U.S. flag

An official website of the United States government

Skip to main content
Return to topReturn to top



caution and careful observation) no evidence exists that this method has saved limbs after wounds of the major arteries. Indeed its prolonged use, as has already been indicated, can lead to damage of the tissues that it is expected to preserve. 

While the direct application of cold to the injured limb was not advocated, it was recommended that the temperature of an extremity to which a tourniquet had been applied should be lowered as much as was feasible short of actual freezing.

  The Use of Heat. While the application of cold to an injured limb was considered unwise and ineffective, the direct application of heat was also considered to be contraindicated. The direct use of heat had several undesirable effects:  It speeded up local tissue metabolism in a part in which the circulation was already inadequate for the demands upon it.  Since the heat could not be carried away by the vessels which were damaged, its effect was cumulative, and the possibility of damage to the tissues was increased.  Finally, heat increased capillary stasis and thereby increased local edema and delayed wound healing. For these reasons it was the practice to leave the injured limb uncovered and exposed to room temperature.  If the patient complained of cold the limb could be lightly covered with a wool or cotton cover, but it was never placed under bed covering.

  Conservation of the body heat was, on the other hand, an important part of treatment. This was accomplished by the use of blankets, or an electric cradle, or hot water bottles. Warming was frequently necessary because of exposure or as part of the treatment of shock, but, in addition, it had a beneficial effect on the wounded limb in that it relaxed the peripheral vessels, favored the development of a collateral circulation, and encouraged the relaxation of vasospasm.



  There was every reason for not omitting debridement in most battle-incurred wounds. No matter how innocent the wound might look from the outside, there might be considerable shattering and destruction of tissue in its depths which could be visualized only through adequate incisions in the skin and fascial planes. When a wound was thus explored it was common to find beneath even a small wound of entrance foreign bodies, bits of clothing, devitalized muscles, hematomas, and major vascular damage. The surgical excision of devitalized tissue was part of the resuscitative program; it minimized the incidence of gas bacillus and other infections, and it had a favorable effect on vasospasm of any degree which might be present in an injured artery which had not lost its integrity. This was first pointed out in Circular Letter No. 178, Office of The Surgeon General, dated 23 October 1943, and was reiterated at intervals thereafter until the end of the war.


Arterial Ligation

  The affected artery had to be ligated in the majority of the vascular injuries which came under the military surgeon's observation in World War II.  It was never the procedure of choice, but one of stern necessity required by the location, type, size, and character of the wound.   Not a great deal need be said about the optimal site of ligation. In World War II, when wound infection was usually a controllable complication and secondary hemorrhage was not the factor of risk which it was in World War I, there was seldom any justification for proximal ligation. While it is desirable theoretically to ligate the vessel at such a level as to avoid the creation of a blind pouch, the deliberate effort to do so frequently involves extensive dissection and may still further jeopardize the circulation of the injured limb.

  Ligation in continuity was not favored in World War II 60 suggested nonabsorbable sutures were placed well above and well below the site of injury or thrombosis and the damaged intervening segment was excised. This technique eliminated the dangers of secondary hemorrhage, thrombosis, and vasoconstrictor influences. Division permitted the ends of the vessel to retract and suppressed generalized vasospasm. Embolism at the site of ligature was also less likely to take place in a divided vessel.

  On the other hand, when it was possible to tie the artery just below a large branch rather than leave a blind end to which the blood flowed with each pulse beat, it was best to do so. Rogers 61 emphasized that there was less interference with the hand and fingers when the brachial artery was tied just distal to the origin of the superior profunda. Rogers also advised the rapid transfusion of 800 to 1,200 cc. of blood during ligation to increase the blood pressure and force the capillaries open.

  All the experience of World War II suggested that the best results were achieved in vascular injuries in which ligation was required if the operation could be delayed for at least several days. In selected cases the possible development of an aneurysm was therefore accepted as a calculated risk if delay could thus be achieved.

  Ligation of the Concomitant Vein. The chief difference in the technique of arterial ligation as it was performed in the two World Wars concerns the ligation of the concomitant vein along with the injured artery. The amount of space devoted in the literature to the discussion of this procedure and the emphasis put upon it in World War I and in the early years of World War II are curiously out of proportion to its value.

 60  See footnotes 22 (1), and (3), p. 70; and 29, p. 81.

61 Rogers, L.: Physiological considerations in vascular surgery; ligature of main arteries to the limbs.  M. J. Australia 1: 517-518, 19 May 45.

62   See footnotes 22 (1), p. 70; and 23, p. 71.


Therapeutic venous ligation has been known since the time of Hippocrates, as Brooks 63 points out in his review of the subject, and has always been widely practiced for venous disease.  Makins seems to have been the first to suggest that it be part of the treatment of acute traumatic arterial lesions.  His experiences in the South African War convinced him that the incidence of gangrene following surgery for traumatic arteriovenous fistula was less when both artery and vein were ligated than when only arterial ligation was done, but he did not record the observation until his Bradshaw Lecture in 1913. 64   Moreover, in his extensive article on vascular injuries of warfare published in 1916 65 he dismissed the procedure rather casually, with the remark: "With regard to the question of the danger of simultaneous ligature of the artery and vein, it may be added that this was done in one of the successful cases."

  In the Hunterian Oration, however, which was delivered in 1917 66 he advocated the deliberate ligature of the uninjured concomitant vein in cases of arterial occlusion, and at this time, as well as in his monograph 67 published after the war, he set forth the evidence on which he based his advocacy.  This evidence may be briefly summarized as follows:

1. The demonstration in varicose veins of the ease with which a compensatory balance is attained when blood is diverted from the larger channels.

  2. The absence of permanent vascular difficulties when the jugular and other large veins are ligated to prevent the diffusion of septic emboli.

  3. The possibility of survival after occlusion of the vena cava.

  4. Personal experience with arteriovenous fistula to the effect that quadruple ligation and excision are less dangerous than simple arterial ligation.

  5. Von Oppel's good results 68 with this technique in 6 cases of senile gangrene resulting from occlusion of the popliteal artery. (Von Oppel used this method because of his observation of the occasional good results which followed arteriovenous anastomosis in this condition and which he attributed to control of the venous circulation and consequent rise in the blood pressure of the limb.)

62 Brooks, B.: Surgical applications of therapeutic venous obstruction.Arch. Surg. 19: 1-23, Jul 1929.

63 Makins, G. H.:  The Bradshaw Lecture on gunshot injuries of the arteries.  Lancet 2: 1743-1752, 20 Dec 13. A1soBrit. M. J. 2: 1569-1577, 20 Dec 13.

64 Makins, G. H.: On the vascular lesions produced by gunshot injuries and their results. Brit. J. Surg. 3: 353-421, Jan 1916.

65 Makins, G. H.: The Hunterian Oration on the influence exerted by the military experience of John Hunter on himself and the military surgeon of today. Lancet 1: 249-254, 17 Feb 17.

67 See footnote 12, p. 64.

68  (1) Von Oppel, W. A.: Zur operativen Behandlung der arterioven?sen Aneurysmen. Arch. f. clin. Chir. 86: 31-52.1908.

  (2) Von Oppel, W. A.: ReduzierteBlutkreislauf. International Congress of Medicine, 1913. London, OxfordUniversity Press, 1913, pt 1, p. 189.

  (3) Von Oppel, W. A.: Wieting's Operation and der reduzierte Blutkreislauf (Wratschebnaja Gaz. 1913, No. 9). Zentralbl. f. Chir. 40: 1241, 2 Aug 13.

  (4) Von Oppel, W. A.: Die Theorie des umgekehrten and perversen Blutlaufs in den Extremit?ten (Russki Wratsch 1913, No. 15).Zentralbl. f.Chir. 40: 1242, 2 Aug 13.


  6. Drummond's experimental demonstration 69 that gangrene follows ligation of the mesenteric artery but does not follow ligation of the mesenteric artery and vein.

  7. Van Kend's experimental studies 70 which showed a local rise of blood pressure in the affected limb when the concomitant vein was ligated subsequent to occlusion of the artery.

This was the evidence on which Makins based his observations, and he advanced, in addition, two other reasons why ligation of the concomitant vein was of distinct advantage:  

  1.  The capacious main vein affords too ready a channel of exit for the diminished arterial supply, as well as an undesirable reservoir of stagnation.

  2. As the result of combined arteriovenous ligation, the smaller amount of blood supplied by the collateral arterial circulation is maintained for a longer time within the limb, with the result that there is an improvement in the conditions necessary to preserve its vitality.

  Aside from Makins' statistics, not a great deal of evidence for or against ligation of the concomitant vein can be found in the literature of World War I. In 1916 (that is, before Makins' observations were published), Sehrt 71 reported that when the artery alone was ligated the incidence of gangrene in the upper extremity was 7.8 percent, but when the concomitant vein was also ligated there was no incidence of gangrene. The corresponding figures for the lower extremity were 20.4 percent and 9.0 percent.  On the basis of this experience, the author concluded that ligation of the concomitant vein was of distinct value and that the  'impounding of venous blood in the extremity" was beneficial. It is not clear on how many cases his observations were based.  Propping 72 in 1917 attempted to provide experimental evidence to support the opinion that concomitant vein ligation is beneficial, and while the experiment was rather naive he concluded from it that ligation of the concomitant vein was beneficial since gangrene of the limb after ligation of an artery is the result of an imbalance between the amount of blood entering the extremity and the amount leaving it through the veins.

  The whole matter was fully discussed at the Inter-Allied Conference of Surgeons held in Paris in May of 1917, and on the basis of Makins' statistics (Chart 13) it was agreed that the concomitant vein should be ligated whenever a major artery was ligated, even if the vein itself was not injured. 73

69   Drummond, Hamilton: Quoted by Makins (footnote 12, p. 64).

70 Vna Kend: Sur les blessures des vaisseaux, 6o S?ance, 17 May 17. Arch. de M?d. et de Pharm. mil. 68:348-349, Jul-Aug Sep,1917.

71 Sehrt, E.: ?ber die kunstliche Blutleere von Gliedmassen and unterer K?rperh?lfte, sowie uber die Ursache der Gangr?n des Gliedes nach Unterbindung der Arterien allein. Med. Klin. 12:1338-1341,17 Dec 16.

72 Propping, Karl:  ?ber die Ursache der Gangr?n nach Unterbindung grosser Arterien. M?nchen. med. Wchnschr. 64: 598-599, 1 May 17.

73 (1)  Depage, A.:  Blessures des Vaisseaux. Comptes-Rendus de la Conf?rence Chirurgicale Interalli?e pour 1?tude des plaies de guerre, 6? Seance, 16 May 17.  Arch. de M?d. et de Pharm. mil. 68:324-328, Jul-Aug-Sep 1917.

  (2)  Jacob: Blessures des Vaisseaux. Comptes-Rendus de la Conference Chirurgicale Interalliee pour l'etude des plaies de guerre, 6o Seance, 16 May 17. Arch. de Med. et de Pharm. mil. 68: 328-340, Jul-Aug-Sep 1917.

  (3) Makins,  G. H.: Blessures des Vaisseaux. Comptes-Rendus de la Conf?rence Chirurgicale Interalli?e pour 1?tude des plaies de guerre, 6o Seance, 17 May 17.  Arch. de Med. et de Pharm. mil. 68: 341-348, Jul-Aug-Sep 1917.


Chart 13. Results of ligation of comparable arteries with and without ligation of concomitan veins in British casualties in World War I. These data include only those cases where for the same artery, there were examples with and without concomitant vein ligation.  

  This conclusion was a complete reversal of the attitude which had prevailed before World War I when the current opinion seemed to be that the prognosis for survival of a limb after interruption of a major artery was worse when the concomitant vein was injured and had to be ligated simultaneously. Under these circumstances, Jacobson 74 wrote, "Leave should be gotten at once for amputation," while Matas 75 declared that "the danger of peripheral gangrene is always made doubly worse by the simultaneous injury of the accompanying or satellite vein."

  Even after World War I, all surgeons did not accept the new point of view unreservedly. In 1921 Punin 76 on the basis of 64 personal and 1,057 collected cases of arterial injury concluded that the incidence of gangrene was no less after the combined procedure than it was when ligation of the vein was omitted. In the same year Maurer 77 stated guardedly that concomitant

74  Jacobson, W. H. A.: The Operations of Surgery, 6th edition by R. P. Rowlands & Philip Turner, London, J. & A. Churchill, 1915, vol 1, p. 843, 894.

75   Matas, R.:   Surgery of the vascular system.  In Keen, William W.:  Surgery-Its Principles and Practice, by Various Authors.Philadelphia and London, W. B. Saunders Company, 1921, vol 5, pp. 17-350.

76 Punin:  Cited by Franz (footnote 17, p. 64).

77 See footnote 14, p. 64.


vein ligation is not harmful and might be advantageous (in 193978 he was even less enthusiastic). The weight of Makins' prestige, however, was so great that his point of view was usually accepted without question, and the practice of ligation of the concomitant vein became the rule when certain arteries, at least, had to be ligated.  

  Brooks,79 in 1929, in his extensive review of the subject decided on the basis of clinical and experimental evidence 80 that the concomitant vein should be ligated when the popliteal or axillary arteries were injured but when the common femoral artery was injured he believed it wiser to close the wound without ligating the vein and to watch the extremity carefully for signs of impending gangrene. The vein should be ligated only if such signs became evident. When the femoral and brachial arteries were damaged he believed that it made little difference whether or not ligation of the concomitant vein was done. In 1933 Wilson 81 published an article which controverted the opinion that ligation of the concomitant vein diminished the incidence of gangrene following ligation of the main artery. If venous ligation was done at a higher level than the arterial ligation, he concluded, both the incidence and extent of tissue death would be increased. This study received indirect support from the observations which Montgomery 82  had published in 1929 which showed that the perminute flow of blood to the extremity was still further reduced if the concomitant vein was ligated in an extremity in which the artery had previously been ligated. Then, in 1934, Brooks 83 and his co-workers published an article on their results of an experimental study of 220 rabbits which showed that massive gangrene of the extremity was 14.5 times less frequent of ter arterial and venous ligation than after arterial ligation alone.

  While the experimental evidence for and against ligation of the concomitant vein is thus rather inconsistent, there seems no doubt, as Wilson 84 pointed out, that Makins' reasoning in favor of the procedure was not based upon sound physiologic concepts. Moreover, Makins' frequently quoted statistics,85 when they are carefully examined, do not seem to warrant the sweeping conclusions

78 Maurer, A.:  Note sur les plaies vasculaires r?centes et leur traitement d'apr?s 444 observations. M?m. Acad. de chir. 65: 1156-1164, 25 Oct-8 Nov 39.

79 See footnote 63, p. 103.

80 (1) Brooks, B., and Martin, K. A.: Simultaneous ligation of vein and artery; an experimental study. J. A. M. A. 80: 1678-1681, 9 Jun 23.

  (2) Holman, E., and Edwards, M. E.: A new principle in the surgery of the large vessels; ligation of vein proximal to site of ligation of artery; an experimental study.J. A. M. A. 88: 909-911, 19 Mar 27.

  (3) Pearse, H. E., Jr.: A new explanation of the improved results following ligation of both artery and vein. Ann. Surg. 86: 850-854, Dec 1927.

  (4) Theis, F. V.: Ligation of artery and concomitant vein in operations on large blood vessels. Arch. Surg. 17: 244-258, Aug 1928.

81  Wilson, W. C.: Occlusion of the main artery and main vein of a limb. Brit. J. Surg. 20: 393-411, Jan 1933.

82  Montgomery, M. L.: Effect of therapeutic venous ligation on blood flow in cases of arterial occlusion. Proc. Soc. Exper. Biol. & Med. 27:178-179, Nov 1929.

83  Brooks, B.; Johnson, G. S., and Kirtley, J. A., Jr.: Simultaneous vein ligation; an experimental study of the effect of ligation of the concomitant vein on the incidence of gangrene following arterial obstruction. Surg., Gynec. & Obst. 59: 496-500, Sep 1934.

84 See footnote 81, above.

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


which have been drawn from them. For one thing, it is not possible to separate the cases of acute arterial injury from the cases of arteriovenous fistula, nor is it possible to determine the proportions of aneurysms in the two groups of cases. For another, the. difference between the incidence of amputations in the series in which only arterial ligation was done and the series in which the concomitant vein was ligated also is not statistically significant. Finally, the incidence of amputations in the whole group of wounds of arteries in Makins' collected World War I series (Chart 13) is actually less than the incidence of amputations in the comparable series in which the concomitant vein was ligated, though the former series, on the basis of his theories, should provide the larger number (or at least an equal number) of poor results.

  Makins' concepts were undoubtedly responsible for the persistence into World War II of the idea that concomitant venous ligation was useful in arterial injuries.Hoche,86 in 1940, warned against ligation of certain arteries, such as the popliteal and the femoral artery above the branching off of the deep femoral, without simultaneous ligation of the concomitant vein on the ground that disturbances resulting from arterial injury and ligation seemed to result not so much from failure of the collateral circulation as from a disparity between the collateral arterial inflow and the venous discharge. Decker,87 in 1941, recommended ligation of the jugular and popliteal veins when the corresponding arteries were ligated. He suggested, however, when other vessels were concerned, that temporary pressure be applied to the vein after the artery had been ligated and that the subsequent procedure depend upon developments:  If after a few minutes the peripheral circulation was found to be favorably affected by this maneuver, ligation of the appropriate vein should be done, while if no effects were observed it should not be done.

  Although ligation of the concomitant vein was recommended in circular letters and other material issued to American medical officers in World War II as late as D-day in Europe,88 the American experience with the method was not extensive (Chart 14). The majority of surgeons did not use it routinely, if at all, and the experience of no single surgeon was large enough to permit valid conclusions. The collected figures seem to indicate that ligation of the concomitant vein does not in any way increase the chance of survival of the limb. The difference between the incidence of amputations in the cases in which the vein was ligated and in the cases in which it was not ligated is not statistically significant. As in the British figures for World War I, the incidence of amputation in the total group of arterial wounds was considerably less than it was in the comparable group in which concomitant vein ligation was done. The difference, in fact, is statistically significant, although if the procedure were of

86 Hoche, O.:  Die wehrchirurgische Behandlung der Verwundeten. Die Blutung and ihre Behandlung. Med. Klin. 36: 919-922, 16 Aug 40.

87  Decker, P.: L'h?nostase d'urgence en chirurgie de guerre. Helvet. med. acts 8: 3-21, Apr 1941.

88  See footnote 23. p. 71.


Chart 14.   Results of ligation of comparable arteries with and without ligation of concomitant veins among British casualties in World War I and American casualties in World War II. These data include only those cases where, for the same artery, there were examples with and without concomitant vein ligation.

definite value one would expect the resultsof venous ligation to be at least as good, if not better.

  The conclusion therefore seems warranted that ligation of the concomitant vein in combat-incurred arterial wounds furnishes no protection whatsoever against the development of gangrene after acute arterial occlusion treated by ligation.

Primary Amputation

  Circular letters and directives issued in the Mediterranean and other overseas theaters 89 and information disseminated from the Surgeon General's Office 90 limited primary amputation to cases in which the extremity was irreparably damaged or devoid of circulation. Primary amputation was also permitted in an occasional case for the control of hemorrhage or as the first

89 (1) See footnote 23, p. 71.

(2) Cir Ltr 101, Off of Chief Surg ETO, 30 Jul 44, sub: Care of battle casualties.
  (3) Cir Ltr 23, Off of Chief Surg ETO, 17 Mar 45, sub: Care of battle casualties.

90 See footnotes 28 and 29, p. 81.


step of treating a traumatic amputation whichis not complete. The site of amputation was thelowest possible level of viable tissue regardless of the utility of thestump. The warning wasfrequently issued that the line of demarcation in an infected extremitywhich was the site of avascular occlusion did not mark the level at which an amputation stumpcould be maintained,since circulation sufficient only to maintain tissue viability couldnot cope with infection.

  When amputation was necessary the entirely proper desire to delay it in order to permit the establishment of an adequate collateral circulation had to be weighed against the possible systemic reaction, the local evidence of infection, and the practicability of keeping the patient under observation. In an active theater of war the latter consideration was of great importance. At echelons in which evacuation might be necessary on short notice, it was sometimes better to amputate a limb in which gangrene had developed as promptly as possible rather than risk complications during the time of evacuation when the patient was not being closely observed.

Suture Repair

  World War I.  On theoretical grounds, repair of a damaged artery by suture offers the chief hope of survival of the limb but practically, as has been pointed out, reparative measures are seldom applicable in military surgery. The value of suture was recognized during World War I but, as in World War II, the number of cases to which it could be applied was very small. It is the only method, Makins 91 stated, which provides ideal results, but he added that it is applicable only in the primary state of vascular injuries and only if infection can be avoided. He regarded lateral wounds of the carotid, femoral, and popliteal arteries as the most suitable sites for its performance and placed the axillary and brachial arteries in the next category. Sencert 92 also wrote that ligature is the method par excellence for the arrest of hemorrhage from recent vascular wounds, but he added, like Makins, that the indications for its performance are few.  
  Bernheim's experiences 93 in World War I were illuminating. He had enthusiastically practiced the Carrel 94 method of suture in his civilian practice and went to France with elaborate personal equipment in order to use the method in military surgery. In nearly 2 years overseas, however, he never saw any other surgeon suture an artery and he himself discontinued it in the few cases in which he attempted it because of loss of supporting tissue as the result of necessary debridement, and because of the unjustifiable amount of time which the operation required. Even in cases in which infection was

91  See footnote 3, p. 60.

92  See footnote 38, p. 84.

93 (1) Bernheim, B. M.:  Blood-vessel surgery in the war. Surg., Gynec. & Obst. 30: 564-567, Jun 1920.

  (2) Bernheim, B. M.:  Symposium on emergency treatment; management of vascular injuries. S. Clin. North America 22: 1417-1426, Oct 1942.

94 Carrel, A.: Surgery of blood vessels. Bull. Johns Hopkins Hosp. 18: 18-28, Jan 1907.


absent he thought that military circumstanceswere unpropitious for vascular suture, while "onlya foolhardy man," he remarked, "would have assayed suture of arterialor venous trunks in thepresence of infections such as were the rule in almost all the injured."

Goodman's experience 95 with vascular suture made him, like Bernheim, enthusiastic about the method, and after a months stay on the British Front in 1917, he felt "enabled to refute the deductions made by the other surgeons present," these being that the risk of gangrene after arterial ligation was sufficient to justify immediate amputation in injuries of the femoral and popliteal arteries, and even the posterior tibial artery. Goodman's personal experience with vascular suture, however, was limited to 5 cases, in 1 of which amputation was later required because of gangrene. In addition to his own cases, he collected a number of others, but the large proportion of aneurysms included limits their values for comparative purposes. Many cases in the collection were of German origin and there is no doubt that German surgeons used suture more frequently than either British or American surgeons in World War I, though many series, such as that reported by von Haberer, 96 consisted entirely or chiefly of aneurysms. Makins 97 was able to collect only 39 cases of arterial suture from the British experience. Three patients died, all from infection, and ideal to good results were obtained in about one-half of the remaining cases, the results being "in no way inferior to those of ligature," which seems somewhat faint praise. Gnilorybov, 98 writing in 1944, stated that during World War I he had performed vascular suture in a number of cases (he did not cite exact figures) and that he saw no instance of gangrene after its use, in contrast to ligation, after which half of his patients developed gangrene.

  Decker 99 furnishes an interesting and probably correct interpretation of why arterial suture was practiced in the early part of World War I by the surgeons of the Central Powers but infrequently during the second half. This was not the result of the tactical situation, although mobile warfare gave way to stationary trench warfare during the course of the war, but of a change in the character of the missiles used. At the beginning of World War I projectile wounds were usually inflicted from a distance, the explosive effect in the tissues was minimal and vascular injuries were therefore frequently suitable for repar-

95 (1)  Goodman, C.:  A histological study of the circular suture of the blood vessels. Ann. Surg. 65: 693-703, Jun 1917.

  (2)  Goodman, C.:  Suture of blood-vessel injuries from projectiles of war. Surg., Gynec. & Obst. 27: 528-529, Nov 1918.

  (3)  Goodman, C.:  Surgery of the Heart; Blood Vessels, Thrombosis and Embolism and Blood Transfusion.Cinq. Cong. de Soc. Internat. de Chin, Paris, 19-23 Jul 20, p. 244-263. Brussels, M. Hayez, 1921.

  (4)  Goodman, C.: Suture repair of war injuries to the blood vessels.  J. Internat. Coll. Surgeons 6: 127-131, Mar-Apr 1943.

96 (1) Von Haberer, H.: Diagnose and Behandlung der Gef?ssverletzungen. Munchen. med. Wchnschr. 65: 363-367; idem, ibid.: 405-409, Apr 1918.

  (2) Von Haberer, H.: Surgery of the blood vessels in wartime. Munchen. Med. Wchnschr. 87: 849-851, 9 Aug 40. Abstr., Internat. S. Digest 31: 21-24, Jan 1941.

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

98  Gnilorybov, T. E.: One Hundred and Thirty Operations on the Blood Vessels. Khirurgiya No. 8, 40-35 (sic), 1942. Moscow. Abstr., Bull. War Med. 4: 452-453, Apr 1944.

99 See footnote 87, p. 107.


ative surgery. Later in the war when bombs,mines, and similar weaponsprevailed and weredischarged at close range, this favorable circumstance ceased to exist.

  World War II.   In World War II the circumstances were propitious for a trial of arterial suture in that thorough debridement, supplemented by chemotherapy and antibiotic therapy, greatly reduced the risk of infection. Nevertheless, this method was not very widely employed. One reason was that it is not a technique suitable for all types of wounds. Although it can be employed in lateral injuries of some size, or in incomplete or complete transections, the most favorable type of injury is the small lateral wound. A second reason was the frequently uncontrollable length of the time between wounding and treatment, an unavoidable military circumstance.

  Case 7. A soldier who was wounded by enemy shell fragments in Italy, 5 January, incurred a wound of the right popliteal fossa with laceration of the popliteal area. There was no fracture. The laceration was sutured on 7 January at an evacuation hospital and right lumbar sympathetic block was done.

  Between 11 January and 20 January four different operations were performed on the patient for local excision of gangrenous skin and muscle. On 2 February, at a general hospital, a guillotine-type of amputation was done for infection and loss of tissue caused by ischemia.

  Comment.  The details are lacking in this case, the history of which was secured from proceedings of a disposition board, but the most significant fact is known, that arterial suture was done more than 24 hours after wounding. This is well past the optimum time period for repair and a good result could scarcely have been expected.

  Arterial suture was performed in only 81 of the 2,471 arterial wounds analyzed in this chapter (Chart 15). Included in the group are 3 end-to-end anastomoses, which were performed on the common femoral, femoral, and popliteal arteries. Most of the other wounds were small lateral lacerations, involving a third or less of the circumference of the vessel. One was a wound inflicted accidentally with a bayonet, a type ideally suited for suture because the incision is clean and there is no great loss of tissue.

  The results of suture in these 81 cases are significantly better than the results of ligation and vein graft and probably better than the results of tube anastomosis (Chart 15). That fact, however, does not offer any great encouragement, for these 81 cases were a highly selective group of minimal wounds without extensive tissue destruction. It would not be possible to duplicate, or even approach, these results in the usual run of arterial wounds.

  Killian's observations100 on arterial suture in the German Army during World War II are not helpful in this connection because, while it was performed in 52 of 72 surgical cases, an unspecified number of these were aneurysms. Recovery occurred in 50 cases, in only 5 of which amputation was necessary. The favorable results in these instances of arterial suture and early operation for aneurysms showed, according to Killian, that suture is a method which can

100  Killian, H.: Uber die Indikation zur Fruhoperation von Gef?ssverletzungen and Aneurysmen. Arch. f. klin. Chir. 204: 355-410, 26 Mar 43.


Chart 15.   Results of various therapeutic measures on incidence of amputation in arterial wounds of the extremities in American battle casualties in World War II, with special reference to the type of arterial repair. 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. Detailed treatment of the remaining 697 American cases shown in Table 8 consisted of amputation as a primary procedure, or is unknown.

be practiced by the field surgeon at thebattlefront. Unfortunately, he does not define his terms,and this fact, together with the apparently large proportion ofaneurysms included in the series,makes his observation of little value.

  Complications. Thrombosis and embolism are immediate, but infrequent, complications of arterial suture. They did not appear in this series nor did any patient present signs of either arterial stricture or aneurysm during the period he was under observation. The most important immediate complication of arterial suture is hemorrhage. The records are incomplete on this point but it is known to have occurred in 2 of 24 cases in which it was possible to determine its presence or absence. It usually occurs 6 to 8 days after operation by which time a collateral circulation has developed and ligation can be done with much less risk than when it is a primary procedure. In the following case the hemorrhage which occurred shortly after operation can be considered a complication of heparinization rather than of the arterial suture.


  Case 8. A soldier wounded by enemy mortar shell fragments in Italy, 17 April 1945, incurred penetrating wounds of the face, both thighs, and the right inguinal region. The latter wound caused a laceration of the common femoral artery, 4 to 5 mm. long, at the junction of the femoral and profunda femoris arteries and near the origin of the circumflex iliac artery. There were no fractures. Sulfanilamide powder was dusted into the wounds and compression bandages were applied. When the patient reached an evacuation hospital about 4 ? hours after wounding, penicillin was administered promptly (25,000 units intramuscularly at 3-hour intervals).  

  The operation was performed, using drop-ether anesthesia, 10 hours after wounding and lasted 3 hours. The blood pressure was normal throughout and the pulse rate varied between 100 and 120 beats per minute. Debridement and exploration of the right groin revealed a large hematoma and the laceration of the femoral artery already described. It was sutured with fine silk. During the procedure hemorrhage was controlled by means of pressure supplemented by a tape placed under the common femoral artery above Poupart's ligament and kept in place for about an hour.  The patient received 1,000 cc. of whole blood before operation and another 1,000 cc. during operation. The soldier had been wounded before, but had never received a transfusion. A plaster spica was applied with pressure over the wound. At the end of the operation the dorsalis pedis and posterior tibial pulses were good.

  Heparin in Pitkin's menstruum, 2 cc., 200 mg. heparin, was given 5 hours after operation and later at an unrecorded hour 1 cc. was given. The following day it was given in 2-cc. amounts at 0200 and again at 0900 hours. A transfusion of 1,000 cc. of whole blood was started 5?hours after operation and completed after midnight.  Heparin was discontinued when bleeding occurred from the wound in the evening of the day after operation.

  The urine voided on the day of operation was described as clear, but was not examined. Thirty-one hours after operation, however, the patient voided bloody urine (Table 5). Urinalysis at 0900 and 1000 revealed protein to be present to which a 4+ rating was ascribed. When the patient was questioned he stated that he had never been ill in the past, but he thought that during the past year he might have had "kidney trouble" because he had suffered from lumbar pain and nocturia.  Examination of the urine showed no urobilinogen or urobilin. The serum urea was 98 mg. per 100 cc. and the urea nitrogen 46 mg. per 100 cubic centimeters. The blood pressure was 120 mm. of mercury systolic and 90 diastolic.

  The third day after operation (20 April) the blood pressure was 130 mm. of mercury systolic and 84 diastolic. The patient was drowsy but was easily aroused.  The urine voided was clear (Table 5) .Circulation in the right foot was normal.

  Recovery thereafter was smooth. Transfusions of 500 cc. each were given without incident on 21 and 22 April. On 27 April the wounds were sutured. When the patient was evacuated to the Zone of Interior 29 May, his wounds were well healed and the circulation in the right leg appeared entirely normal.

  Comment.  It was stated in this case that at no time, before, during, or after operation was it ever impossible to feel pulsations in the dorsalis pedis and posterior tibial arteries. There seems no doubt that suture of the lacerated artery was efficacious. Of special interest are the appearance of probable myoglobin in the urine and the transient azotemia (Table 5).

Sutured Vein Grafts

The extensive loss of substance caused by the extremely destructive weapons used in World War II meant that in certain of the few vascular injuries in which arterial ligation did not seem to be immediately indicated and in which suture repair was clearly impossible, some method of bridging a sizeable gap in the wounded artery had to be used in order to reestablish continuity of the artery.




  Sutured, as opposed to nonsutured, vein grafts had been used with considerable success both experimentally 101 and clinically 102 before World War I and the method was used with surprisingly good results for traumatic aneurysms in that war,103 though apparently it was not used for acute arterial occlusion. The operation was successful in 40 of the 47 traumatic aneurysms reported by Warthm?ller,104 though, as Matas105 emphasized, there was small reason for its use in aneurysms in the light of the successful results achieved by his own technically simpler procedure. Suture vein grafts do not seem to have been used by American surgeons in World War II but a few successful cases (chiefly aneurysms) were reported by Rehn,106 Murray,107 Killian,108 Schneider and Batzner,109 and Khenkin.110

Maintenance of the Main Arterial Channel by Prosthetic Devices  

  Bridging of the arterial gap by intubation to provide for temporary maintenance of the blood flow seems first to have been employed clinically in 1915 by Tuffier.111  He had used silver tubes in performing direct blood transfusion and, on the basis of his experience with them, proposed their use for bridging arterial defects. He later reported a number of successful cases. Makins112

101 (1)  See footnote 95 (1), p. 110.

(2)  Guthrie, C. C.:  End-results of arterial restitution with devitalized tissue.  J. A. M. A. 73:186-187, 19 Jul 19.

(3)  Moure, P.: Technique des greffes vasculaires appliqu?es au r?tablissement de la continuit? des art?res.J. de chir. Paris 9: 541-560, 1912.

(4)  Moure, P.: Greffe vasculaire; preuve de la vitalit? du greffon par une double greffe. Bull. et m?m. Soc. anat. de Paris 88: 489-492, 1913.

102 (1)  Lexer, E.: Die ideate Operation des arteriellen und des arteriell-ven?sen Aneurysma. Arch . F. klin. Chir 83:459-477,1907.

(2)  Warthmuller, Hans. ?ber die Bisherigen Erfolge der Gefasstransplantaion am menschen. Jena, G. Neuen bann, 1917, p. 39.

103 (1)  Alessandri, R.: Chirurgia del cuore e Dei Grossi Vasi. Cinq. Cong. de Soc. Internat. De Chir., Paris, 19-23 Juillet 1920, p. 139-241. Brussels, M. Hayez, 1921.

(2)  Von Bonin, G.: Aneurysmen durch Schussverletzungen and ihre Behandlung. Chir. Hefte 2) 97: 146-176, 1915.

(3)  Hotz, G.: Zur Chirurgie der Blutgefasse. Beitr. z. klin. Chir. 97: 177-188, 1915.

(4)  Renaux, L.: Reconstruction of Popliteal Artery. Brazil-med. 34: 509, 7 Aug 20. Abstract  J. A. M. A. 75: 1756, 18 Dec 20.

(5) Sencert, L.: La Chirurgie des gros vaisseaux: 1. Les Blessures des vaisseaux. Cinq.  Cong. de Soc. Internat. de Chir., Paris, 19-23 Juillet 1920, p. 77-137. Brussels, M. Hayez, 1921.

(6)  Zahradnicky, F.: Die Bebandlung der unecbten Aneurysmen. Wein. klin. Wchnschr. 28: 999-1007,1915.
  (7) See also footnotes 95 (1) and (4), p. 110; and 102 (2), above.

104  See footnote 102 (2), above..

105 Matas, Rudolph: Military surgery of the vascular system.  In Keen, W. W.:  Surgery-Its Principles and Practice, by Various Authors.Philadelphia and London, W. B. Saunders Company, 1921, vol 7, p. 713-819.

106(1)  Rehn, E.: Das Aneurysma der Arteria subclavia durch Schussverletzung and seine Ideal Operation Venenuberpflanzung.Zentralbl. f. Chir. 69: 1262-1275, 1 Aug 42.  Abstr., Bull. War Med. 3: 264, Jan 1943.
(2) Rehn, E.: Grunds?tzliche Bemerkungen zur rekonstruktiven Chirurgie des Schussaneurysma. Zentralbl. f. Chir. 70: 957-965, 3 Jul 43.

107  Murray, G. D. W.:  Heparin in thrombosis and embolism.  Brit. J. Surg. 27: 567-598, Jan 1940.

108   See footnote 100, p. 111.

109  Schneider, H., and B?tzner, K.: Die Behandlung der Aneurysmen der Arteria poplitea und ihrer ?ste. Zentralbl. f. Chir. 70: 965-975, 3 Jul 43.

110  Khenkin, V. L.: Experience in treatment of injuries of blood vessels in the forward area and at the rear. Khirurgiya, Moscow 2: 50-57, Oct 1944. Abstr., Bull. War Med. 6: 75,1945-46.

111  Tuffier: De 1'intubation dans les plaies des grosses arteres.  Bull. Acad. de m?d., Paris 74: 455-460, Oct 1915.

112   See footnote 3, p. 60.


reported 12 operations by this technique in 1922, 4 of which were nonacute lesions. One patient died of sepsis and another of gas gangrene, but results in the other cases were good. Makins also mentioned in this report Cowell's case, in which, when signs of gangrene followed ligature of a completely divided femoral artery, the limb was saved by removal of the ligatures and the use of a Tuffier tube. A number of other observers during this period reported favorable results with this method.113

  Early in World War II British and Canadian surgeons attempted the use of glass tubes for arterial injuries, but their results are not known.114  Their potential clinical value was indicated by the successful experience of Murray and Janes115 in their experiments with heparin on dogs.

  Plastic tubes were similarly used by American surgeons, though the results in the 14 cases in this series were not outstanding (Chart 15).  Plastic prostheses have certain advantages over other prostheses in that they are apparently well tolerated by the tissues and can be altered in size and shape to fit the necessities of the special case merely by soaking the basic material in warm water.  A supply of various sizes therefore need not be kept on hand. The technique of repair is simpler than when vitallium tubes are used, though the possibility of thrombosis is theoretically greater than when vein grafts are used.

  Whether or not a permanent circulation is maintained through these tubes is not a matter of extreme importance if in the interim a collateral circulation has developed, since gradual occlusion of a vascular channel is always less deleterious than abrupt occlusion. This result cannot be achieved, however, unless operation is performed early, and the timelag offers an obstacle to this desideratum in most combat-incurred vascular injuries.  The importance of some form of bridging of the arterial defect to permit at least a minimal circulation while collateral vessels are developing is illustrated by the following case report.  Collaterals were adequate by the time the nonsuture graft thrombosed.

  Case 9. This soldier, wounded by an enemy bomb fragment on shipboard off Southern France 18 August 1944, incurred a perforating wound of the right popliteal space with a laceration of the popliteal artery at its bifurcation into the anterior and posterior tibial arteries. He also received penetrating wounds of the right thigh and hand. There was no fracture. First aid consisted of morphine, tetanus toxoid, dusting of the wound with sulfanilamide powder, and the administration of sulfadiazine by mouth (2 gm. at once and 1 gm. every 4 hours). The following day, after the patient was transferred to a hospital ship, penicillin therapy was begun.

113 (1)  Donati, M.: Ferite del vasi sanguigni degli arti. Chir. d. organ. d. mov. Bologna 1: 191-254,1917.

  (2) Tuffier: A propos des plaies des art?res. Bull. et m?m. Soc. de chir. de Paris 43: 1469-1741, 1917.

  (3) Tuffier:French surgery in 1915. Brit. J. Surg. 4: 420-432, Jan 1917.

(4)  See also footnotes 3, p. 60; 13, p. 64; 95 (3), p. 110; and 103 (3), p. 115.

114 Matheson, N. M., and Murray, G.:  Recent advances and experimental work in conservative vascular surgery.  In Bailey, Hamilton: Surgery of Modern Warfare.Baltimore, Williams and Wilkins Co., 1941, vol 1, p. 324-327.

115 Murray, G., and Janes, J. M.:  Prevention of acute failure of circulation following injuries to large arteries; experiments with glass cannulae kept patent by administration of heparin.  Brit. M. J. 2: 6-7, 6 Jul 40.


  When he was admitted to a general hospital in the base on 21 August his right leg was nearly twice normal size and the calf was tense. The toes and ankles were discolored and mottled. The foot was numb, cold to touch, and motionless as far as the ankle. The skin of the foot was mottled but blanched fairly easily. No pulsations could be felt in the leg or foot. There was moderate pain in the leg. The popliteal space was full but soft. Auscultation revealed a systolic bruit in this area; diastole was clear. The hematocrit value was 30 percent and the total plasma protein concentration 7.15 gm. per 100 cubic centimeters.  

  A transfusion of 1,000 cc. of whole blood was given and conservative treatment was decided upon for the time being.

  The following day (22 August) there was a sudden, brisk hemorrhage from the small lateral wound behind the head of the fibula at the knee with an estimated lossof about 1,000 cc. ofblood. Shock was moderate. Bleeding was controlled by the applicationof a temporarytourniquet which stopped the bleeding in 25 minutes. Seven hours afterthe hemorrhage the footwas ice cold, the mottled areas were more prominent, and it wasapparent that in all likelihoodthe limb would not swas 35 percent anurvive. The bruit in the poplitealspace could nolonger be heard. Thehematocrit valued theplasma protein concentration7.5 gm. per 100 cubiccentimeters.

  The vascular operation was preceded by lumbar ganglionectomy, with resection of the second, third, and fourth lumbar ganglia. When the right popliteal space was explored the gastrocnemius muscle herniated through the incision in the fascia. The fibers of the heads of the muscle were separated, very pale, and putty-like in consistency. Hematoma formation was evident beneath the heads. The tibial nerve appeared contused, but was intact. The popliteal vein had been severed, and the popliteal artery was also severed except for a narrow strand of the wall on the deep aspect which kept the ends from retracting. The artery was open for a distance of about 1? inches, including the area of its bifurcation. A thrombus was aspirated from the posterior tibial artery. The severed popliteal vein was ligated at each end, and a nonsuture anastomosis was done between the popliteal and posterior tibial arteries bridging a gap of about 2 inches. A segment of the left saphenous vein was used. The foot was warmer and of better color after operation than before. Heparin was not available.

  The day after operation the entire right leg was edematous and there was some redness in the upper half of the leg. Both foot and leg were warm and blanched fairly well. The patient's temperature was 102 o F. and his general condition was fair. A transfusion of 1,000 cc. of whole blood was given.

  By 30 August, the sixth day after operation, the acute edema of the leg had subsided and the foot was warm. On the external lateral aspect of the heel was a small round area of discoloration. All sutures were removed from the right flank and left thigh. The wounds of the right popliteal space and right space were sutured. There was still a small amount of slough on the surface of the lateral head of the gastrocnemius, but beneath it was normal-looking muscle. The skin edges came together easily.

On 2 September an arteriogram with diodrast revealed no flow through the anastomosis but an excellent collateral circulation was demonstrable (Fig. 5).  On 5 September faradic stimulation of the leg muscles was done. Reactions were obtained only from the gastrocnemius and posterior tibial muscles, but edema still present in the leg made interpretation of the negative results uncertain. On 6 September all sutures were removed.  Healing was satisfactory. The foot was warm, pink, and dry. The small round spot of necrosis on the heel was thought to be the result of pressure while the posterior plaster shell was in place.
  On 17 September the right calf was still indurated and enlarged. There was no motion in the toes but the patient had begun to feeltingling in the dorsum of the foot. There was anesthesia to 3 inches above the lateral malleolus.  No further observations were possible after this date.


Figure 5.  Laceration of popliteal artery treated by nonsuture anastomosis. Arteriogram 9 days after operation showing excellent collateral circulation. Thrombosis now blocks the anastomosis, which may have been patent earlier, however.

Comment. In this case ischemia, which had been only partial for 3 days after wounding suddenly became complete and it seemed that the limb would not survive. The circulation improved following nonsuture anastomosis (vitallium tubes and vein segynent), though an arteriogram 11 days after operation showed thrombosis of the anastomosis. It is quite possible, however, that the anastomosis was open earlier and that the temporary circulation which it afforded, together with the fact that for the initial 3-day period the ischemia was only partial, allowed the development of a collateral circulation which was sufficient to ensure survival of the limb.


Nonsuture Anastomosis

  The principle of nonsuture anastomosis originally developed by Payr in 1900116 was tested experimentally by Hopfner in 1903 117 and was successfully applied clinically by Lexer in 1907.118  Jeger, in 1913,119 advocated this method for military surgery but it does not seem to have been employed in World War 1. In 1942 Blakemore, Lord, and Stefko120 in effect revived the technique of these earlier investigators as a method of restoring the blood flow through severed arteries. They used vitallium tubes, instead of the magnesium alloy tubes which Payr and H?pfner had used, lined them with vein grafts, and tied the cut ends of the artery over the ends of the connecting cannula. Somewhat later they modified their method, so that it. resembled Hopfner's technique, by using two tubes bridged by a vein graft.

The results secured by Blakemore and his associates 121 with this technique were so encouraging that hope was expressed that the method would be equally effective in military surgery. The  vitallium tubes required became available in the Mediterranean theater in August 1944. 122  They were in limited supply and therefore were used only in carefully selected critical cases cared for by experienced vascular surgeons.123 It is known that the double-tube-vein-graft technique was employed in 40 cases (Chart 15).

  Case 10.  When this soldier was wounded by enemy machine pistol bullets 11 March 1945 near Lucia, Italy, he incurred penetrating wounds of the right upper arm, a compound comminuted fracture of the right humerus, a perforating wound of the right forearm with a compound comminuted fracture of the ulna, and a perforating wound of the right thigh where the femoral artery and vein were both divided in the adductor canal by the missile. The sciatic nerve was partially divided. The femur was not fractured. First aid consisted of sulfanilamide powder and dressings applied to the wound, 4.0 gm. of sulfadiazine given orally during evacuation, and 4 units of plasma. No tourniquet was applied. The patient reached an evacuation hospital 5 hours after injury. As preparation for operation he received 1 unit of plasma and 2,000 cc. of whole blood.  At the conclusion of resuscitation the hematocrit value was 40.5 percent and the blood volume, determined by the T-1824 dye method, was high normal (6,160 cc. for a body weight of 68 kg.). The blood pressure was 144 mm. of mercury systolic and 100 diastolic.

116 Payr, E.: Beitr?ge sur Technik der Blutgef?sse and Nervennaht nebst Mittheilungen ?ber die Verwendung eines resorbirbaren  Metalles in der Chirurgie.  Arch. f. klin. Chir. 62: 67-93, 1900.

117 H?pfner,E.:  UberGef?ssnaht, Gef?sstransplantationen and Replantation von amputirten  Extremit?ten. Arch. f. klin. Chir. 70: 417-454, 1903.

118  See footnote 102 (1), p. 115.

119  Jeger, Ernst: Die Chirurgie der Blutgefaesse and des Herzens.  Berlin, Hirschwald & Co., 1913.

120  Blakemore, A. H.; Lord, J. W., Jr., and Stefko, P. L.:  The severed primary artery in the war wounded; a nonsuture method of bridging arterial defects.  Surgery 12: 488-508, Sep 1942.

121  (1) Blakemore, A. H.; Lord, J. W., Jr., and Stefko, P. L.:  Restoration of blood flow in damaged arteries; further studies on a nonsuture method of blood vessel anastomosis.  Ann. Surg. 117: 481-497, Apr 1943.

  (2) Blakemore, A. H., and Lord, J. W., Jr.: A nonsuture method of blood vessel anastomosis; review of experimental study; report of clinical cases.  Ann. Surg. 121: 435-453, Apr 1945.

  (3) Blakemore, A. H., and Lord, J. W., Jr.: A nonsuture method of blood vessel anastomosis; experimental and clinical study. J.A.M.A. 127: 685-691; idem, ibid.: 748-753, Mar 1945.

122  Later, an ordnance company copied the tubes in stainless steel and they were used with great satisfaction at the 8th Evacuation Hospital.

123  See footnote 50, p. 99.


  The operation, which was performed with the patient under ether-oxygen anesthesia with thiopental sodium induction, lasted 5 hours and 15 minutes. A primary incision was made in the femoral triangle and a tape was passed under the common femoral artery to permit emergency control of hemorrhage. All wounds were debrided. The femoral vein was ligated in the adductor canal. Anastomosis of the severed femoral artery was accomplished by means of stainless steel tubes fashioned after those described by Blakemore and his associates. A segment of the right long saphenous vein was used to bridge the gap.

Before the anastomosis was completed a clot was detected in the end of the artery just above the proximal tube. Heparin was injected subcutaneously (2 cc. in Pitkin's menstruum) and when the clotting time (determined by the Lee and White method) was 10 minutes, the clot was removed and the anastomosis was completed. Good pulsation was noted in the femoral artery distally, beyond the lower tube. The primary incision was closed, but the other wounds were left open for later delayed primary suture. No untoward effects followed procaine hydrochloride block of the third and fourth lumbar ganglia.

  The patient's general condition during operation was satisfactory.  The systolic blood pressure remained in the neighorhood of 120 and the diastolic between 70 and 80 mm. of mercury.  The pulse rate ranged between 110 and 120 beats per minute.  In the course of the operation the patient received 500 cc. of whole blood, 1,000 cc. of 2-percent sodium bicarbonate, and 1,000 cc. of 5-percent glucose in physiologic saline solution.

  On 12 March, the day after  the operation , the patient received 500 cc. of 2-percent sodium bicarbonate by vein and 1,000 cc. whole blood. He was also give, subcutaneously, 2 cc. (200mg.) of heparin in Pitkin's menstruum. Examinations of the urine showed a 2+ to 4+ reaction for benzidine-positive substance but on other abnormality.  The sympathetic block was repeated.

The right foot was cooler than the left; the capillary circulation in the right foot was sluggish. There was neither sensation in the toes nor motor power in the left foot. When the effect of spinal anesthesia on the circulation of the leg was tested, the level of anesthesia reached the crest of the ilium. There was no immediate change in the temperature of the right foot.


A posterior fasciotomy was done from the popliteal space to just above the ankle. The gastrocnemius and soleus group of muscles bulged through the incision. They appeared edematous. There were areas of pallor in the gastrocnemius muscle but some pulsating vessels were seen in it. The soleus muscle was relatively more anemic. A small fasciotomy incision over the anterior tibial group of muscles revealed normal muscles with no bulging. After the posterior fasciotomy the foot appeared much improved in color; it was also warmer and the capillary circulation was better.

  On 13 March the patient received 2 units of plasma and 1,000 cc. of whole blood. The foot was warm, the capillary circulation normal, and the posterior tibial pulse palpable and forceful. The dorsalis pedis pulse was not palpable. There was a moderate amount of serous drainage from the calf.

  The patient received 2 units of plasma and 500 cc. of whole blood on 14 March. Two 2-cc. doses of heparin in Pitkin's menstruum (200 mg.) were given subcutaneously at 11-hour intervals. The foot and leg were edematous to the knee. The anterior tibial group of muscles, though rather tense, was resilient. Exudate continued to drain from the fasciotomy wound. The foot of the cot was elevated at half-hour intervals for 30-minute periods.

  On 15 March the patient received 1 unit of plasma and 1,000 cc. of whole blood, together with 1 dose of heparin (2 cc.).  Heparin thereafter was discontinued because the supply was exhausted. A unit of plasma and 500 cc. of whole blood were given the following day. Weak pulsation was felt in the posterior tibial artery and edema began to subside. Exudation from the fasciotomy wound also began to decrease. Two days later (18 March) both posterior tibial and dorsalis pedis arteries were readily palpable. Penicillin, which had been given daily in the amount of 200,000 units, and sulfadiazine, which had been given in 6-gm.


amounts, were discontinued. On 15 March the urine was free of benzidine-positive substance which had been present in small amounts since it first appeared 12 March.  By 16 March the phenolsulfonphthalein excretory capacity of the patient was greatly improved. It had been 50 percent below normal on 12 March.

  The patient was transferred to a general hospital on 19 March. When the dressings were changed in the operating room 22 March the gastrocnemius and soleus muscles showed some separating slough. All wounds were closed except the wound on the posterior thigh and the fasciotomy wound.

  The hip spica was changed on 5 April and the sutures were removed. The fasciotomy wound showed superficial suppuration and a small amount of slough was still found separating from the gastrocnemius-soleus group of muscles near the tendinous portions. The right foot was warm, though not as warm as the left. The sole of the foot and the toes were anesthetic <>and hypesthetic. No motion was detectable in the toes or the ankle. An arteriogram showed complete patency of the anastomosis (Fig 6A, B)  When the patient was evacuated to the Zone of Interior 13 April he stated that he was beginning to "feel" a little motion in the toes and ankle, though none<> could be detected objectively.  

Figure 6. (Case 10.)  Laceration of femoral artery and vein treated by nonsuture anastomosis. A. Roentgenogram 25 days after operation, before injection of diodrast.B. Arteriogram after injection of diodrast into common femoral artery.

  Comment. The result in this case would probably not have been as satisfactory if the severed blood vessel had not been repaired by anastomosis. The appearance of a benzidine-positive substance (probably myoglobin) in the urine for several days after the operation is of interest in that it may indicate relief of ischemia. It is a physiologic fact that this substance does not appear in the blood and urine until the circulation has become reestablished.


  The proportion of amputations for the double-tube-vein-graft technique was significantly higher than for suture repair. However, the difference was not considered to be due to the techniques used but to the different characteristics of the injury for which the type of treatment was selected.

Case 11. When this 22-year-old soldier was wounded by enemy rifle fire 19 September 1944 in Italy, he incurred a severe wound of the left knee with a compound comminuted fracture of the left femur extending into the knee joint. The popliteal artery and vein were completely transected in the midpopliteal space. When the patient was admitted to an evacuation hospital 20 September, 9? hours after wounding, the blood pressure was unmeasurable. He had already received 2 units of plasma. Three hours later, after resuscitation with whole blood and plasma, the blood pressure was 104 mm. of mercury systolic and 58 diastolic.

  The operation was performed 14 ? hours after injury with the patient under thiopental sodium anesthesia. The wound was debrided, the popliteal vein ligated, and the popliteal artery suture-ligated. A lumbar sympathetic block was done at the conclusion of the operation and was repeated 8? hours later (1.5-percent procaine hydrochloride was used). No improvement was noted after either block.

On 21 September the left foot and leg were cold up to the level of the tibial tubercle. Areas of purple discoloration on the dorsum of the foot and in and around the great toe did not blanche. No pulsations could be felt in the foot or at the ankle. No oscillations were observed in the calf. The foot was anesthetic to just above the ankle. There was no motion in the toes. Although little hope was felt for preservation of the extremity it was decided, as a last resort, to do a nonsuture anastomosis with the hope of improving the circulation sufficiently to save part of the leg. After transfusion of 500 cc. of whole blood a left lumbar ganglionectomy was done, with excision of the second, third, and fourth ganglia. The ends of the popliteal artery were exposed and anastomosis by the nonsuture technique was performed (Fig. 7A). Vitallium tubes and a segment of the right long saphenous vein were used. A gap 2 inches long was thus bridged (Fig. 7B). A posterolateral fasciotomy was done in the calf. A bivalved cast was applied, using only the posterior half.Intravenous administration of heparin was begun at once at the rate of 600 mg. every 24 hours, and the foot was kept in a dependent position.

  The day after operation there was no edema of the foot; the foot was, however, cold and anesthetic to just above the ankle. There was no motion in the toes or at the ankle. The calf was swollen to almost twice normal size. A transfusion of 1,000 cc. of whole blood was given. On 24 September the appearance of the left calf was unchanged. During the previous 48 hours the area from the level of the tibial tubercle to about 6 inches above the malleoli became warmer. The foot was cold and dusky and the terminal phalanges of the toes were purple. The following day the foot was cold but the calf was warm to within 3 inches of the malleoli. The clotting time (by the capillary tube method) was 22 minutes.

By 27 September, the seventh day after injury, when no improvement in the condition of the limb had occurred (Fig. 7C), amputation was recommended. It was carried out the following day, with the patient under ether anesthesia. An attempt was made to perform the amputation at the midleg, but at this level, while there was bleeding from the skin, there was none from the muscles. Upon examination the muscles proved to be pale and nonviable (Fig. 7D).  Amputation was therefore done at the supracondylar level. Recovery was uneventful.

  Comment. When the anastomosis was exposed in the amputated specimen both the cuffs were found in place but the artery was thrombosed proximal to the anastomosis. Thrombosis extended distally into the leg and the venous segment was filled with blood clot. The anastomosis in this case was probably of no value because it was done after extensive thrombosis had occurred in the leg.


Figure 7. A. Details of nonsuture technique. a. Method of making anastomosis, this is essentially the technique described by Blakemore, Lord, and Stefko, except that the vein is not ligated onto the tube. Note that one of the ties is looser than the other and that the direction of the valves in the vein is reversed. b. Completed anastomosis.


Figure 7-Continued.  B. Transection of popliteal artery treated by nonsuture technique.  The arterial injury was complicated by a severe compount fracture of the tibia and the dislocated tibial fragments are still attached in the wound. C. Appearance of foot 7 days after anastomosis.  D. Dissection of amputated leg.  The muscles were pale and swollen and  the subcutatneous tissues were edematous.

In 8 of 13 cases observed in Italy (9 of which one of us personally operated on) amputation was necessary later, in 1 instance limited to the toes, in 2 through the lower leg, and in 5 through the thigh. All of these patients had severe injuries and it is doubtful that the limbs could have been saved under any method of treatment. The cases were deliberately selected for trial because the use of this method under battle conditions was considered purely experimental and therefore justified only in critical cases.

  All of the patients in this series, as just noted, had severe injuries. In 3 of the 5 cases in which amputation had to be done above the knee, severance of the popliteal artery was complicated by a compound fracture of the knee joint. While it must be granted that this is the type of case in which experience has shown that survival of the limb is unlikely, it must also be emphasized


that this is exactly the type of case in which it was hoped that nonsuture anastomosis would improve the results if it could be done early enough for diffuse thrombosis and infection not to have become established.   Failure in 2 cases in the series just mentioned could be attributed to technical errors. In 1 instance re-anastomosis after bleeding from a tiny venule had become apparent was not successful because peripheral thrombosis had already developed. In the other case the edge of the vitallium tube cut through the arterial wall on the 11th day after operation. When the anastomosis was performed there had been a good deal of difficulty in inserting the tube into the lumen of the vessel.

What is more important from the standpoint of military surgery than the exact results secured in this series is the character of the technique. The ease or difficulty with which an operation can be performed is an important consideration in military surgery, and the experience in World War II suggested that nonsuture anastomosis is neither as simple nor as easy as its proponents, whose extensive experience with the method was confined to civilian practice, indicated that it was. One operation in Italy, for instance, performed by a better than-average surgeon, took 3 1/2 hours; in the course of the procedure the ligature twice slipped off the tube and had to be tediously reapplied. A forward hospital, when casualties are heavy, is not the place for such time-consuming surgery if it produces no better results than can be secured by other, simpler methods. It is not known how many times the method was attempted and failed, though this is known to have happened 5 times in a series of 23 cases. Another possible disadvantage of the method is that when it is not successful it may result in the destruction of additional arterial substance. This is a particularly undesirable result and the sacrifice of functional collateral arteries was felt to be a definite contraindication to the use of the method in any case in which there was not a reasonable chance of success.

  It must be granted that although these operations were done by capable surgeons, they involved a new technique, and better results can probably be expected as experience increases.

The various series of reparative operations performed for acute arterial injuries in World War II were all too small to permit definite statements concerning the value of any one of these methods as compared with other methods (Charts 15, 16). In individual cases it is possible to say that one method or another was responsible for the saving of all or part of a limb, but the overall figures permit no such conclusions. It must be emphasized again, however, that most of the cases in these series were cases in which, because of the size of the defect, suture repair was not practical, in which the proportion of critical vessels was high, and in which the prognosis was always grave.

  Sympathectomy was usually performed when a major artery of the lower extremity had been treated by nonsuture anastomosis.


Chart 16.   Incidence of amputation in arterial wounds in American battle casualties in World War II according to method of management. 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 difference between total cases and breakdown in this chart represents 670 cases for which therapeutic measures could not be determined or are not included in the two categories. Type of repair was not known in 9 cases, of which 5 involved amputation,and these are not included inthe types of repair shown in Chart 15.



  As the war progressed disastrous effects of circulatory constrictions, no matter what the cause, became recognized more and more.  Constriction might be caused by action of the missile itself, by bacterial toxins and proteolysis, by vasospasm, and by effect of compound fractures or multiple injuries which impaired circulation directly or indirectly. Unwisely applied emergency measures were an unfortunately frequent cause. They included application of tight bandages, tight packs (particularly in deep wounds), tourniquets, and plaster casts.  Plaster casts were particularly dangerous when applied under conditions which did not permit close observation of the patient afterward or when the cast had not been bivalved to allow for subsequent swelling of tissues. Dangerous ischemia could follow either localized or homogeneous compression.


  A circular bandage inside a cast could give rise to dangerous ischemia and simple linear incision of a circular cast did not provide sufficient safeguard against edema of the limb and embarrassment of the circulation. Even prolonged elevation of a limb with a vascular injury might hasten ischemia which was impending or accentuate the condition if it had already developed.

  Exactly why, when all circumstances seemed comparable, some wounds healed without complications, and in others local infection and gas gangrene developed, was never clearly understood. What was understood, however, was that certain factors favored the growth of clostridia and enhanced the development of infection, and that loss or impairment of circulation in the involved tissues was the most important of these causes.

Eventually these dangers were warned against in directives issued in the Mediterranean 124 and European theaters,125 and the Office of The Surgeon General.126 One of the precautions against the development of serious ischemic complications was retention of a patient with an injured extremity at the evacuation hospital whenever the blood supply seemed to be jeopardized. He was not permitted to be evacuated further until it was clear that the collateral circulation was adequate or until amputation had been performed.

Pulsating Hematoma

  Although aneurysms and arteriovenous fistulas seldom required treatment in forward hospitals, false aneurysms (pulsating hematomas) were in a different category.127  Many surgeons took the position that the risk of hemorrhage or rupture was so great in this type of lesion that evacuation to the Zone of Interior was unsafe until operation had been performed. In theory, this was a complication which could be prevented by proper treatment of vascular injuries. In practice, planned repair of the arterial defect was often not possible because of the circumstances of war.  Frequently, extensive damage to muscles, nerves, and bones did not, even permit an exact determination of the vascular damage.

Pulsating hematomas were usually encountered in cases in which it was not possible to ligate the main vessels securely and immediate hemorrhage had to be controlled by packing. They were also encountered in cases in which, because the wound was small and there was little external loss of blood, massive hemorrhage into the tissues had occurred and had been overlooked. In some cases infection and necrosis produced secondary hemorrhage into the limb, which was not obvious externally. In all such cases the blood extravasated into the part and clotted. Later, if the condition was not remedied, a true endothelial sac was partially or completely organized. Expansile pulsation could be observed as long as blood continued to enter the mass. When the area had become over-distended and no further blood could enter the sac, all

124 See footnote 22 (3), p. 70.

125 See footnote 89 (3), p. 108.

126 (1) Gas gangrene prophylaxis. Bull. U. S. Army M. Dept. 75: 26-28, Apr 1944.
  (2) See also footnote 29, p. 81.

127 See footnotes 22 (3), p. 70; 23, p. 71; 29, p. 81; and 89 (2), p. 108.


pulsation ceased. This was the time during which the blood supply to the limb could be totally occluded both by an intraluminal clot and by secondary compression of the artery. As a result the part below the lesion began to swell and the patient complained of pain, or of more pain if pain had already been a feature of his clinical course.   While a pulsating hematoma could develop at any time following injury, it most often became evident within the first 2 to 8 weeks. It was also particularly easy to overlook it during this period because the patient was being transported to the rear, and examinations along the way were apt to be superficial. If the limb was in a cast, examination was frequently omitted.

  The diagnosis of pulsating hematoma usually was not difficult if it was remembered that it was a possibility in any instance of penetrating trauma in the vicinity of large arteries, and that the possibility was strong if there was a history of initial profuse hemorrhage, or if an initial hemorrhage had had to be controlled by packing. When the packing was removed, a cavity often remained which was an ideal site for the escape of blood. Undue swelling of the part and pain and discomfort, especially after a change of cast, were always indications for careful local examination. Characteristically, the pain was out of proportion to that previously experienced. Sometimes a mass was not palpable, but if hemorrhage had occurred recently, or if rupture was impending, it usually was present and might be of considerable size. If the sac was well developed, pulsations could be seen and were likely to be associated with a systolic bruit.

  Differentiation of a pulsating hematoma from an abscess offered the chief diagnostic problem. If the hematoma was well developed and presented the classical signs there was seldom any difficulty. Otherwise, the differentiation had to be made on the basis of the systemic manifestations likely to be present in abscess formation. It was a sound rule not to incise a fluctuant mass near a blood vessel, even though evidence of infection was present, until the possibility of pulsating hematoma had been eliminated.

  In small pulsating hematomas which had become infected, an occasional spontaneous cure followed the occlusion of the affected vessel by the so-called currant-jelly clot and subsequent fibrosis:

  Case 12.   A 36-year-old soldier wounded by aerial bomb fragments at Anzio, Italy, 28 January 1944, incurred multiple penetrating wounds of the left hand, left wrist, and both thighs. The right femoral artery was involved in the wound on that side. The wounds were debrided at an evacuation hospital the same day, and a foreign body was removed from the left thigh.

  Roentgenograms taken at a general hospital 3 February showed metallic fragments in the upper third of the right thigh; there was an incomplete fracture of the femur in the same area. A moderately low-pitched bruit (systolic) and an expansile pulsation could be demonstrated just proximal to the middle of the right thigh medially. The bruit did not have a "machinery" quality. There was no evidence of nerve damage and the pulses in the right foot were normal. The erythrocytes numbered 3,280,000 and the leukocytes 19,600 per cubic millimeter of blood. Transfusions of 500 cc. of whole blood were given 5 February and 9 February.


On 11 February all wounds were closed secondarily, and subsequent healing was satisfactory. At this time the patient was complaining of periods of numbness in the dorsum of the right foot. The following day the systolic bruit over the left femoral artery was recorded as softer and less shrill.It was still present on 14 February, but on 15 February when the patient was shown to a surgical consultant to secure advice concerning the management of the aneurysm, there were no signs of the lesion nor was there any further evidence of its presence during the week the patient subsequently remained under observation.

  Comment. The case demonstrates one of the advantages of delaying the surgical treatment of traumatic aneurysms and similar lesions, for sometimes, as in this instance, they disappear spontaneously.

In most cases of pulsating hematoma observed in World War II, surgery was necessary, though it was best postponed for 2 to 3 months if pain or evidence of impending rupture did not furnish indications for immediate operation. The longer an operation could be postponed, the more likely was the collateral circulation to be adequate. It was also desirable that the external wound be well healed and that dormant infection be excluded before operation was undertaken, but this was not always possible in the absence of a wellorganized sac. When such a sac was not present, copious bleeding often followed necrosis or slough of extensive soft-tissue wounds and even disruption of recently sutured wounds. In these cases, when pallor of the limb, absence of pulsations distal to the lesion, swelling of the part, or lowering of the surface temperature indicated that the blood supply to the extremity was reduced, it was not safe to postpone operation.

  The most important technical step in operations for pulsating hematoma was control of bleeding from above; a dry operative field was imperative. Frequently this could be achieved with a properly applied tourniquet, though in wounds of the upper thigh this was not possible. A safer method in all cases was to expose the main vessels above the lesion and to ligate them temporarily with rubber bands or tape.An approach through the original wound was usually complicated and unsatisfactory. True arteriorrhaphy or endoaneurysmorrhaphy was not always the wisest procedure, or, indeed, a feasible procedure. The rent in the vessel was frequently too extensive and the separation of the ends too great. Segmental defects in the wall were also frequent. The best plan, after evacuation of the sac, was occlusion of the proximal and distal portions by interrupted silk sutures carefully placed in the ends of the vessel, or by ligation. The temporary bands occluding the blood supply were then loosened to make sure that occlusion was complete. The cavity was lightly packed with gauze and adequate drainage was provided in all cases since most of the wounds were infected. Primary closure was always contraindicated if the operation was done on the indication of rupture or threatened rupture.

Secondary Hemorrhage

  Secondary hemorrhage was a relatively frequent complication of arterial injuries in World War I. In a series of 10,000 patients with wounds in which


the long bones were involved, Waugh128 observed an incidence of 14 percent in 1 year and of 9 percent in the following year. He attributed the improvement in the second year to "improved arrangements for the early excision of wounds."  Tuffier,129 in commenting upon secondary hemorrhage from arterial wounds during the First World War, also related the incidence of the complication to the incidence of infected wounds.

  In World War II the incidence of secondary hemorrhage was strikingly reduced because of improvements in the technique of debridement and immobilization, and the availability of chemotherapeutic and antibiotic agents. Of these reasons, improvement in debridement and the use of chemotherapeutic and antibiotic agents were probably the most important because of their beneficial effect on the incidence of wound infection. The problem was still important, however, because of the disastrous consequences which might follow severe bleeding.

  In all theaters the practice of delay in operation for aneurysm and arteriovenous fistula probably had much to do with reducing the incidence of secondary hemorrhage. It was generally taught that secondary hemorrhage was a possibility, even if a remote one, when arterial or venous ligature was undertaken, and it was recommended that a tourniquet be kept in position about the limb to be tightened immediately if hemorrhage should arise which could not be controlled by simpler measures. In an occasional case the loosening of thrombi several weeks after their original formation resulted in secondary hemorrhage.

  In the North African Theater of Operations not a single secondary hemorrhage was encountered in forward hospitals in the Sicilian campaign, perhaps because the patients were seldom kept long enough for it to occur. On the other hand, the basic reason probably was that the wounds were not, particularly extensive.

  The secondary hemorrhages encountered in the Italian campaign and studied by Capt. J. T. Coyle and Maj. W. D. Thompson 130 were of two types. The first and more frequent type occurred from large muscle areas when a thin layer of necrotic material sloughed en masse. It was readily controlled in every instance by thoroughly cleaning out the slough tissue and applying hot, moist sponges. Petrolatum impregnated dressings were not satisfactory; dry gauze fluffs, removed after 48 hours, worked well. The second type of hemorrhage occurred from large wounds of the calf after sloughing of the wall of the posterior or anterior tibial artery. It was controlled by ligation of the involved artery and no circulatory disturbances were noted subsequently in any of these patients.

128 Waugh, W. Grant: Secondary hemorrhage. In Bailey, Hamilton :Surgery o  Modern Warfare Baltimore, Williams & Wilkins Co., 1941, vol 1, p. 328-332.

129  See footnote 113 (3), p. 116.

130 Coyle, J. T., and Thompson, W. D., Jr.:  Fractures in battle casualties.  Bull. U. S. Army M. Dept. 86: 57-63, Mar 1945.


  The experience reported by Lt. Col. Norman E. Freeman 131 at the 20th General Hospital at Assam, India, may be taken as typical of the World War II experience with secondary hemorrhage. This complication was observed in 23 of 2,168 patients treated in the course of a year for gunshot or shell wounds of the extremities and neck. In 8 of these patients bleeding was from small vessels and was readily controlled by packing or by simple ligation of the bleeding vessel in the wound. In the other 15 patients hemorrhage was from a major vessel and was a serious problem. Freeman personally operated on 12 of the 15 patients and supervised the treatment of 2 others, which makes the report of this particular experience of unusual value.  

  Clinical Observations.  In every instance in Freeman's series the wound was located in the immediate proximity of the vessel from which the hemorrhage occurred, a circumstance which naturally simplified the diagnosis. As in Waugh's series,132 compound fractures were present in more than two-thirds of the cases. Another observation of major importance, though its significance was not always realized, was that the patients in whom secondary hemorrhage occurred had (1) a previous clear-cut story of massive hemorrhage or recurrent bleeding while in the forward area, or (2) a condition of severe anemia at the time of admission to the hospital. Eventually it was realized that injury to a large vessel should always be suspected when a heavy loss of blood was evident.  In 7 patients one or more episodes of bleeding occurred before the final massive hemorrhage. Failure to heed this "red signal" was responsible for the death of 1 patient (Case 13) and would have been responsible for death in others except for the skilled intervention of ward personnel (Case 14). As emphasized by Waugh "a small initial hemorrhage occurs in more than half the total cases and constitutes an inexorable indication for exploration of the wound.'  This lesson was eventually learned in World War II and ward personnel were instructed to report immediately any bleeding, however small.

  Case 13.  A 27-year-old Chinese soldier was struck just below the left shoulder by a bullet from a .25 caliber pistol on 1 March 1944. The wound of entrance was over the lateral surface of the left arm below the acromion and the bullet came to rest in the soft tissues of the axilla. A compound fracture of the humerus was present in addition to the soft-tissue injuries. Debridement had been performed in the forward area and a plaster encasement applied to the left shoulder with the arm in abduction. When the patient reached the 20th General Hospital the radial pulse was normal, normal sensation and motion were present in the fingers, and there was no edema of the extremity. The value for hemoglobin was 8.8 gm. per 100 cc. of blood. When the cast was changed 14 March the wound appeared clean and there had been nothing in the patient's progress to suggest subsequentdifficulties. A hanging plasterencasement was applied to the forearm for traction and a separatesilica was applied to cover the wound.

On 17 March the Officer of the Day noted on the patient's record at 0700 that some bleeding had occurred for a few minutes tinder the encasement while the man was brushing his teeth, but that it had stopped and his condition was good though he would probably

131  Freeman, N. E.: Secondary hemorrhage arising from gunshot wounds of the peripheral blood vessels. Ann. Surg. 122: 631-640, Oct 1945.

132   See footnote 128, p. 130.


need a new encasement.  Six hours later the ward officer noted that there had been no further bleeding, that the patient's condition was satisfactory, and that it could be seen under the encasement that the wound was dry. Two and one-half hours later there was a second small episode of bleeding and the patient vomited and fainted. When the ward officer removed the encasement, the wound was dry. A corpsman reported that he had seen the patient "picking at the wound with his chopsticks," and it was assumed that this action had caused some bleeding from the granulations.  

  For the next 10 days progress was completely satisfactory except that, for the first time, the patient had a higher than normal temperature. There was no bleeding and the alignment of the fracture was satisfactory.  On 27 March, 27 days after wounding and 10 days after the episodes of bleeding, the patient suddenly screamed.  He was found sitting up in bed in a pool of blood and death occurred while the encasement was being cut off and efforts were being made to start a transfusion.

Case 14. A Chinese soldier sustained a shell fragment wound just below the angle of the left mandible 18 March 1944. The missile lodged in the transverse process of the first cervical vertebra.  Debridement was performed in the forward area and was accompanied, according to the operative note, by "much bleeding."  Three days after the patient's admission to the 20th General Hospital, and 4 days after he had been injured, a swelling was noted below the left mandible which continued to increase in size and to cause severe pain.  His temperature was normal.  On 30 March, 12 days after the original injury, with a preoperative diagnosis of cervical abscess the mass was incised.  The "rush of blood" which ensued was controlled with difficulty by suture of the skin and a pressure dressing.  On 7 April, 8 days later, a pulsating tumor with a palpable thrill and continuous murmur was found below the lobe of the left ear.  The wound was healing but serous fluid was weeping from the skin edges.  The anterior jugular vein was distended when the patient was recumbent but collapsed when he sat up. The pulse rate dropped from 72 to 60 beats per minute when pressure over the carotid artery was sufficient to abolish the murmur.  A diagnosis of arteriovenous fistula was made.

  In this case it was considered desirable, if possible, to allow complete healing to occur before the fistula was excised.  The delay was regarded as safe since hemorrhage from an arteriovenous fistula is rare because of the free communication with the venous side of the circulation and the resultant low pressure within the aneurysmal sac.  A tight bandage was applied over the lesion, a procedure which, in retrospect, was realized to be a mistake since compression of the tissues, if it achieved anything, was likely to raise the pressure in the aneurysm because of pressure on the venous outlet.

  The day after the bandage was applied, the nurse on the ward heard the patient cry out.  When she went to him she found him sitting up in bed with blood streaming from his neck.  She immediately applied digital pressure, thus controlling the bleeding and saving his life.  As soon as possible he was transported to the operating room where, with the aid of local analgesia, the external carotid artery was exposed below the digastricus.  The circulation was temporarily occluded by a rubber tube. The arteriovenous fistula was then opened and excised with quadruple ligation of the component arteries and veins.  The hemorrhage had originated from the external maxillary artery and the temperomaxillary vein.

  It is noteworthy that hemorrhage was sometimes preceded by minor activity such as brushing the teeth or moving the arms; however, in other instances it occurred without warning. In any event, such warnings as did occur were often so slight that they were ignored; moreover, the cessation of the initial hemorrhage induced an unwarranted sense of security.

  In 3 of Freeman's patients bleeding into a cast hindered both diagnosis and treatment. Plaster encasement was an essential form of treatment in


both fractures and extensive soft-tissue injuries, but care was not always taken to exclude injuries of major blood vessels before immobilization was instituted.   In World War I secondary hemorrhage most often occurred between the l0th and 16th days after injury. In Freeman's series the range was from 2 days to 3 months. In 10 of the 15 cases he cited, the bleeding occurred within the first 2 weeks after wounding.

  When these cases were reviewed it was found that physical examination at the time the patients were admitted to the hospital had not been helpful.  Although all of the patients had lacerations of major arteries, as their clinical course later proved, the peripheral pulses were absent in only 2 and were diminished in only 2 others.  Most of the patients presented peripheral edema or complained of a sense of numbness of the distal part of the affected extremity.  In 2 patients, although the fingers were warm and the radial pulses palpable, tissue damage had obviously occurred and could have been caused only by severe ischemia of long duration. Since the circulation was adequate at the time of admittance, the observation pointed to a serious arterial injury for which compensation had been provided either by resumption of the blood flow through the injured vessel or by development of a collateral circulation.

  In 4 instances a hematoma was present over the injured vessel, but a systolic bruit was audible in only 2 of these cases. In 4 other cases, although the initial examination was conducted with the possibility of an arterial lesion in mind, no bruit could be heard.

  Character of Injuries.  In 14 of the 15 cases in Freeman's series the arterial injury was a laceration. In the 15th case the vessel was only partially divided.  Injuries of this type are much more likely to lead to severe bleeding than is complete severance in which retraction of the severed ends is a physiologic means of controlling hemorrhage. Waugh,133 in his study of secondary hemorrhage in World War I, cited Makins to the effect that in every secondary hemorrhage an incomplete injury of a blood vessel is preexistent. These cases help to prove that thesis.

  Therapy. In hemorrhage from minor vessels, as already noted, packing of the wound with dry gauze, the application of pressure, or ligation of the bleeding vessel in the wound was sufficient for control.  Immediate arrest of bleeding, by one means or another, was the proper method of treatment in all cases and management included immediate replacement of the blood loss and arrangements for immediate operation.  In the meantime the bleeding was controlled by emergency methods, with a tourniquet the last resort.  In small wounds of the neck digital compression might be the only possible method of temporary control.

  The first principle of surgery was that the incision be long enough for adequate exposure, since circulation through the afferent artery had to be controlled before the bleeding point could be found.  There was never any

133  See footnote 128, p. 130.


hesitation in opening and exposing the original wound.  It was sometimes difficult to find the laceration when the bleeding had been checked by temporary measures, for the laceration was often sealed again by clot formation, and the circulation was depressed from loss of blood.  In 5 of the 15 cases cited by Freeman manipulation was necessary to produce bleeding before the injury could be identified.  It was essential to be quite certain that the precise source of the bleeding had been found.  In 1 patient a small arterial branch of the popliteal artery was thought to be responsible, but recurrent hemorrhage after transfusion necessitated a second exploration and disclosed a laceration of the popliteal artery itself.  

In 1 patient proximal ligation through a separate incision was performed for secondary hemorrhage after previous ligation of the femoral artery.  After the vessel had been exposed, bleeding was controlled by rubber tubes placed proximally and distally and the bleeding point was then sought, in a dry field, without danger of injury to accompanying nerves and other structures.

  Suture of the laceration was attempted twice but was successful in only 1 case.  In the other, a ruptured false aneurysm of the femoral artery, amputation was required 6 days later for ischemic gangrene.  In all the other cases the vessel was ligated and divided.

  Case 15. In this case a shell-fragment wound, 11 July 1944, resulted in a supracondylar fracture of the left femur with little hemorrhage.  When the patient was examined at the 20th General Hospital 12 July, the left foot was warm and the dorsalis pedis pulse was palpable.  Skeletal traction was applied. Infection of the wound followed and was associated with intermittent fever.  Pitting edema of the foot was noted about 4 weeks after injury.  On 15 September, 2 months after injury, about 100 cc. of serosanguineous material containing some small clots was discharged from the wound immediately after the patient had used a bed pan.  Distortion of the thigh was noted and roentgenograms confirmed the clinical impression of displacement at the fracture site.  The ankle pulses were normal, motion and sensation were present in the toes, and there was no pulsation, thrill, or bruit over the femoral vessels just above the knee.  One week later an accumulation of pus on the posterolateral aspect of the lower thigh was drained.  Slight amounts of blood continued to appear on the dressings.  The medial wound was now almost healed.

  On 8 October, almost 3 months after the original injury, a brisk hemorrhage occurred from the posterolateral wound.  It was controlled by packing until the patient could be taken to the operating room.  Then, with the aid of spinal analgesia, the femoral artery was exposed in the adductor canal, the circulation was controlled, and the site of bleeding was sought.  It proved to be a ruptured false aneurysm.  Release of the distal end of the artery allowed no back bleeding.  The laceration in the femoral artery was sutured since it was felt that the collateral circulation was insufficient.  Pulsation of the artery below the line of suture took place after release of the upper rubber tubing applied to control the blood supply.  After operation there was marked ischemia of the tissues below the knee.

  Although the lower leg was refrigerated for 5 days, no improvement was evident and amputation through the fracture site had to be performed 14 October.  Dissection of the arteries of the leg revealed that thrombosis had occurred below the point of suture.

  Although Reid 134 had condemned chromic catgut for these purposes, especially when wound infection was present because its absorbable properties

134  Reid, M. R.:  Partial occlusion of the aorta with silk sutures, and complete occlusion with fascial plugs; effects of ligature on the arterial wall. J. Exper. Med. 40: 293-299, Sep 1924.


made rapid disintegration likely, it was used in the series reported by Freeman without any untoward results.  The factors of safety were probably adequate drainage and the use of sulfonamide drugs.

After operation alcohol was injected into the region of the paravertebral sympathetic ganglia in 4 patients with good results. In a fifth patient, procaine was injected repeatedly.  In the case just described in which ischemic gangrene required secondary amputation, alcohol injection was not employed.


  Vasospasm is well known in civilian practice as accompanying wounds or manipulations of the arteries.  It is, in fact, a natural response to any form of trauma which directly or indirectly affects the vascular structures.  It was therefore a common occurrence in the arterial injuries observed in World War I as well as in those observed in World War II.  Depending upon a number of factors it might involve only a small part of a vessel (as in arterial concussion), might spread to neighboring vessels, or might become sufficiently generalized to involve the entire limb and even larger areas of the body.  Although vasospasm could under certain conditions be considered a compensatory mechanism, prolonged vasospasm could lead to serious consequences. When it was minimal, amounting merely to a localized area of constriction, the resulting ischemia was minimal and the limb was seldom placed in jeopardy. When it was more extensive, and especially when the collateral circulation was involved, ischemia was often sufficient to produce localized gangrene. When it occurred in cases in which trauma to the tissues had already seriously impaired their vitality, vasospasm was often the factor which determined whether the limb would survive or die. An additional factor of danger was that the results of ischemia might cause development of gas bacillus infection.  

In the category of spasm are included the few cases of contusion recorded as such in American battle casualties in World War II because this condition, if it was recognized at all, was not infrequently accompanied by spastic phenomena. This type of lesion was usually slight and its inclusion in a tabulated series therefore alters the results favorably. (This is perhaps one reason why Makins' statistics, in which contusions were included, are so much more favorable than other statistics from which they are omitted.)  Some doubt is felt about the incidence of amputation (25 percent) reported in the 28 cases of spasm in this series (Chart 12).  That the figures are weighted is suggested by the fact that in 1 sample of 6 cases, there were 5 instances of gangrene.  This is so contrary to the usual experience as to suggest that the diagnosis of spasm was probably not correct in 1 or more cases in the group. Furthermore, most cases of spasm were not recorded at all and reports therefore included only those which developed complications.

  Vasospasm, as the following cases show, might be transient or might be of long duration.

Case 16
. A soldier wounded in Tunisia 31 March 1943, incurred a perforating shellfragment wound of the right leg.  When he was admitted to an evacuation hospital the following day the wound was described as located 6 to 8 cm. below the anteromedial lip of the tibia, midway between the tibia and the fibula.  The wound of exit, which was small, was located in the calf posteriorly.  The leg was warm and of good color but was markedly edematous.  Pulsation could not be felt in the dorsalis pedis.   Exploration was undertaken 1 April, 32 hours after wounding.  The missile track passed through the opening above the interosseous membrane in close proximity to the anterior tibial artery.  The peroneal nerve was not injured.  There was no visible injury to the vessel wall, but the vessel was in complete spasm for a distance of at least 10 cm. distally.  Three hours after operation pulsation still could not be felt in the dorsalis pedis artery, but 4 hours later pulsation was easily felt and the volume improved progressively thereafter.  Edema also subsided progressively until the patient was evacuated on the fourth postoperative day.

  Comment.  This case is an illustration of short-lived arterial vasospasm caused by a wound close to the vessel which did not, however, directly involve it.  If lumbar sympathetic block had been done promptly after operation, improvement might have occurred even more rapidly.

  Case 17. A 26-year-old soldier wounded near Cassino, Italy, incurred shell-fragment wounds of the right leg in the upper third and of the right thigh in the lower third.  The wounds were debrided at an evacuation hospital 15 hours later.  A shell fragment was removed from the right leg through an elective incision along the head of the fibula.  Exploration of the posterior tibial artery showed it to be in complete spasm but pulsations were felt in the popliteal artery.  The dorsalis pedis pulse could not be felt and the foot was cold.  Lumbar sympathetic block was done promptly after operation and was repeated three times in the next several days.  The results were not recorded.

  When the patient was admitted to a general hospital 13 days after wounding, the right foot was slightly swollen and was colder than the left. There was some question about whether peripheral pulses were detectable.  Two days after admission secondary closure of the wounds was done and paravertebral procaine hydrochloride block was repeated because the foot was cyanotic and clammy.  The wounds healed normally but the foot continued as described except immediately after sympathetic block; then it was transiently warm and pink.  Eight blocks were carried out over a 14-day period.  Femoral pulsations continued to be normal, but the popliteal, dorsalis pedis, and posterior tibial pulses were never palpable.  A lumbar ganglionectomy was therefore done on the 30th day after wounding.  Eight hours later the foot was described as flushed and warm, and all the peripheral pulses became palpable and so remained.

  Comment.  In contrast to the preceding patient (Case 16) the arterial vasospasm in this patient persisted for an unusually long period of time.  The indications for ganglionectomy were sound and its performance clearly contributed to the good results.

  Therapy. Treatment in vasospasm is always based upon an attempt to counteract the condition and to produce maximum vasodilatation in the involved extremity.  This was the objective in such cases in World War II.  A few attempts were made to induce and maintain vasodilatation by the use of such a vasodilating agent as whiskey, but in general, chemical or surgical means were preferred.  The logic of these methods is evident:  Since the disturbance is apparently the result of a vasomotor reflex initiated in the traumatized tissues, and since vasoconstrictor impulses are transmitted by way of the


sympathetic nerve fibers, interruption of these impulses by the means suggested is the proper mode of treatment.

Interruption of sympathetic impulses, usually by paravertebral injection of 1-percent procaine hydrochloride solution, was widely practiced by American surgeons in World War II, but in the material available for analysis it was possible to determine definitely in only 280 cases that the procedure had or had not been performed.  The results (Chart 17) provide no substantial evidence that the method was of any value.  The incidence of amputation in the group in which sympathetic block was performed was only slightly less than the incidence for the group as a whole, while the incidence in the cases in which ganglionectomy (which was preferred to repeated nerve blocks) was performed was greater than for the entire series.  

  Familiarity with the clinical material permits a different and more accurate interpretation of results than the figures on sympathectomy might suggest. Actually, sympathectomy was used only as a last resort when it had already become apparent that the limb would not survive. Sympathetic block was
Chart 17. Results of sympathectomy and sympathetic block in comparable arterial wounds 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.  Only those cases where, for the same artery, there were examples of management with sympathectomy and sympathetic block, and also examples where neither had been used, are included in this chart.


more frequently instituted as part of theimmediate postoperative routine and was continued untilthe outcome in respect to survival or death of the limb became obvious.

  When these facts are known, the apparently poor results of sympathectomy and the apparently better results of sympathetic block are more readily understood.

  It should also be emphasized that it is extremely difficult to evaluate such procedures as these on a purely statistical basis. First, it was not always possible to determine from the records that the block had been performed effectively.  Second, proper objective methods to determine the efficacy of the procedure are lacking, and suitable controls upon which to base an evaluation of results are also lacking.  Survival or death of the limb is not a critically objective test; the outcome does not permit a clear decision whether the therapeutic measures employed in a given case have influenced the results and it can serve as a criterion only when sufficiently large numbers of cases are available for statistical evaluation.  Third, and perhaps most important of all, the ultimate viability or death of the part was established in most cases at the time of wounding, and the margin within which improvement could be demonstrated was therefore so small that great numbers of cases would be required to establish the efficacy of any single procedure on a statistical basis.

  Despite the lack of statistical proof, there is much clinical evidence in favor of sympathectomy and sympathetic block. The personal testimony of numbers of surgeons indicates the usefulness of these procedures.  In case after case moderate degrees of tense swelling and muscle pain characteristic of ischemia were observed to regress and the temperature of the limb was observed to rise following their use.  Most American surgeons who worked in forward areas regarded interruption of the sympathetic system when practicable as perhaps the most useful adjunct method of treatment available to them.

  Cohen 135 was among the few experienced observers unwilling to accept the theory that arterial spasm is influenced by local or distant autonomic reflexes, his position being that it is myogenic in origin.  He also cited evidence to indicate that blockage of the vasomotor control of a wounded limb was dangerous because vasodilatation of the skin was not accompanied by vasodilatation in the muscles, and harm was done by diverting blood from the muscles into the skin. For this reason he deprecated the use of sympathetic block or sympathectomy as a therapeutic measure to combat traumatic vasospasm associated with direct injury.  His reasoning, however, seems inconsistent since he was willing to use sympathectomy to control vasospasm associated with a crushing injury or following the prolonged application of a tourniquet, on the ground that the vasospasm observed under these circumstances is of a reflex nature.  It seems highly unreasonable to assume that vasospasm can be initiated reflexly by one type of trauma and not by another, especially since the end results of both

135  See footnote 48, p. 98.


varieties take the form of ischemia from tissue damage.  It seems equally unreasonable to assume that sympathetic block or sympathectomy can be injurious in one of the varieties of vasospasm and beneficial in the other.

Aneurysmsand Arteriovenous Fistulas

  As has been pointed out, the nature of the missiles employed in World War II altered in many respects the character of the wounds produced by them. This was particularly true of vascular injuries. In addition to wounds caused by machinegun and rifle bullets and high explosive shells, multiple injuries were frequently caused by fragmentation of land mines and grenades.  It was not uncommon to find as many as a hundred small, separate wounds scattered over the surface of the body, none of them serious enough to cause death but many of them capable of producing aneurysms.  The incidence of aneurysms and of multiple aneurysms was therefore greater in World War II than it had been in any previous war. The same was true of the incidence of arteriovenous fistulas.  

  Since, as will be pointed out shortly, the policy of management of aneurysms and arteriovenous fistulas was entirely conservative, the majority of patients with these conditions received their definitive treatment in Zone of Interior vascular centers. (See Chapter IV.)  As a matter of fact, when these complications were seen overseas it was almost without exception in hospital installations of the base sector.  They usually took weeks to become manifest and would scarcely have had time to develop in field and evacuation hospitals of corps and army areas.  The diagnosis was probably missed in some cases overseas, but in many others the lesions still had not developed when the patients were evacuated.  Not uncommonly their presence was first detected by the patients themselves who days or weeks after the injury noticed a "buzzing" in the region of the wound. Traumatic aneurysms and arteriovenous fistulas were practically always observed in injuries which caused only partial interruption of the flow of blood through the vessel.

  The plan of management of arterial aneurysms and arteriovenous fistulas was basically conservative for several reasons: (1) Delay permitted whatever infection might be present from the original wound to subside completely. (2) Delay lessened the chances of secondary infection when operation was done. (3) Delay diminished the chances of secondary hemorrhage which was a possibility in any case in which there was extravasation of blood into the tissues. (4) Delay provided for the occasional case in which a traumatic aneurysm might heal spontaneously. (5) Delay allowed time for an adequate collateral circulation to develop.

  Of all the reasons listed in favor of conservative treatment of aneurysms and arteriovenous fistulas the last-named was the most valid.  Major vessels could not be ligated and divided with safety until such a circulation had developed. In the natural course of events when the blood flow through the main artery of a limb was impeded or partially diverted by trauma, a collateral cir-


culation developed and took over the function of supplying the limb with blood. This development, however, was a matter of weeks, not of days.  It was thought to reach its maximum not less than 10 weeks, and more often not less than 12 to 16 weeks, after the need for it first arose.  It was therefore the established policy in all theaters of operations that in the absence of indications to the contrary, surgery for traumatic aneurysms and arteriovenous fistulas should be deferred for 3 to 4 months after injury and should be carried out preferably in a vascular center in the Zone of Interior.

  In a certain number of cases, operation for arterial aneurysms and arteriovenous fistulas had to be undertaken in overseas theaters because the circumstances did not permit delay.  Surgery was then more or less an emergency matter. The chief indications were impending rupture, hemorrhage, infection, threatened gangrene of the extremity because of circulatory deficiency, increase in the size of the mass, signs of pressure on neighboring vessels and nerves, severe pain, and evidences of strain on the cardiovascular system.  These indications were  more likely to appear in pulsating hematomas (false aneurysms) than in true aneurysms and arteriovenous fistulas.  As the following cases show, more than one indication was frequently present in a single case.

  Case 18. A 22-year-old soldier was wounded accidentally in Italy by a bullet from a .45 caliber pistol 20 September 1943. The bullet entered just beneath the middle third of the left clavicle and emerged from the thorax through the left scapula.  The subclavian artery was damaged.  The wounds of entrance and exit were debrided in an evacuation hospital about 3 ? hours after injury.  It was noted that the pulse was less strong in the left wrist than in the right, which suggested that the subclavian artery was in spasm.  Roentgenograms revealed fragmentation of the scapula but no injuries of the thorax.

The day after operation examination showed paralysis of the left arm, forearm, and hand except for slight deltoid, triceps, and supinator action. Sensation was present in the axillary area and in the area supplied by the medial and dorsal antebrachial cutaneous nerves.  It was concluded that the injury involved the inner cord of the brachial plexus completely and at least half of the posterior cord below the level of the axillary nerve.

  On 23 September, when the patient was examined at a general hospital to which he had been evacuated by air, the left arm was found flushed and slightly edematous throughout its length.  By 26 September the edema had subsided to a considerable degree but the radial pulse remained weaker in the left arm than in the right.  On 28 September, when he sat up for breakfast, about an ounce of blood oozed through the dressing over the left clavicular region.  Examination showed edema of the entire left arm with distention of the veins.  The palm of the left hand was moist but no voluntary motor activity could be elicited.  Operation was undertaken 48 hours later because of the complaint of pain and the continuing evidence of compression of the brachial plexus by the aneurysm.  The aneurysm arose from the first portion of the axillary artery.  After a rather difficult operation the subclavian and axillary arteries were ligated in continuity proximal to the origin of the subscapular artery.  A transfusion of 1,000 cc. of whole blood was given during the procedure.

  At the end of the operation the hand was warm and of good color. Recovery was uncomplicated and the patient was in good condition when he was evacuated farther to the rear 26 October.

  Case 19. A 23-year-old soldier was wounded in action by enemy shell fragments on the Anzio beachhead 20 February 1944.  He incurred multiple wounds of the left buttock, the left thigh and leg posteriorly, and the left popliteal area.  There were no fractures.  He was given first aid, including a unit of plasma, within 3? hours after wounding and was admitted


to an evacuation hospital shortly afterwardwhere all his wounds were debrided.Roentgenograms showed foreign bodies beneath the base of the left fifthmetatarsal and in theleft thigh medially and also medial to the lesser trochanter.

  When the patient was received at a general hospital in the base 23 February after evacuation by hospital ship, examination showed impending gangrene in the toes of the left foot which was slightly edematous.  He complained of pain in the foot which he said was "worse" at night.  The calf was swollen and tense.  The skin over the lateral aspect of the leg was hot and reddened.  No pulsations were felt in the popliteal space or below it.  The site of the arterial lesion responsible for the vascular insufficiency was not determined.  After two lumbar sympathetic blocks with procaine hydrochloride, pain lessened and swelling decreased in both the leg and the foot.

On 25 February edema of the extremity was more prominent.  The first four left lumbar sympathetic ganglia were blocked with procaine hydrochloride.  On 27 February the toes were dark, the calf was tense, and the foot was edematous.  The patient was able to move the ankle and toes.  By 1 March the foot was still swollen and there was definite gangrene of the great toe and the fourth and fifth toes.  The following day the erythrocytes numbered 2,570,000 and the leukocytes 11,900 per cubic millimeter.  By 4 March all of the toes of the left foot were undergoing the changes of dry gangrene and there was gross discoloration of the skin of the distal half of the foot.  On 5 March the erythrocytes numbered 3,410,000 and the leukocytes 12,500 per cubic millimeter; gangrenous changes (dry) continued to progress slowly proximally on the foot.  Pain was limited to the foot.

  On 14 March the patient himself noted a peculiar sensation in the thigh near the left inguinal region.  Examination showed the characteristic "machinery" murmur of an arteriovenous fistula.  The blood pressure in the arm was essentially unchanged when the aneurysm was occluded. On 17 March an electrocardiogram was reported as being normal.  The pulse rate was 120 beats per minute before the arteriovenous fistula was occluded by pressure and 96 and 100 beats per minute while it was occluded.  A firm, diffuse, tense swelling appeared in the peroneal group of muscles and was progressive.

  By 30 March the dry gangrene in the foot was well demarcated.  On the sole it extended almost to the heel.On 1 April left lumbar sympathectomy was performed with the aid of spinal analgesia.  By 6 April there was noticeable improvement in the circulation of the left leg and the tense swelling of the peroneal group of muscles had somewhat receded.

  On 8 April the arteriovenous fistula, which was located about 5 cm. below Poupart's ligament, was excised.  Thereafter there was continued improvement in the circulation.  Before the patient was evacuated to the Zone of Interior in May (the exact date is not known) a guillotine amputation was done just above the ankle.

  Comment.  The cause of the early vascular insufficiency in this case is not clear. It may have been caused by arterial spasm, or it may have been the result of a small arteriovenous fistula. The outcome might have been different if it had been recognized earlier and if the circulation had been aided promptly by sympathetic block or sympathectomy. When the aneurysm was clearly established, 3 weeks after wounding, the pathologic changes in the leg had become progressive. The course in this case seemed to furnish an indication for excision of the aneurysm rather than for temporizing measures.

  Case 20.   This 25-year-old soldier was wounded by shell fragments in Italy 7 December 1943. He incurred multiple wounds of the abdomen, chest, and extremities.  The following day in a field hospital all wounds were debrided, a colostomy was established because the colon was the site of several injuries, and "a diffuse axillary hematoma was drained through a stab wound in the axilla."  This hematoma was related to an anterior wound of the right shoulder.


  When the patient was evacuated to a general hospital 16 days after wounding, a mass was palpable on the right side in the subpectoral region. Three days later the mass suddenly increased in size and the patient complained of severe, burning pain which involved the right upper extremity.  Examination showed the mass to be expansile.  A systolic bruit was heard over it.  The right arm was weak and there was hypesthesia over the radial and median nerve distribution.  The blood pressure was 104 mm.of mercury systolic and 26 diastolic on the right side, and 130 mm. of mercury systolic and 78 diastolic on the left.   Because of the increase in the size of the mass and the pain, which was difficult to control even with morphine, operation was carried out 31 December.  The first portion of the axillary artery was exposed by division of the clavicle and was controlled by means of rubber tubing passed under it.  The aneurysm was then approached through a separate incision.  It was opened and cleared of blood clot.  The arterial wound was at the level of the anterior circumflex humeral branch which was ligated.  The axillary artery was then ligated and divided across the wounded area. The lesion was thought to be proximal to the subscapularis artery.  The missile had also lacerated a small trunk of the median fasciculus of the brachial plexus as it lay anterior to the axillary artery.  The circulation appeared satisfactory at the conclusion of the operation, in the course of which the patient received 1,500 cc. of whole blood.

  The day after operation the function of the right hand was better than it had been before operation. The right hand was cooler than the left. The radial and ulnar pulses were not palpable but there was good capillary circulation. The sympathetic nerve supply of the right arm was blocked twice with procaine hydrochloride.

  When the patient was evacuated 46 days after operation he was showing satisfactory progress.  The function in the right hand had improved progressively though there was still no radial pulse.

  It was usually a simple matter to determine the status of the circulation in the limb. When interruption of the blood flow was only partial, the limb was cool but not cold. The muscles distal to the wound were essentially normal and were not tense or swollen. Oscillations were detectable with a standard oscillometer or with a sphygmomanometer. Vascular spasm was likely to be present in these cases and to respond well to repeated sympathetic block with 1-percent procaine hydrochloride. When emergency operation was necessary sympathectomy was frequently performed either before or after operation to assure maximum vasodilatation and thus increase of circulation in the involved part. The results of surgery for arterial aneurysms and arteriovenous fistulas, whether the operation was done overseas or in the Zone of Interior, were generally excellent.


  Volkmann's Ischemic Contracture.  In many cases of vascular injury a curious phenomenon became evident after interruption of the blood supply, sometimes as soon as 12 hours afterward, in the form of a tense, hard swelling of the muscles (Fig. 8).  It was particularly likely to occur after wounds of the popliteal artery, but also occurred after wounds of the brachial and femoral arteries. When the femoral artery was injured the swelling was likely to be limited to groups of muscles in the leg, such as those in the anterior tibial compartment. The involved extremity became progressively larger, the maximum size being reached in 36 to 48 hours. It was characteristically hard and tense

143 Figure 8. Tense swelling of muscles of calf, without significant subcutaneous edema, 12 hours after transection of popliteal artery. Fasciotomy, useful in some cases of this kind, was not successful in wounds involving the popliteal artery in this experience.

on palpation. Subcutaneous edema eventuallyappeared, but was often absent at first and mightnot appear until 3 to 5 days after injury.

  The nature of this brawny swelling in an ischemic extremity was never entirely clarified. An occasional biopsy specimen showed no evidence of extracellular edema or infection. The swelling was sometimes compared to the changes observed in rigor mortis, but the comparison was actually not valid since in ischemic swelling the muscles always appeared to be contained under considerable pressure by the enveloping fascia. It was sometimes explained as the result of generalized intracellular edema of the ischemic muscle cells, while the subcutaneous edema which appeared later was explained as the result of venous stasis and of inflammatory changes in the leg. More probably the chain of events was as follows: When the main arterial supply to the muscles of the affected extremity was suddenly interrupted, a vicious circle promptly developed in which impairment of the capillary circulation was increased by swelling of the muscular tissue and the pressure of the enveloping fascia. The swelling and pressure, in turn, were increased with increasing impairment of the capillary circulation until a stage was eventually reached at which the circulation ceased entirely.


  On the basis of these assumptions, fasciotomy 136 was recommended in these cases, both for those in which the condition seemed incipient and progression likely and for those in which the patient was not seen until it had become fully developed. Longitudinal incisions were used for both upper and lower extremities. In the leg the incision was made posteromedially to decompress the muscles of the gastrocnemius-soleus group, and anterolaterally to decompress the anterior tibial compartment. In the forearm the incision was made on the volar aspect. When the fascia was incised, it spread widely and the pale underlying muscles bulged forcibly through the incision.

Fasciotomy was open to the criticism that the incision might destroy the collateral circulation from the skin and thus further compromise the regional circulation by introducing the risk of infection. On more theoretical grounds, however, these objections were superseded by the consideration that the operation might permit the reestablishment of circulation through the ischemic musculature. Certain British surgeons 137 in the Mediterranean theater were enthusiastic advocates of fasciotomy and reported excellent results with it.  American surgeons in the area were much less favorably impressed by its possibilities. In the limited number of cases in which it is known to have been used, results were occasionally good (Cases 10, 21) but they were uniformly poor in wounds of the popliteal artery (Case 11).

Case 21.   A 25-year-old soldier incurred an accidental wound of the right arm 21 November when two carbine bullets perforated the arm through the biceps and lodged in the cervical region and the superior mediastinum.  The brachial artery and vein were severed and the median nerve was partially severed and badly traumatized.  There were no fractures.  A tourniquet was applied to the right upper arm 5 minutes after wounding and was released and reapplied in about an hour.

  When the patient was admitted to a station hospital 1 hour and 45 minutes after the accident there was a large, tense hematoma in the right arm extending into the axilla.  The arm was cold and limp and was without sensation.  There was no sensation in the forearm; its muscles were soft and relaxed.  Radial and ulnar pulses were absent.

Operation was performed at once with the patient under thiopental sodium anesthesia.  The incision was made over the medial edge of the biceps muscle.  Clotted blood was evacuated.  The brachial artery and vein were ligated above the profunda branch.  The ulnar nerve was not identified.

  Ice bags were applied to the arm and forearm as soon as the operation was concluded.  About 11 hours later swelling and stony hardness were observed in the uninjured forearm, and it was also noted that the superficial veins were dilated.  No pulses were palpable at the wrist.  The wrist and fingers were held stiff in flexion.  When the ice bags were discontinued 12 hours after operation the patient said that the arm had been more comfortable when they were in place.  No other effects were noted from their use.

  Twenty-four hours after the first operation (about 26 ? hours after wounding), with the patient under thiopental sodium anesthesia, the forearm was decompressed by 3 incisions, 1 dorsal and 2 ventral, which were carried throughout the length of the arm and through the deep fascia.  The surgeon described the tissues as "wet."  Shortly after the

136 See footnotes 22, p. 70 and 126 (1), p. 127.

137 (1)  Clark, R.:  Forward surgery of injuries to major blood vessels. Proc. Cong. C. M. F. Army Surgeons, Rome, 12-19 Feb 45, p. 107-111.

(2)  See also footnote 42 (1), p. 95.


operation had been completed, the wrist and fingers which had been held in flexion could be extended passively.  The forearm and hand were put up in a plaster splint with the hand in the position of function.   Recovery was uneventful.  On 28 November, although the pulses were still not palpable at the wrist, the fingers were warm.  By 9 December the swelling of the forearm was less, but the tissues were still not soft to palpation.  There was beginning ability to extend the wrist and some return of sensation in the radial nerve distribution on the hand, but the median and ulnar nerves showed no return of motor or sensory power. The patient was not observed after this date.

Comment.  It is doubtful that the brief refrigeration served any useful purpose in this case. On the other hand, there seems little doubt that the arm would have been lost if fasciotomy had not been performed promptly.

  The condition just described may be classified as an early phase of the type of contracture known in civilian practice as Volkmann's ischemic contracture and usually observed in injuries associated with fractures. Contractures involving the forearm and hand, and less frequently the leg and foot, constituted one of the most crippling sequels of acute ischemia consequent upon arterial wounds. They apparently developed as a result of severe ischemia which stopped just short of causing actual gangrene. The incidence of this complication in World War II is not known, but it was apparently relatively infrequent. In 1 series of 35 arterial injuries observed at a vascular center in the Zone of Interior there were 4 instances of contractures, and in another series of 77 injuries there were 20 instances. These figures are, of course, weighted:  Patients who were hospitalized in vascular centers were sent there because of complications.

  The management of these contractures was not very satisfactory.  In spite of every effort to improve the circulation, including sympathectomy, the results were only moderately good at best.  Other attacks upon the problem took the direction of orthopedic plastic operations and physiotherapeutic measures designed to make the best use of whatever functioning muscle tissue might remain.

  Causalgia.  Causalgia was not a frequent complication of the vascular injuries observed in World War II. If only the figures from vascular centers are considered this observation might seem to be incorrect; there were 17 instances in a series of 77 cases observed at a vascular center. As just pointed out these figures are weighted in that they include only patients in whom complications required their reference to the vascular centers. In a series of 75 cases of causalgia reported by Ulmer and Mayfield, 138  the etiologic factor in no instance was an arterial injury.  In the cases observed at the vascular centers the incidence of causalgia was always higher when nerve injuries were associated with the vascular injuries.

  The great majority of patients with causalgia responded well to sympathetic block and sympathectomy.

138  Ulmer, J. L., and Mayfield, F. H.: Causalgia; study of 75 cases.Proc. Neuropsychiat. Conf., Sixth Serv. Command, 179-184, Jul 1945.


  Circulatory Insufficiency.  Clinical evidences of circulatory insufficiency were seldom apparent soon after wounding in the army areas or even in the communications zone. They became manifest when the wounded became ambulatory and were rehabilitated for either return to duty or discharge from the Service. Since a detailed discussion is given in Chapter XV, only brief mention is necessary here.  Clinical manifestations included color changes, intermittent claudication, and, in the more severe cases, partial paralysis.  Some patients experienced only mild discomfort even after exertion, while others had pain even when resting.  Similar observations had been made during World War I and have been described by numerous observers in civilian practice.139

  No accurate data are available concerning the exact incidence of this clinical manifestation of acute arterial injuries in World War II.  In one sample of 88 cases, however, vascular insufficiency severe enough to be manifested clinically was observed in 68 percent.  Two patients had had early sympathectomy and in 49 the operation was done late. With the exception of 1 patient, all were definitely improved.

  The concept of sympathectomy in the management of these cases was developed by Leriche140 in World War I and this procedure apparently provided the best results reported in the management of this condition in World War II, though it was, of course, not always successful.

Case 22. This 23-year-old private in the Infantry was accidentally shot by another soldier 15 May 1944 on the Anzio beachhead.  The wound was debrided at a field hospital 4 1/2 hours after wounding.  The left femoral artery and vein were ligated in the midthigh distal to the profunda femoris branch.  Sulfadiazine therapy was instituted and lumbar sympathetic procaine hydrochloride injections were done daily for 3 days.  On 23 May the wounds were closed and healing progressed satisfactorily.

  When the patient was evacuated to a general hospital 26 May, examination revealed the left foot to be colder than the right  No pulsations could be felt in it.  The foot was not swollen and the color was good. Both feet were sweaty. Two days later a transfusion of 1,000 cc. of whole blood was given.

  On 20 June, which was a cold day, both feet were cold and moist.  The left foot was slightly cyanotic and no pulses were demonstrable.  On 8 July the posterior tibial pulse could be felt without difficulty.  The dorsalis pedis pulse was weak.  The temperature of both feet was about the same; both were cold and sweaty.  The patient could walk about 400 yards slowly and without pain, but climbing two flights of stairs at moderate speed caused pain in the left calf.  On 31 July (a cool day) the left foot was cooler than the right;

139  (1)  Burrows, H.: Paralysis following arterial injuries. Brit. M. J. 1: 199-203, 16 Feb 18.

  (2)  Leriche, R.:  De la sympathectomie p?ri-art?rielle et de ses r?sultats.  Presse m?d Paris 25: 513-515, Sep 1917.
(3)  Leriche, Ren?: La Chirurgie de la Doulcur.  Paris, Masson et Cie, 1940.

(4)  Leriche, Ren?: Physiologie Pathologique et Chirurgie des art?res. Principes et m?thods de la Chirurgie Art?rielle. Paris, Masson et Cie, 1943.

  (5)  Leriche, R., and Heitz, J.: R?sultats de la sympathectomie p?ri-art?rielle dans la traitement des troubles nerveux post-traumatiques d'ordre r?flexe (type Babinski-Froment).  Lyon Chirurg. 14: 754-792, Jul-Aug 1917.

  (6)  Leriche, R., and Werquin, M. G.: Effects of arterial ligature on the vasomotor system. Lancet 2: 296-297, 7 Sep 40.

 (7)  Mitchiner, P. H.:  Injuries of bloodvessels and their treatment.  St. Thomas's Hosp. Gaz. 38:92-96, Jul 1940.
(8) Mitchiner, P. H.: Treatment of wounds of blood vessels. Post-Grad. M. J. 16: 157-161, May 1940.

(9)  See also footnotes 12, p. 64; 38, p. 84; 103 (5), p. 115; 113 (2), and (3), p. 116.
 140 See footnotes 139 (2) and (5).


both were sweaty.  At this time, 2 ? months after injury, the patient could walk only 200 to 300 yards at moderate speed without complaining of pain in the left calf.  

  On 10 August left lumbar sympathectomy was performed, with the aid of spinal analgesia, through an anterior muscle-splitting extra peritoneal approach.  The second and third lumbar ganglia and the connecting sympathetic trunk were excised.  By 26 August; 16 days after operation, the patient could walk at least 10 times the distance he could walk prior to sympathectomy at the same rate of speed without cramping the calf.  On 20 September, 41 days after sympathectomy, he walked 1? miles without pain.  The foot and leg were warm and dry, and the left foot was now warmer than the right.  It was not swollen.  The posterior tibial pulse was fairly strong.

  As of this date the patient returned to limited duty in the Mediterranean theater.

  Case 23.   This 28-year-old Infantry sergeant was wounded in action in Southern France 19 August.  He received penetrating wounds of the left hip and thigh and right heel, and lacerations of the right femoral artery and vein.  The wounds were debrided at a field hospital (apparently soon after wounding, though the exact hour is not available) and the femoral vessels were ligated at the site of injury below the profunda femoris branch.

  Penicillin therapy was begun immediately after operation and lumbar sympathetic injection of procaine hydrochloride was done daily for 3 days.  When the patient was evacuated to a general hospital in Italy 25 August, he was quite pale.  The plasma protein concentration was 6.7 gm. per 100 cc., the hemoglobin 8.2 gm. per 100 cc., and the hematocrit value was 24.1 percent.  The right foot was pale, slightly cyanotic, and cooler than the left foot, but was not swollen.  No pulses could be felt in it.

  The patient was given 1,000 cc. of whole blood 26 August.  When he had received about 100 cc. of a second transfusion the following day he had a severe reaction with back pain, numbness, tingling of the extremities, dyspnea, and cyanosis.  The symptoms were promptly relieved with adrenalin.  When the blood was rechecked it was found compatible and the reason for the reaction could not be determined.

  Secondary closure of the wounds was done 27 August.  A transfusion of 1,000 cc. of whole blood was given during the operation.  Healing was satisfactory.

  On 2 October, 2 months after injury, the patient complained of cramping pain in the right calf after walking 500 to 600 yards at moderate speed. The pulses were now palpable, although weak.  The foot was cold and pale on cold days.  On 28 October right lumbar sympathectomy was done, with the aid of spinal analgesia, through an extraperitoneal muscle-splitting anterior abdominal incision.  The second and third lumbar ganglia and the connecting trunk were excised.  Soon after operation the right foot was found to be dry and to be warmer than the left foot.  By 6 November the pulses in the foot were of fairly good volume.  A week later the patient was able to take daily walks without cramping in the calf. By 25 November, 3 months after injury and 1 month after sympathectomy, he could walk 2 miles at moderate speed without pain in the calf.  The right foot was dry, not edematous, and warmer than the left.  The color was good and the pulses of good volume.

  Soon afterward the patient was returned to limited duty in the Mediterranean theater.

  Case 24.  A 22-year-old soldier was accidentally wounded when one of his own mortar shells exploded prematurely in Italy 23 November.  The shell fragment caused a complete interruption of the right femoral artery and vein.  There were no fractures.  A tourniquet was applied promptly above the wound and firm admitted to a field hospital platoon 2 hours after injury.  Resuscitation was accomplished with 5 units of plasma (in addition to a unit of plasma station) and 1,000 cc. of whole blood.

  Operation was performed with the patient under ether anesthesia 5 ? hours after injury.  The foreign body was removed and the severed femoral artery and vein were ligated. right lumbar ganglionectomy was then done with removal of the second, third, and fourth


lumbar ganglia.  Four units of plasma and 1,000 cc. of whole blood were given during operation and gas gangrene antitoxin was given immediately afterward.  A well-padded circular cast was applied and bivalved.

  On 24 December, the day after operation, the right leg was warm to 8 inches below the knee. The remainder of the leg and the foot were cool. On 27 December the circulation in the right leg was good and it was warm down to the junction of the middle and lower thirds.  Below this level the leg was cyanotic and cold (Fig. 9).  It was thought that demarcation would take place low enough to save the knee.

  No particular change was noted in the extremity until 2 January.  The muscles continued to be tense and on this date pitting edema was first observed.  The following day, although there were no systemic signs or symptoms, supracondylar amputation of the leg was decided on because of the absence of apparent improvement in the foot and leg.  Amputation was done above the knee because of poor circulation in the muscle of the leg.

Figure 9. (Case 24) Complete severance of right femoral artery and vein treated by ligation belowe profunda femoris. Dry gangrene of great toe and second two 4 days after operatoin with line of demarcation at jumction of middle and lowe third of leg. Supradondylar amputaion was required in this case because of muscular ischemia.

  Comment. This is a case in which sympathectomy failed to achieve the desired results. It also illustrates the possible discrepancy between the appearance of the skin and the condition of the underlying muscles. The appearance of the skin, and the apparently excellent circulation and healthy subcutaneous tissues, suggested that the muscles would be in equally good condition. Actually, although the skin was viable, the muscles, as frequently happens in this type of case, were found to be necrotic. In retrospect, it would have been wiser to delay amputation as long as there was no evidence of toxicity from the ischemic leg in the hope of saving the stump below the knee.