U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 10

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

The Primary Repair of Wounds of Major Arteries: An Analysis of Experience in Korea in 1953*

Lieutenant Colonel Carl W. Hughes, MC, USA

Damage to blood vessels of the extremities constitutes a small but important component of war wounds. Of 2,471 such wounds collected by De Bakey and Simeone from World War, 1 49.6 percent of those with involvement of a major artery had to be amputated. Ligation of damaged vessels was the accepted practice. They further reported 81 cases in which suture of the artery was performed. In these 81 cases, the amputation rate was 35.8 percent.

Shortly after the beginning of the Korean conflict, Walter Reed Army Hospital was designated as the peripheral vascular surgery center for the Army. This center received a substantial number of Army personnel who had sustained traumatic, arteriovenous fistulas and aneurysms. The results of reparative and reconstructive surgery of many of these lesions have been reported. 2-4 On the basis of this experience, primary repair of damaged major blood vessels of extremities was considered practicable.

Personnel trained in the technic of blood vessel repair were sent to Korea as members of the Surgical Research Team of the Army Medical Service Graduate School. The results of their study established the practicability of repair of major arteries of extremities as part of the definitive surgery of war wounds. A total of 130 major vessel injuries were repaired and follow-up data were reported by three members of the Surgical Research Team (Table ). 5-7  The average amputation rate for these 130 vascular repairs was 10.8 percent.

Ziperman 8  analyzed the results of 234 major and minor arterial wounds treated in Korea during the first 9 months of 1952. His report included the work of two members of the Surgical Research Team. He compared the over-all findings in his group to the results reported by De Bakey and Simeone and he reported 127 major arteries repaired by suture, with an amputation rate of 20.5 percent. This represents a 42.7 percent decrease in amputation rate. Both of these series, however, are based on relatively small numbers.

 *Previously published in Annals of Surgery 141: 297, 1955.


This study includes an analysis of all major vascular wounds repaired in Korea from January 1 through July 1953 (exclusive of Navy and Marine personnel), during which time 211 major arterial injuries were detected in 205 patients. Two patients had three vascular injuries each and two other patients bad two injuries each. The author treated 72 of these vascular injuries (see Table 1); and they are reported in detail in Chapter 11 of this volume.5

Table 1.  Controlled Follow-up Studies on Major Arterial Repairs in Korea, 1952-1953

Materials and Methods

Reports on all cases in the Korean theater were centralized in the Office of the Surgeon of the Eighth United States Army. They emanated from the forward hospitals in which the surgery was performed and from the evacuation hospitals through which these patients passed.

The results are tabulated as short-term and late follow-up studies. Such a division was necessary because this series included 61 patients who were Korean nationals or prisoners of war and other United Nations' personnel, many of whom were not available for later observation. Late follow-up studies were accomplished on other Allied patients as they returned to vascular centers in Japan and on American patients who were returned to Walter Reed Army Hospital where they were subsequently evaluated by arteriography, oscillometry, skin temperature and exercise tolerance studies.

In this over-all theater total of 211 major vascular injuries, deaths occurring from causes other than arterial injury were reported in 15 patients with 18 arterial repairs. Arteries involved in these patients were the carotid 1, axillary 1, common iliac 1, femoral (unqualified) 3, common femoral 2, superficial femoral 3, and popliteal 7. Since an adequate evaluation of the arterial repairs in these patients


was not possible, these 18 repairs have been eliminated, leaving 193 repaired major vascular injuries in 190 surviving patients.

Regardless of efforts made to follow through on every patient, 62 of the 193 repairs in surviving patients were lost to long-term follow-up studies because of evacuation after operation to Korean or United Nations hospitals. Thus, 193 arterial injuries were available for short-term studies; and 131 of these were included in the follow-up studies.

Approximately 85 percent of the vascular wounds were caused by fragmenting missiles and 15 percent by nonfragmenting missiles. Of all the vessels damaged, the majority were lacerated although an almost equal number were severed. Only a few vessels were thrombosed or in spasm (Table 2). The lower extremity was the most prevalent site of vascular injury (55.5 percent), with the upper extremity a close second (38.4 percent), while the neck and trunk received a very low percentage of the injuries. A distribution of injuries is shown in Table 3.

Table 2.  Major Vascular Injuries in Korea, 1953

Table 3.  Major Vascular Repairs in Korea, 1953


Data available from approximately one-third of the patients in this study showed that 40 percent of the patients arrived in shock of varying degree, and 47 percent had a tourniquet applied for an average of 4 hours.

Vascular injuries received a high, but not first priority evacuation unless there was uncontrolled hemorrhage or severe shock which could not be treated prior to evacuation. The time lag from injury to operation for all cases of vascular injury in 1953 averaged 9.8 hours, almost identical to that reported by Ziperman in 1952. Approximately 6 hours of this time was spent reaching the hospital, and 4 hours in preparation for surgery. This is a reflection of the degree of injury and shock in these patients.

An anatomic, surgical approach was used regardless of location of the wound. An adequate incision was always made and proximal control of the damaged vessel secured first. After excision of the damaged portion of the vessel, repair was accomplished by a continuous mattress suture with 00000 braided arterial silk everting the edges of the artery, apposing intima to intima.

Table 4.  Types of Repair in Major Vascular Wounds in Korea, 1953*

Most of the repairs (64.8 percent) were accomplished by direct anastomosis, with autogenous vein grafts (14.5 percent) the second most common type of repair. Lateral suture repair followed in third place, and homologous arterial grafts in fourth place (Table 4). Conservative nonoperative treatment was practiced in some instances of nonexpanding pulsating hematomas, especially of the carotid artery. Ligation was utilized only where repair was not feasible in a noncritical artery or when the patient's condition did not permit further surgical procedures.

After careful débridement, care was taken to cover the repaired vessel in order to nourish and protect it; but the wound was left open


for drainage and to minimize the risk of infection. Penicillin and streptomycin were used routinely in all patients. Sympathectomy was not practiced, and anticoagulants were not utilized. Following operation, only those extremities with complicating fractures were placed in casts. For a period of 2 weeks, only limited active motion was permitted at the vascular repair site. If the wound remained clean, delayed closure was accomplished on the fourth to sixth postoperative day.


Of the 193 vascular repairs in surviving patients, 26 required amputations. These amputations were performed following repairs of the axillary artery in 2 cases, brachial artery (unqualified) 2, femoral (unqualified) 4, common femoral 5, superficial femoral 2, and popliteal 11, most of these being considered critical arteries. The percentage of amputations was higher in those cases requiring grafts for repairs (see Table 5).

Causes for amputation were determined where possible. Seven patients were reported as having good blood flow through necrotic muscle at the time of amputation. The muscle had undergone irreversible changes prior to reconstruction of the artery. Thrombosis was responsible for three amputations; compression of the repaired vessel by a displaced open, comminuted fracture was responsible for one amputation, and venostasis for another. The reason for amputation in the remaining 14 was not recorded. Of these, five limbs were amputated over varying periods of several days to 3 months following repair of the damaged artery. In one case reported here as an amputation, the patient lost only four toes; while another had a transmetatarsal amputation.

An attempt was made to correlate the rate of amputation with the presence of open, comminuted fractures, but the findings are not significant. As may be expected, however, there was a direct correlation between rate of amputation and size of the wound.

Complications were encountered much less frequently than was expected. Hemorrhage from the suture line was quite rare. Infection at the repair site, which may result in hemorrhage or thrombosis, was rarely a problem. Latent thrombosis probably occurs more frequently than is realized ; but the thrombus is often slow in forming, which allows time for the collateral circulation to increase and compensate for the major artery thereby preventing limb loss.

One of the greatest determining factors in the final results of arterial repair is the time lapse from arterial injury to repair. Even this period can be quite variable depending on many factors such as size of the wound, number of collaterals involved, level of the artery involved, ambient temperature, severity of shock and anatomic variations. Although it has been shown that results are proportionally


better when arterial repair is done within 10 hours after injury, an extremity may undergo irreversible muscle changes much earlier or remain viable much longer, depending on the factors mentioned. The author previously reported five cases with major arterial injury repaired from 11 to 24 hours after injury (an average of 16 hours). At the time of amputation, all five extremities exhibited good blood flow through necrotic muscles. It was considered that time-lag from injury to operation was a significant factor in those five patients.

The time from injury to repair in the 26 patients whose extremities subsequently required amputation varied from 1 to 24 hours, with a mean of 10.5 hours, which was almost identical with the average time-lag of 9.8 hours recorded for the entire series.

Numerous patients were seen with the injured limb cold, ischemic, anesthetic, paralytic, and with the joints fixed. As these limbs became warm and sensation and motion returned after arterial repair, they often began to swell and required a fasciotomy. When fasciotomy was delayed, all degrees of muscle necrosis occurred, varying from microscopic areas of focal necrosis to loss of complete compartments. The flexor compartment of the forearm and the anterior tibial compartment of the leg seemed to be most vulnerable.

Short-Term Follow-up

If analogous groups of cases are compared after follow-up studies limited to the theater, findings reported by Ziperman showed that there were 127 major arterial repairs in 1952, with 26 cases (20.5 percent) resulting in amputation. In this study for 1953, of the 193 major arterial repairs with limited follow-up studies there were 26 cases (13.5 percent) which resulted in amputation, an improvement of 34.1 percent in saving of limbs during the last year of the Korean conflict.

Thrombosis was reported as occurring in 29 cases (17.4 percent) of the total surviving unamputated patients with limited studies (Table 5). No arteriograms were made during these short-term studies. Thrombosis was considered as occurring when no pulse returned following arterial repair or when a pulse had been present postoperatively and later disappeared. Failure of a pulse to return following operation may have been the result of faulty technic rather than thrombosis, but the consequences are the same. Excluding the patients who underwent an amputation or died, thrombosis occurred in 15.2 percent of the vessels repaired by direct anastomosis, in 30.0 percent. of those repaired by autogenous vein grafts, and in 62.5 percent of those repaired by homologous arterial grafts. Lateral repair resulted in only 5.6 percent thrombosis.

Late Follow-up Studies

Since 62 of the 193 repairs in surviving patients could not be followed, only 131 cases in which there were late follow-up studies are


Table 5.  Short-Term Follow-up on Major Vascular Repairs in Korea, 1953*

included (Table 6).  Counting the same amputations reported in the short-term group which remain unchanged for this latter group of 131, the amputation rate is increased to 19.8 percent.

Table 6.  Late Follow-up on Major Vascular Repairs in Korea 1953*


An attempt to determine the exact number of thromboses was difficult even in this late group, because it was not possible to do an arteriogram on every patient. The volume of the pulse distal to the repair site is not always indicative of the condition of the repair. At some levels, collateral vessels are quite adequate or develop quite rapidly. Even though some patients with a poor pulse have been shown by arteriogram to have a patent but constricted repair, all patients with a poor pulse or absence of pulse following operation have been recorded in this study as thrombosed.

Thrombosis was reported as occurring in 19 (18 percent) of the 105 surviving, unamputated patients on whom there were late follow-up studies (Table 6). Of the unamputated patients whose vessels were repaired by anastomosis, 13.9 percent thrombosed. Excluding the case amputated, none with lateral repair were reported as thrombosed. In the 21 patients whose vessels were repaired by grafts, 14 autogenous vein grafts and 7 homologus arterial grafts were used. In this group, 35.7 percent of the vein grafts thrombosed, as compared to 57.1 percent of the arterial grafts. From previous experience with homologous arterial grafts, this percentage is not surprising; it was not expected, however, that it would be as high with the autogenous vein grafts.


Admittedly the Korean conflict offered many advantages over World War II in regard to vascular repair. During the latter phase of the conflict, the front line in Korea was relatively stable. The surgical hospitals were within 6 to 12 miles of the front and there was ample air cover, plus the advantage of more rapid evacuation by helicopter. The expanded antibiotic armamentarium and availability of new vascular clamps, plus the experience gained in vascular surgery since World War II, all contributed to the success of vascular surgery in the Korean conflict. The Potts ductus and coarctation clamps contributed immensely to the success of the entire vascular surgery program.

Obviously, once an amputation was reported, the late follow-up concerning that patient's extremity was known. By far the highest percentage of amputations occurred at the initial installation; thus they were reported. While there were 26 amputations (20 percent) reported in 131 patients in the late follow-up group, this percentage is believed to be too high. If all of the 62 patients whose cases were not followed had viable limbs, then the amputation rate was 13.5 percent. This rate was certain to be low because amputations were recorded outside of the Korean theater. The actual amputation rate for the entire series probably ranged between 13.5 percent and 19.8 percent, and possibly near the mean of approximately 16.5 percent.


In order to present an accurate picture of the rate of thrombosis, findings were included on both the entire group of 193 patients with limited studies and on the late follow-up group of 131 patients. Ten thromboses are known to have existed in the 62 patients lost to follow-up. The number of thromboses in the amputated patients is unknown; hence the rate of thrombosis was figured for only the living unamputated patients. When rates of thrombosis of the short-term follow-up group and the late follow-up group were compared, they were found to be almost identical (Tables 5 and 6). If these findings are any criteria, for the most satisfactory methods of repair, then the preferred methods, in order, are: lateral repair, direct anastomosis. autogenous vein graft, and homologous arterial graft. However, it must be borne in mind that lateral repair was used for only the simplest. lacerations, while the more extensive wounds required a more complicated repair.

Lateral repair should be reserved for only minor, clean-cut lacerations of an artery. Large irregular lacerations are better excised with repair by direct anastomosis. By the same token, if the damaged area of the vessel is large, it is often wiser to débride thoroughly and insert a graft, rather than to sacrifice important collaterals in order to perform an anastomosis. An anastomosis under undue tension tends to separate or to create spasm with resulting thrombosis.

Even though a number of limbs had to be amputated and others were crippled by loss of muscle tissue or had complications of nerve injury or open fractures, there was a significant saving of limbs in Korea by repair of acute vascular injuries.


1 An attempt was made in Korea during 1953 to carry out follow-up studies on every soldier having a major artery repaired. These findings were compared with findings reported from Korea during 1952.

2. During 1953 there were 211 major arterial repairs in 205 patients reported in Korea. Death occurred in 15 patients with 18 repairs. Of the surviving patients, 26 required limb amputation.

3. Of the 193 repairs in surviving patients with short-term follow-up studies during 1953, amputation resulted in 13.5 percent of the cases. This is an improvement of 34.1 percent over an analogous series of 127 cases with 20.5 percent amputations reported for 1952.

4. Sixty-two of the 193 patients were lost to follow-up studies, leaving 131 with the late follow-up studies of which 26 required amputation (19.8 percent).

5. The correct amputation rate for the entire series ranged between 13.5 percent and 19.8 percent, or possibly near the mean of approximately 16.5 percent.


6.  A comparison of the rates of thrombosis in repairs with short-term follow- up studies and those with late follow-up studies showed the findings to be almost identical. If absence of thrombosis is an indication of superior methods of repair, then in order of preference the methods are: lateral repair, direct anastomosis, autogenous vein graft, and homologous arterial graft. The results of the study indicated a close correlation with the severity of the vascular wound.


1. De Bakey, NI. E., and Simeone, F. A.: Battle Injuries of the Arteries in World War II. Ann. Surg. 123: 534, 1946.
2. Seeley, S. F.; Hughes, C. NV.; Cooke, F. N., and Elkin, D. C.: Traumatic Arteriovenous Fistulas and Aneurysms in War Wounded. Am. J. Surg. 83: 471, 19-52.
3. Cooke, F. N.; Hughes, C. NV.; Jahnke, E. J., Jr.; and Seeley, S. F.: Homologous Arterial Grafts and Autogenous Vein Grafts Employed in Order to Bridge Large Arterial Defects in Man. Surgery, 33: 183, 1953.
4. Seeley, S. F.; Hughes, C. NV.; and Jahnke, E. J., Jr.: Direct Anastomosis Versus Ligation and Excision in Traumatic Arteriovenous Fistulas and Aneurysms. Surgical Forum, Clinical Congress of the American College of Surgeons, 1952, pp. 152-154. W. B.. Saunders Company, Philadelphia, 1953.
5. Hughes, C. NV.: Acute Vascular Trauma in Korean War Casualties. Surg., Gynec. & Obst. 99: 91, 1954.
6. Jahnke, E. J., Jr., and Seeley, S. F.: Acute Vascular Injuries in the Korean War. Ann. Surg. 138: 158, 1953.
7. Jahnke, E. J., Jr., and Howard, J. M.: Primary Repair of Major Arterial Injuries. Arch. Surg. 66: 646, 1953.
8. Ziperman, H. H.: Acute Arterial Injuries in the Korean War. Ann. Surg. 139: 1, 1954.