U.S. flag

An official website of the United States government

Skip to main content
Return to topReturn to top



AcuteBattle-Incurred Arterial Injuries

Michael E. DeBakey, M. D, *and Fiorindo A. Simeone, M. D.**

  Acute injuries of the major arteries have always constituted one of the most serious problems of military surgery. They are associated with a high immediate death rate and not infrequently result in permanent disability. It is curious, therefore, that so little information is available concerning them.

When the United States became actively involved in World War II, the majority of medical officers knew practically nothing about vascular injuries.  Few factual data were available because this type of injury occurs infrequently in civilian practice and because the recorded experience of previous wars was meager. In the official American history of World War I are contained data on the gross incidence of arterial wounds 1 and brief general statements regarding the management of the injured artery and vein,2 but no information is provided regarding the results of arterial wounds in relation to the characteristics of the injury and the type of treatment. Very few medical officers were familiar with the British experience concerning arterial injuries in World War I as recorded by Makins. 3

  In the years which elapsed between World War I and World War II numerous advances were made in vascular surgery, but they were confined to the small groups of surgeons who were interested in this field and who developed new methods or applied to it developments which had been made in other fields. The principles of debridement were clarified and standardized. Chemotherapeutic agents of great effectiveness were introduced, as were vasodilating and anticoagulant agents. Sympathectomy had become a recognized therapeutic procedure in selected cases of vascular insufficiency. Yet in spite of these advances the results secured in arterial injuries in World War II, at least in the first months of American participation, were no better than those

 *  Professor and Chairman, Department of Surgery, Baylor University College of Medicine.  Formerly Colonel, MC, AUS.

** Professor of Surgery, Western ReserveUniversity School of Medicine. Formerly Colonel, MC, AUS.

1 Love, Albert G.:  Statistics.  The Medical Department of the United States Army in the World War. Washington, Government Printing Office, 1927, vol 11, pt 1, p. 57-85.

2  Pool, Eugene H.:  Wounds of soft parts.  In The Medical Department of the United States Army in the World War. Washington, Government Printing Office, 1927, vol 11, pt 1, p. 302-303.

3 Makins, G. H.: Injuries to the blood vessels.  In Official History of the Great War Medical Services. Surgery of the War London, His Majesty's Stationery Office, 1922, vol 2, p. 170-296.


reported by Makins in World War I and in somerespects were actually worse. This anomaloussituation is explained, at least in part, by the general ignorance ofmedical officers concerning theprinciples of management of vascular injuries.

  It is only fair to point out that the unhappy experiences of American surgeons with arterial injuries were not unique. In a conference of British surgeons held in Cairo in August 1943, Ogilvie4 reported that in the course of the war he had observed no instance of ligation of the popliteal artery not followed by gangrene.  Blackburn 5 in 1944 made substantially the same statement. British experiences in World War II thus did not support the observations of Makins in World War I.  He had reported that amputation was necessary in only 43.1 percent of 144 casualties with wounds of this artery.

  When medical units were activated in the United States after the attack on Pearl Harbor much of their basic training was devoted to such matters as the use of the Lyster bag, the construction and disposition of slit trenches, map reading, military protocol, and other nonmedical subjects.  Medical instruction was limited chiefly to hospital administration.  It included very little concerning the management of combat-incurred wounds.  Medical officers learned practically nothing of the American medical experiences of World War I as they were recorded in the official history of that war or of the valuable experiences of Allied medical personnel in the early years of World War II.  As a result, American medical officers were sent to the various fronts with a certain amount of information concerning court-martial proceedings, sanitation, and similar matters, but with almost no factual information upon the pathology and therapy of war wounds.

  The general ignorance concerning the management of combat-incurred vascular injuries and the complacency felt concerning them by those who were familiar with Makins' observations and conclusions were in no wise altered by the appearance, in 1943, of a Military Surgical Manual 6 containing a section on vascular injuries. Valuable as was some of the information it contained, other material in this manual was quite misleading. In one table, for instance, the source of which was not stated, the expected percentage of gangrene or other disturbance was set forth for sudden ligation of various arteries. Ligation of the aorta was stated in this table to be associated with a 100-percent incidence of gangrene, and that of the common iliac artery and of the posterior tibial artery (when both bones of the leg were fractured additionally) with an incidence of 50 percent. But ligation of the popliteal artery with its concomitant vein, according to this table, was never followed by gangrene. The figures presented in the table, the text stated, ". . . are rather out-of-date. Today, by simultaneous division of the companion vein and by performing

Ogilvie, W. H.:  War surgery in Africa.  Brit. J. Surg. 31: 313-324, Apr 1944.

5  Blackburn, G. :   Surgery in the field. Lancet 1: 361-365, 18 Mar 44.

6  National Research Council, Division of Medical Sciences. Burns, Shock, Wound Healing and Vascular Injuries.  Prepared under the auspices of the Committee on Surgery of the Division of Medical Sciences of the National Research Council.  Military Surgical Manuals, vol 5. Philadelphia, W. B. Saunders Co., 1943.


sympathetic procaine block, several times ifnecessary, to avoid vasoconstriction and toencourage the collateral circulation, the chances of gangrene are muchsmaller."

Later in the text, in a detailed discussion of ligation of the popliteal artery, the following statements were made: "The artery can be doubly ligated and divided with impunity in the upper part of its course but with less safety in its lower portion, close to its division.  Apparently the most danger is encountered in those cases in which hematomas have been present for many hours or days before treatment is instituted. Among such, gangrene of the foot is rather common."

  Medical officers who relied upon these data were naturally disappointed when their results did not approximate those given. Furthermore, in some series of arterial wounds observed during the Tunisian campaign, gangrene followed injury to the popliteal artery in almost all cases; a finding in direct contrast to the results cited in the manual which was in press at the time.

Constant attention to details of wound management, constant dissemination of information, repeated practical demonstrations of techniques by the surgical consultants and their assistants in the Mediterranean and European Theaters of Operations were eventually supplemented by information from the Office of The Surgeon General.  As the concept of arterial injuries became more nearly correct there was some corresponding improvement in therapeutic results. That the improvement was not more pronounced is an index of the essential seriousness of this type of injury.  Toward the close of the war the amputation rate in interruptions of the popliteal artery, for instance, was closer to 75 percent than to the 100 percent it had been in the beginning, but it cannot truthfully be said that the results in this or in most other acute arterial injuries were ever really satisfactory.


  The incidence of acute arterial injuries is admittedly low in all recorded wars, but such statistics as are available generally seem to underestimate their frequency.  Even in World War II, when a real effort was made to report the figures accurately, the data were still deficient in certain respects. This lack of information is not difficult to explain:  For obvious reasons, vascular injury is seldom recorded as a primary diagnosis and in many instances, including both the most serious and the least serious cases, it is probably not recorded at all.  In many reports, cases which came to amputation as the result of such complications of arterial wounds as life-endangering infection, such as gas gangrene, are not recorded as arterial injuries.  In addition, analysis of the available statistics requires considerable caution.  Most series are actually or relatively small, and many from previous wars include both acute and chronic conditions.  In drawing conclusions from them, therefore, many qualifying circumstances must be taken into account.



  Available statistics (Chart 2) suggest that the incidence of arterial injuries among battle casualties of previous wars was extremely low.  It was approxiinately 0.29 percent in the Civil War, 7  and 0.8 percent, according to LaGarde, in the 1,400 injuries sustained at Santiago during the Spanish-American War.8   In the Russo-Japanese War,9 a few years later, it was reported as 2.4 percent.

Chart 2. Incidence of arterial wounds among battle casualties in various wars.  Nature and source of these data indiated in text, pp. 63-65.

Otis, George A., and Huntington, D. L.:  The Medical and Surgical History of the War of the Rebellion. Surgical History. Washington, Government Printing Office, 1883, vo1 2, pt 3, pp. 763, 765.   In the official history are recorded 485 cases of primary lesions of blood vessels out of 2,235 cases 'in which the bleeding vessels were definitely ascertained.'   It should be understood that some of these may have been cases of "secondary hemorrhage" in which the artery may not have been directly wounded.  The incidence reported by DeBakey and Simeone in Annals of Surgery 123: 534-579, Apr 1946, omitted such possible cases.  On the assumption that the same proportion held for the remaining 1,010 cases in which the blood vessels involved were not named, the 485 would have been expanded to 704.

8 LaGarde, Louis A.:  Gunshot Injuries:  How They Are Inflicted, Their Complications and Treatment.New York, William Wood & Co., 1916.

  9 Macpherson, W. G.:  Report 14.  Port Arthur: The Russian Red Cross Society's Hospital: Statistical Analysis of 2,080 Wounds Treated During the Siege.  The Russo-Japanese War. London, His Majesty's Stationery Office, 1908, p. 281-291.


  Despite the lack of clinical discussion of vascular wounds in the official American history of World War 1, the available information indicates that the incidence of this type of wound among American troops was about 0.4 percent of battle casualties.10 The British official history,11 in spite of the extensive discussion of the subject which it contains, supplies no data concerning incidence and Makin's classic monograph12  is no more helpful.  His material, in fact, is quite misleading unless it is remembered, as will be emphasized later, that it includes nonacute as well as acute injuries.  It is limited to 1,191 cases and consists of 2 series.  One comprises 668 cases handled by numerous individual surgeons in the British Isles and overseas and obviously includes nonacute as well as acute injuries.  The other consists of 523 cases observed in France by a single surgeon. Makins himself warned that only modified reliance could be placed on any deductions drawn from this material. Almost the only data available on the subject of arterial injuries per se among British troops in World War I are in Bowlby and Wallace's report 13 of 277 ligations of major arteries (1.3 percent) in 20,589 casualties treated at a single casualty clearing station.

According to Maurer,l4 443 wounds of the blood vessels, 17 percent of which involved only the veins, were observed among 8,000 wounded treated in a French ambulance service in World War I.  When the wounds which are exclusively venous are excluded, the incidence of arterial injuries in this series was 4.6 percent. Mignon's figures 15 for 6 French ambulances (so stated that a tabulation is possible) show 399 vascular injuries, 2.0 percent, among 19,734 wounded. The French official history of World War I 16  lists 6,397 vascular injuries (0.3 percent) among 2,052,984 casualties. The German incidence, according to Franz, 17 was three times as great, 0.99 percent.

  Matas,18  in his treatise on vascular injuries, stated that 24.7 percent of all battle casualties treated in forward hospitals had injuries of the blood vessels and that 2 percent of all wounded admitted to base hospitals presented traumatic aneurysms. He does not indicate the source material for this incidence, which is the highest recorded, nor does be identify the casualties involved.

World War II

  Statistics for World War II are considerably more complete. Figures for some of the armies during certain portions of the fighting in Italy, in Western

10  See footnote 1, p. 60.

11  See footnote 3, p. 60.

12  Makins, G. H.: On Gunshot Injuries to the Blood Vessels.  Bristol, John Wright & Sons, Ltd., 1919.

13  Bowlby, A., and Wallace, C.:  The development of British surgery at the front. Brit. M. J. 1: 705-721, 2 Jun 17.

14  Maurer, A.:  Wounds of the vessels of the face, neck and limbs; their surgical treatment, based upon four hundred and forty-three cases. Internat.  Clin. 1: 164-195, 1921.

15  Mignon, A.: Le Service de Sant? pendant la Guerre, 1914-1918.  Paris, Masson & Cie., 1926, Tome 2.

16  Minist?re de la Guerre.  Direction du Service de Sant?.  ?tude de Statistique Chirurgicale Guerre de 1914-1918.  Paris, Imprim?rie Nationale, 1924.

17  Franz, Carl: Lehrbuch der Kriegschirurgie.  Dritte Auflage.  Berlin, Springer-Verlag, 1942.

18 Matas, R.: Endo-aneurismorrhaphy.  I. Statistics of operation of endo-aneurismorrhaphy.  II. Personal experiences and observations on the treatment of arteriovenous aneurisms by the intrasaccular method of suture; with special reference to the transvenous route. Surg., Gynec. & Obst. 30: 456-459, May 1920.


Europe, and in the Pacific and China-Burma-India theaters, reveal 1,570 vascular injuries among 163,980 battle casualties (Chart 3).19  At the time these data were analyzed, figures from First and Ninth Armies were not available.  Experience cited for Third, Fifth, and Seventh Armies is that for which break down of arterial injuries was available. The annual reports of the surgeons of those armies indicate that the total numbers of battle casualty admissions to hospital and quarters during the entire war were 91,000, 73,000, and 84,000, respectively. The total incidence of 0.96 percent in the samples is remarkably close to the incidences reported from the separate theaters and armies. Similar correspondences noted in the distribution of vascular wounds (1,564) among wounds of the extremities (108,405) observed in the separate theaters and armies suggest that the overall incidence of 1.4 percent is fairly representative (Chart 4).

Chart3. Incidence of arterial wounds among American battle casualties inWorld War II.  Figures are based on samples as available and arenot to be considered as complete. See text, pp. 64-65, for details ofselectivity. 

10 Preliminary data based on sample tabulations of individual medical records indicate that during 1944 and 1945 (the only years for which such data are presently available), in the total Army, there were 12,600 battle wound and injury admissions with artery involvement, or about 1.7 percent of the approximately 760,000 battle wound and injury admissions to hospital and quarters.  Expansion of the data to include arterial injuries among battle and nonbattle wounds and injuries produces a percent of 0.8 for the 2-year period.


Chart 4.  Incidence of arterial wounds among wounds of extremities in American battle casualties in World War II. Figures are based on samples as available and are not to be considered as complete.  See ext, pp. 64-65, for details of selectivity.

  As these statistics show, the total incidence of vascular wounds among all casualties has been relatively and absolutely small in all wars.  Even the large numbers of wounded in World War II do not make the absolute number of vascular wounds very large. The significance of the figures lies not in the actual incidence but in the threat which these wounds carry to life and limb. Only by examining the frequency with which these injuries appear as the indication for amputation will their importance be realized. (Chart 5.)

  Incidence in Relation to Amputation.  No significant statistics compiled from this point of view seem to have been recorded, but accurate data for World War II are available from the Mediterranean and European Theaters of Operations (Chart 5).  Among 3,177 major amputations performed in these theaters, 2,179 (68.6 percent) were the result of extensive trauma, 380 (11.9 percent) resulted from clostridial myositis or other serious infections, and 618 (19.5 percent) followed major arterial injuries.  Records which became available with the capture of a German amputation center showed that among 1,359 major amputations, 64.3 percent were the result of trauma, which is close to the American incidence of 68.6 percent. Thereafter the German figures are not in


Chart 5.  Causes of amputation among Ameridan battle casualties in World War II.  Figures are based on samples as available and are not to be considered as complete.  See text, pp. 64-65, for details of selectivity.

agreement with the American; 29.7 percentof the German amputations were the result ofclostridial myositis and only 6 percent followed vascularinjuries. Corresponding figures on theRussian experience reported by Kramarov 20showed trauma to be thecause of amputation in 16percent of the cases, gas gangrene and other infections in 79 percent,and vascular injury in 5percent. The striking difference between the incidence of infection asa cause for amputation inthe American series and in the other series may be interpreted asreflecting the excellence of thesurgery done by American surgeons in forward areas. The percentages ofamputations because ofvascular trauma provide a true concept of the magnitude of the problemand also indicate theinevitability of the great majority of amputations for these injuries.

Regional Distribution.   An analysis of the relative frequency with which various arteries are involved in battle casualties (Chart 6, Table 3) shows that the brachial, tibial, femoral, and popliteal arteries are involved far more often

20  Kramarov:  Quoted by Aird, I.:  A review of recent Russian studies of gunshot wounds of the thigh.  Bull. War Med. 4: 619-620, Jul 1944.


than any others. They accounted for 70percent of the total vascular injuries reported by Makins21 in World War I and for 85 percent ofthose sustained byAmerican troops in 3 overseastheaters of operations in World War II. The explanation of thispreponderance is simple:Casualties suffering from injuries of larger arteries frequently didnot live long enough to reachmedical installations and therefore do not appear under precisediagnoses in medical statistics.Furthermore, injuries of less critical arteries were frequently notdiscovered or, if discovered,were not recorded.

A comparison of the frequency of wounds of the brachial, tibial, femoral, and popliteal arteries in the combined British and American statistics for World War I and in the American statistics for World War II shows that the incidence of wounds of all four arteries as a group was about the same in World War

Chart 6. Distribution of arterial wounds according to anatomic location in wounds of the extremities in various wars. Figures are based on samples as available and are not to be considered as complete. See text, pp. 63-65, for details of selectivity. Incidences in this chart, it should be emphasized, are based only on arterial wounds of the extremities, whereas those shown in Table 3 are based upon all arterial wounds. Arteries of the head and neck and renal arteries have been excluded.

21   See footnote 3, p. 60.



II as in the previous war.However, in WorldWar II the incidence of wounds of the poplitealartery was higher than the incidence recorded previously. The reasonfor these differences canonly be surmised. Methods of recording were probably more exact inWorld War II. Certainly inthis war there was a greater


and more selective interest in vascularinjuries than had been exhibited in previous wars. Finally,in World War II there was greater emphasis on conservation of limbsthan in other wars ofhistory.


NorthAfrican and Mediterranean Theaters of Operations

  In the North African and Mediterranean Theaters of Operations specific directives were not issued for the treatment of vascular injuries, but no opportunity was lost to disseminate information concerning such injuries.22  Case teaching and technical demonstrations were also carried out by the surgical consultant for the theater and by his assistants on their visits to both forward and base hospitals. The following principles were emphasized:

1. Careful debridement in all cases with excision of all devitalized tissue and preservation of all undamaged vessels however small.

  2. Adequate incision to permit clear visualization of all structures implicated in the wound and to avoid out-of-sight manipulations.

  3. Exploration of all major vessels whenever there was evidence of circulatory insufficiency in an extremity whether the artery lay directly in the track of the missile or at some distance from it.

  4. Procaine block of the sympathetic nerve supply to the extremity to relieve spasm of the major vessels whenever it was suspected, or to improve the collateral circulation to the limb when the main arterial supply had to be ligated or manipulated.

  5. Lumbar ganglionectomy for serious lesions of the major arteries of the lower extremities whenever conditions permitted this procedure. For injuries of vessels of the upper extremity sympathetic denervation other than by means of procaine block was not recommended. This operation is too specialized to warrant its performance in the field, and the incidence of gangrene following interruption of major arteries of the upper extremity was so much less than the incidence in the lower extremity that it was not regarded as necessary.

  6. Suturing the arterial lesion whenever conditions were favorable.

  7. Arteriectomy, thrombectomy, and other operations according to the indications of the individual case.

  Part of the teaching, in circular letters and in person, was concerned with methods for controlling hemorrhage. Warnings were issued against the use of tourniquets without special precautions, and the dangers of gauze packs were also emphasized. If a pack was necessary, the desideratum was that it be removed at the earliest opportunity and that a notation of its use be made

22  (1) Cir Ltr 13, Office of the Surgeon, NATOUSA, 15 May 43, sub:  Memo on forward surgery. HD.

(2) Cir Ltr 46, Office of the Surgeon, NATOUSA, 29 Aug 44, sub:  Surgical management of the wounded (111). HD.

(3)  Cir Ltr 8, Office of the Surgeon, MTOUSA, 10 Mar 45, sub:  Notes on care of battle casualties.HD.


on the soldier's record. The preferablemethod of controlling hemorrhage from major vessels wasstitch ligature.

In the hope that an adequate collateral circulation would develop, early operation for aneurysm or arteriovenous fistula was not recommended. The only justification for early operation in this overseas theater was the arrest of hemorrhage or some similar emergency. A deep hematoma, however, which was impeding the circulation, was to be dealt with immediately.

EuropeanTheater of Operations

  In the Manual of Therapy 23 prepared before D-day, for the European Theater of Operations, the following instructions were given for the management of wounds of the large blood vessels:


  1. Diagnosis is made by bright blood escaping from wound or an increasingly tense hematoma.

  2. Remember that the skin need not be broken in order for an injury to an artery from a contusion, a fracture or a dislocation to be present.


1. Control hemorrhage in cases in which blood is escaping from the wound by any of the following methods:

a.  Elevation of the part. This should be done with due care of associated fractures, regardless of the local methods used.

b.  Direct pressure on the wound with a bulky dressing.

c.  Secure the bleeding vessel with a clamp and ligature, and ligate the companion vein. This is only to be done when the open vessel can be easily seen in the wound, lest blind clamping in the depths of the wound cause nerve injury.

d.  Direct control of the artery by digital pressure at the root of the limb.

e.  Ligation of the artery and companion vein at a distance from the wound.

f.  The use of a tourniquet is to be discouraged except as a temporary emergency measure to facilitate control by other means, i. e., ligature, packing, and pressure dressing. Evacuation with a tourniquet in place means the almost certain loss of the extremity distal to the tourniquet. However, during evacuation, an unapplied tourniquet should accompany patients with injuries to major blood vessels.

  2.  In cases in which the bleeding is chiefly subcutaneous, local pressure over the hematoma is the only emergency local measure indicated.

  3.  Cases with severe bleeding into fascial planes of the extremity have a greater chance of gangrene than those in whom the blood can escape.  Priority evacuation is indicated in these cases.

  4.  Replace lost blood, whether external or into tissues, with plasma and blood transfusion.

  5.  For transport, keep the body warm but apply no artificial heat to the limb.

  6.  Surgical technique for the ligation of important arteries. The following technical steps should be employed in approaching individualblood vessels. The incisions depicted are thosesuitable for reaching the vessels at a point where they are undamaged.Where the approach isthrough a wound directly onto the lesion in the vessel,incisionsshould be much larger than thosehere shown, in order to give enough exposure for adequate control ofhemorrhage. The decision as to the site for ligation is best made by constriction of the arteryin the wound with observationof the color, pulse, and temperature in the hand or foot duringconstriction. This is safer than relying on a constant pattern of collateral circulation in a system sosubject to anomalies as thearterial tree.

23  Manual of Therapy, European Theater of Operations, 5 May 44.


  The Manual of Therapy 24  for the European Theater of Operations also set forth the following general principles for the definitive surgical treatment of vascular injuries:

  1. Cases of large increasing hematoma not controlled by pressure should be explored. Evacuate the hematoma and ligate the involved vessel.

  2. Divide the artery between two ligatures and, for additional security, transfix both stumps with a ligature. Division of the artery permits more accurate and secure ligation as well as providing the beneficial effects of periarterial sympathectomy to the extremity. When ligation of a major artery is required, the companion vein should be ligated as well, since this decreases the likelihood of gangrene.

  3. If the artery has not been opened, but is contused, and thrombosis has taken place, it should be ligated and divided, and the exposed, thrombosed portion excised, and the same procedure performed on the companion vein.

  4. When approach to the artery through the wound appears too difficult, expose the artery believed to have been injured so far proximal to the point of injury as to secure a clean, dry field. For temporary arrest of the hemorrhage, pass a small rubber tubing or tape around the vessel.

  5. A pulsating hematoma increasing in size and progressively interfering with the circulation of the extremity distal to the hematoma, requires surgical intervention.

  6. Cases of arteriovenous fistulas are rarely detected in the first day or two after wounding. They are not per se an indication for operation and are better left until a later date unless there is a complication, such as leakage or deterioration of the arterial supply to the limb.

  7. Supportive treatment for the patient in whom there is anxiety about the circulation to a limb should be as follows:

  a. Heat the individual's body and the root of the injured limb, protect the limb from heat loss by woolen coverings, if possible, but never heat the injured limb itself.

  b. Perform paravertebral novocain block, if feasible, to overcome spasm of the collateral blood supply. This procedure to be repeated every 12 to 24 hours if a beneficial response is obtained.

  c. If gangrene threatens, keep limb wrappedin sterile towel and observe closely for the possibledevelopment of gas gangrene.


  Any discussion of the therapy of vascular injuries must begin with the premise that the restoration of the flow of blood through the original channel is the desideratum. Unfortunately, even in civilian practice this can be accomplished in only a limited number of cases, while in military surgery, for a variety of reasons, this number is still more limited. These reasons, which have been present in all the wars of history, may be divided into two categories:  Those in which the factors are of such vital significance that they seal the fate of the limb regardless of the form of therapy; and those which jeopardize the effects of ideal therapy or preclude its institution.

24  See footnote 23, p. 71.


Time Lag

The first factor which influenced the outcome in vascular injuries in World War II, as in previous wars, was the time lag between wounding and surgery. Therapy, designed to reestablish the circulation of the limb, must be carried out within a limited period after wounding. This limit was generally (if arbitrarily) set at 6 to 8 hours. While this criterion can usually be observed in civilian practice, in military surgery it is frequently impossible because the circumstances of combat often do not permit prompt rescue and treatment. In a sample group of 104 first-priority patients studied in the Mediterranean theater, the time lapse from wounding to arrival at a field hospital ranged from 1 to 34? hours and averaged about 12? hours. The time lag between wounding and tagging (first aid) ranged from a few minutes to 25 hours and averaged about 5 hours. The time lag from arrival in the hospital to operation ranged from 1 to 10? hours and averaged about 3?hours. This delay was caused chiefly by the necessity for resuscitation before surgery. In a sample of 58 casualties with vascular injuries from the same theater, the time lag between wounding and surgical treatment averaged about 15 hours, which is almost twice the maximum period for safe arterial repair. After such a length of time, regardless of the nature of the original wound, thrombosis had usually occurred in the vascular tree distal to the injury and the tissues of the wounded extremity had been deprived of oxygen and nutrition too long for the changes to be reversible.  From the military standpoint, it is doubtful that the time lag between wounding and treatment can be reduced much lower than it was in World War II.  As the figures just cited show, nearly half of it was taken up by the period between wounding and first aid and was therefore a strictly military consideration. Furthermore, while in World War II every effort was made to bring expert surgical care as near the frontlines as possible, it was impractical to institute it at the level of the battalion aid station.

  The establishment of vascular wounds as a special category, to be handled by a special routine, was also not feasible.25 For one thing, vascular injuries, in spite of their seriousness, constituted such a small proportion of the total wounded that the imposition of another special category on the already overburdened military organization could not have been justified. For another, the effective operation of such a classification would have required of medical corpsmen a degree of differential diagnostic skill which they could not be expected to possess.

To preserve the circulation in the affected limb until patients with vascular injuries could reach installations at which specialized treatment was available numerous methods were suggested. Most of the suggestions were made by physicians and surgeons without experience in the hard facts of military surgery and were not practical. Certain advocates of refrigeration, for instance,

25 Odom, C. B.: Causes of amputations in battle casualties with emphasis on vascular injuries.  Surgery 19: 562-569, Apr 1946.


proposed its use under combat conditions.Aside from other considerations, discussed later in thechapter, this method was completely impractical for the battlefield.

The use of anticoagulant therapy (heparin) before the patient reached a hospital installation with laboratory facilities was also impractical. This form of therapy can never be employed with safety unless the patient is under constant observation and in an installation where repeated determinations of the clotting time of the blood can be made.

The same charges of impracticability, though perhaps to a somewhat lesser degree, can be made against sympathectomy and even sympathetic block. Installations on the level of a field hospital are not designed for such purposes and the performance of these operations in them can not be justified in any but occasional cases.

  In view of these various factors, there was little to do but make the best of the situation and to keep the time lag between wounding and treatment as low as the military circumstances permitted. Admittedly if this lag could have been reduced much better results could have been obtained. The importance of the time factor can best be revealed by an analysis of 473 cases from the American material, in 207 of which amputation had to be performed (Chart 7). Only a negligible number of the casualties in the 1- to 10-hour group were seen much earlier than 10 hours, which explains why the results in that category are not much better than the results in the whole group. When debridement had to be delayed overlong the limb was often lost because of anaerobic infection, though it is doubtful that this cause explained many of the cases in the 1- to 10-hour group. It played an increasingly important role, however, as time progressed, and it probably accounted, at least in part, for the unhappy results in the group observed 20 hours or more after wounding. On the other hand, from the standpoint of possible reestablishment of the circulation, it is questionable whether a time lag of more than 10 hours was of special significance, since, as already pointed out, corrective measures must be instituted well within the upper limits of this period if good results are to be secured.

Practical Difficulties

  Reparative vascular surgery is always difficult surgery. It is time consuming, it requires special and delicate equipment, and it must be performed by a surgeon of highly specialized experience and great technical dexterity.26  When casualties reached battalion aid stations they were usually observed by nonspecialized medical personnel or, occasionally, by military administrative officers.  Men assigned to these stations were neither trained nor equipped to institute specialized treatment. Even at field hospitals casualties were not usually seen by experienced vascular surgeons. The infrequency of vascular injuries would not have warranted the assignment of the limited number of

26  De Takats, G..:  Vascular surgery in the war. War Med. 3: 291-296, Mar 1943.


Chart 7. Effect of time lag between wounding and operation on incidence of amputation in arterial wounds of the extremites among American battle  casualties in World War II.  The data depicted are based on a selected group  for which the necessary information was available.  They do not include all arterial wounds during World War II. .

medical officers in this group toinstallations on this level. Finally, when casualties with vascularinjuries did reach installations at which definitive surgery could bedone, the ablest surgeons, asalready intimated, were working against factors which were beyond theircontrol and underconditions which had usually ceased to be favorable. Outstandingresults could not, therefore, beexpected.

< Status of the Casualty

  In addition to such external factors as the time lag and the lack of experienced vascular surgeons in forward areas, certain other factors inherent in this type of injury precluded generally good results. One was the predominant necessity of saving the casualty's life before any endeavor was made to save his limb. Even when the vascular injury was the only injury, the patient was often in such poor condition from exposure, loss of blood, and other causes when he reached the hospital that operation had to be postponed until he had been sufficiently resuscitated to withstand it.


  An important consideration in this connection was the amount of blood lost. In the majority of vascular wounds considerable loss occurred before first aid could be instituted. A sample group of 27 patients studied during the first week after wounding was found to have an average erythrocyte count of only 2,700,000 per cubic millimeter. The hemoglobin concentration in 24 of the 27 patients was still only 60 percent of normal, though all of the patients had been given whole blood in amounts which were thought to be adequate. One patient in the group, with a wound of the femoral artery, had lost 40 percent of his normal blood volume when first seen.

Naturally, if the volume of circulating blood was reduced, the amount which passed through the peripheral arteries was also reduced and the circulation in the distal portion of the extremity in which a vascular injury had occurred was still further curtailed and nutrition of the limb suffered accordingly. From both the systemic and local standpoints prompt restoration of the circulating blood volume and hemoglobin concentration was essential. In fact, the patient with a wound of the extremity often required as much blood, and sometimes even more blood, than one with a wound of the chest or abdomen.


Another consideration which contributed to the seriousness of vascular injuries was the presence of associated wounds which were often extremely serious.  In not more than a third of all cases was the vascular wound the only wound. Associated wounds of the abdomen, chest, or head sometimes required attention far more urgently, as a lifesaving matter, than did the vascular wound, and vascular surgery had to be deferred in their favor. Other local wounds might further impair, or altogether destroy, the regional circulation and make salvage of the limb entirely impractical. Compound fractures, for instance, frequently tore collateral vessels and favored the development of inflammation and thrombosis. Patients were seen in whom plaster casts were not adequately padded under the distal fragment of a fracture of the lower third of the femur. Damage to the femoral and popliteal arteries resulted, and these cases almost invariably terminated in gangrene.  

  The series of vascular injuries which form the basis of this chapter show how seriously the presence of associated fractures affected the outcome of such injuries (Chart 8). The incidence of amputation was significantly higher when the vascular injury was complicated by a fracture.  The observation is significant in itself and also suggests the importance, when studied in connection with the British figures, of analyzing all the circumstances in any given series of cases before drawing conclusions from comparative studies.

  Nerve injuries were frequently associated with vascular injuries, as they naturally would be, since with few exceptions all large blood vessels are accompanied by peripheral nerve trunks. This is especially true of the neck, and of the brachial, femoral, and popliteal regions.  The sciatic nerve was likely to be involved in traversing wounds of the thigh, and the peroneal and tibial


Chart 8. Effect of presence orabsence of associated fractures on incidence of amputation in arterialwounds of the extremities among American battle casualties in World WarII. The data depicted are based on a selected group for which thenecessary information was available. They do not include all arterialwounds during World War II. 

components in large open wounds of the popliteal space. Nerve injuries often had a large share of the responsibility for the permanent disability which followed some vascular wounds, but occasionally, and paradoxically, they exerted a transient beneficial effect in that they produced vasodilatation peripherally in the area supplied by the affected nerve.

Site andType of Injury

On the basis of actual experience of surgeons in the field, and under the conditions and limitations imposed by military practice, the categoric statement can be made that the site and type of a vascular wound determine the therapeutic procedure and therefore predetermine, so to speak, the end results. The British experience in World War I, whenever detailed statistics are available for analysis (Chart 9), bear out this generalization. It is also borne out by the American experience in World War II (Chart 9).  Almost invariably it could be anticipated that wounds of certain vessels, such as the popliteal artery, were much more likely to be followed by ischemic gangrene than were those of certain other vessels, such as the brachial artery. Indeed, for practical pur-


Chart 9. Effect of site of arterial wounds of the extremities on incidence of amputation in American battle casualties in World War II and in British battle casualties in World War I.  The data depicted are based on a selected group for which the necessary information was available and are not to be considered as complete. See text, pp. 63-65,  for details of selectivity.

poses it was quite possible to make up categories of critical and noncritical arteries. This observation held, however, only for single wounds.  In the material studied, a wound of either the anterior or the posterior tibial artery was relatively noncritical, but when both arteries were damaged the injuries resulted in the second highest proportion of gangrene in both the British and the American material.   Generally speaking, wounds of the lower extremity were more serious than those of the upper extremity (Chart 10).27 The incidence of amputation was considerably greater in wounds of the lower extremity, from the iliac through the popliteal arteries, than in those of the upper extremity, from the subclavian through the brachial arteries. Other considerations also entered into the amputation rate.  In injuries of the brachial and femoral arteries, for instance, it made a difference whether the injury was above or below the profunda branch (Chart 11).

27   The figure 2,453 is used instead of 2,471 as shown in Table 3, because included in the latter are 18 cases: aorta, 3; carotid, 10; external carotid, 3; and renal, 2.


Chart 10.  Effect of site of arterial wounds of the extremities on incidence of amputation in American battle casualties in World War II. The data depicted are based on a selected group for which the necessary information was available. They do not include all arterial wounds during World War II.  

  The role of special vessels in producing the end results of traumatic vascular surgery can be arrived at in another way.  If the injuries of single vessels in the forearm and leg (the radial, ulnar, anterior and posterior tibial, and peroneal arteries) are excluded from the 2,453 arterial wounds of the extremities (see footnote 27, p. 78) collected from American armies in various theaters of operations (Table 3), the incidence of amputation after vascular. injury rises to 49.6 percent.  If only multiple arterial wounds of the leg and wounds of the iliac, femoral, and popliteal arteries are considered, the incidence of amputation rises to 62.6 percent (Table 3). A similar selective analysis of the British material reveals comparable conditions.

What these figures show is this:  In any unselected group of vascular injuries the proportion of poor results will be high or low according to the number of important or unimportant vessels involved.  All recorded military experience is to this effect. The most common reason for the poor results of vascular surgery in World War II was that the surgeon, no matter how great his skill, bad no factor of choice, and therefore no responsibility in regard to the site of the wound. He had to institute therapy in accordance with its location, and


Chart 11.  Effect of site of wounds of the brachial and femoral arteries on incidences of amputation in American battle casualties in World War II.  The data  depicted are based on a selected group for which the necessary information as available. They do not include all arterial wounds during World War II. The 27 cases involving the profunda femoris shown in Table 8 have been excluded as not pertinent to this consideration.

its location was the determining factor in whether or not the limb could be saved.   The extent of the arterial wound also played a part in the end results. Generally speaking, the larger the injury, (1) the greater was the chance that the collateral circulation was also damaged, (2) the more extensive was the necessary debridement, and (3) the greater was the chance of infection if debridement was not adequate.

The actual type of injury was another factor of dominant importance in determining the end results of arterial injury. Although in the American material from World War II it was possible from the records to separate only 620 of the 2,453 cases into distinct categories on this basis (Chart 12), the incidence of amputation seemed definitely related to the type of lesion. The figures merely corroborate the clinical opinions of the surgeons who handled these patients.

In those injuries in which thrombosis occurred, since this type of lesion was always likely to cause widespread interference with the flow of blood through collateral vessels, the worst results ensued, as might have been expected. Circular Letter No. 178, dated 23 October 1943, from the Office


of The Surgeon General, 28 recommended that thrombotic segments of damaged vessels be excised, not ligated, and later directives 29  from the Surgeon General's Office and other theaters of operations gave the same instructions.

  Arterial lacerations usually represented a serious type of injury. Minor lateral lacerations, in which conservative surgery was possible, were seldom encountered. In the present series (Chart 12) the category of lacerations and transections includes 200 lacerations, 253 transections, and a number of cases in which it could not be determined which of these lesions was present.  In the latter group the incidence of amputation, as might be expected in view of its composition, was midway between the incidence of lacerations in which interference with the circulation might be limited and the incidence in transections in which there was abrupt interruption of the blood supply. Clean-cut

Chart 12.   Effect of type of arterial wounds of the extremities on incidence of amputation in American battle casualties in World War II. The data depicted are based on a selected group for which the necessary information was available.  They do not include all arterial wounds during World War II.  The total number of lacerations and transections is greater than the sum of the individual lesions in this category because it could not be determined from 85 of the records which of these particular lesions was present.

28  Cir Ltr 178, SGO, 23 Oct 43, sub:  Care of the wounded in theaters of operation.

29  TB Med 147,  Notes on care of battle casualties, Mar 1945.


transections were seldom observed.  In most cases tearing of the tissues resulted in extensive loss of substance, usually associated with extensive damage to the collateral circulation.

The category of arterial compression (Chart 12) includes only a small number of cases (10), but its importance is indicated by the fact that in 6 of the 10 cases amputation was necessary. Experience showed that this was a very important group because compression injuries commonly occurred in association with fractures of the lower end of the femur with posterior displacement. No difficulties arose when the situation was corrected promptly. When it was not, the consequences were usually disastrous.

  A category of injuries which does not appear in this classification because it could not be identified from the records, but which is known to have been observed in World War II, is the so-called concussion of the artery (stupeur art?rielle).30This was a localized segmental type of spasm of the artery which followed various forms of trauma. A traumatizing agent which did not injure the artery itself but passed near the vessel, or even at some distance from it, was usually the cause. In such cases the limb was pale, cold, and pulseless, but there was nothing to suggest hemorrhage, hematoma formation, or other results of laceration. In differential diagnosis it had to be distinguished from thrombosis. The spasm which followed arterial concussion usually lasted only for some hours, but might last for many days (see Case 16).  It was sometimes sufficiently intense and prolonged to result in death of the limb, though as a rule it responded well to debridement of adjacent lacerated tissues, conservation of body heat, or sympathetic block.  The latter procedure, to be successful, had to be accomplished as soon as practicable after the diagnosis was made. If it did not achieve the desired results, then sympathectomy was performed.

In one case of arterial concussion observed in the Mediterranean theater, a shell fragment produced wounds of entrance and exit on the lateral aspect of the right calf at the midpoint.  Although exploration revealed the track of the missile to be between 6 and 8 cm. from the posterior tibial artery, this vessel was in complete spasm.


  In World War II infection was perhaps the least important of the limitations on therapy imposed by the circumstances of warfare.  Battle wounds were always potentially infected, it is true, but when adequate debridement was carried out surgical procedures directed toward the treatment of the vascular injury could be done at the same time with a high degree of safety.  It was not infection but other circumstances (time lag and type and location of wound) which prevented reparative procedures in most battle injuries of the blood vessels. The majority of the wounds in World War II were produced by shell fragments; many of them were of considerable size, were associated

30 Seefootnotes 22 (3), p. 70; 28 and 29, p. 81.


with extensive destruction of tissue, and were not favorably located (Charts 9, 10, 11).  As a result, the vitality of the limb was usually gravely impaired before the surgeon ever saw the patient.


The results of acute occlusion of the major arteries in combat-incurred wounds, whether caused by the injury itself or by ligation of the injured vessel, have not been clearly established.

Part of the confusion, as has already been intimated, lies in the statistics which for so many years were accepted as representing the incidence of gangrene after acute traumatic vascular occlusion. Actually they represented no such thing, since most of them included instances of aneurysm as well. The two lesions are, of course, totally dissimilar.  In acute occlusion, whether produced by injury or ligation, the blood flow is cut off abruptly.  In aneurysm the occlusion occurs gradually, if at all, and by the time it has developed or the aneurysm has been excised a more or less adequate collateral circulation exists.

Furthermore, in many series gangrene is the only unfavorable result reported. The percentage of cases in which primary amputation had to be done because of arterial injury is completely disregarded and such series there fore do not represent the incidence of poor results in acute traumatic arterial occlusion.  As for the various series in which amputation is included-in popliteal arterial occlusion, for instance, it varies from 0 to 100 percent 31 -it is difficult to determine on what possible basis the calculations have been made.

Makins' statistics, 32  which represent the largest collective experience in World War I, furnish an excellent example of the confusion which results from the inclusion of both acute and nonacute (aneurysmal) lesions in a single series.  Actually, 49 percent of the cases in the series are aneurysms, and in such vessels as the axillary and subclavian arteries the proportion is more than 70 percent.  When, so far as possible, the favorable influence of aneurysm is excluded, the proportion of poor results in Makins' series rises from 18.1 to 26.5 percent.

  When Makins' figures are thus corrected, the American results in World War II, which at first glance seem much worse than those of the British in World War I, are probably better. The American series does not include aneurysms, the influence of which on statistics is always favorable, because, as the experience at the vascular centers in the Zone of Interior clearly showed, gangrene almost never occurs after operation for this type of lesion.

  The confusion apparent in Makins' series is almost always evident in other reported series, e. g., Salomon,33 and Soubbotitich.34  On the other hand, from

31   See footnotes 4 and 6, p. 61.

32 See footnotes 3, p. 60; and 12, p. 64.

33   Salomon, R.:  Beitrag zur Lehre der arteriovenosen Schussaneurysmen.  Beitr. z. klin. Chir. 113: 369-410, 1918.

34 Soubbotitich, V.:  Quelques consid?rations sur les an?vrismes traumatiques.  Bull. et m?m. Soc. de chir. de Paris 42:698-708,1916.


every series in which aneurysms are excluded, the proportion of amputation in vascular injuries is high.  In 74 cases collected by Franz 35  from German World War I statistics the percentage of gangrene was 70.4 percent, and Mocquot and Fey 36 reported an amputation incidence of 44 percent in 61 vascular injuries observed in a French surgical ambulance.

The reported series which concern arterial ligation are no less confused than those which deal with acute vascular injuries in which ligation has not been done. From the period prior to World War I, no series large enough to be representative is available, but Sencert, 37 who collected a small number of cases, reported that the incidence of gangrene ranged from 5 percent when ligation of the subclavian and brachial arteries was done, to 50 percent when the common iliac artery was ligated. The author himself explained this remarkably wide range by the inclusion of aneurysms in the series.

  On the basis of his personal military experience, Sencert 38 reported in 1918 that ischemic gangrene had occurred only twice in 70 cases of vascular injury in which ligature was done a few hours after injury and in which no hematoma of any significance was present.  The figures, again, cannot be accepted at their face value because 4 cases are included in which the injury was venous, not arterial.  When a diffuse hematoma had formed, the results were much less satisfactory; of 20 cases in which the axillary, femoral, and popliteal arteries were involved, gangrene developed in 6.  The figures are perhaps somewhat selective because of Sencert's own pronounced views concerning the important role of hematoma formation in arterial injuries.

  Amputation after arterial wounds in World War II was usually done for one of two reasons, gangrene or infection.  The so-called toxic absorption, which is assumed to be the result of autolysis of muscle in the ischemic limb and which gives rise to systemic manifestations, should probably be classified as infectious in origin.

  Among American troops in World War II, ischemic gangrene, not infection, was the chief reason for amputation after vascular injuries (Table 4).  Even when clostridial myositis is combined with other infections it is seen that not more than a fifth of the amputations were performed because of infection. This is in decided contrast to what happened in some other armies. Turovets, 39 for instance, in an analysis of the 12 amputations required in 49 vascular injuries sustained in the Russo-Finnish War of 1939-1940, found that 10 of the 12 were for gas gangrene or other infection while only 2 were for ischemic gangrene.

35   See footnote 17, p. 64.

36  Mocquot, P., and Fey, B.:  Gravit? des l?sions arterielles du membre inf?rieur dans les plaies de guerre.  Rev. de chir. 53: 241-267, Mar-Apr 1917.

37   Sencert, L.: Le traitement des plaies vasculaires ? 1'avant. Lyon chir. 14: 638-686, Jul-Aug 1917.

38 Sencert, L.:  Wounds of the Vessels.London, University of London Press, Ltd., 1918.

39 Turovets, I. G.:  Gunshot wounds of the blood vessels.  Khirurgia No. 5-6, 1942, p. 35-46. Abstr. Am. Rev. Soviet Med. 1: 41, 1943, and Bull. War Med. 4: 628-629, Jul 1944.



  There are other reasons, in addition to the inclusion of aneurysms in Makins' series, why the American results of arterial injuries in World War II seem worse than the British results in World War I. In 86 percent of the American cases the wounds involved important or critical arteries. This was true of only 70 percent of the British cases (Chart 6). The American figures, furthermore, include among the poor results all cases in which amputation was necessary after vascular injuries regardless of the cause.  It is possible, though exact statements cannot be made on this point, that there are excluded from the British figures some cases of gangrene and amputation which were classified as infections rather than as vascular injuries, as well as those in which amputation was necessary for extensive trauma and other complications of vascular injuries. Finally, because of the more destructive weapons used in World War II, it may be assumed that tissue destruction was greater among American wounded in the Second World War than in the British cases of the First World War. As a result, in addition to the injury to the main arterial circulation, the collateral circulation was frequently and seriously impaired. Further, debridement, which was probably more commonly practiced in the Second World War than in the First, was more extensive. Debridement was, of course, an essential procedure.  Its omission in cases of arterial injuries would undoubtedly have resulted in a higher mortality rate, as well as in the loss of more limbs from infection. When debridement is properly performed, however, it involves loss of tissue and in extensive wounds its thorough performance inevitably entails additional damage to the collateral circulation.



In the control of hemorrhage, packs were avoided as far as possible. Generally speaking, seepage was less harmful than the damage resulting from their use, for gauze packs inserted to check bleeding inevitably became tight and when allowed to remain in place during a prolonged evacuation eventually


caused a good deal of damage. Quite properly,packs were used as a temporary expedient forcontrolling hemorrhage when they could predictably be removed within afew hours at a hospitalinstallation. The use of a tourniquet was thereby avoided.


  Tourniquets surely saved the lives of many soldiers in World War II who otherwise would have died of hemorrhage. On the other hand, they were undoubtedly applied in some cases in which they were not really needed and were responsible for the loss of some limbs which could have been saved. In all theaters of operations surgical consultants and their assistants gave instruction in the correct use of the tourniquet, emphasizing when its use was and was not indicated, proper methods of application when its use was justified, and correct management of a casualty with a tourniquet in place. Only by careful evaluation of the patient's blood loss, state of shock, facilities available for blood volume replacement therapy, time interval likely to elapse before primary surgery could be done, and the extent of damage to an extremity, could a proper decision be made as to the application or removal of a tourniquet.   Intelligent handling of this life- and limb-saving device demanded that each casualty be treated as an individual problem.  The application of a tourniquet for every wound of the extremity, with its release every half hour by rule of thumb, was always a potentially dangerous method.

  Observations on the correct use of the tourniquet were best made in forward installations, chiefly in field hospital units, which accepted extremity wounds in which hemorrhage was a factor.  It was in these units that most patients who had had tourniquets applied were seen and it was therefore in these units that most of their life-saving and life-endangering results could be observed. The following observations on tourniquets are based on the conclusions reached concerning them by Maj. Luther H.Wolf and Capt. Trogler F. Adkins,40 who were attached to the 2d Auxiliary Surgical Group. This team operated for more than a year in one of the "first priority" surgical hospitals on the Italian front. It will be noted that certain of their data and conclusions are at variance with practices accepted as correct before World War II, though they are in agreement with the data and conclusions of most other surgeons who had similar frontline experiences.

Choice of Tourniquets.  Because complete control of bleeding is the primary objective of tourniquet application, any result less than this must be considered inadequate. For that reason the strap-and-buckle type of tourniquet issued by the Army and generally used early in the war was not a satisfactory piece of equipment for first aid on the battlefield: It was so narrow that it cut into the tissues, and it seldom controlled bleeding completely no matter how tightly it was applied. The failure of this piece of equipment to produce desired results

40  Wolff, L. H., and Adkins, T. F.:  Tourniquet problems in war injuries. Bull. U. S. Army M. Dept. 87: 77-84, Apr 1945.


could be explained in part by the usualpractice of placing a gauze roll or some similar firmobject over the affected vessel before it was applied. The vessel,being cordlike and elusive,frequently slipped out from under the roll and bleeding continued. Ifthe principal vessel wassuccessfully occluded the collateral blood supply parallel to it wasnot, since the pad so elevatedthe tourniquet on either side that no pressure was applied to thevessels on either side of theprincipal vessel. When the strap-and-buckle tourniquet was the onlytype supplied, better resultswere obtained by substituting mechanic's waste or a Carlisle dressingfor the gauze pad, or byusing no pad whatsoever.

  The most satisfactory tourniquet was the pneumatic type in which the pressure could be accurately adjusted and maintained. This type, however, had two disadvantages:  Pressure leakage was always a possibility, and it was too cumbersome and complicated for quick application on the field of battle. However, when circumstances permitted, the pneumatic tourniquet which records pressure in pounds per square inch or in millimeters of mercury was safer. One pound of pressure in the model supplied was equivalent to 51.7 mm. of mercury. When the tourniquet was applied to the arm, 5 pounds of pressure (250 mm. of mercury) was usually sufficient; in the leg the pressure seldom had to exceed 6 pounds (300 mm. of mercury).

  The simplest and most effective type of tourniquet for frontline use was not a manufactured article but a 32-inch soft rubber tubing (U. S. Army Medical Supply Catalog Item No. 3879000) in 6-foot lengths. When this size was not available ?-inch rubber tubing (Item No. 3878000) was an acceptable substitute. Care had to be taken that only moderate tension was applied because complications which followed the use of rubber tubing were nearly always the result of too much tension.

Technique of Applying Rubber Tubing Tourniquet. The bleeding extremity was padded at the level at which the tubing was to be applied with a towel, shirt sleeve, trouser leg, or any similar available material. The tubing was then wound around the limb in 4 parallel turns under moderate tension only. The end first applied was overlapped and anchored by the second turn, and the last turn was anchored by the next-to-last turn.  If the smaller-size tubing was used more turns were necessary from 8 to 10 on the thigh, from 4 to 5 on the arm.  It was most important that tension be moderate. If several turns were applied moderate tension was sufficient to occlude bleeding vessels completely and the dangers inherent in a tightly-applied tourniquet (a general fault) were thus avoided.

  The tourniquet was placed about the thigh or upper arm as close to the site of injury as was feasible.  If large defects were present care had to be taken to apply it sufficiently far above the injury to prevent its slipping down into the defect. Tourniquets placed about the forearm or leg were frequently not effective since bleeding from incompressible interosseous vessels was likely to continue and to lead to a considerable loss of blood.


  Selection of Cases.  Both judgment and experience were required to determine when a tourniquet should be applied and when it could be omitted safely. Many severe wounds of extremities, especially badly contused or avulsed wounds not involving major blood vessels, required no tourniquets. An inadequate tourniquet could actually increase bleeding from this type of wound.  On the other hand, tourniquets were not used early enough and were insufficient in many serious extremity wounds involving large blood vessels or in patients with traumatic amputations. There may be some bias attached to this observation, since the only casualties admitted to hospitals on this level were those in shock or with tourniquets.

  Statistical data are not available to show what proportion of casualties bled to death on the battlefield from wounds of the extremity, nor is it known how many died from hemorrhage after they had received first aid. No comments on this point would therefore be justified. It was common experience, however, to encounter in forward hospitals patients with wounds of the extremities to which tourniquets had not been applied and who survived the transportation. Although their wounds on gross inspection did not appear extensive, they had lost large volumes of blood. Studies of circulating blood volume and other data obtained from many such individuals often showed that they had lost from 33 to 50 percent, and sometimes more, of their circulating blood volume. This blood loss might have been avoided by proper use of the tourniquet.

  It is true that when a wounded soldier reached the field hospital, usually within 6 to 8 hours after wounding, bleeding was likely to have stopped, but this did not always prevent his condition from being critical from loss of blood. Spontaneous arrest of hemorrhage took place as a physiologic process in many wounds of the extremity, but frequently it did not occur until the systolic pressure had fallen and sufficient spasm had developed in the vessel for natural processes to effect a cessation of the blood flow.

  It was an essential precaution to apply a tourniquet to an extremity in which there had developed, or was developing, a large, pressure-producing hematoma as the result of injury to a major artery.  Experience showed that more damage could be done to the limb as the result of the pressure and hemorrhagic infiltration caused by this type of lesion than by the application of a tourniquet for a reasonable time. It was highly unlikely that in such cases the collateral circulation present about the hematoma immediately after injury was sufficient to affect materially the viability of the distal tissues. It was also thought that the early application of a tourniquet increased the chances of firm, occlusive clotting.  It was shown repeatedly that operation for the release of tension and for repair or ligation of the damaged artery was simpler, and clots and infiltrated blood could be handled more readily, if bleeding had not been permitted to progress unchecked.

  That reliance on natural methods is a dangerous way to secure control of hemorrhage is shown by the following case:


  Case 1. A sergeant was seen immediately after admission to a field hospital, 5 hours after a shell fragment had perforated the middle third of his left arm.  No tourniquet had been applied.  Although the initial blood pressure was 115 mm. of mercury systolic and 60 diastolic he was extremely pale, restless, and apprehensive and had obviously lost a large quantity of blood. The pulse rate (in the carotid artery) was 128 beats per minute.  Generalized vasoconstriction was intense.  Examination of the wound showed no bleeding at this time.

Within 10 minutes of admittance the patient's condition began to deteriorate rapidly, his blood pressure fell to 72 mm. of mercury systolic and 40 diastolic, and a transfusion of low-titer group O blood was started immediately.  The blood was forced in under pressure, approximately 300 cc. being given in 15 minutes. In spite of the transfusion and other standard methods for the treatment of shock, the man died 25 minutes after he had been brought into the hospital.

A complete necropsy revealed his only wound to be a perforating one of the middle third of the left arm, which had caused a transection of the brachial artery and a compound comminuted fracture of the left humerus. The arterial wound showed no evidence of intraluminal clotting. The patient had had no signs of symptoms of fat embolism before death and no signs of this process were observed at necropsy. The cause of death was recorded as hemorrhagic shock.

  Comment.  The details of the circumstances preceding the admittance of this patient to the field hospital could not be obtained, and there are a variety of possible reasons why no tourniquet had been applied. Whatever they may be, the statement is believed warranted that had a tourniquet been applied in this case, death from blood loss would probably not have occurred. As this case, and numerous others, showed, an adequate tourniquet should be applied proximal to any wound of the extremity which is bleeding freely.  It should be applied, moreover, at the earliest possible moment. Many of the fears of damage from tourniquets are probably unfounded to a large extent, and certainly the saving of every possible ounce of blood is the more important consideration from the standpoint of the patient's general condition.

  Release and Removal of the Tourniquet.  It is unfortunately true that in World War II some limbs were lost because tourniquets were left in place too long and gangrene developed. There was some disagreement, however, over how long "too long" was. Wolff and Adkins' experience in Italy 41 showed that a tourniquet could be left in place from 2 to 6 hours without clinically detectable damage depending on the temperature of the atmosphere and of the extremity. During the winter of 1943-44 they observed tourniquets which had been left on as long as 8 hours without apparent deleterious effects. Surgeons of an auxiliary surgical group performed  amputations on approximately a thousand patients in a forward field hospital. These surgeons all observed extremities on which tourniquets had been left applied up to 4 hours, yet no cases of gangrene solely from a tourniquet were noted.  In no case was there an unusual degree of edema, a slough of the skin edges, or any ischemia of the muscles which could be attributed solely to the use of the tourniquet. Moreover, during the period of observation which lasted from 5 to 10 days in most

41  See footnote 40, p. 86.


cases, no instance was observed in which theapplication of a tourniquet had caused either nerveinjury or thrombosis.

About the development of gas bacillus infection it is necessary to be less positive. This type of infection was observed in patients who had worn tourniquets for various periods of time, but it was also observed in others who had not worn them. Gas bacillus infection seemed somewhat more prone to occur in the bloodless extremity to which a tourniquet had been applied, but no definite opinion was formed on this point. It was the consensus, however, that even if these complications did occur more frequently, the role of the tourniquet as a lifesaving measure would still make its use imperative.

  There are cases in which the removal or loosening of a tourniquet is unwise and unnecessary; in fact, the results may be disastrous. Experience has led to the conclusion that under no circumstances should a tourniquet be loosened on a patient in shock or in incipient shock unless means are present and immediately available to control any hemorrhage that may occur and to replace rapidly the volume of circulating blood. The presence of a medical officer was considered necessary. The following case histories illustrate the widsom of these precautions.

  Case 2.  An officer was admitted to a field hospital 3 hours after he had been wounded by a shell fragment which traversed the right arm and produced a compound comminuted fracture of the humerus.  This was his only wound.  When he was admitted to the field hospital he was in mild to moderate shock.  The blood pressure was 110 mm. of mercury systolic and 66 diastolic and the pulse rate was 112 beats per minute.  The blood vessels were in a state of vasoconstriction.  Although he had obviously lost considerable blood his condition was recorded as fair.

  Examination of the wounded arm showed an improvised splint, a tourniquet of the strap-and-buckle type applied tightly at the level of the insertion of the deltoid muscle, and a fairly extensive perforating wound of the inner aspect of the lower arm.  A slight but persistent oozing from the wound was thought to be of venous origin, the result of back stasis from a nonocclusive tourniquet.  The tourniquet was therefore loosened and slipped off after a blood pressure cuff had been placed about the arm above it.  Before the cuff could be tightened, the severed brachial artery broke away and in the few seconds required to pump up the blood pressure cuff enough blood was lost (not more than 100 cc.) to cause a drop in the blood pressure from 110 mm. of mercury systolic and 66 diastolic to 80 systolic and 40 diastolic and to deepen materially the state of shock. It is not hard to imagine what might have happened, since the circulatory volume was at a critical level, if measures for the prompt control of hemorrhage had not been available.  The patient was immediately given 2,000 cc. of whole blood and after resuscitation had been accomplished a successful operation was performed. Recovery was uneventful.

  Case 3. A private was admitted to a field hospital 6 hours after he had sustained a traumatic amputation of the left leg, just below the knee, from a German "Schu" mine.  He was in severe hock with a blood pressure of 86 mm. of mercury systolic and 56 diastolic, and a pulse rate of 116 beats per minute. Examination disclosed a persistent trickle of blood from the badly contused stump although there were two tightly placed strap-and buckle tourniquets about the thigh.  A tourniquet of soft rubber tubing was substituted for these tourniquets and bleeding was promptly controlled.  Vigorous resuscitative measures were undertaken, including infusion of 2,000 cc. of whole blood.  At the end of 3 hours, however, the patient was still in such poor condition that operation could not be considered.???


  Ordinarily a patient with a wound of this sort could be prepared for operation within an hour and one-half at the most; if he could not be, a gas bacillus infection or continued bleeding was the usual explanation.  Examination of this man's wound showed that the dressings which had been applied over the stump after the original examination were saturated with blood and that a pool of blood had collected under his buttocks. Questioning disclosed that the tourniquet had been loosened by a ward attendant at 30-minute intervals ever since it had been applied, and as a result the patient had been losing as much blood by its periodic release as he was gaining by transfusion.  The introduction of another 1,000 cc. of whole blood with the tourniquet tightly in place produced the desired results and rendered the patient a safe risk for surgery within a short time.

It was repeatedly observed in the course of World War II that patients who had lost small to moderate amounts of blood could tolerate further rapid loss of small additional amounts without placing their chances of recovery in jeopardy.  It was also repeatedly observed that many patients who had lost blood up to the critical level could maintain a semblance of circulatory equilibrium only by intense vasoconstriction, rapid heart action, and other physiologic compensatory mechanisms.  It was in these patients that urgent care had to be used to prevent any additional decrease in blood volume. The rapid loss of even 100 cc. of blood could completely break down the delicately balanced compensatory mechanism and result in profound and often fatal shock through failure of the vasoconstrictor apparatus.  It was never safe to loosen a tourniquet in such cases, no matter how long it had been applied, until the blood volume had been restored to a safe level by plasma transfusions or, better, by whole blood transfusions.

  It also had to be borne in mind that patients may lose more blood than is first apparent on loosening a tourniquet, particularly a tourniquet about the thigh. The arterial system fills up more or less uniformly on release of a tourniquet, and if the vascular injury is some distance below the tourniquet the injured area continues to drain off the blood in the intervening vessels, even though the tourniquet is promptly reapplied upon the first sign of hemorrhage.

  Principles of Tourniquet-Application.  On the basis of observations made in the early months of the war, the following ideal plan was substituted for the former plan of releasing a tourniquet every 20 to 30 minutes:

  1.  An adequate tourniquet should be placed on an actively bleeding extremity at the earliest possible moment after injury.

  2.  A patient with a tourniquet in situ should be given the highest priority for transportation to the nearest available hospital and the presence of the tourniquet should be plainly indicated on his record.

  3.  At the end of 2 hours, providing the patient is not in shock and depending on the circumstances of the temperature and tactical situation, the tourniquet should be cautiously loosened under the supervision of a medical officer. The precaution of professional supervision should be omitted only in very exceptional circumstances.  If bleeding recurs, the tourniquet should be reapplied at once. If hemorrhage apparently has been controlled, the tourni-


quet should be removed, but it should be keptavailable for instant reapplication and the patientshould be kept under constant observation.

4. If a patient is in shock from hemorrhage, the tourniquet should on no account be removed within the first 4 to 6 hours of its application unless the blood volume has been at least partially replaced by plasma or whole blood. This plan involves a calculated risk, but the possible loss of a few inches of a badly damaged extremity cannot be permitted to compromise a patient's life.

5. At the end of the 4- to 6-hour period, removal or loosening of the tourniquet should be a matter of individual judgment on the part of the medical officer. After 8 to 10 hours have passed, it should be removed for it can be assumed that sufficient spasm and clotting have developed to prevent further bleeding. Even when the extremity is chilled, it is unwise to leave the tourniquet in place longer.

6. The temperature of an extremity about which a tourniquet has been applied should be lowered as far as possible, short of actual freezing. Numerous experimental observations have shown that the temperature of a bloodless extremity has much to do with the speed with which tissue necrosis occurs, and that the metabolic demands of the tissues vary directly with the temperature. This observation was of importance in tourniquet application. A tourniquet applied in the heat of the tropics, or in the desert, had to be loosened at shorter intervals than a tourniquet applied in a cool or cold climate. While care was taken to prevent frostbite and other cold injuries in cold and freezing weather, it was the rule to leave uncovered an extremity about which a tourniquet had been applied, and never to employ artificial means to warm it.


  The salvage of limbs which were the site of vascular injuries depended quite as much upon a multitude of details as upon the major surgical method used. Careful wound debridement minimized the chances of infection, as did chemotherapy and antibiotic therapy.  Every possible collateral vessel was spared. Such circulation as was present was maintained by proper posturing and by the correction of oligemia and anemia. Circulation through the collateral channels was encouraged by the release of the vessels from vasoconstricting influences by means of sympathetic block or sympathectomy. Heparin was employed when circumstances permitted the proper precautions, to prevent thrombosis. No single one of these measures, all designed to improve the circulation, could guarantee survival of any limb, but the careful use of all of them, or of as many of them as possible, frequently made the difference between a good and a poor result.

  Some surgeons felt that in a small and carefully selected group of vascular injuries either uncomplicated healing might take place without any surgery at all, or, if it did not, that aneurysm formation would take place. It is true that there is an almost negligible incidence of loss of limb after excision of aneurysms,


but it is also true that a traumatic arteriallesion is not necessarily followed by aneurysmformation. On the other hand, when it is possible to delay surgicalligation for a period of time,the status of the limb was frequently better than when it was necessaryto perform it immediately.

Case 4. A soldier wounded by enemy shell fragments in Italy on 1 June 1944, at 1100 hours, incurred multiple penetrating wounds of the right hand, wrist, arm and chest wall, and of the right knee in the popliteal space where he sustained a through-and-through wound of the popliteal artery.  He suffered no fractures in the lower extremity.  All wounds were debrided in a forward hospital on 2 June between 2100 and 2135 hours, with the patient under thiopental sodium anesthesia.

  On 6 June, en route in a hospital ship to a general hospital in the base, the patient complained of pain in the right knee and during ambulance transportation he bled from the wound in the popliteal space.  Bleeding ceased spontaneously after loss of about a pint of blood.

  He was taken promptly to the operating room where exploration revealed that the popliteal artery was enclosed in a sac which was pulsating strongly.  The sac was not disturbed in the hope that an aneurysm would form.  Since the wound looked clean, partial closure of the muscle was done.  A firm dressing was applied down to the muscles and the leg was partially immobilized in a posterior plaster shell.

  Roentgenograms the day after operation revealed a metallic foreign body 3 by 5 mm. in the right popliteal space. The total plasma protein concentration was 4.75 gm. per 100 cc. and the hematocrit value was 26.9 percent.  By 9 June transfusions had amounted to 1,850 cc. of whole blood.

  The following day the foot was warm.  The patient could move his toes but he complained of pain in the right popliteal area radiating into the toes which were hypesthetic but not anesthetic.  Recovery proceeded smoothly until 19 June, when there was a sudden hemorrhage from the right popliteal space. A pressure dressing was reapplied.  The foot was warm and motion in the toes was normal.  On 20 June, the total plasma protein concentration was 6.5 gm. per 100 cc. and the hematocrit value was 33 percent.

  Bleeding recurred 23 June and was again controlled by a pressure dressing.  Two days later the toes were still warm and there was less complaint of pain.  On this day, however, there was another episode of bleeding which required the application of a tourniquet as well as of a pressure dressing.  The total plasma protein concentration was 5.8 gm. per 100 cc. and the hematocrit value was 35.5 percent.  On 26 June a right lumbar ganglionectomy, with excision of the second, third, and fourth lumbar ganglia, was done with the aid of spinal analgesia.

On 27 June, the right popliteal artery was explored with the patient under ether anesthesia.  After removal of a large organized clot with a pseudosac from the popliteal space, a through-and-through wound of the artery was found (Fig. 4).  After the artery was


Figure 4.   (Case 4.) Through-and-through woundofpopliteal artery. Partially excisedwall of hematoma is beingretracted.

freed it was evident that suture of the injured vessel was not possible. The artery was therefore ligated above and below the injuries with No. 1 chromic catgut, the damaged segment (about 1.5 inches) was excised, and the foreign body removed. The segment which was excised lay between the outlet of the adductor canal and the middle geniculate artery.  The vein was not thrombosed and was not ligated.  Five hundred cubic centimeters of blood were given in the course of the operation and 500 more after operation.

Before the operation the right foot was definitely warmer than the left, but on palpation neither the dorsalis pedis nor the posterior tibial artery was palpable.  Motion and sensation were normal except for mild hypesthesia to pin prick in the toes.  The day after operation the right foot was colder and dryer than the left foot.  Motion continued to be normal in the foot and toes.  No pulsations were palpable in the right posterior tibial and dorsalis pedis arteries.  Thereafter recovery was smooth, and by 30 June there was every indication that the injured extremity would survive. By 7 July when the patient was evacuated, the temperature of both feet was the same. The popliteal wound was well healed. The right calf was very slightly larger than the left, but no induration was palpable.

  Comment. There are a number of lessons to be derived from this case. One is that it is possible for pulsation to be present in the dorsalis pedis and posterior tibial arteries even when there is a through-and-through wound of


the popliteal artery.  Another is the advantage of delayed ligation of an injured artery. This patient would almost certainly have lost his leg if the popliteal artery had been ligated immediately after injury. On the other hand, this case clearly shows that evacuation of a patient with a wound of this character is hazardous. Early evacuation was permitted in this case because the wound was not recognized. Had it been recognized, and had expectant treatment been deliberately instituted in order to delay ligation of the artery, the patient would have been kept under continuous observation until the vessel had been ligated or the wound had healed, and the repeated hemorrhages could probably have been avoided.

Conservative therapy of vascular wounds was advocated as a deliberate policy by some British surgeons at the Congress of Central Mediterranean Forces Army Surgeons held in Rome in February 1945, 42 but only a few American casualties were deliberately treated in this manner. Sandzen and Evans 43 treated 3 cases conservatively with excellent results in a series of 89 vascular wounds, 64 of which involved major arteries, and Rose, Hess, and Welch 44 treated 8 of 100 cases conservatively, also with good results.  In 4 of the 8 cases the wounded vessels were exposed during debridement, but in the other 4 cases, in all of which the popliteal artery had been injured, they were not visualized.

  The selection of cases for conservative management required expert surgical judgment, as well as a good deal of courage. The chief objection to the method was that it usually implied the omission of debridement which was so essential a part of the management of all war wounds that exceptions to its routine performance had to be made with the greatest caution. An additional reason for carrying out complete debridement in any case was that it afforded an opportunity to repair a laceration by suture or to perform nonsuture anastomosis. If the artery was intact its continuity could be determined without dislodging it entirely from its bed. Generally speaking, it was best to explore even trivial wounds with the idea of performing remedial or reparative surgery if there was evidence of complete interruption of the circulation. The following case histories point out the advantages of exploration:

  Case 5. A soldier wounded by enemy shell fragments 20 May 1944 suffered a 2-cm. penetrating wound of the lateral aspect of the right knee with a laceration of the popliteal artery just below the geniculate branches.  Only a quarter of the anterior arterial wall remained intact. Examination in a forward hospital revealed the popliteal area and calf to be tense and swollen.  The right leg and foot were cold and motionless. No pulsations could be felt in the popliteal, dorsalis pedis, or posterior tibial vessels.

42 (1)  Brown, J. J. M.:  A plea for conservatism in the primary surgery of wounds of the main arteries of the limbs.  Proc. Cong. C. M. F. Army Surgeons, Rome, 12-19 Feb 45, p. 103-106.  Also discussions by Major Robb and Major Cleland.

  (2)  Erasmus, J. F. P.:  The problem of wound excision in the treatment of vascular injuries.  Proc. Cong. C. M. F. Army Surgeons, Rome, 12-19 Feb 45, p. 112-113. 
43 Personal communication to F. A. Simeone from S. C. Sandzen and B. H. Evans.

44 Rose, C. A.; Hess, O. W., and Welch, C. S.:  Vascular injuries of the extremities in battle casualties. Ann. Surg. 123: 161-179, Feb 1946.


  The patient was operated upon 7 hours after wounding. The arterial laceration was sutured after removal of a large hematoma and the deep fascia over the upper half of the swollen calf muscles was incised. The right lumbar sympathetic trunk was blocked with procaine hydrochloride. The right thigh and leg were enclosed in plaster with the knee flexed to 30 degrees.  No pulsations were felt in the posterior tibial and dorsalis pedis arteries at the end of the operation.

  The following day the ankle and dorsum of the right foot were warmer though the toes remained cold and blanched.  No pulse was palpable in the foot.  Right lumbar sympathectomy was done with no obvious immediate improvement.  Two days after operation the right foot and toes were warmer and there was partial return of sensation and motion.  No pulse was palpable in the foot.  Thereafter improvement was continuous until the patient was evacuated to the base 11 days after injury.  At this time, although pulsations had not returned in the dorsalis pedis or posterior tibial arteries, the circulation in the injured extremity appeared satisfactory.

  Comment.  The result achieved in this case would be good for any type of acute injury of the popliteal artery. Before operation the circulation in the extremity was apparently seriously impaired. The collateral vessels were intact, and it is impossible to be certain whether the improvement which began a day or two after operation could be attributed to the function of the sutured popliteal vessel or to the function of the collateral vessels which had been released from compression by removal of a constricting hematoma.

  Case 6.  A soldier was wounded by mine fragments 20 October 1944.  He incurred a wound with its point of entrance just above the head of the left fibula; the wound of exit was on the posteromedial aspect of the left thigh at a level 6 cm. above the wound of entrance. The popliteal artery was severed below the geniculate branches.  At the evacuation hospital the popliteal space and upper calf were found to be swollen and firm.  The left foot was cold and pulseless and only slight motion was possible.

  With the patient under endotracheal ether anesthesia, operation was performed 6 hours after wounding.  A large hematoma was evacuated from the popliteal space.  The retracted ends of the popliteal artery were identified and were approximated by slight traction and 30-degree flexion of the knee.  Thrombi were removed from both ends of the severed vessel and an end-to-end anastomosis was done with interrupted sutures of fine silk.  All wounds were debrided.  A long leg plaster was applied with the knee in 30-degree flexion. The left lumbar sympathetic chain was blocked with procaine hydrochloride.  At the conclusion of the operation the condition of the foot was unchanged.

  The day following operation the foot was still cold and pulseless but motion in the toes had improved.  Sympathetic block was repeated. On 22 October the foot and toes were warm and free, and motion was possible although the dorsalis pedis pulse was still not palpable.  By 2 November, when the patient was evacuated to a general hospital, all the toes were warm and sensation and motion in the whole extremity were practically normal.

Comment. The result in this case, in which the location of the lesion was approximately the same as in the preceding case, is also good for a laceration of the popliteal artery. In both cases the location was the best possible for maintenance of function of the collateral vessels. Since, however, improvement in the foot was delayed for 2 days after operation, it is impossible to be certain whether the popliteal artery functioned after suture or whether the good result is to be attributed to an adequate collateral circulation.


SupplementalTherapeutic Measures

  Posture.  The position of an extremity in which the blood supply is embarrassed as the result of disease or injury has long been a controversial matter.  Directives and circular letters issued for the care of arterial injuries reflected the conflicting opinions. Circular Letter No. 13, dated 15 May 1943, 45 included elevation of the part among the emergency measures for the control of hemorrhage, but advised that following ligation of a vessel the extremity should not be elevated but should be slightly depressed.  In December 1943, 46the Surgeon of the Fifth Army issued a directive in which it was stated that in all cases in which the blood supply to an extremity is impaired the extremity should be depressed. The Manual of Therapy issued in the European Theater of Operations, 5 May 1944, listed as a measure for the control of hemorrhage the following: "Elevation of the part. This should be done with due care of associated fractures, regardless of the local methods used.'  Circular No. 8, dated 10 March 1945, issued from the office of the Surgeon, Mediterranean Theater of Operations, stated: "The position of an extremity is important, as elevation may accentuate ischemia. A dependent position is preferable even if a moderate degree of edema appears to be the result."  'This warning was also repeated elsewhere."

  These altered instructions reflected the conflict of opinion on this subject which existed in civilian practice and can be attributed to failure to distinguish between peripheral edema per se and peripheral edema associated with acute arterial injuries. In these injuries, particularly in acute occlusions of the main blood supply to a limb, the blood flow to the part was impeded and had to be maintained through the collateral circulation. Elevation of the extremity above the level of the heart accentuated the ischemia by forcing the blood flow to overcome the amount of gravity pull created by this position. Against this argument was the fact that edema in the tissues, which under ordinary circumstances would be treated by elevation of the limb above the level of the heart, was undesirable in arterial injuries for a number of reasons:  It disrupted tissue and predisposed to fibrosis. It compressed and occluded small vessels.  It increased the barrier to the transport of metabolic products to and from the blood. The edema which occurred in battle injuries was therefore as undesirable as edema always is, but it was the result, in part at least, of ischemia which would be increased and not reduced by elevation of the limb above heart level. When confronted with this choice it therefore seemed more reasonable to maintain the extremity in a slightly dependent position, or possibly at heart level, and to regard the edema as the lesser of two evils. Alternate raising and lowering of the limb for brief periods was suggested as a solution of the problem, but careful judgment was essential to be certain that elevation did not do more harm than good by increasing the period of ischemia.

45  See footnote 22 (1),  p. 70.

46   Essential Technical Medical Data from Overseas Forces, Hq NATOUSA, 23 Dec 43.  HD: 350.05.

47  See footnote 28, p. 81.


  Some authorities were not in agreement with this point of view.  Cohen,48   for instance, advocated elevation of the limb on the ground that this position does not empty the arterial tree and cause capillary anoxia. He stated that elevation by diminishing venous pressure and increasing lymph flow prevents edema which will compress the capillaries. The experience of American surgeons, however, confirmed the desirability of placing the extremity in a slightly dependent position.

  Physiologic rest of the injured limb was essential both to reduce to a minimum the nutritional needs of the tissues and to limit infection and absorption of toxic by-products. These objectives were best accomplished by placing the extremity in a well-padded posterior plaster shell made by splitting a long tubular plaster cast. The limb could then be lowered onto a support off the side of the bed or kept in a dependent position by some other means.

  Oxygen Therapy.  Since lack of available oxygen is one of the factors responsible for death of tissue after vascular occlusion, oxygen inhalation seemed a reasonable method of supplying the lack in acute vascular injuries. An increase in the oxygen saturation of the blood could be easily effected by this means when oxygenation was impaired by thoracic wounds or similar causes.  Under normal respiratory conditions, however, the increase in the partial pressure of oxygen in the inspired air had little effect upon the oxygen content of the blood. The slight increase (15 percent) was made possible by an increase in the amount of oxygen dissolved in the plasma. Whether or not this increase resulted in a sufficient increase in the oxygen gradient between capillary blood and the rest of the tissues to affect the outcome in arterial injuries seemed doubtful, especially in view of the difficulties inherent in the use of the method under combat conditions. It was the consensus that better results would be achieved by devoting the same amount of effort to correction of oligemia and anemia with the objective of improving the oxygen-carrying capacity of the circulating blood.

  Anticoagulant Therapy. Anticoagulant therapy, in spite of its value in selected vascular surgery cases in civilian practice, had an extremely limited application in the military setup in overseas theaters. Its use, as already pointed out, was impractical immediately after wounding because the criteria of its employment, close clinical observation and repeated laboratory studies, could not usually be met.  Mere arrival of the patient at a hospital did not meet these criteria. For one thing, the vascular injury had to be the only injury; the use of an anticoagulant agent if the patient had other injuries was contraindicated for obvious reasons.  Moreover, while it may be theoretically possible in a civilian hospital or a Zone of Interior installation to employ heparin or some similar agent before operation and to control the clotting time during the procedure, this could not be regarded as either practical or safe in an overseas forward area.

48   Cohen, S. M.: Traumatic arterial spasm.  Lancet 1: 1-6, 1 Jan 44.


  Another, and perhaps the chief, reason for not employing anticoagulant therapy in most combat-incurred vascular injuries was that by the time the patient could be transported to a hospital equipped to administer it, too much time had elapsed for the measure to be of any benefit. Thrombosis usually had occurred, and the distal portion of the extremity had been deprived of blood for too long a period for the pathologic changes which follow in the absence of circulation to be reversible.  

  In spite of these obstacles to its routine employment, in a few cases of battle injuries in which anticoagulant therapy was indicated and in which conditions permitted careful, continuous observation and adequate laboratory checks, the method had a definite field of usefulness and was employed.

  Data are not available as to how often anticoagulant therapy was used in vascular wounds by American surgeons in World War II, although it is known that it was employed only rarely.  In 1 sample of 12 cases in which it was instituted (it was usually administered in Pitkin's menstruum) 49 as early as was considered feasible under the military conditions and in which careful studies were made, no significant advantages were observed from its use.50  E Langley 51 suggested a continuous saline drip as a more effective mode of administration than intermittent intravenous injection.  He also regarded the generally accepted dosage as insufficient.

  The Use of Refrigeration.  The concept of reducing the metabolism of an injured part by cooling to make metabolic activity commensurate with whatever circulation remains in the limb is theoretically sound and had attracted some attention before World War II.52 On the other hand, the wisdom of actual refrigeration of a limb in which the circulation is for any reason impaired is still open to question unless the objective is to control infection and diminish lymphatic absorption prior to amputation, or to permit amputation without anesthesia in an aged and debilitated subject.53

  Bruneau and Heinbecker 54 showed that under experimental conditions the clinical application of cold (6E C.) to infected tissues was of no therapeutic value per se, and when the application was continued for periods of 24 to 96 hours, there were definite changes in tissue hydration, pronounced vasodilatation, and a decrease in the growth-resisting action of the tissues toward bacterial

49 Loewe, L., Rosenblatt, P., and Lederer, M.:  A new method of administering heparin. Proc Soc. Exper Biol & Med. 50: 53-55, May 1942.

50 Simeone, F.:  Wounds of the arteries of the extremities.  Report to the Surgeon, Mediterranean Theater of Operations, Aug 1945. HD: 350.05.

51  Langley, G. F.:  Repair of ruptured popliteal artery, with note on heparin therapy after arterial suture. Surg. 31: 161-165, Oct 1943.

52 (1)  Blalock, A.:  Effects of lowering temperature of injured extremity to which a tourniquet has been applied. Surg. 46: 167-170, Feb 1943.

  (2)  Freeman, N. E.:  Influence of temperature on development of gangrene in peripheral vascular disease. Surg. 40: 326-333, Feb 1940.

53 (1)  De Takats, G., and Miller, D. S.: Use of direct heat and indirect heat to increase blood flow to the extremities. War Med. 2: 429-436, May 1942.

  (2)  Richards, V.:  Refrigeration anesthesia in surgery. Ann. Surg. 119: 178-200, Feb 1944.

54   Bruneau, J., and Heinbecker, P.: Effects of cooling on experimentally infected tissues. Ann. Surg. 120: 716-726, Nov 1944.


organisms. These changes became more severe as the period of refrigeration was prolonged. They concluded their article by stating:  "While the exact clinical limitations of the procedure remain to be determined, this would contraindicate, except for brief periods, the cooling to levels around 6? C. of infected limbs which one aims to save by conservative measures."  Large and Heinbecker 55 noted in 1944, as Brooks and Duncan 56 had previously noted, that healing progressed less satisfactorily after refrigeration, and they also called attention to the possibility of nerve injury following prolonged cooling of an extremity.

Throughout the war some observers continued to advocate the refrigeration of the wounded extremity. They argued that application of ice packs would gradually produce anesthesia in the injured part and allow painless transportation of the wounded soldier which is a considerable factor in the prevention of secondary collapse from continued pain, bacterial growth would be inhibited by refrigeration therapy because of decreased oxidation, and enzymes and toxins would be temporarily inactivated. A dirty wound would, therefore, be maintained in a stationary condition until debridement could be carried out. It was also pointed out that when refrigeration was continued following definitive treatment because of irreparable vascular damage it had to be continued until the collateral circulation was reestablished, vasospasm was overcome, thrombosed vessels recanalized, or until failure was evident. In any event, it had to be withdrawn slowly to prevent the rapid spread of gangrene or infection.

  Regardless of the logic orthe fallacy ofthis line of reasoning, refrigeration, as already pointedout, was only occasionally a practical procedure in hospitals of acombat zone, while thetransportation of a wounded soldier through channels of evacuation withrepeated replacement ofice packs on his wounded limb was a concept which no experiencedmilitary surgeon couldpossibly contemplate.

  Refrigeration was employed in 3 cases in the Mediterranean theater in which circumstances were favorable for a trial of this method. The femoral artery was the site of the wound in 1 case and the popliteal in 2 cases. Amputation was required in all 3 instances. These poor results are in accord with those reported by Auster' 57 and by Ottaway and Foote, 58 who gave the method a trial in naval casualties in the South Pacific, and by Snyder,59 who used it in the Mediterranean and European theaters. While the amputations necessary in these cannot be attributed to the method (it was employed with great

55 Large. A., and Heinbecker, P.:  Nerve degeneration followingprolongedcooling of an extremity.  Ann. Surg. 120:742-749, Nov 1944.

56  Brooks,B., and Duncan, C.:  The influence of temperature on wounds. Ann.Surg.114: 1069-1075, Dec 1941.

57  Auster,L. S.:  Combat injuries in the South Pacific. U. S. Nav. M. Bull.42:1009-1018, May 1944.

58 Ottaway, J. P., and Foote, J. J.:  Refrigeration of wounded extremities. U. S. Nav. M. Bull. 43:1041-1043, Nov1944.

59Snyder, J. M.: Vascular wounds; report of a series of 108 casesencountered in a forward evacuation hospital.Surgery 21: 77-85, Jan 1947.