- Email: [email protected]
Civilian arterial trauma of the upper extremity
Civilian Arterial Trauma of the Upper Extremity An 11 Year Experience in 267 Patients
Karen R. Borman, MD, Dallas, Texas William H. Snyder, III, MD, Dallas, Texas John A. Welgelt, MD, Dallas, Texas
Nearly half of civilian arterial injuries occur in the upper extremity [l-4]. No large report of violent civilian injuries describes the presentation, diagnosis, management, and morbidity of this site of arterial trauma. This study was undertaken to characterize these injuries and their management. Material and Methods Acute injuries of the axillary, hrachial, radial, and ulnar arteries treated at Parkland Memorial Hospital were reviewed. Iatrogenic injuries were excluded. All patients with such injuries operated on between February 1972 and April 1983 are included. Complete records were available for 267 patients who ranged in age from 3 to 68 years (average 29 years). Two patients had bilateral injuries, so 269 extremities are included. Outpatient follow-up averaged 6 months and was available for 218 patients. (82 percent).
Results Penetrating agents accounted for 250 injuries (93 percent) and blunt trauma for 19 (7 percent). Sharp objects produced the most injuries (119), but gunshot wounds were the single most common cause of injury (85). Gunshot wounds predominated in axillary trauma and sharp instruments in more distal arteries. The incidences of right and left limb trauma were almost equal. Injuries involved a single artery in 243 extremities, 2 arteries in 23, and 3 arteries in 3 extremities (total 298 arterial injuries). There were 59 axillary (20 percent), 126 brachial (42 percent), 65 radial (22 percent), and 48 ulnar (16 percent) arterial injuries. Both radial anduhrar arteries were involved in 17 limbs. One hundred one patients (38 percent) arrived at the hospital in shock. Eighty-five of these patients (84 percent) responded to fluid resuscitation; 16 were persistently hypotensive and required urgent operations. Initial physical examinations demonstrated FromtheDepartment of Surgery, Southwestern Medical School, The Unlversity of Texas Health Science Center, Dallas, Texas. Requests for reprints should be addressed to Karen R. &man, MD, Departmant of Surgery, Southwestern Medical School, The University of Texas Health Science Center, Dallas, Texas 75208. Presented at the 36th Annual Meeting of the Southwestern Surgical Congress, Honolulu, Hawaii, April 21-28. 1984.
796
definitive evidence of arterial injuries in 224 limbs (84 percent). Abnormal distal pulses were the most common definite sign and were noted in 197 extremities (73 percent). Arterial bleeding occurred in 85 extremities (32 percent) and bruits were heard in 9 (3 percent). Forty-four extremities (16 percent) had equivocal evidence of vascular injury. Twenty-eight (10 percent) had only a peripheral nerve deficit, a stable, nonpulsatile hematoma, or both. Sixteen limbs (6 percent) had normal pulses and no findings except penetrating trauma proximate to major vessels. Site-specific presenting signs revealed some anatomic differences (Table I). Definite signs of injuries were present in almost 90 percent of brachial, radial, and ulnar arteries, but in only two thirds of axillary injuries. Pulses were abnormal in more than 75 percent of the extremities with distal injuries, but in only 54 percent of those with axillary trauma. Bruits were prominent signs only in axillary injuries, and in five of eight patients they were the only definitive evidence of injuries. Preoperative arteriography was performed in 106 extremities (39 percent), most often to prove injuries in patients with equivocal findings. Thirty-three (70 percent) of 47 patients with equivocal signs had preoperative angiography. Thirty patients with blunt or multiple missile trauma were studied to define precise injury sites. Isolated arterial injuries occurred in only 49 patients (18 percent). Associated injuries involved adjacent nerves, bones, veins, or soft tissue in 195 patients (73 percent). Nerve injuries were the most common and were identified perioperatively in 152 extremities (57 percent). Nonpaired axillary or brachial veins were injured in 58 limbs (22 percent). Skeletal trauma occurred in 42 extremities (16 percent), comprising 39 fractures and 3 dislocations. Seventy-two patients (27 percent) had remote injuries; chest or abdominal trauma predominated. Standard arterial reconstructions were performed, most often resection of the injury (220 reconstructions) with end-to-end anastomosis (155 reconstructions) or interposition graft (75 reconstructions).
TheAmarlcan
Journal of Surgery
Arterial Trauma of the Upper Extremity
TABLE I
Bite-Bpeclfk Presenting $igns
TABLE II
Axillary
Brachial
Radial or Ulnar
Total
Injured(n)
59
123
87
269
Definite signs (96) Pulse deficit Arterial bleeding Bruit
68 51 20 14
88 82 28 1
89 76 45 0
84 73 32 3
Equivocal or absent Mns (%) Hematoma Nerve deficit No sians
32
12
11
16
49 53 5
21 50 8
17 38 3
26 46 6
Primary anastomosis was achieved whenever possible with division of arterial branches to permit mobilization. Autogenous vein was the preferred conduit and saphenous was most commonly used (62 reconstructions); arm veins were successfully employed in 11 reconstructions. Prosthetic material was necessary in two patients with axillary injuries and inadequate saphenous veins. Lateral repair was possible in 24 tangential wounds, most often in proximal, larger vessels. Ligations were performed only for radial or ulnar arterial injuries with adequate collateral circulation. Systemic heparin anticoagulation was used after operative arterial control in 127 patients (48 percent). Temporary arterial shunts were employed in 14 patients. Twenty-one extremities (8 percent) required fasciotomies, most often in combined forearm arterial injuries (29 percent). Intraoperative arteriography was performed in 71 patients, usually (75 percent) to confirm satisfactory reconstructions and distal arterial patency. Intraoperative technical problems complicated 24 arterial repairs (9 percent). Twenty-two were resolved initially and 2 required reoperations. The cause of these difficulties was incomplete distal thrombectomies or inadequate reconstruction in 22 patients and the causes were unknown in 2 patients. Twenty-eight (48 percent) of 58 injured axillary and bra&al veins were repaired and 30 were ligated (52 percent). Sharp trauma permitted primary neurorrhaphies in 23 of 152 nerve injuries (15 percent). Twenty-six extensively damaged nerves (17 percent) were tagged for subsequent repair. Eighteen (46 percent) of 39 fractures were operatively reduced and stabilized, including 12 of 19 humeral and 6 of 12 radial-ulnar fractures. Ten humeral and 4 forearm fractures were unstable, and in 11, bone fixations preceded arterial repair. Arterial shunts were initially inserted in seven limbs with unstable humeral fractures. Two patients (0.7 percent) died from axillary artery injuries complicated by multiple organ failure.
voiumo148, Dacembw 1904
Neurokgk MorMdlty
Minor Partial sensory deficit Moderate Complete sensory deficit (1 nerve) Partial motor deficit Reflex sympathetic dystrophy Major Total motor deficit (1 nerve) Paralysis Resolved Total
n
%
49
37
45
34
23
17
3 13
2 10
133
10
Four amputations (1.5 percent) followed repairs of axillary (two), brachial (one), and combined radial and ulnar (one) arteries. Three of four injuries were from severe trauma, including a high-velocity bullet, a motor vehicle accident causing axihary and bra&al artery disruption, and a partial amputation by a corn shredder. Distal radial and ulnar arterial thromboses from handcuffs led to the fourth amputation. All four extremities were cadaveric and arterial reconstructions were attempted for limb salvage. Arterial flow was reestablished in three arms, but muscle necrosis and infection necessitated amputation. A forequarter amputation, two shoulder disarticulations, and a wrist disarticulation were performed. Minimal morbidity resulted from the arterial injuries in the 261 surviving patients with intact limbs. At most recent follow-up, pulses were palpable in 239 extremities (92 percent), absent in 14 (5 percent), and not recorded in 8 (2 percent). Major morbidity was from nerve injuries (Table II). One hundred sixty arms (59 percent) had nerve deficits, and follow-up was sufficient for evaluation in 133. The functional importance of these deficits is expressed by a crude grading system (Table II). Persistent deficits seriously limited function in 71 patients (27 percent).
Comments Arterial injuries of the upper extremities are almost as common as those of the lower extremities but have received’much less attention. Notable differences relate to the mode and signs of injury, the risk of limb loss, and the prognosis for return of normal function. Penetrating trauma predominates, but lacerations and industrial accidents are more frequent in the upper extremity. In the present study, broken glass caused 20 percent of the injuries with only 7 percent from blunt trauma. Signs of arterial injury differed with the site (Table I). Definite signs of injury comprising pulse deficits, arterial bleeding, or bruits were absent in a third of patients with axillary trauma compared with 10 percent of those with more distal injuries. Normal pulses were present in 49 percent of patients with axillary injuries but in only 12 percent of those with
797
Barman et al
brachial injuries. Pulses distal to forearm arterial injuries may be more reliable if palmar arch refill is prevented [5]. The frequent absence of definite signs of arterial trauma underscores the importance of angiography to exclude injuries [6]. Thirty of 47 patients with equivocal or absent signs were diagnosed arteriographically. Hypovolemic shock was initially present in 101 patients (38 percent), documenting the importance of controlling hemorrhage and repleting intravascular volume in these patients. Afterial accessability usually allows hemostasis by direct pressure; tourniquets or clamps may cause further damage and are not appropriate methods. Expedient diagnosis and operative repair are important, but abundant collateral vessels make successful revascularizations less time dependent. Operative fields should include the chest for proximal trauma, although only four patients with axillary injuries required thoracotomies. Abundant collateral flow makes proximal and distal control of axillary injuries incomplete, and a direct approach to the injury is usually more expedient. Arterial reconstruction is seldom difficult, applying the standard principles of arterial surgery. Arterial resection is usually advisable, and adequate vessel mobilization, despite the need to interrupt collateral branches, often allows primary anastomosis. However, vein interposition grafts are used liberally, and upper extremity veins deserve consideration to preserve the saphenous veins. The importance of complete resection of injured arterial wall, thorough proximal and distal catheter thrombectomies, and meticulous anastomotic technique are reemphasized by the intraoperative problems in 24 patients. Insufficient arterial debridements, anastomotic stenoses, and small diameter autografts caused anastomotic failures. Ligation is only considered for isolated radial or ulnar arterial injuries with adequate collateral flow. Ischemic risk is increased by incomplete palmar arches, anatomic anomalies being found in 9 percent of extremities [7]. The safety of forearm artery ligation should be confirmed by an Allen test or arteriography. Management of isolated radial or ulnar arterial injuries defined preoperatively is a dilemma. Unrepaired forearm arterial injuries cause consistent alterations in hand vascularity, but signs of ischemia or cold intolerance are infrequent [8]. If palmar arch patency is unknown, restitution of flow is advisable [9]. The indications for operation in single and asymptomatic forearm arterial injuries are unclear and relate to potential late complications. Nine percent of a large series of traumatic aneurysms and arteriovenous fistulas involved forearm arteries
[101.
The frequency of injured adjacent structures requires some consideration of operative management. Axillary or brachial venous injuries were repaired in
798
27 patients (46 percent) and ligated in 31 (54 percent). One patient had postoperative edema after each procedure. Several reports document infrequent complications after venous ligation in the upper extremity, but if simple repair is feasible, it is probably appropriate [11,12]. Postoperative extremity elevation is important for any venous injury, regardless of its operative treatment. Unstable fractures jeopardize arterial repairs more in the upper extremities than in the lower extremities because of less adjacent muscle mass. Our experience supports priority skeletal fixation, particularly for unstable humeral fractures, after distal catheter thrombectomy and insertion of an arterial shunt. Death from these upper extremity injuries is unusual (0.7 percent in our study) and usually relates to initial blood loss. Amputations less often result compared with arterial injuries of the lower extremities, follow massive trauma, and are infrequently related to failure of arterial reconstruction. The major disability is from nerve injuries, as depicted by our crude grading system (Table II). Serious functional limitations persisted in 71 patients (27 percent). This has been previously documented in smaller series and treatment of these injuries is the crucial failure in upper extremity trauma [13,14]. Our data allow no insight into means of decreasing the failure rate except to emphasize its existence. Summary Two hundred ninety-eight arterial injuries in 269 upper extremities were reviewed. Penetrating agents accounted for 250 injuries (93 percent) and blunt trauma for 19 (7 percent). Fifty-nine axillary, 126 brachial, 65 radial, and 48 ulnar arteries were damaged. Twenty-six extremities had more than one artery injured. The initial vascular examination revealed no abnormalities or was equivocal in 16 percent of all patients and in 32 percent of those with axillary artery injuries. Adjacent upper extremity structures were injured in 195 limbs (73 percent). Resection and primary anastomosis (54 percent) or vein interposition grafting (26 percent) were the most frequent methods of repair. Two deaths (0.7 percent) occurred and four amputations (1.5 percent) were required. Distal pulses were present at discharge in 93 percent of the evaluable extremities. Despite excellent success with arterial reconstruction, functional results were limited by associated nerve injuries. One hundred fifty patients (49 percent) had nerve deficits at discharge, and 71 (27 percent) had serious functional limitations. References 1. Burnett HF, Parnell CL, Williams
GD, Campbell GS. Peripheral arterial injuries: a reassessment. Ann Surg 1976;163: 701-g. 2. Perry MO, Thal ER, Shires GT. Management of arterial injuries.
The American Journal ol Surgery
Arterial Trauma of the Upper Extremity
Ann Surg 1971;173:403-8. 3. Smith RF, Elliott JP, Hegsmsn JH. Szifagyi DE, Xavier AO. Acute penetrating arterial injuries of the neck and limbs. Arch Surg 1974;109:198-205. 4. Sturm JT, Bodily KC, Rothenberger DA, Perry JF Jr. Arterial injuries of the extremities following blunt trauma. J Trauma 1980;20:933-6. 5. Gelberman RH, Menon J, Fronek A. The peripheral pulse following arterial injury. J Trauma 1980;20:948-50. 6. Snyder WH, Thal ER, Bridges RA, Gerlock AJ, Perry MO, Fry WJ. The validity of normal arteriography in penetrating trauma. Arch Surg 1978;113:424-8. 7. Gelberman RH, Biasingame JP. The timed Alien test. J Trauma 1981;21:477-9. 8. Gelberman RH, Blasingame JP, Fronek A, Dimick MP. Forearm arterial injuries. J Hand Surg 1979;4:401-8. 9. GekIetman RH, Nunley JA, Komsn LA, et al. The results of radial and ulnar arterial repair in the forearm. J Bone Joint Surg 1982;64A:383-7. 10. Rich NM, Hobson RW, Collins &I Jr. Traumatic arteriovenous fistulas and false aneurysms: a review of 558 lesions. Surgery 1975;78:817-28. 11. Agsrwal N, Shah PM, Clauss RH, Reynolds BM, Stahl WM. Experience with 115 civilian venous injuries. J Trauma 1982;22:827-32. 12. Hardin WD Jr, Adinolfi MF, O’Connell RC, Kerstein MD. Management of traumatic perfphsral vein injuries: primary repair or vein ligation. Am J Surg 1982;144:235-8. 13. Ashbell TS, Kleinert HE, Kutz JE. Vascular injuries about the elbow. Clin Drthop 1967;50:107-27. 14. Visser PA, Hermreck AS, Pierce GE, Thomas JH, Hardin CA. Prognosis of nerve injuries incurred during acute trauma to peripheral arteries. Am J Surg 1980;140:596-9.
Discussion Robert W. Barnes (Little Rock, AR): Do you have experience with free flaps in patients with major soft tissue destruction? Kenneth R. Sirinek (San Antonio, TX): The precision with which a correct judgment can be made concerning the existence of a peripheral vascular injury is often the result of clinical experience and not of diagnostic modalities. The present report on upper extremity vascular injuries is a good example of that principle and representa one of the many fine contributions by Dr. Snyder and his colleagues at Parkland Memorial Hospital. With only two deaths and four amputations occurring in 267 patients, it is readily apparent that the major sequela of this injury is a functionally impaired upper extremity. With 73 percent of patients having an associated extremity injury, primarily to an adjacent nerve, a successful vascular repair alone saves the anatomic structure, although function is lost. Routine exploration of all penetrating injuries in anatomic proximity to a major vascular structure has been replaced by the selective use of preoperative arteriography in patients with equivocal signs of injury, multiple missile injuries, blunt injury, or the presence of a bruit. The presence of distal pulses in 40 percent of patients with axillary artery injury and in 25 percent of those with distal injury attests to the need for angiography in this group of patients. However, the authors have questioned the propriety of routine diagnostic arteriography for all penetrating wounds without signs of arterial injury. It would be unwise to switch
the burden from the operating room to the angiography suite. Dr. Snyder, would you ever consider emergency room arteriography as recommended by Dr. Feliciano and his colleagues? Is concomitant venous repair really necessary or successful beyond the axillary vein? What are your indications for fasciotomy? William I-I. Snyder, III (closing): One of the major things that continued to show up in reviewing charts was the frequent absence of signs of arterial injury. Diminished pulses and arterial bleeding were disturbingly infrequent, particularly in axillary artery injuries. Dr. Sirinek’s question about emergency room arteriography has been discussed before, as he said, at the meeting of the Western Surgical Association last fall. Dr’Feliciano and I had some disagreement about it, and this will probably surface again. In 1976, we reviewed our patients in whom arteriography had been performed to exclude injuries. I reviewed all the films and was very disturbed with the number that had very minor radiographic findings. These were arteriograms obtained in the radiology suite with several views and often several injections. The frequent minor findings with complete arteriography causes me concern about singleshot arteriograms in the emergency room to exclude injuries. Using arteriography in the emergency room to locate an injury site when the extremity is going to be explored seems reasonable. I continue to think its use is unreasonable to exclude injuries, since minor arteriographic findings often represent major injuries. Concomitant venous repair, as Dr. Borman noted, was conducted in only about half the venous injuries, and we do not have data describing how many remained patent. It did not appear that upper extremity venous ligation, even in axillary injuries, caused difficulties with the patency of the arterial repair or the survival of the extremity. There are a number of series that document relatively minor morbidity related to upper extremity venous ligation. Postoperative elevation of the extremity is important no matter what is done to venous injuries in the upper extremity. We repair venous injuries when it can be performed simply. If venous hypertension is obvious when the vein is clamped, we repair it even if it requires a more complicated reconstruction, assuming the patient is stable. However, with numerous upper extremity venous collateral vessels, this happens infrequently. Indications for fasciotomy are also infrequent in the upper extremity. Only 21 of these limbs required fasciotomies. They were usually performed after arterial or venous reconstruction. All but two were forearm fasciotomies and most often were required with combined radial and ulnar arterial injuries. Fasciotomy seems to play a minor role, and the indications are more obvious than is the case with leg injuries. Regarding Dr. Barnes question about soft tissue reconstruction, this was a problem in only 12 patients, and in several of them, free flaps were used. It did not seem to present the problem that it does in the leg especially with popliteal arterial injuries. I am not sure I understand why, but possibly it is because of the good collateral flow and the less frequent problems with muscle necrosis in the arm.
799
Karen R. Borman, MD, Dallas, Texas William H. Snyder, III, MD, Dallas, Texas John A. Welgelt, MD, Dallas, Texas
Nearly half of civilian arterial injuries occur in the upper extremity [l-4]. No large report of violent civilian injuries describes the presentation, diagnosis, management, and morbidity of this site of arterial trauma. This study was undertaken to characterize these injuries and their management. Material and Methods Acute injuries of the axillary, hrachial, radial, and ulnar arteries treated at Parkland Memorial Hospital were reviewed. Iatrogenic injuries were excluded. All patients with such injuries operated on between February 1972 and April 1983 are included. Complete records were available for 267 patients who ranged in age from 3 to 68 years (average 29 years). Two patients had bilateral injuries, so 269 extremities are included. Outpatient follow-up averaged 6 months and was available for 218 patients. (82 percent).
Results Penetrating agents accounted for 250 injuries (93 percent) and blunt trauma for 19 (7 percent). Sharp objects produced the most injuries (119), but gunshot wounds were the single most common cause of injury (85). Gunshot wounds predominated in axillary trauma and sharp instruments in more distal arteries. The incidences of right and left limb trauma were almost equal. Injuries involved a single artery in 243 extremities, 2 arteries in 23, and 3 arteries in 3 extremities (total 298 arterial injuries). There were 59 axillary (20 percent), 126 brachial (42 percent), 65 radial (22 percent), and 48 ulnar (16 percent) arterial injuries. Both radial anduhrar arteries were involved in 17 limbs. One hundred one patients (38 percent) arrived at the hospital in shock. Eighty-five of these patients (84 percent) responded to fluid resuscitation; 16 were persistently hypotensive and required urgent operations. Initial physical examinations demonstrated FromtheDepartment of Surgery, Southwestern Medical School, The Unlversity of Texas Health Science Center, Dallas, Texas. Requests for reprints should be addressed to Karen R. &man, MD, Departmant of Surgery, Southwestern Medical School, The University of Texas Health Science Center, Dallas, Texas 75208. Presented at the 36th Annual Meeting of the Southwestern Surgical Congress, Honolulu, Hawaii, April 21-28. 1984.
796
definitive evidence of arterial injuries in 224 limbs (84 percent). Abnormal distal pulses were the most common definite sign and were noted in 197 extremities (73 percent). Arterial bleeding occurred in 85 extremities (32 percent) and bruits were heard in 9 (3 percent). Forty-four extremities (16 percent) had equivocal evidence of vascular injury. Twenty-eight (10 percent) had only a peripheral nerve deficit, a stable, nonpulsatile hematoma, or both. Sixteen limbs (6 percent) had normal pulses and no findings except penetrating trauma proximate to major vessels. Site-specific presenting signs revealed some anatomic differences (Table I). Definite signs of injuries were present in almost 90 percent of brachial, radial, and ulnar arteries, but in only two thirds of axillary injuries. Pulses were abnormal in more than 75 percent of the extremities with distal injuries, but in only 54 percent of those with axillary trauma. Bruits were prominent signs only in axillary injuries, and in five of eight patients they were the only definitive evidence of injuries. Preoperative arteriography was performed in 106 extremities (39 percent), most often to prove injuries in patients with equivocal findings. Thirty-three (70 percent) of 47 patients with equivocal signs had preoperative angiography. Thirty patients with blunt or multiple missile trauma were studied to define precise injury sites. Isolated arterial injuries occurred in only 49 patients (18 percent). Associated injuries involved adjacent nerves, bones, veins, or soft tissue in 195 patients (73 percent). Nerve injuries were the most common and were identified perioperatively in 152 extremities (57 percent). Nonpaired axillary or brachial veins were injured in 58 limbs (22 percent). Skeletal trauma occurred in 42 extremities (16 percent), comprising 39 fractures and 3 dislocations. Seventy-two patients (27 percent) had remote injuries; chest or abdominal trauma predominated. Standard arterial reconstructions were performed, most often resection of the injury (220 reconstructions) with end-to-end anastomosis (155 reconstructions) or interposition graft (75 reconstructions).
TheAmarlcan
Journal of Surgery
Arterial Trauma of the Upper Extremity
TABLE I
Bite-Bpeclfk Presenting $igns
TABLE II
Axillary
Brachial
Radial or Ulnar
Total
Injured(n)
59
123
87
269
Definite signs (96) Pulse deficit Arterial bleeding Bruit
68 51 20 14
88 82 28 1
89 76 45 0
84 73 32 3
Equivocal or absent Mns (%) Hematoma Nerve deficit No sians
32
12
11
16
49 53 5
21 50 8
17 38 3
26 46 6
Primary anastomosis was achieved whenever possible with division of arterial branches to permit mobilization. Autogenous vein was the preferred conduit and saphenous was most commonly used (62 reconstructions); arm veins were successfully employed in 11 reconstructions. Prosthetic material was necessary in two patients with axillary injuries and inadequate saphenous veins. Lateral repair was possible in 24 tangential wounds, most often in proximal, larger vessels. Ligations were performed only for radial or ulnar arterial injuries with adequate collateral circulation. Systemic heparin anticoagulation was used after operative arterial control in 127 patients (48 percent). Temporary arterial shunts were employed in 14 patients. Twenty-one extremities (8 percent) required fasciotomies, most often in combined forearm arterial injuries (29 percent). Intraoperative arteriography was performed in 71 patients, usually (75 percent) to confirm satisfactory reconstructions and distal arterial patency. Intraoperative technical problems complicated 24 arterial repairs (9 percent). Twenty-two were resolved initially and 2 required reoperations. The cause of these difficulties was incomplete distal thrombectomies or inadequate reconstruction in 22 patients and the causes were unknown in 2 patients. Twenty-eight (48 percent) of 58 injured axillary and bra&al veins were repaired and 30 were ligated (52 percent). Sharp trauma permitted primary neurorrhaphies in 23 of 152 nerve injuries (15 percent). Twenty-six extensively damaged nerves (17 percent) were tagged for subsequent repair. Eighteen (46 percent) of 39 fractures were operatively reduced and stabilized, including 12 of 19 humeral and 6 of 12 radial-ulnar fractures. Ten humeral and 4 forearm fractures were unstable, and in 11, bone fixations preceded arterial repair. Arterial shunts were initially inserted in seven limbs with unstable humeral fractures. Two patients (0.7 percent) died from axillary artery injuries complicated by multiple organ failure.
voiumo148, Dacembw 1904
Neurokgk MorMdlty
Minor Partial sensory deficit Moderate Complete sensory deficit (1 nerve) Partial motor deficit Reflex sympathetic dystrophy Major Total motor deficit (1 nerve) Paralysis Resolved Total
n
%
49
37
45
34
23
17
3 13
2 10
133
10
Four amputations (1.5 percent) followed repairs of axillary (two), brachial (one), and combined radial and ulnar (one) arteries. Three of four injuries were from severe trauma, including a high-velocity bullet, a motor vehicle accident causing axihary and bra&al artery disruption, and a partial amputation by a corn shredder. Distal radial and ulnar arterial thromboses from handcuffs led to the fourth amputation. All four extremities were cadaveric and arterial reconstructions were attempted for limb salvage. Arterial flow was reestablished in three arms, but muscle necrosis and infection necessitated amputation. A forequarter amputation, two shoulder disarticulations, and a wrist disarticulation were performed. Minimal morbidity resulted from the arterial injuries in the 261 surviving patients with intact limbs. At most recent follow-up, pulses were palpable in 239 extremities (92 percent), absent in 14 (5 percent), and not recorded in 8 (2 percent). Major morbidity was from nerve injuries (Table II). One hundred sixty arms (59 percent) had nerve deficits, and follow-up was sufficient for evaluation in 133. The functional importance of these deficits is expressed by a crude grading system (Table II). Persistent deficits seriously limited function in 71 patients (27 percent).
Comments Arterial injuries of the upper extremities are almost as common as those of the lower extremities but have received’much less attention. Notable differences relate to the mode and signs of injury, the risk of limb loss, and the prognosis for return of normal function. Penetrating trauma predominates, but lacerations and industrial accidents are more frequent in the upper extremity. In the present study, broken glass caused 20 percent of the injuries with only 7 percent from blunt trauma. Signs of arterial injury differed with the site (Table I). Definite signs of injury comprising pulse deficits, arterial bleeding, or bruits were absent in a third of patients with axillary trauma compared with 10 percent of those with more distal injuries. Normal pulses were present in 49 percent of patients with axillary injuries but in only 12 percent of those with
797
Barman et al
brachial injuries. Pulses distal to forearm arterial injuries may be more reliable if palmar arch refill is prevented [5]. The frequent absence of definite signs of arterial trauma underscores the importance of angiography to exclude injuries [6]. Thirty of 47 patients with equivocal or absent signs were diagnosed arteriographically. Hypovolemic shock was initially present in 101 patients (38 percent), documenting the importance of controlling hemorrhage and repleting intravascular volume in these patients. Afterial accessability usually allows hemostasis by direct pressure; tourniquets or clamps may cause further damage and are not appropriate methods. Expedient diagnosis and operative repair are important, but abundant collateral vessels make successful revascularizations less time dependent. Operative fields should include the chest for proximal trauma, although only four patients with axillary injuries required thoracotomies. Abundant collateral flow makes proximal and distal control of axillary injuries incomplete, and a direct approach to the injury is usually more expedient. Arterial reconstruction is seldom difficult, applying the standard principles of arterial surgery. Arterial resection is usually advisable, and adequate vessel mobilization, despite the need to interrupt collateral branches, often allows primary anastomosis. However, vein interposition grafts are used liberally, and upper extremity veins deserve consideration to preserve the saphenous veins. The importance of complete resection of injured arterial wall, thorough proximal and distal catheter thrombectomies, and meticulous anastomotic technique are reemphasized by the intraoperative problems in 24 patients. Insufficient arterial debridements, anastomotic stenoses, and small diameter autografts caused anastomotic failures. Ligation is only considered for isolated radial or ulnar arterial injuries with adequate collateral flow. Ischemic risk is increased by incomplete palmar arches, anatomic anomalies being found in 9 percent of extremities [7]. The safety of forearm artery ligation should be confirmed by an Allen test or arteriography. Management of isolated radial or ulnar arterial injuries defined preoperatively is a dilemma. Unrepaired forearm arterial injuries cause consistent alterations in hand vascularity, but signs of ischemia or cold intolerance are infrequent [8]. If palmar arch patency is unknown, restitution of flow is advisable [9]. The indications for operation in single and asymptomatic forearm arterial injuries are unclear and relate to potential late complications. Nine percent of a large series of traumatic aneurysms and arteriovenous fistulas involved forearm arteries
[101.
The frequency of injured adjacent structures requires some consideration of operative management. Axillary or brachial venous injuries were repaired in
798
27 patients (46 percent) and ligated in 31 (54 percent). One patient had postoperative edema after each procedure. Several reports document infrequent complications after venous ligation in the upper extremity, but if simple repair is feasible, it is probably appropriate [11,12]. Postoperative extremity elevation is important for any venous injury, regardless of its operative treatment. Unstable fractures jeopardize arterial repairs more in the upper extremities than in the lower extremities because of less adjacent muscle mass. Our experience supports priority skeletal fixation, particularly for unstable humeral fractures, after distal catheter thrombectomy and insertion of an arterial shunt. Death from these upper extremity injuries is unusual (0.7 percent in our study) and usually relates to initial blood loss. Amputations less often result compared with arterial injuries of the lower extremities, follow massive trauma, and are infrequently related to failure of arterial reconstruction. The major disability is from nerve injuries, as depicted by our crude grading system (Table II). Serious functional limitations persisted in 71 patients (27 percent). This has been previously documented in smaller series and treatment of these injuries is the crucial failure in upper extremity trauma [13,14]. Our data allow no insight into means of decreasing the failure rate except to emphasize its existence. Summary Two hundred ninety-eight arterial injuries in 269 upper extremities were reviewed. Penetrating agents accounted for 250 injuries (93 percent) and blunt trauma for 19 (7 percent). Fifty-nine axillary, 126 brachial, 65 radial, and 48 ulnar arteries were damaged. Twenty-six extremities had more than one artery injured. The initial vascular examination revealed no abnormalities or was equivocal in 16 percent of all patients and in 32 percent of those with axillary artery injuries. Adjacent upper extremity structures were injured in 195 limbs (73 percent). Resection and primary anastomosis (54 percent) or vein interposition grafting (26 percent) were the most frequent methods of repair. Two deaths (0.7 percent) occurred and four amputations (1.5 percent) were required. Distal pulses were present at discharge in 93 percent of the evaluable extremities. Despite excellent success with arterial reconstruction, functional results were limited by associated nerve injuries. One hundred fifty patients (49 percent) had nerve deficits at discharge, and 71 (27 percent) had serious functional limitations. References 1. Burnett HF, Parnell CL, Williams
GD, Campbell GS. Peripheral arterial injuries: a reassessment. Ann Surg 1976;163: 701-g. 2. Perry MO, Thal ER, Shires GT. Management of arterial injuries.
The American Journal ol Surgery
Arterial Trauma of the Upper Extremity
Ann Surg 1971;173:403-8. 3. Smith RF, Elliott JP, Hegsmsn JH. Szifagyi DE, Xavier AO. Acute penetrating arterial injuries of the neck and limbs. Arch Surg 1974;109:198-205. 4. Sturm JT, Bodily KC, Rothenberger DA, Perry JF Jr. Arterial injuries of the extremities following blunt trauma. J Trauma 1980;20:933-6. 5. Gelberman RH, Menon J, Fronek A. The peripheral pulse following arterial injury. J Trauma 1980;20:948-50. 6. Snyder WH, Thal ER, Bridges RA, Gerlock AJ, Perry MO, Fry WJ. The validity of normal arteriography in penetrating trauma. Arch Surg 1978;113:424-8. 7. Gelberman RH, Biasingame JP. The timed Alien test. J Trauma 1981;21:477-9. 8. Gelberman RH, Blasingame JP, Fronek A, Dimick MP. Forearm arterial injuries. J Hand Surg 1979;4:401-8. 9. GekIetman RH, Nunley JA, Komsn LA, et al. The results of radial and ulnar arterial repair in the forearm. J Bone Joint Surg 1982;64A:383-7. 10. Rich NM, Hobson RW, Collins &I Jr. Traumatic arteriovenous fistulas and false aneurysms: a review of 558 lesions. Surgery 1975;78:817-28. 11. Agsrwal N, Shah PM, Clauss RH, Reynolds BM, Stahl WM. Experience with 115 civilian venous injuries. J Trauma 1982;22:827-32. 12. Hardin WD Jr, Adinolfi MF, O’Connell RC, Kerstein MD. Management of traumatic perfphsral vein injuries: primary repair or vein ligation. Am J Surg 1982;144:235-8. 13. Ashbell TS, Kleinert HE, Kutz JE. Vascular injuries about the elbow. Clin Drthop 1967;50:107-27. 14. Visser PA, Hermreck AS, Pierce GE, Thomas JH, Hardin CA. Prognosis of nerve injuries incurred during acute trauma to peripheral arteries. Am J Surg 1980;140:596-9.
Discussion Robert W. Barnes (Little Rock, AR): Do you have experience with free flaps in patients with major soft tissue destruction? Kenneth R. Sirinek (San Antonio, TX): The precision with which a correct judgment can be made concerning the existence of a peripheral vascular injury is often the result of clinical experience and not of diagnostic modalities. The present report on upper extremity vascular injuries is a good example of that principle and representa one of the many fine contributions by Dr. Snyder and his colleagues at Parkland Memorial Hospital. With only two deaths and four amputations occurring in 267 patients, it is readily apparent that the major sequela of this injury is a functionally impaired upper extremity. With 73 percent of patients having an associated extremity injury, primarily to an adjacent nerve, a successful vascular repair alone saves the anatomic structure, although function is lost. Routine exploration of all penetrating injuries in anatomic proximity to a major vascular structure has been replaced by the selective use of preoperative arteriography in patients with equivocal signs of injury, multiple missile injuries, blunt injury, or the presence of a bruit. The presence of distal pulses in 40 percent of patients with axillary artery injury and in 25 percent of those with distal injury attests to the need for angiography in this group of patients. However, the authors have questioned the propriety of routine diagnostic arteriography for all penetrating wounds without signs of arterial injury. It would be unwise to switch
the burden from the operating room to the angiography suite. Dr. Snyder, would you ever consider emergency room arteriography as recommended by Dr. Feliciano and his colleagues? Is concomitant venous repair really necessary or successful beyond the axillary vein? What are your indications for fasciotomy? William I-I. Snyder, III (closing): One of the major things that continued to show up in reviewing charts was the frequent absence of signs of arterial injury. Diminished pulses and arterial bleeding were disturbingly infrequent, particularly in axillary artery injuries. Dr. Sirinek’s question about emergency room arteriography has been discussed before, as he said, at the meeting of the Western Surgical Association last fall. Dr’Feliciano and I had some disagreement about it, and this will probably surface again. In 1976, we reviewed our patients in whom arteriography had been performed to exclude injuries. I reviewed all the films and was very disturbed with the number that had very minor radiographic findings. These were arteriograms obtained in the radiology suite with several views and often several injections. The frequent minor findings with complete arteriography causes me concern about singleshot arteriograms in the emergency room to exclude injuries. Using arteriography in the emergency room to locate an injury site when the extremity is going to be explored seems reasonable. I continue to think its use is unreasonable to exclude injuries, since minor arteriographic findings often represent major injuries. Concomitant venous repair, as Dr. Borman noted, was conducted in only about half the venous injuries, and we do not have data describing how many remained patent. It did not appear that upper extremity venous ligation, even in axillary injuries, caused difficulties with the patency of the arterial repair or the survival of the extremity. There are a number of series that document relatively minor morbidity related to upper extremity venous ligation. Postoperative elevation of the extremity is important no matter what is done to venous injuries in the upper extremity. We repair venous injuries when it can be performed simply. If venous hypertension is obvious when the vein is clamped, we repair it even if it requires a more complicated reconstruction, assuming the patient is stable. However, with numerous upper extremity venous collateral vessels, this happens infrequently. Indications for fasciotomy are also infrequent in the upper extremity. Only 21 of these limbs required fasciotomies. They were usually performed after arterial or venous reconstruction. All but two were forearm fasciotomies and most often were required with combined radial and ulnar arterial injuries. Fasciotomy seems to play a minor role, and the indications are more obvious than is the case with leg injuries. Regarding Dr. Barnes question about soft tissue reconstruction, this was a problem in only 12 patients, and in several of them, free flaps were used. It did not seem to present the problem that it does in the leg especially with popliteal arterial injuries. I am not sure I understand why, but possibly it is because of the good collateral flow and the less frequent problems with muscle necrosis in the arm.
799