REPORT OF 10 OCTOBER 1942, ON COOLING EXPERIMENTS ON HUMAN BEINGS

I. Problem of the Experiment

Up to the present time there has been no basis for the treatment of shipwrecked persons who have been exposed for long periods of time to low-water temperatures. These uncertainties extended to the possible physical and pharmacological methods of attack. It was not clear, for example, whether those who had been rescued should be warmed quickly or slowly. According to the current instructions for treating frozen people, a slow warming up seemed to be indicated. Certain theoretical considerations could be adduced for a slow warming. Well-founded suggestions were missing for a promising medicinal therapy.

All these uncertainties rested in the last analysis upon the absence of well-founded concepts concerning the cause of death by cold in human beings. In the meantime, in order to clarify this question, a series of animal experiments were started. And, indeed, these officials who wished to make definite suggestions to the doctors in the sea-rescue service had to assume a great deal of responsibility if it came to a question of convincing and consistent results in these animal experiments. At this particular point it is especially difficult to carry the findings in animals over into the human field. In the warm-blooded, one finds a varied degree of development in the heat-regulating mechanism. Besides this, the processes in the skin of the pelted animals cannot be carried over to man.

II. General Procedure of the Experiment

The effect of water temperatures of 2°, 3°, to 12° C. [34°, 37°, to 54° F.] were investigated. A tank 2×2×2 m. [6-2/3×6-2/3×6-2/3 ft.] served as an experimental basin. The water temperature was attained by addition of ice, and remained constant during the experiment. The experimental subjects were generally dressed in equipment such as the flier wears, consisting of underclothing, uniform, a one piece summer or winter protective suit, helmet, and aviators fur-lined boots. In addition they wore a life preserver of rubber or kapok. The effect of additional protective clothing against water-cold was tested in a special series of experiments, and in another series the cooling of the unclothed person was studied.

The bodily warmth was measured thermoelectrically. Following preliminary experiments in which gastric temperatures were measured by a thermic sound, we adopted the procedure of continuously registering rectally the body temperature [Kerntemperatur]. Parallel with this, the recording of the skin temperature was undertaken. The point of measurement was the skin of the back at the level of the fifth thoracic certebral process. The thermoelectrical measurements were controlled before, during, and after the experiments by thermometric tests of the cheek and rectal temperature.

In severe cooling, checking of the pulse is difficult. The pulse becomes weaker, the musculature become stiff, and shivering sets in. Auscultation during the experiment by means of a tube stethoscope fastened over the tip of the heart proved effective. The tubes were led out of the uniform and made possible the continuous listening to the heart during the stay in the water.

Electrocardiographic controls were not possible in the water. After removal from the water they were possible only in those cases in which, a too severe muscle shivering did not disturb the electrocardiograph records.

The following chemical studies were carried out: following up of the blood sugar picture (continuous); the sodium chloride picture in the serum; the nonprotein nitrogen (Rest-N); the alkali reserve; the alkali reserve of the venous and arterial blood and sedimentation rate (before and after the experiment). Besides this the general blood condition and viscosity were followed during the experiment, and before and after the experiment the resistance of the red blood cells and the protein content of the blood plasma (this refractometrically) were measured.

The following urinalyses were made regularly: sediment, albumen, sugar, sodium chloride, acetone, acetic acid, as well as a qualitative albumen determination.

In part of the experiment lumbar and suboccipital punctures were made as well as corresponding spinal fluid studies.

Among physical and therapeutic measures the following were tested:

In addition the following drugs were given: Strophanthin i. v.; Cardiaz 1 i. v. and i. c.; Lobelin and Coramin i. v. and i. c. In other experiments alcohol or grape sugar was given.

A part of the experiments was begun under narcosis (8 cc. Evipan i. v.).

III. The Clinical Picture of Cooling

The clinical picture as well as the behavior of the body temperature showed certain regularities in the general course; the time of appearance of certain phenomena was, however, subject to very great individual variations. As one might expect, a good general physical condition delayed the cooling and the concomitant phenomena. Further differences were conditioned by the position of the subject in the water and the manner of clothing. Furthermore, differences showed up between experiments in which the subject lay horizontally in the water so that the nape of the neck and the back of the head were splashed with water, and others in which neck and head protruded freely out of the water.

Peculiarly, the actual water temperatures between 2° C. and 12° C. [35° and 54° F.] had no demonstrable effect upon the rate of the cooling. Naturally such an effect must exist. But since besides the already mentioned individual differences and those due to experimental conditions, the various subjects cooled on different days at different rates of speed, the effect of the actual water temperatures between 2° and 12° disappears behind such variations.

If the experimental subject was placed in the water under narcosis, one observed a certain arousing effect. The subject began to groan and made some defensive movements. In a few cases a state of excitation developed. This was especially severe in the cooling of head and neck. But never was a complete cessation of the narcosis observed. The defensive movements ceased after about 5 minutes. There followed a progressive rigor, which developed especially strongly in the arm musculature; the arms were strongly flexed and pressed to the body. The rigor increased with the continuation of the cooling, now and then interrupted by tonic-clonic twitchings. With still more marked sinking of the body temperature it suddenly ceased. These cases ended fatally, without any successful results from resuscitation efforts.

In the course of the narcosis experiments the evipan effects in a few cases went directly over into a cold narcosis; in other cases one could determine a transitory return of consciousness, immediately following the awakening effect already described; at any rate, the experimental subjects were dizzy. Cold pain was not expressed.

Experiments without narcosis showed no essential differences in the course of cooling. Upon entry into the water a severe cold shuddering appeared. The cooling of the neck and back of the head was felt as especially painful, but already after 5 to 10 minutes a significant weakening of the pain sensation was observable. Rigor developed after this time in the same manner as under narcosis, likewise the tonic-clonic twitchings. At this point speech became difficult because the rigor also affected the speech musculature.

Simultaneously with the rigor a severe difficulty in breathing set in with or without narcosis. It was reported that, so to speak, an iron ring was placed about the chest. Objectively, already at the beginning of this breathing difficulty, a marked dilatation of the nostrils occurred. The expiration was prolonged and visibly difficult. This difficulty passed over into a rattling and snoring breathing. However, the breathing at this point was not especially deep as in Kussmaul’s breathing nor were any Cheyne-Stokes breathing or Biot’s breathing to be observed. Not in all subjects, but in a great number, a simultaneous hindering during this breathing through very profuse secretion of mucous could be established. Under these conditions sometimes a white, fine-bubbled foam appeared at the mouth which reminded one of an incipient lung oedema, though it was not possible to determine this symptom with certainty by clinical auscultation; only a sharpened unclean breath sound was audible. This foam might occur early, that is, at rectal temperatures of 32° C. to 35° C.; [90°-95° F.]. No special significance was to be attributed to this regarding the outcome of the experiment which is the opposite of the described relaxation of rigor. The rate of breathing increased at the beginning of the experiment, but after about 20 minutes it decreased to something like 24 per minute with slight variations.

In general a definite dulling of consciousness occurred at the dropping of the body temperature of 31° C. [88° F.] rectal temperature. Next, the subjects still responded to speech but finally answered very sleepily. The pupils dilated markedly. The contraction under light became increasingly weaker. The gaze was directed overhead with a compulsive fixation. After withdrawal from the water an increase in the reflexes was evident in spite of the rigor, and regularly a very marked drawing up of the testicles occurred which practically disappeared into the abdomen. Early in the experiment the face was pale. After 40 to 50 minutes cyanosis appeared. With this the face appeared redder, the mucous membrane bluish-red. The skin veins were not maximally collapsed and were virtually always penetrable.

The heart activity showed a constant change independent of all other individual variations, which was noticeable in all subjects. Upon introduction into the water with narcotized subjects as well as nonnarcotized subjects, the heart rate went suddenly to about 120 per minute. At a rectal body temperature of about 34° C. [93° F.] it then began to become increasingly slower and to sink continuously to about 50 per minute.

The bradycardia at a body temperature of about 29° to 30° C. [84° to 86° F.] changed suddenly to an arrythmia perpetua or, as the case may be, to a total irregularity and this began with a slow form of about 50 beats per minute; this slow form of irregularity could be transformed into a faster one. The transformation to the faster form was not an unfavorable sign regarding life.

When an electrocardiographic control after the experiment was possible, it regularly showed a Vorhof flutter. Let it be anticipated that this irregularity could continue to exist after the cessation of the cooling and a recovery of the body temperature to 33° or 34° C. [91° or 93° F.] 1½ to 2 hours after removal from the water, but then customarily changed of itself and without therapeutic aids into a coordinated heart activity. In the same way let it be anticipated that in all cases with a lethal termination, a sudden cessation of the heartbeat ensued upon an irregularity of the slow type.

A check of the blood pressure was attempted, but was in no case satisfactory since an exact measurement was not possible in the decisive stage of the experiment because of the severe rigor and muscle fibrillation.

Reference has already been made to individual differences in the behavior of the rectal temperatures. Figure 4 gives an example which includes four experiments, in which four different experimental subjects were cooled at identical water temperatures and with identical clothing. It was shown that in water at 4.5° C. [40° F.] temperature the time required for reaching a rectal temperature of about 29.5° C. [85° F.] varies between 70 and 90 minutes. But nevertheless the diagram shows that in spite of these individual differences, it is observable that the progress of the rectal temperature proceeds according to rule. The body temperature begins to sink rapidly from about 35° C. [95°-97° F.].

It is of very great practical significance at this point that the body temperature continues to sink virtually lineally for a considerable time after removal from the water. This continued drop can last 20 minutes or more. During this drop an after-drop of 4° C. [7° F.] could be observed, and indeed not only at temperatures under 30° C. [86° F.]. In one case it was observed that an interruption of the experiment at 35° C. [95° F.] after a further lapse of 20 minutes the rectal temperature had fallen 4° to 5° C. [8° F.] more. We will later discuss the “arresting” of this after-drop by physical measures.

In our experimental series, the lowest rectal temperatures which could be survived varied individually just as did the progress of the temperature drop. In general (in six cases) death occurred with a drop in temperature to values between 24.2° and 25.7° C. [75.6° and 77.6° F.]. In one case, however, a drop to 25.2° C. was survived. This experiment fell outside the typical picture insofar as after 90 minutes at 26.6° C. [79.9° F.] a virtually stationary condition of the rectal temperature had become established for 85 minutes. We will come back again to this special experiment.

The skin temperature sinks or drops much more rapidly than the rectal temperature. Within a minute there occurs a thorough saturation of the articles of clothing. Correspondingly the skin temperature falls already within 5 minutes to values between 24° and 19° C. [75° and 66° F.]. After 10 minutes it may have already dropped to 12° C. [54° F.]. Within 10 to 20 minutes more after the beginning of the experiment the steepness of the drop changes considerably. The curve of the skin temperature runs for some time, that is, for 15 to 30 minutes virtually horizontal. After this time there follows a further but now slower drop to the lowest figures, which may lie below 15° C. [59° F.] at the close of the experiment.

Parallel experiments which compare the course of the rectal temperatures and the cooling of the body with and without submersion of neck and back of head showed great difference in temperature drop. The curves pertain to the same experimental subject. The one with the deep fall to 26° C. [79° F.] in 70 minutes was obtained with a water temperature of 12° C. [54° F.] the other with a drop to 32.5° C. [90.4° F.] in the same time resulted from a water temperature of 5.5° C. [41.9° F.]. The very marked difference cannot be explained by a variation in resistance of the particular person, but is to be attributed to the position of the subject in the water and his head covering. In the experiment with the water at 12° C. [54° F.] the subject, in a kapok life preserver, lay flat in the water so that his neck and the back of his head were well submerged; beyond this he did not wear a flier’s helmet. In the other experiment with water at 5.5° C. [41.9° F.] the head was covered with an aviator’s summer helmet without headphones. The subject wore a rubber life preserver open at the back; with this, the head is somewhat out of the water.

In order to follow up the effect of isolated cooling of the neck and the back of the head on consciousness, body temperature, and circulation, this was undertaken in three special experiments. The experimental subject lay horizontal; the back of the head and the neck were dipped into a receptacle through which water of corresponding temperature was continuously run. In an experiment of 3 hours duration there occurred small temperature drops of not more than 0.8° C. [1.4° F.]. The water temperature was 1° to 2° C. [34° to 35° F.]. In one case after 50 minutes a marked sleepiness occurred which changed over into a deep narcosis. The heart activity was variable, and obvious bradycardia could not be observed. Irregularity never developed. Changes were not seen in the electrocardiograph. On the other hand in all three subjects the spinal fluid pressure was markedly increased after the ending of the experiment to maximal values of 300 mm. After the experiment, ataxia and definite Romberg phenomena were observed, as well as exaggeration of the normal reflexes; pathological reflexes were absent.

IV. Blood, Spinal Fluid, and Urine During Freezing

The differential blood smears showed no special features during cooling. On the other hand the number of white and red blood corpuscles shows a regular change. The number of leukocytes rapidly increases, roughly with the beginning of the steeper temperature drop at about 35° C. [95° F.] rectal temperature to values of from 25,000 to 27,000 per cu. mm. After one hour a maximum may be reached and a falling-off begins in the number of leukocytes, while the body temperature falls still further. The number of red corpuscles undergoes an increase, though to a relatively small degree, which in its course resembles the change in the number of leukocytes. We saw increases up to 20 percent. This increase is interrupted even earlier than the increase in the number of leukocytes, so that both curves give no reflection of the temperature curve. The increase of the erythrocytes corresponded to the increase of the hemoglobin of from 10 to 20 percent. A reduction of the fragility of the red corpuscles could not be demonstrated with certainty, on the other hand, although in three experiments a definite hemolysis occurred.

The viscosity regularly increases with the beginning of the fall in temperature. The rise can reach values up to 7.8. This rise occurs very early, indeed, already at body temperatures of 35° C. [95° F.]. After that the values remain relatively constant with further temperature falls. The albumen content of the plasma was likewise increased after the experiment, on the average by 1 percent of the absolute value. Since these measurements could not be made as often as those of viscosity for technical reasons, the connection with the progress of the viscosity remained unclear. Such a connection could not be recognized from the absolute values obtained.

With the acceleration of the temperature drop, there always occurs a more marked increase of the blood sugar to maximal values which may attain an average increase of 80 percent and in a few cases may reach an increase of over 100 percent. According to that, the maximal value of about 27.5° C. [81.5° F.] is reached and is maintained for some time. It is to be observed that as long as the temperature drop continues, in no experiment was it possible to observe a decrease in these high blood sugar values. It is usually to be observed that a relatively rapid drop of the blood sugar values sets in when, after removal from the water, the temperature drop ceases and goes over into a temperature rise. We consider these findings to be of theoretical significance. During the isolated cooling of the neck and back of the head which was described in section III the blood sugar remained constant.

In striking contrast to the increase of the blood sugar, there was never established a corresponding glycosuria in the urine collected immediately after the experiment or withdrawn through a catheter, although considerable quantities of urine averaging 500 cc. were found in the bladder; in only two cases could traces of sugar (0.5 percent) be demonstrated. This paradoxical behavior can, perhaps, be explained in this manner: during the time of great blood sugar increase, a blocking of the kidneys had occurred, and that the associated urine quantities were formed before or after this blocking under reflex polyuria. Acetone and acetic acid, likewise, could not be demonstrated in the urine.

The alkali reserve in the arterial and venous blood was regularly very much reduced at the end of the experiments. Experiments concerning oxygen saturation could not be carried out. According to the color of the venous blood withdrawn from the arm veins, the saturation of this blood must have been very greatly reduced; the blood was virtually black as it came into the syringe. Noteworthy in this connection are the autopsy findings which were undertaken directly after death. In these, the blood in the right heart appeared very dark, and in the left heart very bright red. According to this, one must calculate upon an increase in the saturation differential between the arteries and veins.

Sodium chloride and nonprotein nitrogen in the blood were not clear in the blood at the end of the experiments or increased within the limit of error. Sodium chloride in the urine was generally less, corresponding to a reduction of the specific gravity. On the other hand at the end of the experiments traces of albumin could regularly be demonstrated in the urine and moderately increased leukocytes, occasional erythrocytes, and epithelial cells in the sediments. In particular cases, albumin casts were also observed. The reaction of the urine remained identical before and after the experiments virtually without exception. The studies of the bile yielded no results.

Lumbar and suboccipital punctures immediately after the experiments showed a considerable increase in fluid pressure. On the average it amounted to between 50 and 60 mm. In one case, an increase to 420 mm. was seen. The protein values were always normal. Cell increases did not appear, likewise no abnormal deviation of the colloidal gold curve was observed. The meaning of these findings for therapy is still to be discussed later.

V. Recovery After Cooling and Its Dependence Upon Physiotherapeutic Measures

The important fact has already been referred to that after rescue from the cold water, the body temperature sinks further and so a further temperature reduction of 4° C. [7° F.] may take place. As was likewise emphasized, this may occur as a postphenomenon not only when low temperatures have been obtained already during the experiments, but it can be noted also at final temperatures of 35° C. [95° F.]. A dependence of this after-drop on the duration of the experiment could not be established; as a result it is difficult to calculate in advance. This fact becomes of great importance for practical measures; on the other hand it makes it difficult to gain an insight into the manner in which various physiotherapeutic measures affect the arresting of this after-drop and the recovery of the body temperature. Only because of the large number of the experiments was it possible to obtain well-founded concepts of this.

The flattest rise of the body temperature was to be observed when the subject was merely dried off, wrapped in warm cover, and left to himself after removal from the water. The recovery is greatly accelerated if the subject is placed in a hot bath as soon as possible after the removal of the wet articles of clothing. Warming under a light cradle assisted the temperature rise. Vigorous massage had a favorable effect, however, only if it was preceded by treatment in a hot bath or light cradle. In no case was it established that there was any indication of bad effects from the hot water or the light cradle, or that the subject had been harmed in any way. On the other hand, it was observed in three cases that a hot bath had doubtless a life saving effect. In two of these cases there had been complete cessation of heart and breathing action, and in one case the heart had stopped for several seconds after a markedly slackened irregularity before the subject was placed into water of not more than 50° C. [122° F.]. As a result of this we can discard all traditional objections to a sudden rewarming.

The favorable effect of a hot bath is still clearer in the observation of the general condition of the subject than in the temperature curves, although it cannot be presented objectively. The breathing very often becomes “freer” immediately upon introduction into the hot water. The hot water releases a strong stimulus; the unconscious subject often reacts with an outcry. Soon thereafter there occurs a distinct lessening of the severe rigor. The return of consciousness occurs sooner, and indeed at temperatures at which it did not usually happen under other methods of treatment.

In the first experiments with hot water treatment, this was continued only for 10 minutes; after that the subjects were removed and vigorously massaged. Under these circumstances it could be established that the temperature rise continued during the rubbing, indeed in one experiment the rise became steeper. As already indicated, this favorable effect of dry rubbing was not so pronounced without preliminary treatment by heat. It is important, too, that the rubbing be done when the severe spasm of the peripheral vessels has already passed.

In view of this, the hot hath is the best method of treatment of the severely cooled person. However, in the practice of sea rescue service it will not be possible to carry out this method, since the necessary means are not available in aircraft and boats. Under these circumstances we must consider next only the rapid rewarming with light cradle or electrically heated sleeping bag. Therefore a sleeping bag as now used in the sea rescue service was also tested. It was evident that the temperatures which can be developed by this means are not sufficient for heat therapy. With those it was possible to reach a temperature of only 32° C. [90° F.] over the skin, with the heat turned on fully. Besides this, the wall of the foot-section of the sleeping bag is only partly heated; on the outer sides it remains completely cold. As long as no improvement and strengthening of the heating equipment of the sack is carried out, the sleeping bag can be considered only as a substitute for wrapping in warm covers.

The warming by means of the light cradle is more uneven than with a hot bath. With warming by light one might expect severe local vessel expansion with danger of collapse. Actually the subjects often complained of dizziness and nausea after reaching consciousness if the treatment lasted longer than 15 minutes. Occasionally vomiting occurred. In these cases it is indicated to switch off the light cradle and to pack the subject with covers. Apart from this it must be remembered that during unconsciousness the subject should be protected against direct contact with the lamps by means of covers, otherwise burns could occur during clonic-tonic convulsions.

This suggests that “short waves” be employed to supply heat, since it was shown in animal experimentation that by this means it is possible to bring about a thorough warming of the whole animal, which leads to a recovery of the animal with puzzling rapidity. We did not have the proper equipment for a thorough warming of a human being by this means. For this reason the short wave therapy of the heart was tried. This did not have any demonstrable effect. Above all, it is necessary to advise against a practical application of this method, since there exists the danger of prolonged burning even in full consciousness, as the result of cold anaesthesia, even if the treating physician carefully tries to avoid this.

The severe difficulty in breathing as well as the formation of foam before the mouth, which reminded one of incipient lung oedema, seemed to indicate oxygen therapy. Therefore this therapy was tried in four experiments. It showed no effect on either the breathing or the heart action. It has been pointed out that the arterial blood appears especially light red.

VI. Death After Cooling in Water

Practical and Theoretical Considerations

Reports to the effect that those who have been rescued at sea are imperilled for a considerable time after rescue has aroused special attention. It has been reported especially that sudden cases of death occurred as much as 20 minutes to 90 minutes after rescue, and that in mass catastrophes these sudden deaths could amount to mass-dying (rescue collapse). These observations have set off far-reaching discussions. Bleeding in the rewarming periphery, break-downs of neural and humoral correlations and similar ideas have been brought up.

In contrast to this our experiments give a relatively simple explanation of cold-death under these conditions. With the exception of a single case, a total irregularity of the heart chamber could be definitely demonstrated in all cases of cooling under 30° C. [86° F.], (50 experiments), when the rectal temperature reached 29° C. [84° F.] and usually already at a cooling of 31° C. [88° F.]. The exception was an experiment on an intoxicated subject, which is to be gone into more fully below (see sec. VII).

Furthermore heart-death was established clinically in all cases of death observed by us. In two cases breathing ceased simultaneously with the heart activity. These were cases in which it was specially noted that the neck and the back of the head lay deep in the water. In all remaining cases breathing outlasted the clinical chamber cessation by as much as 20 minutes. In part this was “normal, much decelerated breathing,” in part an angonal form of gasping. As already referred to, an auricular flutter could be demonstrated cardiographically during the irregularity.

In cases in which a special cooling of neck and back of head had existed before death, the autopsy showed a marked brain oedema, a tight filling of the general brain cavity [Hirngefaesse] blood in the spinal fluid as well as blood in the Michaelisrhomboid.

The heart findings warrant our taking a certain attitude toward the question of rescue collapse. Death occurred relatively quickly after removal from the water, which may be compared with rescue. The longest interval involved was 14 minutes. It is to be noted, however, in the first place, that almost certainly a much larger number of deaths would have been observed if an active heat therapy had not almost regularly been coupled directly with the completion of the experiment; in the second place, that in such cases there would have been very much longer intervals. We have already called attention repeatedly to the after-cooling following the experiment. In every case where this had proceeded to a certain point, countermeasures were taken, since the experiments were never planned to end in death. One may well imagine, however, that in mass catastrophes, in which almost exclusively rescue collapse has heretofore been described, the therapeutic measures were confined to an undressing and drying off of the rescued together with a subsequent wrapping in covers. Under these conditions after-drops of great magnitude and long duration were to be expected. In the course of this delayed fall in temperature, a heart-death might occur as in our experiments.

We should like to emphasize that the irregularity per se is not to be regarded in our experiments as a symptom of danger to life any more than in the clinic, but rather as a sign of direct heart damage, which increases continuously with further falling off of temperature, until finally the heart fails. If the temperature drop is arrested, the slow form of irregularity passes over into a rapid form. This transition is a favorable sign for survival; for this irregularity virtually always passes over of itself after a time averaging 90 minutes into normal heart activity. It continues therefore for a long time after the body temperature has already risen markedly. A danger to the circulatory system could not be demonstrated at this stage. In three cases the return of the heart action to normal occurred in spite of simultaneous energetic physical work.

With the demonstration that cold-death of man is primarily a heart-death, the essential points for therapy are also cleared up. The cause of the severe heart damage is another question. Since our studies were primarily aimed at the development of practical methods of treatment, we will not go very far into the theoretical concepts which may be developed in this connection. Still, several hints may be drawn from the blood studies:

1. The great increase of the viscosity causes an increased loading upon the heart.

2. The choking of peripheral vessel areas by the severe vessel contraction leads to an over-filling of the central areas. This appears not only from our autopsies. In all available records of autopsies which pertain to cases of death from cold in the water after sea disaster, we find uniformly a severe over-filling of the right heart.

3. It is to be calculated that, under the effect of the low blood temperature, the heart itself becomes severely hypodynamic. It has been proved long ago in animal experimentation that a Vorhof flutter can be developed by the overloading and cooling of the isolated heart.

Besides a physical damaging of the heart musculature by the cold, we must also keep in mind the damaging by pathological products of metabolism. Next, the sharp increase in blood sugar may be connected with the increased outpouring of adrenalin. The constancy of this increase of blood sugar during the temperature drop is, however, remarkable. One may well assume that this flow of adrenalin exhausts itself with the continuance of the temperature drop. With this there would have to be a rapid decrease in the blood sugar if the oxidation processes were to continue undisturbed. The decrease in the alkali reserve or the development of an acidosis argues strongly for an injury.

Animal experiments, with general cooling, give grounds for believing that the intermediary metabolism is disturbed during drops in temperature; but this change is also discussed in connection with local freezing of the human being and has been proved to a certain extent. Furthermore, not only this disturbance shows a transition between general and local damage by cold. In both cases there occurs an increase in viscosity, which points to a change in the capillary walls and indicates the conclusion that there is a change in the permeability of those walls for protein and water.

The heart-death remains prominent, the regular increase of spinal fluid pressure with severe cooling of the neck and back of the head leaves it unsettled whether, in addition, this has pathognomonic significance for the outcome. With a fluid pressure of 420 mm. it must in fact be assumed that this participates in the development of bradycardia.

The detection of an increase in fluid pressure is also not without significance for therapy. One may think of a lumbar or suboccipital puncture as a measure to be prescribed. After a lumbar puncture there occurs a transformation of the slow form of arrhythmia into the rapid form. It must remain undecided whether such measures, which delay a rapid, active rewarming, are to be recommended for practical application in the sea-rescue service.

The idea that cold-death in water depends upon failure of the heart, accompanied or unaccompanied by breathing, is subject to limitation. One experiment among fifty-seven was typical. This involved survival of a cooling to 25.2° C. [77.4° F.] during a stay of 3 hours in water of 5.5° C. [41.4° F.]. The rectal temperature under these conditions remained constant within slight variations between 27° and 25° C. [81° and 77° F.] for the last hour and a half. Likewise, quite irregularly, no increase in blood sugar occurred. But most striking was the fact that until the end of the experiment and after its termination consciousness was undisturbed. The course of the experiment reminded one of the behavior of certain experimental animals which can withstand extremely low body temperatures for long periods of time. Lower, warm-blooded animals (for example, rats) can endure rectal temperatures of 20° C. [68° F.] for several hours. It is conceivable that this atypical experiment, had it been continued, would have shown also an atypical cause of death. Against this we have the fact that an irregularity had already set in but not before a temperature of 30.1° C. [86.2° F.] had been reached.

Also, aside from the fluid pressure increase, the part which the central nervous system plays in the outcome of the experiment seems to us to be secondary. The experiments with simultaneous cooling of the neck of course showed how the cooling of the neck and back of the head speeds up the lowering of temperature. This is to be explained by the fact that the counter-controls which are relayed from the temperature center to the periphery, either cannot exist further because of hypofunction of the centers (effect of oedema and cooling), or are no longer transmitted because of cold-blocking of the pathways. But likewise central counter-controls for the areas of the peripheral capillaries may fall; thus delaying the overloading of the heart by extended periphera vasco friction.

VII. The Influence of Pharmacology and the Question of Alcohol

Now experiments by Jarisch have shown that heart drugs like strophanthin and stimulants like cardiazol and coramine in therapeutic doses may react toxically upon cooled animals. These findings are a warning to be most careful in the medicinal treatment of severely cooled persons, though strophanthin and cardiazol have heretofore been expressly recommended in such cases.

In experiments with fatal outcome, the stopping of the heart occurred either in the water or after an interval of not more than 14 minutes after removal from the water. With such a rapid course of events it is unlikely that one can favorably influence the heart action by intravenous injections of strophanthin, especially because the circulation is at a very low ebb before the heart-death. For this reason, in a case whose condition was already very dangerous, strophanthin was given intracardially in a dose of 0.25 mg. Thereupon the heart condition grew still worse and after 5 minutes the heart stopped. One had the impression that the heart action was made worse by the intracardial injection of strophanthin. This is, however, the only case which left the possibility of damage by strophanthin in doubt. No such damage could ever be established in the intravenous injection of strophanthin. On the other hand no therapeutic effect, even with maximal doses of 0.5 mg., could be detected. Figure 11 [not reproduced], last section, shows the total duration in 10 cases of the irregularity observed without strophanthin dosage. This varies between 25 and 200 minutes. On the other hand in Figure 13 in the last section, first five cross-rows there are corresponding time values of 175 to 360 minutes. At various experimental time points during these experiments 0.25 to 0.5 of strophanthin were given. Accordingly, a shortening of the duration of the irregularity cannot be established. Furthermore no improvement of the pulse or general condition was ever noted. Obviously these experiments are too few to rule out a possible favorable effect in all cases. Several hundred experiments would be necessary to obtain statistically reliable data on this point. And so, since contrary to animal experimentation, we could not unquestionably establish damage following intravenous strophanthin dosage, we may leave it to the treating physician whether or not he may still want to make an experiment with strophanthin. To be sure, such an employment of it must be advised against in case of a very much decelerated form of irregularity. This will be observed when there is the greatest danger; under such circumstances time should never be lost by experimenting with drugs, but every effort should be made in the direction of intensive heat therapy.

Also in the experiments with cardiazol, coramin and lobeline we restricted ourselves primarily to determining whether injurious effects occurred in the case of relatively large doses. Four cc. of 10 percent coramin as well as 2 cc. of 1 percent lobeline were injected intravenously at various stages of recovery without any marked objective and subjective deterioration of the state of the heart, the breathing, and the general condition. But just as with strophanthin, it is impossible to rule out a possible therapeutically favorable effect because of the small number of experiments. We never observed such an effect. Especially the marked deepening of breathing and of the irritability of the trigeminal nerve which usually sets in very suddenly after coramin (for example, sneezing immediately after the injection) were always missing. Contrary to strophanthin, in the case of which we cannot advise against experimentation by intravenous injection under certain conditions, we believe on theoretical grounds that such experiments with peripheral circulatory drugs which may heighten the vessel tonus are not indicated because of the following considerations: The damage to the heart is to be attributed, among other things, to an overloading, which is caused by a blocking of enlarged vessel areas, aside from an increase in viscosity. If the vessel tonus is further increased in the areas which have remained unimpeded, the conditions for the heart are thereby made worse.

The sceptical attitude toward the effect of drugs is strengthened above all by the observation that in the majority of the experiments in which no drugs were given, even the most severe disturbances of the peripheral circulation were reduced remarkably rapidly under intensive heat treatment. In this connection it must be emphasized that besides the recovery of body temperature through heat therapy an unloading of the heart takes place because the blocked areas open up. Contrary to earlier concepts, according to which there was danger of hemorrhage into the periphery during rapid rewarming, and according to which one sought to avoid this hemorrhage by wrapping up the extremities as well as by very slow warming, the “venalous bleeding into the periphery” may be life-saving under some circumstances. An exception, namely, loval pyperacmia after considerable rise in temperature and corresponding reestablishment of circulation has already been described in the reference to the danger in some cases of very prolonged treatment in the light cradle.