Acute Vascular Trauma in Korean War Casualties: An Analysis of 180 Cases*
Lieutenant Colonel Carl W. Hughes, MC, USA
Acute injuries of major arteriesal ways constitute a threat to life and limb. During time of war, these injuries assume even greater importance because of the epidemic increase in numbers of such lesions. The Korean conflict provided the opportunity for the observations recorded in this communication. This analysis, which embodies 79 injuries of major arteries, 30 of minor arteries, and 71 of major veins, comprises the personal studies of one of three observers working independently in Korea as members of the Surgical Research Team of the Army Medical Service Graduate School, Washington, D. C. Findings of the other two observers have already been reported.1 2 At the same time, a similar program was under way by Navy medical officers in the Marine hospitals in Korea.3
The most complete review of acute vascular injuries is the De Bakey and Simeone analysis of 2,471 cases from World War II.4 They found that the incidence of amputations from injury to major vessels of the extremities was 49.6 percent. Of 81 suture repairs of these lesions, the amputation rate was 35.8 percent. In the initial phase of the Korean conflict, attempts at anastomosis of damaged arteries were only slightly successful.5
Major Artery Injuries
During a 6-month period in Korea, 79 consecutive injuries to major arteries in 76 unselected patients were treated by the author at the 8055th Mobile Army Surgical Hospital (later named the 43d Surgical Hospital). The distribution of these injuries is presented in Table 1.
The 79 vascular injuries encountered were broken down as follows: lacerated 42, severed 31, contused with spasm 2, and contused with thrombosis 4. Most patients had multiple injuries. Complicating these were severe open, comminuted fractures present at the site of the arterial injury in 12 cases, with less severe open fractures present in 5 other cases (Fig. 1). All but 10 wounds (12.7 percent) were caused by fragmenting missiles.
*Previously published in Surgery, Gynec. & Obst. 99: 91, 1954.
Table1. The Distribution of the Major Arterial Injuries
Thirty of the patients arrived in shock of varying severity. Thirty-seven were admitted with a tourniquet in place which had been applied from 40 minutes to 14.5 hours previously, the average time being 4 hours. The over-all time from injury to surgery averaged 10 hours with extremes from 3.5 to 28.5 hours. Of this interval, evacuation time to the hospital averaged 5.8 hours and time from admission to surgery an additional 4.2 hours. Resuscitation was often a problem and, although the patients were sometimes admitted soon after injury, the time lost during resuscitation sometimes militated against recovery of the limb. This is a reflection of the severity of injury and shock in these patients.
Concomitant nerve injury, not necessarily at the site of the vascular injury, was evident in 36 of the 79 vascular injuries. Of these, there were 20 major nerve injuries, 9 minor nerve injuries, and 7 instances of nerve involvement believed due to anoxia, tourniquet, or mild contusion. Because of the multiplicity of wounds, prolonged use of the
FIGURE1. Successful vascular repairs were performed in the presence of such open, comminuted fractures
tourniquet, and periods of anoxia, it was difficult to determine the cause of nerve injury in some cases when the actual damage was not visualized.
Major vascular injuries were given priority evacuation along with other priority cases. High priority is important for resuscitation and surgery since every minute may be vital in saving a limb.
Preoperative oscillometric studies were made on limbs showing absence of a pulse, unless the limb was in traction or a tourniquet was in place. Twenty-two of the 32 patients admitted without tourniquet had no palpable pulse in the injured extremities. Oscillometric readings in these limbs were almost universally zero, with only three showing a quiver of the oscillometer needle. Immediately following arterial repair, some cases showed better oscillometric readings on the repaired side than on the normal side but the group as a whole averaged 0.67 unit less on the repaired side.
An anatomic surgical approach was used regardless of location of the wound. An adequate incision was always made and proximal control of the vessel secured first. All repairs were by suture with 00000-braided arterial silk, continuous everting mattress technic, intima to intima placement.
Two Pott`s ductus or coarctation clamps plus arterial tissue forceps, a fine needle holder, and 00000-braided arterial silk were the only additional equipment found necessary for peripheral vascular surgery. Each hospital receiving such patients should be so equipped.
All wounds were adequately débrided, but all collateral vessels were preserved where possible. Even after arterial reconstruction, it often took the damaged muscle some time to bleed, making it difficult to evaluate properly the viability of the muscle distal to sites of arterial injury.
The repaired vessels were covered for nourishment and to avoid secondary infection. Coverage should be loose but adequate, and without primary closure of the skin. If infection occurs, it will often result in thrombosis at the site of repair.
To be sure there was no undue tension on the suture line, repairs at joint sites were inspected at completion of surgery by fully extending the joint and visualizing the repair site. These limbs were then flexed 10 to 15 degrees for 2 weeks, being restrained only with adhesive tape to allow flexion but to limit extension. Limited active motion to improve the circulation was encouraged as soon as possible after surgery. When active motion was not possible, passive motion was utilized.
Delayed closure of clean wounds was carried out on the fourth to sixth postoperative day when the patient workload permitted. If
wounds remained clean and healing uneventful, joints were gradually, fully extended and the patient ambulated after 2 weeks.
When bed space permitted, these patients were observed at the surgical hospital for 8 to 10 days, then evacuated to Army hospitals in Japan where primary follow-up studies were initiated. Those eligible for evacuation to the United States were subsequently evaluated further by arteriography, oscillometry, and skin temperature studies at Walter Reed Army Hospital. Inasmuch as some of the casualties were Ethiopian, Korean, Colombian, and Commonwealth patients, long-term follow- up studies were possible only on the Americans.
It is fully realized that long-term follow-up studies will be necessary for a complete evaluation of acute arterial repairs. However, it appears at this time that a very high percentage of limbs with arterial damage can be saved if the patient can be evacuated to a hospital and resuscitated before irreversible muscle changes take place.
Six patients with 7 arterial repairs (popliteal 4, superficial femoral 2, common femoral 1) died shortly after operation from causes other than arterial injury. For this reason, adequate evaluation was not possible and these cases are eliminated from the analysis, except to mention any complication occurring before death. Summaries of these are given in the addendum. Of the 72 remaining patients who survived, 8 subsequently came to amputation of the involved limb. Thus, 11.1 percent of the limbs were lost and 88.9 percent were saved.
The indications for the 8 amputations warrant some explanation. Five of these were performed within the first postoperative week and three subsequently. Seven were in lower extremities and one in an upper extremity. Five amputations resulted from irreversible muscle changes which probably had taken place prior to vascular repair. These muscle changes were obvious at the time of reparative surgery in three cases, and the viability of the muscles was questionable in the two others. Muscle biopsies at the time of repair confirmed the gross observations. Cold injury was a complication in two of the amputees. In these five cases, the time from injury to repair varied between 11 and 24 hours, with a mean of 16 hours. Restoration of blood flow was effected in the hope of saving the limb or, at least, providing the longest possible viable stump. At the time of amputation there was a brisk flow of blood in the already devitalized muscles. If these five cases with irreversible muscle damage at operation are excluded, repair in this series was successful 95.8 percent and amputation was required in 4.2 percent.
A sixth patient required amputation because of venostasis, with ensuing liquefaction of the muscles. The seventh amputation resulted from slippage of the fractured femur with compression and thrombosis of the repaired artery. The eighth patient required a transmetatarsal amputation 3 months later for dry gangrene of the toes; probably the result of a small embolus from the repair site and not directly due to initial failure of the repair.
Table2. Resulting Amputations, Vessels Injured, and Causes
Of the repairs in limbs subsequently amputated (Table 2), grafts were employed in 37 percent of the cases, as compared to 19 percent for the whole series. This is a further indication of the extent of the trauma in these patients. It should also be noted that the injuries were to arteries at critical levels which may be expected to result in gangrene in a high percentage of cases if ligated.
Factors Influencing Saving of Limb Following Blood Vessel Surgery
It is not possible to establish a specific time interval after which repair is no longer practicable because of muscle necrosis. There are many variables which make it impossible to determine clinically when muscle changes are irreversible. Factors to be considered are the length of time after injury, severity of the wound, the artery injured, the level of the injury, the number of collaterals damaged, the outside temperature, exposure, the state of shock, and anatomic variations. A decrease in the surrounding temperature (short of causing frost-
bite) appears to be advantageous. The converse also seems to be true. During very hot weather, muscle changes and gangrene have been observed to occur much more rapidly. Acute vascular injuries should be repaired in borderline cases of irreversible muscle damage, even though the limb is expected to be lost. Restoration of blood flow may preserve some length of the limb or even a joint before a line of demarcation appears.
The application of a tight pressure dressing was not always adequate for control of hemorrhage as evidenced by 47 percent of these patients admitted with a tourniquet in place. Without control of the hemorrhage, resuscitation cannot be accomplished. Prolonged use of the tourniquet for resuscitation may result in a nonviable extremity, so a vicious cycle is created. In such cases, the bleeding artery was exposed and a clamp was applied to the very end of it in order to preserve all of the artery possible for repair. A pressure dressing was then usually adequate, and resuscitation could be continued.
Many physicians are still not tourniquet-conscious. Tight Esmarch tourniquets have been found covered by external dressings. It is believed that a tourniquet should be applied in the field and left in place until removed by a medical officer at a time when blood is available to resuscitate the patient. A pressure dressing will usually suffice if it is applied tight enough to control the hemorrhage. If watched carefully, the pressure dressing can be loosened after a matter of 20 to 30 minutes, but continued observation is necessary to prevent recurrence of hemorrhage as blood pressure rises.
Many patients were admitted with muscle hardened by ischemia with contractures and fixed, frozen joints. These extremities were cold, spastic, anesthetic and paralytic. After restoration of blood supply, the muscles swelled and fasciotomies were performed. At fasciotomy, the muscle was often found to be pale, swollen, and ischemic, and a blood-tinged transudate often ran out as the muscle bulged from the compartment. It was observed that muscles of the anterior tibial compartment of the leg and the flexor compartment of the forearm were the most vulnerable to swelling and broke down more quickly than others if fasciotomy was not done early. As more experience was gained, fasciotomy was performed prophylactically in borderline cases.
After restoration of blood supply and the initial swelling, such extremities became warm; the swollen muscles softened, relaxed, and sensory and motor function gradually returned. This process sometimes took 2 to 3 days. In the lower extremities, peroneal nerve function was invariably slow to return.
Serial muscle biopsies were taken in many cases showing various stages of degeneration, fragmentation, hyalinization and regeneration of muscle. In other instances, microscopically normal muscle was seen after prolonged periods of anoxia. This project was undertaken to study the condition of muscle and its blood supply at the time of surgery and to observe the progression of muscle degeneration and regeneration. It was hoped that the cause of irreversible muscle damage might be delineated as being due to muscle fiber death or secondary to venous, capillary, and arterial thrombosis.
In some of the above cases, changes developed varying from microscopic areas of focal necrosis to loss of whole muscle compartments. Some of these muscles underwent complete sterile liquefaction. On dissection of these extremities after amputation, many small veins were found thrombosed while the arterioles were patent. One of these patients came to amputation 18 days after repair because of slow muscle changes. One compartment after another was débrided until only skin, bone, nerve and vessels remained.
It is possible in some cases that restoration of the blood flow hastened muscle degeneration. Experimental work has shown that rapid revascularization of ischemic muscle hastens degeneration.6 It has been suggested that there may be some similarity to frostbite where a slow, rather than rapid, return of blood flow and warmth may be important to survival of damaged muscle.7
In the lower extremity, when muscle degeneration had taken place before restoration of blood supply, it was invariably found at amputation that the soleus muscle had undergone more extensive degeneration than the gastrocnemius.
These muscle studies are still in progress and will be reported separately and in detail at a later date.
Types of Repair
While simple suture repairs of lacerated arteries were held to a minimum in this series, some were done by others in the Theater. Except for the small, clean, knife-like laceration, it is difficult to evaluate, débride and suture the damaged artery without constriction of the lumen. It is usually better to exercise the damaged portion and do an end-to-end anastomosis. It was possible to do a direct anastomosis in 65 percent. of the cases (Table 3). The portion of artery excised averaged 0.75 inch. With dissection of the artery for approximately 2 inches in each direction, an anastomosis could usually be done without undue tension (Figs. 2 and 3). Major branches and collaterals were preserved, although minor branches were at times sacrificed to gain arterial length.
Table3. The Frequency and Types of Repair Utilized
FIGURE2. A brachial artery anastomosis and arteriogram showing the repair site several months
FIGURE3. A popliteal artery anastomosis and arteriogram showing the repair site several months
Excised sections of damaged arteries were studied histologically for adequacy of débridement. Several sections taken from the cut ends of the artery adjacent to the anastomosis have shown microscopic changes including polymorphonuclear infiltration and irregularities of the internal elastic membrane, which might be interpreted by some as being due to inadequate débridement. Follow-up studies have shown no correlation with postoperative thrombosis in these cases.
Some anastomosed segments removed from amputated limbs and at autopsy have shown microscopic areas of infarction at the suture line. It has not been determined whether these changes represent inadequate débridement or changes secondary to the stripping of adventitia and vasa vasorum incident to suture of the vessel. It is possible that such infarctions may occur in many cases and that the infarcted area is replaced in the same manner as a graft. Details of histologic studies of arteries will be reported in a separate communication.
Grafts were required in 19 percent of the cases (Table 3). Arterial defects up to 4.5 inches in length were encountered but, as already mentioned, the defects averaged 0.75 of an inch. Both autogenous vein grafts and homologous arterial bank grafts were utilized. Autogenous vein grafts from 0.75 inch to 3.75 inches and homologous artery grafts from 1 inch to 4.25 inches were used (Fig. 4). The saphenous vein was most often used as a graft; however, the cephalic, basilic and others were also used. The saphenous vein is usually of adequate size for most peripheral vessels and appears to be stronger than veins of upper extremities. Vein grafts are always turned end for end, when inserted, so the valves do not obstruct blood flow. Grafts usually are not taken from the involved limb, especially if other major veins of the limb are damaged. Sometimes the companion vein of the artery may be utilized, if already damaged, but usually it is too large. Vein grafts should be chosen slightly smaller than the artery into which they are to be inserted to prevent overdilatation.
Homologous arterial bank grafts were preserved in a modified Tyrode`s nutrient graft solution, up to 30 days.8 Availability of bank grafts saves time and is advantageous in cases in which an autogenous vein graft is not always readily accessible because of the patient`s position at surgery.
Autogenous vein grafts have been more satisfactory than homologous arterial grafts. This has borne out our previous experience.9 Although no limbs were lost, as a result of it, two of the four homologous arterial grafts in this group thrombosed within 4 to 8 weeks. The other two, utilized in extremities which came to amputation, were patent at the time of amputation 6 days after insertion.
FIGURE4. An Example of a massive avulsing wound of the upper arm prior to surgery and after debridement and insertion of a 3-inch saphenous vein graft into the brachial artery.
Ligation was reserved for non-critical arteries in which it was deemed unwise to do a repair, and for patients who could not tolerate prolonged surgery. Of the 3 major arteries ligated in this series, the terminal brachial, involving ulnar and radial bifurcations, and the terminal popliteal, below the origin of the anterior tibial artery, were involved. Two of these patients were in such poor condition that they could not tolerate further surgery. In each case it was determined at surgery that the limb had adequate circulation for viability.
Ligation of the concomitant vein at the time of artery ligation was not practiced and it is believed that it may even be a harmful procedure.
Previously, a very high percentage of injuries of the carotid arteries, ligated or repaired immediately, resulted in death or hemiplegia. It has been found in partially severed major arteries of extremities that, if ligation is delayed several hours or days, a much lower incidence of gangrene occurs. Such arteries may be ligated weeks later and gangrene is unusual. On this basis, it was decided that when the hematoma was self-contained and the hemorrhage controlled, all carotid injuries would be treated conservatively, particularly if an arteriovenous fistula was present. The wounds were débrided superficially and great care was taken not to dislodge the clot and to avoid wound infection. It was believed that even if the patient bled 1 or 2 days later, his chances of survival were much better after delayed rather than immediate surgery. There has been no occasion to regret this decision. Of the four cases in this group treated conservatively, three have since had successful definitive surgery and the fourth apparently had spontaneous closure of the arterial defect.
The same conservative treatment was applied to self-contained, high subclavian injuries because of the difficult surgical approach, long operation, and dangers of massive hemorrhage and infection. Such lesions were allowed to develop an arteriovenous fistula or false aneurysm and then were operated on subsequently.10
While follow-up studies sometimes show some constriction at the suture line, actual thrombosis has been quite rare. These studies have also shown the repaired limb to average 1.6 oscillometric units less than the normal side. Several factors may influence these oscillometric changes--such as the multiplicity of wounds, nerve injury, scar tissue, or any constriction present at the repair site. These same changes may also alter skin temperature changes which have averaged one degree less in the limb with the repaired vessel in the follow-up studies.
There must be close observation for postoperative complications of infections and hemorrhage. Infection may result in thrombosis or dehiscence of the arterial suture line with hemorrhage. Penicillin and streptomycin were used prophylactically in these cases and usually the infection was not a problem. In fact, only one severe infection occurred; this was after a vein graft and resulted in thrombosis of the graft. Dehiscence and hemorrhage may also occur from a poor anastomosis, inadequate arterial débridement, inadequate arterial coverage, or too much tension.
Hemorrhage from the suture line in the immediate postoperative period was not observed. It was reported in three cases in follow-up studies. One such patient who was anuric, and who was dialyzed, had repeated convulsions and bled from his suture line into a nonhealing wound 12 days following operation. This patient died. A second patient bled from his wound following delayed skin closure 21days after operation. The peripheral pulses were unaltered. After a second hemorrhage 36 days postanastomosis, the pulses gradually disappeared. There was no surgical intervention or proof of the site of the hemorrhage, but this patient had a severe open, comminuted fracture of the femur and had been in traction during this period. The artery may easily have been lacerated by a bone fragment. This patient was still in traction 9 months later. A third patient bled from his suture line 14 days postoperatively. The vessel was ligated without limb loss. Two other patients had hemorrhage as a complication but not from the suture line. One of these, the third fatality listed in this series, had an open, comminuted fracture of the femur which lacerated the artery above the repair site. The other case of hemorrhage occurred 11 days following injury with hemorrhage from the artery one inch above the anastomosis, presumably from a damaged area in the artery, overlooked at the time of surgery; this necrosed later.
Thrombosis was not a problem in arterial repair. Anticoagulants were not used and are not considered necessary.
Although amputation in one case resulted from arterial spasm alone (see Table 2), this was rarely a problem. Spasm was treated variously at the time of surgery with 2.5 percent papaverine hydrochloride, warm saline packs, segmental periarterial sympathectomy, and 25 to 50 mg. of intra-arterial priscoline. Postoperatively, repeated sympathetic blocks with 1 percent procaine were used as indicated. Sympathectomy was not done after repair of these acute injuries. A second patient, treated for severe segmental spasm only, still has a patent artery but a slight constriction throughout the length of the site of the original spasm. It is not known whether this is clue to changes in the arterial wall, a thrombus within the artery, or to scar tissue and callus formation from the associated open fracture.
The trauma causing the arterial damage, the vasoconstriction of the collaterals, plus shock and its accompanying peripheral vasoconstriction, all combine to decrease the blood flow to a limb with arterial injury. As a result of this chain of events, some patients having major arterial damage, even for a short period of time, may exhibit muscle changes resulting from anoxia. Open, comminuted fractures did not cause as high a percentage of complications as expected. While severe open, comminuted fractures occurred in ony 15 percent of the arterial injuries, direct complica-
tions could only be attributed to the fractures in two cases previously mentioned, In one, the fracture slipped, compressing the repaired artery and requiring amputation. In the other, laceration of the repaired artery was produced by the bone fragments, requiring ligation of the artery.
Such fractures complicate vascular repairs by slipping and compressing at lacerating the vessel. Tissue pulled between the artery and bone is protective. To avoid damage to the repaired artery, pin fixation in a cast may be necessary in cases such as fractures of the distal femur. Fixation in casts should be reserved for those with concomitant fractures. Such casts should be bivalved immediately and taped together to allow immediate access to the repair site in the event of hemorrhage.
Minor Artery Injuries
Thirty injuries to minor arteries were studied in other patients. Minor arteries injured included the profunda brachii, profunda femoral, radial, ulnar, anterior and posterior tibial. When only one of two minor arteries to an extremity was involved (such as an injured radial but intact ulnar), the injured vessel was ligated. In no case did this procedure lead to complication. In instances in which both vessels were involved, the limb was invariably so macerated and shattered that amputation was necessary. In such cases, one or both of the arteries would have been repaired had there been viability.
Concomitant major vein injuries were found in 50 cases with major arterial injuries. Other major vein injuries were added to this group to bring the total to 71. Fifty-eight of these veins were ligated. Complications ranged from none to various degrees of venostasis. In one patient, loss of an extremity resulted from severe venostasis. The popliteal, greater and lesser saphenous veins had been severed. While there was good arterial inflow, the muscles gradually liquefied and required amputation. On dissection of the leg, the arteries were all patent but the veins were all thrombosed. This observation has also been recognized by other investigators; 3 and is similar to a massive venous thrombosis.
After complications were observed following venous ligation, major vein injuries, especially those at critical levels, were carefully evaluated for repair. An evaluation of an extremity should be made at the time of surgery to determine the extent of other vein injuries before a repair is performed. While the author has performed numerous venous repairs at elective surgery,10 it was realized there was
more danger of thrombosis and embolism in acute trauma. However, venous repair is believed to be a safe procedure if the clots are carefully removed and the vein is properly débrided and sutured. The size and number of wounds, size and level of vein injured, and the number of collateral veins injured are all factors to be considered in determining whether or not a vein should be repaired.
Venous repairs were performed by simple over-and-over suture except for one case in which anastomosis was carried out by a continuous everting type suture. Only 13 major veins were repaired in this group, 12 by lateral suture and 1 by end-to-end anastomosis (Table 4). No anticoagulants were used in these venous repairs.
Table4. Locations of Major Veins Repaired
Two patients early developed a thrombophlebitis of the superficial veins of the leg, but in each case the greater saphenous system had been previously ligated at the time of surgery. Follow-up studies have shown that when repair resulted in decrease in size of the lumen, a high percentage of the veins thrombosed. Thrombosis occurred after the initial 8- to 10-day period of observation. Venostasis has not been a major problem at follow-up. It is difficult to place the cause of persistent swelling in some cases because many have neurologic, venous, lymphatic, and multiple muscular
1. One hundred eighty cases of acute vascular injuries to soldiers wounded in the Korean conflict were studied. These consisted of 79 major artery injuries, 30 minor artery injuries, and 71 major vein injuries.
2. After repairs of the major arterial injuries, 88.9 percent of the limbs were saved and 11.1 percent amputated.
3. The amputations included five limbs which had undergone irreversible muscle changes prior to vascular repair. The other three amputations resulted from complications of repairs.
4. Fasciotomies were found to be valuable adjuncts after arterial repair in limbs with borderline muscle changes.
5. A specific time interval after which arterial repair could not be performed was not established.
6. Tourniquets should be applied and left in place until they can be removed by a doctor who has adequate blood available for resuscitation.
7. All degrees of muscle degeneration to regeneration were observed in cases of delayed arterial
8. The major methods of arterial repair were end-to-end anastomosis and use of autogenous vein grafts and homologous artery grafts. Autogenous vein grafts in arteries have been far more satisfactory than homologous artery grafts.
9. Complications were rare, and included postoperative hemorrhage, infection, spasm and thrombosis. Antibiotics were used. No anticoagulants were administered.
10. Thirty minor arteries were ligated without complications.
11. Seventy-one injuries to major veins were treated with some complications which led to the successful repair of 13 major veins.
12. Immediate repair of injuries to major arteries is practical. All major arteries should be repaired except when the additional time required for surgery endangers the patient`s life. While some limbs survive major artery ligation, they are not functionally equal to the limb with a successful vascular repair. If thrombosis should occur at the repair site, the process will often be slow enough so that a collateral blood flow will develop and provide adequate circulation to the limb. If patients with vascular injuries can be brought to surgery before irreversible muscle changes take place, a very high percentage of limbs can be saved.
1. The first death occurred in a patient who was admitted in deep shock with an injury of the common femoral artery. Cardiac arrest occurred soon after induction of anesthesia; and he was resuscitated by cardiac massage. While his chest was being closed; an artery graft was inserted into the defect of the femoral artery. He died 4 hours later.
2. A second patient had a severed popliteal artery plus injuries to all extremities, abdomen, and chest. He was in deep shock for many hours. The artery was anastomosed while abdominal surgery was being performed. Postoperatively, the patient lived only 38hours.
3. A third patient with superficial femoral artery injury and multiple other injuries was admitted in severe shock. His condition was so poor that, after anastomosis of the artery, other débridements were not completed. He continued to be irrational after surgery and, although placed in a spica cast, the open fracture of the femur lacerated the artery above the repair site on the third postoperative day.
Since his condition did not permit prolonged surgery, the artery was ligated. The leg was amputated 9 days later. The patient died of cerebral damage a few days after evacuation.
4. The fourth patient was admitted in severe shock with a severed superficial femoral artery and cerebral anoxic changes. After a quick, easy arterial repair, the patient continued to be mentally hazy and became anuric. He was evacuated to the Renal Insufficiency Center. He had repeated convulsions, and 12 days after surgery he hemorrhaged from the arterial repair. The artery was ligated. Two days later the patient died of cardiac arrest during an attempted further débridement.
5. A fifth death occurred in a patient with a popliteal artery injury. The arterial injury was easily and successfully repaired but the patient died following a subsequent negative abdominal exploration at another hospital.
6. The sixth death occurred in a patient admitted in severe shock with extensive injuries to both popliteal arteries, buttocks, back, all extremities, and open, comminuted fractures of the femur. One popliteal artery was anastomosed while a colostomy was being carried out. The other popliteal artery was ligated in its terminal portion below origin of the anterior tibial artery. The patient`s condition did not permit completion of his débridement. Later at a second operation, he developed pulmonary edema and died.
1. Jahnke, E. J., Jr., and Seeley, S. F.: Acute Vascular Injuries in the Korean War. Ann. Surg. 138: 158, 1953. (Chapter 8 of this volume.)
2. Jahnke, E. J., Jr.. and Howard, J. M.: Primary Repair of Major Arterial Injuries. Arch. Surg. 66: 646, 1953.
3. Spencer, F.: Personal communication.(Johns Hopkins Hospital, Baltimore, Maryland.) 4. De Bakey, M. E., and Simeone, F. A.: Battle Injuries of the Arteries in World War II. Ann. Surg.123: 534, 1946.
5. Hughes, C. W., and Bowers, W.F.: Observations on Blood Vessel Surgery. Am. J. Surg. 85: 174, 1953.
6. Harman, J. W., and Gwinn, R.P.: The Recovery of Skeletal Muscle Fibers from Acute Ischemia as Determined by Histologic and Chemical Methods. Am. J. Path. 25: 741, 1949.
7 .Brooks, B.: Pathologic Changes in Muscle as a Result of Disturbances of Circulation. Arch. Surg. 5: 188, 1922.
8. Cooke, F. N.; Kuhns, D. M.; Elston, J. T.; Dozier, S. M.. and Fusillo, M. H.: Blood Vessel Bank. U. S. Armed Forces Med. J. 2: 1779, 1951.
9. Cooke, F. N.; Hughes, C. W.; Jahnke, El. J., Jr., and Seeley, S. F.: Homologous Arterial Grafts and Autogenous Vein Grafts Employed in Order to Bridge Large Arterial Defects in Man. Surgery 38: 183, 1953.
10. Seeley, S. F.; Hughes, C. W.; Cooke, F. N., and Elkin, D. C.: Traumatic Arteriovenous Fistulas and Aneurysms in War Wounded. Am. J. Surg: 83: 471, 1952.