of heating used. With radiant heat the respirations were regular 90 and only slightly decreased in depth. When the 80 fever was produced with short wave diathermy, 70 - respirations became shallow and grossly irregu- 60 lar. The hyperpnea was probably due to the in- -I C.O. 50 creased CO2 production in the tissues, the in- 37 38 39 40 41 42 41 40 39 38 creased activity of the respiratory center itself, SODY TEMPERATURE-S'C and the initiation of thermolytic reflexes. It has FIG. 1. A COMPOSITE GRAPH OF THE AVERAGE VALUES been reported that extreme tachypnea and shallow THE HEATING AND COOL- OF VARIOUS FACTORS DURING respiration may seriously interfere with the ex- ING OF SIX DOGS change of alveolar air, and thus decrease the arte- A-V Diff. represents the arteriovenous oxygen dif- rial oxygen content (3, 5). However, the slope ference (vols.%); QO2, the oxygen consumption in cc./ of the oxygen consumption curve in our experi- kg./min.; Resp. and Pulse are, respectively, the respira- ments indicated that the diffusion of oxygen in the tory and pulse rates per minute; B.P., the mean arterial was not when radiant heat was blood pressure in mm. Hg; C.O., the cardiac output in lungs impaired cc./kg./min.; and R.A.P., the mean right auricular pres- employed. With diathermy, the slope of the sure in cm. saline. curve was highly irregular. It was because of this

TABLE I Oxygen content of arterial and venous blood (volume per cent)

Body Dog no. Hi Dog no. H2 Dog no. H3 Dog no. H4 Dog no. H5 Dog no. H6 temperature Art. Ven. Art. Ven. Art. Ven. Art. Ven. Art. Ven. Art. Ven. 0 C. 36 12.6 6.3 24.1 20.8 37 20.6 13.6 25.1 16.2 - 38 - 11.6 3 12.8 7.5 -- 14.9 12.1 39 19.7 10.4 11.7 2.4 13 5.5 23 8 15.5 11 16 12 41 19 8 12 2 12 2.7 24.5 9.6 19.4 6.3 16.2 11.4 42 19.7 5.7 12 1.9 12.1 25.3 10.1 - 16 10 40 19 8.7 - 13 5 - 20.3 7 16.4 8.7 39 - 24.8 6.3 -- - 38 18.9 8.7 - 12.7 6.0 - 20 4.6 15.4 7.9 37 I - - - 24.6 I3.2 -- - - Art. = arterial. Ven. = venous. CIRCULATORY RESPONSES TO HYPERTHERMIA 303 irregularity that the results obtained with dia- it that the utilization of oxygen increased pro- thermy were not treated in the present study. portionately to the rise in body temperature de- The arterial oxygen content varied only slightly spite the progressively reduced cardiac output during our experiments with an average decrease (Figure 1). The average arteriovenous oxygen of 0.8 vol.%o. The large difference in dog H5 difference was 6.1 vol.% at 36° to 380 C and in- (Table I) was associated with a simultaneous increased to 11.3 vol.% at 420 C, remaining high in crease in hematocrit. No significant changes were some instances or even increasing during the re- noted in the control series. Others have reported cooling period. The increase in A-V oxygen dif- decreases in arterial oxygen content during hy- ference was relatively greater than the increase in perpyrexia (3, 6) with a relatively greater de- oxygen consumption. These findings are in gen- crease in venous blood oxygen content (3). eral agreement with those of Uyeno (3). The effect of hyperpnea on the CO2 content of the arterial and venous blood should be empha- Hematocrit sized. Figure 2 shows that as the respiratory rate Approximately 100 cc. of blood were removed increased, the CO2 content of the arterial blood and replaced by isotonic saline solution during decreased. The curve for venous blood is similar each experiment. Hematocrits of arterial blood to that of arterial blood but shifted to the right. samples showed no consistent trends. During The primacy of the thermolytic mechanism which fever it has been reported (7) that there is usu- is effected chiefly by panting in the dog is seen by ally an initial hemodilution with increased blood the marked reduction of blood CO2 and the conse- volume. quent alkalotic tendency. Despite this tendency to alkalosis, the respiratory rate continued at high Cardiac output levels. Two of the most important factors which must be assayed in order to understand the circulatory Arteriovenous oxygen difference changes which occur during exposure to radiant The oxygen content of the arterial and venous heat include the simultaneous measurement of the bloods is given in Table I. It can be seen from cardiac output and the blood pressure. In our experiments the cardiac output decreased con- RESPIRATORY TE /MIN sistently with fever (Table II) in all but two ani- T70 42 70 mals in which a transient rise occurred at the be-

65 ginning of the warming period. The output be-

60 - gan to decrease early or immediately after the heat was applied. In the two dogs H2 and H4, the 55- cardiac output decreased markedly while the temperature was rising from 360 to 380 C (from 43 1396 to 821 cc., and from 1300 to 712 cc./min., 40- respectively). The relatively greater increase in

320 \ the arteriovenous oxygen difference compared to

TABLE II 25 Cardiac output cc./kg./min. and body temperature

20 Body temperature (0 C.) 5t ARTERIAL VENOUS Dog no. a5 2D 25 30 35 40 45 36 37 38 39 41 42 40 39 38 37 00a O0-0VOLUMES % HI -78- 95 77 64 54 -53 - FIG. 2. A GRAPH SHOWING THE RELATIONSHIP BE- H2 89 52 64 - - 57 diedat TWEEN THE BODY TEMPERATURE, RESPIRATORY RATE, AND 430 C. THE CARBON DIOXIDE CONTENT (VOLS.%o ) OF ARTERIAL H3 - 136 82 86 - 85 -77 - H4 111 64 - 41 49 50 - 20 - 12 AND VENOUS BLOOD DURING THE WARMING PERIOD FROM H5 ---74 33 129 - 16 - 370 rO 420 C H6 --344 166 163 129 81 - 75 - T represents temperature. 304 0. PREC, R. ROSENMAN, K. BRAUN, R. HARRIS, S. RODBARD, AND L. N. KATZ that of the oxygen consumption accounted for the basis of regional differences in vascular tone. decreased cardiac output which occurred in the This mechanism deserves further analysis. In face of increases in oxygen consumption and pulse the control series, a slight decrease in blood pres- rate. sure occurred probably due to the anesthetic and Electrocardiographic studies (2, 8) and the to the spontaneous fall in body temperature (1). transitory character of the changes during fever suggest that the conduction system of the heart is Pulse rate apparently not damaged in short-lasting fever. A close correlation between body temperature We may therefore assume that the changes in car- and pulse rate has been shown (2). Acceleration diac output during fever were determined pri- of the pulse rate is effected by the direct action of marily by a decreased venous return rather than the increased temperature on the sinus node which by cardiogenic factors. The decreased venous re- remains the pacemaker even during very rapid turn is probably an early result of redistribution rates (10). The rate increases as much as 92 of blood which may later be accentuated by the beats per minute at 420 C, averaging 39 (Figure markedly reduced blood volume resulting from the 1). No significant changes were noted in the con- fluid loss consequent to heating. Some decrease trol series. An acceleration of the pulse rate is in cardiac output was noted in the controls due obviously not necessarily associated with an in- partly to the anesthetic and partly, perhaps, to crease in cardiac output since other factors play more dominant roles the spontaneous fall in body temperature of one (11). to two degrees (1). Intracardiac pressures During recooling the cardiac output continued The pressure in the right auricle was measured to decrease significantly, even after body tempera- in four dogs and showed steady and substantial de- ture had fallen to control levels (Table II). This creases (Figure 1). The pressures continued to would suggest that the febrile period sets up fall during the cooling period and at the end of mechanisms which reduce the cardiac output, and the experiments were as low as 4 cm. saline below that these mechanisms may persist for a time after the control levels. Large variations in both di- the cessation of the heating period. rections were seen in the pulmonary arterial pressure in both the heated and control series. The Blood pressure changes in the right auricular pressures in the Blood pressure changes seen during the induced control group were inconstant, varying only fever were variable (Figure 1). In four dogs an slightly from control levels. initial and progressive fall occurred. In one, no change in pressure occurred and in one there was DISCUSSION an initial rise followed by a fall. The greatest fall The data collected during our experiments in pressure was 57%o of the original level. In two represent an analysis of the circulatory responses of the three dogs heated with short wave diathermy to exposure to an intense radiant heat. The the blood pressure tended to rise and then re- mechanism of the rise in body temperature induced mained at the higher levels, findings similar to by exposure to radiant heat is essentially that of those reported by Wiggers and Orias (9). a prevention of heat loss (12). There is no doubt Since the blood pressure is the resultant of a that adaptation will be different following ex- number of factors, the effect of heating on each of posure to different types of heat. them must be considered in the interpretation of The generalized application of intense radiant results. Among the more important are the de- heat was followed shortly by a progressive fall in creased cardiac output, an increase in the vascular cardiac output, regardless of the elevated body bed due to vasodilation in various portions of the temperature, oxygen consumption and the heart body, and neurogenic factors which tend to in- rate. It has too often been assumed that cardiac crease the peripheral resistance and thus maintain output invariably increases during fever (2, 10). the blood pressure. The apparent contradiction While the output may increase in certain circum- in the latter two factors may be resolved on the stances (3, 4, 6) recent ballistocardiographic stud- CIRCULATORY RESPONSES TO HYPERTHERMIA 305 ies have failed to demonstrate an increase during When the febrile animal was permitted to lose fever (13) and others have suggested that the out- heat, the body temperature rapidly returned to put may be reduced in hyperthermic states (9, normal but the cardiac output continued to de- 14). Our results demonstrate that in induced crease. Evidently mechanisms initiated during fever due to radiant heat, the cardiac output is the hyperthermic phase which reduced the cardiac usually diminished and that this is a dominant output persisted during the phase of return to factor in the ensuing circulatory changes. normal body temperature. Rational therapy In hyperthermia produced by high external should therefore be directed to measures which temperature, marked vasodilatation occurs in the would increase the cardiac output to normal values cutaneous and other vascular beds (2). Large during the cooling phase. Simple reduction of the amounts of blood may therefore be shifted to the body temperature to normal obviously would be periphery, resulting in a decreased volume of blood inadequate since the cardiac output remains mark- available for venous return to the heart. This is edly reduced and peripheral circulatory failure may indicated by the falling venous pressure which we result in death. observed in our hyperthermic animals. The increased respiratory activity associated with an SUMMARY excessive internal temperature and the loss of The changes in the cardiovascular system were body fluids both contribute to decrease further the studied in anesthetized dogs during acute hyper- effective blood volume. These changes also lead thermia induced by intense radiant heat and dur- to a reduction in venous return and so to a further ing the return of body temperature from 420 C to reduction in cardiac output, in proportion to the normal. duration and intensity of exposure to radiant heat. During the warming period the oxygen con- Many explanations have been advanced to ex- sumption increased markedly, rising to about plain the circulatory collapse in heat stroke and 150%o of normal at 42° C. The cardiac output these have been reviewed recently by Daily and was progressively reduced. As a consequence the Harrison (4). These include hemoconcentra- arteriovenous oxygen difference was increased. tion, reduction of blood volume, increased intra- The blood pressure was maintained at relatively capillary pressure in normal levels in the face of an apparent increase resulting increased fluid in the capacity of the transudation and cardiogenic vascular collapse. vascular bed and a coincident and decreased cardiac output, indicating an active Wiggers Orias (9) have demonstrated a re- vasoconstriction duction in systolic discharge and a decrease in the process. The heart rate increased force of ventricular contraction when the body progressively with hyperthermia and the right auricular pressure tended to fall. temperature rises above 410 C due to exposure to During the radiant heat. Our results suggest that the normal recooling period the oxygen consumption, respiratory rate and heart rate re- heart is not significantly injured by hyperpy- turned toward normal levels. The cardiac output rexia and appears to be able to cope adequately and right auricular pressure continued to decline with the venous return from the periphery. significantly even after the body temperature had Previous experiments in this laboratory (1) returned to control levels. The blood pressure have re-emphasized that both man and animals was maintained in most of the animals. No sig- may undergo prolonged exposure to cold with re- nificant changes were found in the pulmonary ar- duction in body temperature as great as 120 C in terial pressure. the dog, with complete recovery in most instances During hyperthermia the heart appears to be following rewarming. Tolerance to elevated body able to cope with the venous return. Our ex- temperature is therefore considerably less than periments therefore suggest that the circulatory tolerance to cooling. Thus, in the dog, survival changes occurring in acute hyperthermia depend after a rise in temperature of 50 C will be rare upon (1) an increased demand for blood by the and even lesser elevations of body temperature tissues which leads to (2) an increase in the vol- may be sufficient to embarrass the cardiovascular ume of the vascular bed. Since (3) the circulat- system. ing blood volume and venous return do not in- 3b6' 0. PREC, A. R6SE14MAN, k. BRAUN, R. RARRIg, S. R6DBAkb, AND L. S. IArz crease proportionately and may even decrease, a 6. Hartman, F. W., Lesions of brain following fever marked reduction in cardiac output occurs. (4) therapy. J. A. M. A., 1937, 109, 2116. If the blood pressure and the circulation are to be 7. McIntosh, R., Kajdi, L., and Meaker, D., The re- maintained vasoconstriction must occur. Failure sponse of plasma, water and electrolytes to elevation of body temperature. Bull. Johns Hopkins of the vasomotor centers to compensate for the Hosp., 1930, 47, 61. progressively reduced cardiac output prepares the 8. Simpson, W. M., Studies on the physiology of fever. ground for peripheral vascular collapse. J. A. M. A., 1936, 106, 246. 9. Wiggers, C. J., and Orias, O., Circulatory changes BIBLIOGRAPHY during hyperthermia produced by short radio waves 1. Prec, O., Rosenman, R., Braun, K., Rodbard, S., and (radiothermia). Am. J. Physiol., 1932, 100, 614. Katz, L. N., The cardiovascular effects of acutely 10. Cheer, S. N., Effects of high temperatures on heart induced hypothermia on the circulatory system. and circulation in intact animals. Am. J. Physiol., J. Clin. Invest., 1949, 28, 293. 1928, 84, 587. 2. Altschule, M. D., and Freedberg, A. S., Circulation 11. Knowlton, F. P., and Starling, E. H., The influence and respiration in fever. Medicine, 1945, 24, 403. of variations in temperature and blood on 3. pressure Uyeno, K., Studies on the respiration and circulation the performance of the isolated mammalian heart. in the cat. III. Effect of rise of body tempera- J. Physiol., 1912, ture. J. Physiol., 1923, 57, 203. 44, 206. 4. Daily, W. M., and Harrison, T. R., A study of the 12. Asmussen, E., Cardiac output in rest and work in mechanism and treatment of experimental heat humid heat. Am. J. Physiol., 1940-41, 131, 54. pyrexia. Am. J. Med. Sc., 1948, 215, 42. 13. Starr, I., and Jonas, L., Supernormal circulation in 5. Looney, J. M., and Borkovic, E. J., Changes pro- resting subjects. Arch. Int. Med., 1943, 71, 1. duced on oxygen and carbon dioxide content of 14. Huddleston, 0. L., Baldes, E. J., and Krusen, F. H., arterial and venous blood of brain during dia- Alterations of cardiac cycle and of polygrams pro- thermy therapy for general paresis. Am. J. duced by artificial fever. Proc. Soc. Exper. Biol. Physiol., 1942, 136, 177. & Med., 1939, 42, 1.