INHALATION of carbon oxide is now administered for various thera-

peutic and prophylactic purposes. The benefits afforded depend

chiefly on the part which carbon dioxide plays in the control of respi-

ration. Resuscitation from asphyxia of the newborn,1 prevention of

postoperative atelectasis and pneumonia,2 elimination of ethyl ether,2

of carbon monoxide,4 and of other anesthetic and toxic gases3 from the

lungs and blood, all alike depend upon the increase of pulmonary

ventilation under stimulation of the respiratory center by inhalation

of carbon dioxide.




The equally marked influence of carbon dioxide upon the circulation

has not as yet, or to an equal degree, been exploited for therapeutic

purposes. Yet, simultaneously with the modern development of res-

piration, physiology has produced also observations indicating a pow-

erful influence of carbon dioxide upon the heart. In a long series of

experiments,6 from fifteen to twenty-five years ago, my collaborators

and I demonstrated on dogs, under artificial respiration or breathing

naturally under a slight pressure of air after the opening of the tho-

rax, a condition verging on tetanus of the heart-in the physiological

sense of the word tetanus, that, is a fusion of successive beats.

This state was effectively counteracted and heartbeats of full

amplitude were restored as a consequence of the restoration of a normal,or per-

haps slightly excessive, carbon dioxide content in the blood. The les-

son of these observations is now generally applied in experiments upon

the isolated heart in the Starling7 heart-lung preparation. It is rec-

ognized that the blood for perfusion, in addition to being oxygenated,

must also be supplied with a sufficient amount of carbon dioxide to

overcome the tendency of an exposed or excised heart to develop an

inadequate diastolic relaxation. Otherwise it passes into a continuous

systolic state, an incomplete tetanus or cramp of the heart.


Following these observations upon the heart, it was shown in an-

other series of experiments” that carbon dioxide may exert an equally

strong influence upon the peripheral circulation, and particularly upon

the volume of the venous return to the right heart. Clinically this


*From the Laboratory of Applied Physiology, Sheffield Scientific School, Yale 



effect is best seen as a result of the inhalation of carbon dioxide after

a prolonged and extensive surgical operation under open ether anes-

thesia. The return of blood to the cutaneous vessels, the flushing of

the skin, the refilling of the veins previously collapsed, bear at least a

superficial similarity to the effects induced by inhalation of amyl

nitrite. But the effects, like those of a hot bath and in contrast to

amyl nitrite, are more physiological than pharmacological; they are

lasting and are accompanied by a restoration of a full strong heart

action and a recovery of normal arterial pressure. Never in my expe-

rience have there been any symptoms suggesting an overloading of the



Among the various treatments of heart disease, that at Nauheim is

the most celebrated. It consists in baths in carbonated water. The

good effects, which the treatment, is claimed to have, have never really

been explained. There is a stimulating action upon the skin, but there

is little ground for believing that a slight cutaneous hyperemia can by

itself be of much benefit. It is at least possible that the greater part

of the benefit sometimes derived from the Nauheim treatment is due

rather to inhalation of the carbon dioxide volatilizing from the surface

of the bath.




With these considerations as a physiological background, it has

seemed to me justifiable to try, with all due caution at first, the influ-

ence of carbon dioxide inhalation upon cases of angina pectoris which

are as yet in their earlier stages, but in which moderately severe suf-

fering on exertion is already developing. This is not an emergency

treatment, but a therapy for prolonged application. The inhalations

are not given during an attack of pain, but at regular times every

day, usually before the midday and evening meals, and at bedtime.

The patient lies quietly on his back for a few minutes holding over his

own face a mask which has a sufficiently large opening to the outside

air to offer no resistance to breathing. He is told to keep his mouth

open and to breathe deeply rather than rapidly. Then a stream of

carbon dioxide gas through a small rubber tube from a cylinder of the

pure liquefied substance is fed to the mask. At first the flow is kept

small, but as respiration gradually deepens the amount of the gas is

increased until at the end of two or three minutes a maximal or nearly

maximal depth of breathing is developed. The stimulation is not,

however, pushed to the point of increase of rate. This condition is

maintained for fifteen or twenty minutes continuously. Then the gas

is shut off, and the patient is directed to lie quiet for ten minutes more,

so as to avoid the slight giddiness which occurs if he gets up immedi-




It is to be noted that the technic of this inhalation differs markedly

from the use of a mixture of oxygen and 7 per cent carbon dioxide.

which is best employed for resuscitation from asphyxia. The inhala-

tion used on these heart eases is on the contrary essentially like that

applied by Henderson, Haggard and Coburn, and by White after anes-

thesia and operation.3 A mixture of oxygen and carbon dioxide is

rather expensive, and a cylinder of it, is exhausted in a single inhala-

tion; while on the contrary even a small cylinder of liquid carbon

dioxide lasts for several weeks of this treatment.. so that its cost,

aside from the control apparatus, is slight. But of course pure carbon

dioxide should be used only with such an open mask that the small

volume of the gas supplied is diluted by the patient’s breathing in the

relatively large volume of the inspired air.


The essentials of the control apparatus employed are (a) an open

mask, as above described, and (b) a device for the control of the flow

of the gas to the mask, so that by no possibility can the patient receive

an excess ; it must be fool proof.

For this purpose a water manometer only 100 to 120 mm. long is so

arranged that it not only serves as a gauge but also as a safety valve,

through which any excess supply of the gas blows off into the air

without going to the mask. It is merely a straight glass tube, 5 mm.

bore, which extends down for this distance below the surface of water

in a bottle, flask or test tube. Between this manometer and the tube

leading to the mask is a hole of not more than 1 mm. diameter. Thus

the maximum volume of gas that can go to the mask is that which

under a pressure of 100 to 120 mm. of water column will pass through

a 1 mm. hole. The supply of gas to the manometer and to this milli-



meter hole is controlled either by a finely adjustable needle valve, or

better by a McCaa reducing valve and a needle valve, as shown in

Fig. 1.* If only the needle valve is used, it has to be frequently re-

adjusted by the person administering the inhalation. If the reducing

valve is used also, the flow is so steady that an intelligent patient may

manage the entire procedure for himself. This is a distinct. advan-

tage in a treatment which the patient may find it necessary to con-

tinue one or more times a day for the remainder of his life. For the

utmost that can be expected, or even hoped, in regard to such a dis-

order as angina pectoris is not absolute cure, but, a check upon the

fundamental conditions, prevention of suffering and prolongation of



The effects of this inhalation on the patients thus far treated, as yet

only three in number and none with high arterial pressure, have been,

in addition to the deeper respiration, as follows: There is a distinct

improvement in the color and temperature of the lips and skin, previ-

ously rather bloodless or bluish and cold, but becoming warm and pink

under the influence of carbon dioxide upon the peripheral circulation.

Arterial pressure and the pulse rate are not in appreciable degree in-

creased, although a markedly fuller circulation is evident. The sen-

sation of oppression in the chest, and the pain or “pins and needles”

in the shoulder and arm is considerably decreased, and may cease alto-

gether for some hours after the inhalation. After some weeks of daily

inhalations the capacity to take such exercise as walking uphill is

markedly increased, and the chief difficulty is to prevent the patient

from overexerting his partially restored physical capacity.


This is all that it seems justifiable to report in an initial communi-

cation on a treatment which will certainly need far more evidence

before it can be regarded as of proved general therapeutic value. The

cases thus far treated have, however, appeared so much benefited and

the technic here described appears to be so safe, that it seems best to

put the matter on record, in order that others also may try it for a

class of patients for whom there is otherwise little that can be done to

relieve or even to delay the development of a peculiarly painful, anx-

ious and hopeless form of invalidism, with an ever-present risk of

sudden death from coronary thrombosis.


Mention may also be made here of the effects of this inhalation

upon two cases of intermittent claudication in the lower extremities.

In both of these cases a marked improvement in the circulation of the

ischemic limb was observed under the inhalation. and in one in which

the treatment was continued for some weeks there was a distinct

cumulative beneficial effect. These cases were studied particularly in

the belief that, if the pain in angina pectoris and that in intermittent.


*An apparatus of this type without the McCaa reducing valve may be obtained 

from the Foregger Company, 47 West 42nd Street, New York City. An apparatus 

with the reducing valve from the Mine Safety Appliances Company, Pittsburgh. Pa.




claudication are due to similar local reactions, improvement in both

types of cases would tend to support the probability that in both the

results are real and not imaginary, either in the patient, or in the mind

of him who has applied the treatment.




The general conception under which these observations were made

was as follows: In a normal person muscular exertion induces no

pain either in a limb or in the heart, for the reason that the blood sup-

ply is sufficient to afford all the oxygen needed initially in the working

parts. This supply of oxygen quickly converts a large part of the

fatigue products, especially lactic acid, into carbon dioxide. The car-

bon dioxide then induces a relaxation of the blood vessels and thus

increases the blood supply to the working parts, both heart and limbs.

This is the normal reaction to exercise. The healthy man takes a walk

or plays an athletic game to improve the oxygenation of his tissues.

Thus as Miescher” said forty-five years ago : “Carbon dioxide spreads

its protecting wings over the oxygen supply of the body.”


Quite different, is the reaction in a person in whose heart or limbs

the blood vessels are sclerotic or constantly contracted. The blood

supply and therefore the oxygen supply are insufficient for the initial

requirements of exertion. Lactic acid and other fatigue products ac-

cumulate; for in the absence of a large supply of oxygen they cannot

be burned to carbon dioxide. They tend to induce a cramp of the

musculature, cardiac or striated, in contrast to the influence of carbon

dioxide which, as I long ago demonstrated experimentally on the

heart, promotes relaxation.6 To do effective work a muscle must be

able to relax as well as to contract. From excess of lactic acid and

local deficiency of carbon dioxide come the abnormal reaction to exer-

cise, the ischemia and the cramp.


If now a patient who is liable to such an abnormal reaction receives

an inhalation of carbon dioxide, still a third form of reaction devel-

ops. He experiences the benefits, without the disadvantages, of physi-

cal exercise. He makes no exertion. His muscles are at rest, and his

heart is put, under no additional strain. There is no decrease of the

oxygen supply to any part, but rather an increase, for the carbon di-

oxide inhaled induces a relaxation of the finer blood vessels, a more

ample heartbeat, and a fuller circulation. The balance of supply and

demand for oxygen in the tissues is thus distinctly improved and the

tendency to cramp is diminished. Furthermore as the treatment is

repeated day after day the blood vessels and the heart muscle, under

the influence of an essentially normal physiological agent and an essen-

tially normal reaction, gradually acquire and retain a state of de-

creased habitual strain and more normal behavior. Along these lines


we may figure to ourselves why and how inhalations of carbon diox-

ide may exert a beneficial effect both immediate and to some degree


In support of this general conception mention may here be made

also of the extraordinary observations which were reported by the late

Dr. A. S. Loevenhart10 in which he found that inhalation of carbon

dioxide administered to cases of catatonia caused a temporary resto-

ration of mental responsiveness. The simplest explanation of the

results in these cases is to postulate an habitual contraction of blood

vessels in the brain of the catatonic patient and to assume that the

influence of carbon dioxide upon these vessels is similar to that upon

the peripheral circulation elsewhere in the body. The effects of over-

breathing in inducing, and of oxygen and carbon dioxide in tempo-

rarily inhibiting, fits in epilepties11 are also suggestive of a similar


Finally I have pleasure in acknowledging my indebtedness to my

colleague, Dr. George Blumer, for the opportunity to work on one of

the cases here reported, and to Dr. Samuel C. Harvey for two of the

others. Investigations on the experimental side of this general prob-

lem are now being published from Dr. Harvey’s laboratory, and fur-

ther investigations upon patients are to be conducted in the clinic




Daily inhalations of carbon dioxide appear to offer a possibility of

considerable amelioration of the crippling effects and suffering in cases

of angina pectoris and also of intermittent claudication.



  1. Henderson, Y.: The Prevention and Treatment of Asphyxia in the New-

Born, J. A. M. A. 90: 583, 1928.

Incomplete Dilatation of the Lungs as a Factor in Neonatal Mortality, J. A. M. A. 96: 495, 1931.


  1. Henderson, Y.: Acapnia as a Factor in Postoperative Shock, Atelectasis and

Pneumonia, J. A. M. A. 95: 572, 1930.


  1. Henderson, Y., Haggard, H. W., and Coburn, R. C.: The Therapeutic Use of

Carbon Dioxide After Anesthesia and Operation, J. A. M. A. 74: 783, 1920.

White, J. C.: Deetherization by Means of Carbon Dioxide Inhalations, Arch.

Surg. 7: 347, 1923.


  1. Henderson, Y.: The Dangers of Carbon Monoxide Poisoning and Measures to

Lessen These Dangers, J. A. M. A. 94: 179, 1930.


  1. Henderson, Y., and Haggard, H. W.: Respiration Influencing Their Action. Noxious Gases and the Principles of

American Chemical Society Monograph Series, New York City, 1927, The Chemical Catalog Company.


  1. Henderson, Y.: Acapnia and Shock. I. Carbon Dioxide as a Factor in the Regulation of the

Heart Rate, Am. J. Physiol. 21: 126, 1908.


  1. Starling, E. H.: Linacre Lecture on the Heart, London, 1918. Also Knowlton,
  1. P., and Starling, E. H.: The Influence of Variations in Temperature and

Blood Pressure on the Performance of the Isolated Mammalian Heart, J.

Physiol. 44: 206. 1912. Also Patterson, S. W., and Starling. E. H.: On

the Mechanical Factors Which Determine the Ouput of the Ventricles, J.

Physiol. 48: 357, 1914; with Piper, H. : The Regulation of the Heartbeat, J. Physiol. 48: 465, 1914.


  1. Henderson, Y., and Harvey, S. C.: VIII. The Venopressor Mechanism, Am. J.

Physiol. 46: 363, 1918.

Bryant, J., and Henderson, Y. : Closed Ether and a Color Sign, J. A. M. A.

65: 1, 1915.

  1. Miescher, F.: Bemerkungen zur Lehrc van den Athembewegungen, Archiv. f.

Anat. u. Physiol. 1885. Physiol. Abtheilung, p. 355. Republished in Die

Histochemischen u. physiologischen. Arbeiten von Friedrich Miescher. Verlag

  1. C. W. Vogel, Leipzig, 1897.
  1. Loevenhart, A. S., Lorenz, W. F., and Walter, R. M.: Cerebral Stimulation,
  1. A. M. A. 92: 880, 1929.
  1. Lennox, W. G., and Cobb, S.: Epilepsy, Medicine 7: 162, 1928.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top