U.S. flag

An official website of the United States government

Skip to main content
Return to topReturn to top

Battle Casualties in Korea: Studies of the Surgical Research Team Volume III

Experiences With Intra-Arterial and Rapid Intravenous Transfusions in a Forward Surgical Hospital*

Lieutenant Colonel Curtis P. Artz, MC, USA
Captain Yoshio Sako, MC, USA
Captain Alvin W. Bronwell, MC,USA

Early in the Korean conflict, intra-arterial blood transfusion was used in the resuscitation of hundreds of men critically injured in battle. The resulting impressions of the many surgeons in the theater varied so widely that, after an extensive clinical experience, we initiated a clinical evaluation of transfusion by the intra-arterial route in an attempt to outline the possible indications for the procedure. During the early phase of the study, every patient with severe oligemic shock received an intra-arterial transfusion. Blood was usually transfused into the femoral artery through a 15- or 17-gauge needle. The femoral artery was surgically isolated and the surgeon held the needle in the artery during the administration of the blood. A Rochester plastic needle was used in two patients after cutting down on the femoral artery, and in one patient blood was given directly into the aorta through a 17-gauge needle during a laparotomy.

The technic of surgically isolating the femoral artery and holding the needle in place during the rapid infusion of blood was feasible. Minimal amounts of blood were given simultaneously by intravenous infusion. The maximum amount of blood normally given intra-arterially was 3,000 cc. although in one patient who received 4,500 cc., intra-arterial transfusion was discontinued as soon as the patient`s systolic pressure reached 100 and further replacement therapy was given intravenously. No complications occurred from this therapy.  When the needle was removed from the artery, any bleeding was readily controlled with a small piece of gelfoam and gentle pressure. The large-gauge needle, because of its larger lumen, was preferable to the Rochester plastic needle. Unless the surgeon steadied the needle during the pressure transfusion, it would impinge on the wall of the artery and a free flow of blood would be impeded.


After seven patients in severe oligemic shock had been resuscitated by intravenous blood transfusion through multiple veins, the data were

 *Previously published in U. S. Armed Forces Medical Journal 6: 313, 1955.


Table1.  Patients Receiving Intra-Arterial Blood Transfusions


Table2.  Patients Receiving Rapid Intravenous Blood Transfusions


assembled. Six patients had received intra-arterial transfusions and seven had received rapid intravenous infusions. In the clinical evaluation of the patients in the two groups, the degree of shock was comparable. Four patients in the former group and only one in the latter group were admitted with an unobtainable blood pressure. The degree of injury and amount of hemorrhage, however, was greater in those receiving blood intravenously, and the amount of blood required for resuscitation was slightly higher. The data on these patients are summarized in Tables 1 and 2.

There was no appreciable difference in the rate at which blood was given to patients; via artery it was 88 cc. per minute and via vein, 70 cc. per minute. No attempt was made to achieve an absolute maximum rate of infusion by either method. These data represented the actual accomplishments of blood replacement at one forward surgical hospital in a small, comparable series of patients in deep oligemic shock. In one moribund patient not included in this series, a definite attempt was made to determine how rapidly blood could be infused by the intravenous route. In 30 minutes 5,500 cc. of blood were injected into two veins through 15-gauge needles; 3,500 cc. of the blood were pumped into one vein in 21 minutes. This demonstrated that blood can be given very rapidly by the intravenous route.

After the controlled use of intra-arterial transfusion in six patients, it was our belief that the patients did not show any appreciably improved response as compared with patients who received blood at a comparable rate through multiple, intravenous routes.


Proponents of the belief that intra-arterial transfusion is superior to intravenous transfusion in treatment of hemorrhagic shock have cited the following reasons to account for its supposed superiority: (1)  blood given intra-arterially increases the coronary and cerebral arterial flow; (2) direct infusion into the arterial tree causes an instantaneous rise due to the simple hydrostatic effect; (3) blood given intravenously would tend to pool, whereas intra-arterial blood readily mixes; and (4) blood can be administered more rapidly by the intra-arterial route.1 These four reasons are given as prime factors in the superiority of intra-arterial transfusion over intravenous infusion. At the time most of these statements were made, there was no experimental or clinical evidence to support their contentions.

Recently a number of well-designed experiments were carried out in an attempt to evaluate the role of some of the above-mentioned factors. Kohlstaedt and Page,1 and Case and associates,2 studied the effects on coronary flow and arterial pressure following intra-arterial and intravenous infusion of equal magnitude in dogs in oligemic


shock for varying periods of time. They were able to show in repeated experiments that coronary flow and arterial pressures responded just as rapidly and to the same extent with either intravenous or intra-arterial blood transfusions. In addition, the right and left auricular and pulmonary artery pressures were measured and found to be the same during both types of transfusion. Their conclusion was that their data did not constitute a contraindication to intra-arterial blood transfusion, but that there was no convincing evidence that. it was superior to intravenous blood transfusion.

Patients in shock may not have a discernible vein suitable for giving infusions, but if by surgical means the same search is made to locate a vein as an artery, it could be found. Moreover, when necessary, the femoral vein is always available for cannulation. The statement that blood can be administered more rapidly by the intra-arterial than by the intravenous route is not true. In summarizing their findings, Case and associates stated that the rate of administration is a function of the pressure in the bottle, pressure in the vessel, resistance to the flow of blood in the needle and tubing, and the viscosity of blood. Under these circumstances, blood can be administered more rapidly through the intravenous route.

Another statement frequently made is that blood administered by the intravenous route will produce cardiac failure. There is evidence to show that. this is so only after the blood pressure has returned to normal or when there is an underlying cardiac disease. The contention that direct infusion into the artery at rates as currently practiced will cause a rise in pressure due to hydrostatic effect has not been substantiated.

Maloney and co-workersgave rapid infusions of blood, both intravenously and intra-arterially, and measured cardiac output by the blue-dye injection method. They also measured mean arterial pressure on a series of normal dogs and on dogs in shock due to hemorrhage; they were unable to demonstrate any difference in the rise in arterial pressure by transfusion through either route. Case and associates cited other experimental work along the same line of investigation indicating that an animal in oligemic shock did not respond more favorably to intra-arterial blood transfusion than to intravenous blood transfusion. The volumes of blood concerned when transposed from the rate of a dog to that of a man were in the realm of 445 cc. per minute; this rate is much higher than the rates given in a clinical operation.

Richards and Hansenalso studied the comparative action of intraarterial and intravenous transfusion in the treatment of oligemic shock. They found no demonstrable advantage of intra-arterial over intravenous transfusion when equal infusion rates were used. They stated that rapid intravenous transfusion was well tolerated, particu-


larly when vessels were in a collapsed state due to hemorrhage. They concluded that this was logical because the blood flow to and from the heart per minute is many times the volume of blood which can be forced into the circulation by any type of pressure of transfusion in a similar length of time. Prior to their experimental work, statements regarding the advantages of intra-arterial transfusion over the intravenous route were made without being substantiated by clinical or laboratory experiments.

A situation in which intra-arterial transfusion may be indicated is in the resuscitation of a patient whose heart is in asystole because, once the output of blood from the left ventricle has ceased, there is no other way for the blood to get into the arterial tree. Another situation would be the sudden need for massive transfusion during an operation for mitral stenosis to get blood into the arterial tree. Except for these two unusual circumstances, most of us believe that there are certain disadvantages associated with the process of intra-arterial transfusion. Cases have been reported of arterial insufficiency necessitating amputation of an extremity, tissue necrosis, arterial spasm caused by arterial cannulation, and in some instances a delay in getting the blood infused because f additional time required to begin the intra-arterial transfusion. A major criticism by most investigators on the work of those who present the advantages of intra-arterial transfusion over intravenous transfusion is that their clinical as well as experimental investigations have not dealt with comparable rates of infusion of blood.24   In some instances conclusions were based on patients in whom four times as much blood was given by the intra-arterial route as was given to others by the intravenous route.

Because of the impressions gained from patients in whom blood was given primarily by the intra-arterial route coincided clinically with the above experimental data, intra-arterial transfusions were discontinued in favor of rapid intravenous infusion of blood through multiple, large-gauge needles or intravenous cannulae.


No definite conclusions can be drawn from this very limited experience. Our impressions were that blood given by the intra-arterial route was of no more value in the resuscitation of patients in deep oligemic shock than was blood administered by the intravenous route, if it was given at the same rate. Observations in one forward surgical hospital showed that, in a small series of severely wounded patients of comparable severity, blood replacement was accomplished as rapidly by multiple intravenous routes as it was by the intra-arterial route.



The clinical evaluation of intra-arterial blood transfusions administered to six patients who were wounded in combatwas compared to the results obtained by rapid intravenous transfusions given to seven men who were also combat casualties. The average rate ofblood given intra-arterially was 88 cc. per minute, and intravenously,70 cc. per minute. These were actual rates at which blood was given andnot the maximum rates possible. There was no discernible difference in the rate of response between the patients who received blood intra-arteriallyand intravenously.


1. Kohlstaedt, K. G., and Page, I.H: Hemorrhagic Hypotension and Its Treatment by Intra-arterial and Intravenous Infusion of Blood. Arch. Surg. 47:  178, 1943.
2. Case, R. B.; Sarnoff, S. J.; Waithe, P. E., and Sarnoff, L. C.: Intra-arterial and Intravenous Blood Infusion in Hemorrhagic Shock. J. A. M. A. 152: 208 (May 16), 1953.
3. Maloney, J. V., Jr.; Smythe, C. McC.; Gilmore, J. P., and Handford, S. W.: Intra-arterial and Intravenous Transfusion: Controlled Study of Their Effectiveness in Treatment of Experimental Hemorrhagic Shock. Surg. Gynec. & Obst. 97: 529, 1953.
4. Richards, R. C., and Hansen, F. L.: Comparison of Intra-arterial and Intravenous Transfusion in Treatment of Hemorrhagic Shock. Surgical Forum., Proceedings of the Forum Sessions, pp. 478-484, Thirty-ninth Clinical Congress of the American College of Surgeons, Chicago, Illinois, 1953.  W. B. Saunders Co., Philadelphia, Pa., 1954.
5. Ziperman, H. H.: Personal communication.