Lessons learned in the management of thirteen celiac axis injuries.

Objectives: Celiac axis injuries are rare. The purposes of this study were to (1) review institutional experience, (2) determine additive effect on death of associated vessel injuries, and (3) correlate mortality rates with the American Association for the Surgery of Trauma--Organ Injury Scale (AAST-OIS) for abdominal vascular injury.

Methods: This was a retrospective, 132-month study (January 1992 to December 2002) of patients with celiac axis injuries.

Results: Thirteen patients were included in the study. Mean revised trauma score was 5.35 [+ or -] 2.63; mean injury severity score was 25 [+ or -] 12. The mechanism was penetrating in 12 (92%) and blunt in 1 (8%); 3 of 13 had Emergency Department thoracotomy (100% mortality rate). Treatment included ligation in 11 and primary repair in 1; 1 exsanguinated. Overall survival was 5 of 13 (38%). Adjusted survival excluding patients who had Emergency Department thoracotomy was 5 of 10 patients (50%). Those surviving with isolated injuries included 57% of patients. Mortality rate versus AAST-OIS was grade III, 43% (3 of 7 patients); grade IV, 50% (1 of 2 patients); and grade V, 100% (4 of 4 patients).

Conclusions: Celiac axis injuries are rare. Patients with isolated injuries have better survival rates. Mortality rate correlates well with AAST-OIS for abdominal vascular injury.

Key Words: abdominal vascular injury, celiac axis injury, exsanguinating hemorrhage

**********

Injuries to the celiac axis are uncommon but devastating, incurring high mortality rates. (1-11) The rarity of these injuries prevents many trauma surgeons from developing significant experience with their management. (1-11) There is a relative paucity of articles in the literature dealing with their management. Exsanguinating hemorrhage is the most common cause of early death because these injuries are associated with rapid rates of blood loss, are difficult to expose, and pose challenges in establishing early proximal and distal control. (1-13)

Celiac axis injuries are among the least frequently reported of all arterial injuries. Graham et al (6) in 1978 reported the largest series, consisting of 13 injuries as part of a study describing 66 visceral arterial injuries. To our knowledge, there have been no series specifically devoted to the management of celiac axis injuries. It is the purpose of this study to review our institutional experience, to determine the additive effect on death of multiple associated vascular injuries, and to correlate mortality rates with the American Association for the Surgery of Trauma-Organ Injury Scale (AAST-OIS) (14) for abdominal vascular injury, for this very rare and often lethal injury.

Materials and Methods

Over a 132-month period (January 1992 to December 2002), all patients admitted to the Los Angeles County--University of Southern California Medical Center with a proven celiac axis injury were retrospectively reviewed, and data were entered into a collection sheet. Institutional review board approval was obtained. Data collected included demographics such as sex, age, mechanism of injury, admission vital signs, revised trauma score (RTS), and injury severity score (ISS). All celiac axis injuries were graded by means of the AAST-OIS for abdominal vascular injuries. Other data collected included the amount and type of resuscitation fluids and blood products; surgical findings, including vessels injured and location of retroperitoneal hematomas; and surgical procedures used to deal with these injuries, including need for resuscitative thoracotomy. In addition, the number of associated vascular and nonvascular injuries, mean Surgical Intensive Care Unit length of stay (LOS) and mean hospital LOS, along with complications, mean overall mortality rates, vessel-specific deaths, and additive deaths of combined vessel injuries, were recorded.

Results

During the course of the study, 13 patients were admitted with celiac axis injury. The mean age was 24 [+ or -] 8.07 years. All were male. Twelve were admitted secondary to penetrating injury (92%) and one as the result of blunt trauma (8%). Of the 12 patients admitted with penetrating injuries, 9 (75%) sustained gunshot wounds and 3 (25%) had stab wounds, whereas the only patient with blunt trauma was injured in a motor vehicle collision (see Table 1). All patients had clinical signs of acute peritonitis and/or hemoperitoneum on admission, mandating immediate exploration.

The mean admission systolic blood pressure was 91 [+ or -] 41 mm Hg. Mean admission heart rate was 87 [+ or -] 21 beats/min. Mean volume of resuscitative fluids in the Emergency Department (ED) was slightly more than 3 L (mean volume crystalloids, 2,692 mL; packed red blood cells, 466 mL). Mean RTS was 5.35 [+ or -] 2.63 and mean ISS was 25 [+ or -] 12, denoting a severely injured population. All patients survived ED resuscitation and were rapidly transported to the operating room. Three patients underwent ED thoracotomy, aortic cross-clamping, and open CPR; they reached the operating room (OR) but did not survive. In the OR, all patients underwent exploratory laparotomy. Furthermore, an additional five patients were subjected to resuscitative OR thoracotomy; only one survived.

Operative findings revealed that seven patients had zone I retroperitoneal hematomas (three supramesocolic, two inframesocolic), and the remaining two patients had a combination of both. Of the 13 celiac axis injuries, seven were isolated and six had associated vascular injuries. Five of the patients with combined injuries incurred a celiac axis plus a single associated vessel injury, whereas one patient had three vessels injured (see Table 2). Surgical treatment included ligation in 11 patients and primary repair in one patient; one patient exsanguinated before definitive repair. Of the five surviving patients, four were treated with ligation and one with primary repair (see Table 3). Primary abdominal wall closure was performed in four of the surviving patients; one required damage control and prosthetic abdominal wall closure with an intravenous bag, and eight were not closed because they died in the operating room.

All patients had their abdominal vascular injuries graded by means of the AAST-OIS for abdominal vascular injury. There were seven (54%) patients with grade III, two (15%) with grade IV, and four (31%) with grade V injuries. Mortality rates stratified according to injury grade were grade III, three of seven patients died; grade IV, one of two patients died (50%); and grade V, all four patients died (100%) (see Table 4). Mortality rates correlated well with grade of injury. In this group of patients, there were a total of 33 associated injuries, averaging 2.54 associated injuries per patient. Of these 33 associated injuries, 25 were nonvascular, averaging 1.9 nonvascular associated injuries per patient. There were eight associated vascular injuries, averaging 0.62 associated vascular injuries per patient. The mean estimated blood loss was 5,434 [+ or -] 3,082 mL. The mean total volume of OR fluids administered was 10,299 mL. The breakdown according to the different types of fluids administered included crystalloids, 5,461 [+ or -] 3,338 mL; blood, 3,935 [+ or -] 3,257 mL; fresh frozen plasma, 320 [+ or -] 445 mL; and colloids and other blood products, 583 [+ or -] 874 mL.

Five of the 13 patients in this series survived. Adjusted survival rate excluding the three patients requiring ED thoracotomy, aortic cross-clamping, and open CPR is 50%. The cause of death in the eight nonsurvivors was exsanguination in the OR. Of the seven patients with isolated injuries, four survived (survival rate, 57%). Of the six patients with associated vessel injuries, only one, with an associated renal artery injury, survived (survival rate, 17%). Among the survivors, there were a total of 26 major intraoperative complications. These complications included hypothermia and arrhythmias in seven patients each and coagulopathy and acidosis in six patients each. The mean Intensive Care Unit LOS was 6.33 days (range, 2 to 11 days). The mean hospital length of stay was 15 days (range, 4 to 38 days).

Discussion

Visceral arterial injuries are among the most lethal injuries encountered by trauma surgeons. (1,2,11-13,15,16) The majority of these patients arrive with severe physiologic compromise secondary to massive blood loss and a large number of associated injuries. (1,2,11-13,15-16) Makins, (17) DeBakey and Simeone, (18) Hughes, (19) and Rich et al (20) recorded no wounds of the visceral arteries in their reports of America's wartime experience. In the civilian arena, both the Feliciano et al (21) series reporting a 12-month experience with 456 vascular and cardiac injuries and the Mattox et al (22) large civilian epidemiologic study of 5,760 cardiovascular injuries spanning a 30-year period failed to describe celiac axis injuries. Graham et al (6) reported the largest series in the literature, a 15-year experience with 66 visceral arterial injuries, of which 13 were celiac axis injuries. Recent series dealing with abdominal vascular injuries consistently report fewer than 10 celiac axis injuries. (6-9,11,23) To the best of our knowledge, Patman et al (3) in 1964 reported the first case of a celiac axis injury as part of a series of 256 patients with abdominal vascular injuries. To date, an extensive review of the literature yields only 33 celiac axis injuries reported from 1968 to 2001, (3-10) aside from the 13 patients in this series (see Table 5).

The outcome of visceral arterial injuries has been correlated with the presence of shock at admission, (1,11,16,25) the need for ED or OR thoracotomy, (6,16,24-28) the number of associated injuries, (16) the anatomic location of injuries, (1,2,16,24-28) method of repair, and the presence of "black bowel" on entering the abdominal cavity. (1,2,11,16,24--28)

The visceral arterial circulation consists of the celiac axis, the superior mesenteric artery (SMA), and the inferior mesenteric artery. They arise directly from the anterior aspect of the abdominal aorta and supply the foregut, midgut, and hindgut, respectively. These three vessels supply both the small and large bowel and carry approximately 20 to 25% of the total cardiac output. Specifically, the celiac axis arises from the anterior surface of the aorta just below the aortic hiatus of the diaphragm at the level of T12. It usually measures between 1.0 and 1.5 cm in length. It is well surrounded in its origin by a very dense plexus of neural tissue, including the celiac ganglia. The origin of this vessel is nestled deeply, well posterior into the abdomen, and is surrounded by important organs such as the pancreas, duodenum, and portal venous system. (1)

The celiac axis gives rise to the splenic, left gastric, and common hepatic arteries. This trifurcation is known as the tripod of Haller. According to Strandness (28) and others, (29-34) 90% of patients have all three major branches; however, two branches are found in 10% of the population, with the following distribution: gastric and splenic in 5.5%, hepatic and splenic in 3.5%, and hepatic and gastric in 1.0%. All three visceral arteries are linked by collateral circulation. (30-34) The celiac axis is linked to the SMA by both the superior and inferior pancreaticoduodenal arteries and their rich anastomotic network surrounding the head of the pancreas. In addition, this vessel has connections with vessels supplying the diaphragm, esophagus, and intercostals, plus the abdominal wall. (1,29-34)

The incidence of celiac axis injury has not been previously reported in the literature. To calculate the incidence, we looked at our trauma center's experience. During the same period of time, there were 2,357 patients who required trauma laparotomy. Of those, there were 1,010 laparotomies for gunshot wounds, 498 for stab wounds, and 849 for blunt trauma. The calculated incidence of celiac axis injuries in our trauma center is 0.01%.

Penetrating trauma remains the predominant cause of the majority of celiac axis injuries. (1,2) In our series, however, one resulted from blunt trauma. To the best of our knowledge, this is the only reported case that was caused by this mechanism. Of the 33 celiac axis injuries identified in our literature review, only the case reported by Kavic et al (10) describes its mechanism of injury, although the majority are assumed to occur secondary to penetrating trauma. In the largest study to date, Graham et al (6) reported 66 patients with visceral arterial injuries, of which 13 had celiac axis injuries. In this series, 35 patients presented with blood pressure less than 70 mm Hg. In our series, the mean admission blood pressure was 91 mm Hg; moreover, the mean RTS was 5.35 and the mean ISS was 25, denoting that our patients were a physiologically compromised and severely injured patient population.

The use of resuscitative thoracotomy has been reported in the treatment of these injuries by Graham et al, (6) Mattox et al, (5) Kashuk et al, (7) and Davis et al. (9) In this series, three of our patients underwent ED thoracotomy, which allowed them to reach the OR; however, all died. In addition, five patients underwent OR resuscitative thoracotomy, and only one survived. During surgery, the presence of a zone I retroperitoneal hematoma is usually indicative of injury to the aorta or its branches. In our series, 54% of the patients had zone I retroperitoneal hematoma. Surgical exposure of the celiac axis can be accomplished by medial rotation of the left-sided viscera, including mobilization of the left colon, spleen, pancreas, and stomach. (1,2,11)

The surgical treatment of celiac axis injuries consists either of primary repair or ligation. Most of the series reported in the literature reveal little detail regarding the specific treatment of these injuries. In the Mattox et al (5) series, it appears that all three associated celiac axis injuries were treated by ligation. Of the 13 injuries reported by Graham et al, (6) eight were treated with primary arteriorrhaphy, four were ligated, and one underwent end-to-end anastomosis. Of the six patients reported by Davis et al, (9) four were treated with ligation, one with primary repair, and one exsanguinated without repair. In our series, 11 patients were treated by ligation (one by primary arteriorrhaphy and one exsanguinated before definitive repair).

It appears that surgical technique correlates with survivability. In the Graham et al (6) series, of the eight patients given primary repair, six survived, versus two of the four ligated. In the Kashuk et al (7) series, all six patients with celiac axis injuries survived, although no description of their surgical treatment is presented. Kavic et al (10) reported a single patient who survived with ligation. From the literature, it appears that any patient subjected to complex repairs such as primary end-to-end anastomosis and/or reimplantation into a prosthetic abdominal aortic graft failed to survive. (1-2,5-6) In our series, four of the five survivors were treated with ligation and one with primary repair. From our limited experience, we recommend primary repair if the injury is small and tangential, and would reserve ligation for those with larger and more destructive lesions involving the majority of the lumen of the artery. Our overall survival in this series is 38%; if patients with ED thoracotomy are excluded, the adjusted survival is 50%, which compares well with the literature.

An important predictor of outcome is the number of associated vessels injured with the celiac axis. In the Graham et al (6) series, all three patients who had three associated vascular injuries died. However, for the six patients with one associated vascular injury, the mortality rate was lower (33%), and all patients with isolated celiac axis injuries survived. In our series, the seven patients with isolated celiac axis injuries had a 57% survival rate, whereas the remaining six with associated vessels injured had a 17% survival rate. One patient with an associated renal artery injury survived; this vessel was primarily repaired. Three of our patients had associated portal venous and aortic injuries leading to a 100% mortality rate. Patients with celiac axis injuries clearly incur significant blood losses. Graham et al (6) reported an average of 12.7 U of packed red blood cells administered per patient. In our series, the mean estimated blood loss was 5,400 mL, with a mean replacement of approximately 13 U of packed red blood cells per patient.

According to Kavic et al, (10) ligation has not been the preferred approach to deal with celiac axis injury. This is in clear disagreement with the experience reported in the literature. Feliciano et al (2) suggest that celiac axis ligation be used as a method of treatment for patients in extremis. Although they cite no specific study, they state that celiac axis ligation can be accomplished without morbidity or mortality or the development of bowel ischemia. Ligation is tolerated because of the existing collateral circulation between the celiac axis and SMA. In our review of the literature, there are no reported cases of bowel ischemia secondary to celiac axis ligation, although Kavic et al (10) report a case of full-thickness necrosis of the gallbladder after ligation. In our series, none of the surviving patients had intestinal ischemia.

Little experience is recorded in the literature detailing the use of bailout/damage control for the treatment of this type of injury. (30,31) None of the older series specifically reports its use, other than the single case reported by Kavic et al. (10) In our series, one patient underwent damage control. On reexploration, there was no evidence of intestinal ischemia. No information exists in the literature describing the use of regional heparinization for patients requiring complex reconstruction and/or primary repair. Besides the case reported by Kavic et al (10) and one of our patients, there is little information describing the use of second-look procedures.

Conclusion

Although previous series (11,15-16) describing abdominal vascular and SMA injuries have correlated mortality rates with the AAST-OIS for abdominal vascular injury, (14) none has reported its application to celiac axis injuries. It appears from our series, which matches the largest experience previously reported by Graham et al, (6) that mortality rate correlates well with AAST-OIS injury grade, although the numbers are very small. From our limited experience, we conclude that celiac axis injuries are among the rarest of all vascular injuries reported. They incur high mortality rates, in large part because of the difficulty of obtaining rapid exposure and hemorrhage control. Associated vascular injuries appear to increase mortality rates. In our experience, operative treatment can be achieved with ligation or repair. (1,2,11,16,24)

Table 1. Mechanism of injury

Penetrating  12 (92%)  GSW  9 (75%)
                       SW   3 (25%)
Blunt         1 (8%)   MVA  1 (100%)

GSW, gunshot wound; SW, stab wound; MVA, motor vehicle accident.

Table 2. Celiac axis and associated vessel injuries

                                                   Survival
Vessel name                      Total  Survivors     %

Isolated celiac A                  7        4         57
Celiac A + splenic A               1        0          0
Celiac A + renal A                 1        1        100
Celiac A + portal V                2        0          0
Celiac A + portal V + renal A +    1        0          0
  gonadal V
Celiac axis + aorta                1        0          0
Total                             13        5         38

A, artery; V, vein.

Table 3. Operative management

                              Whole group  Survivors
                              (13) n (%)   (5) n (%)

Ligation                        11 (85)      4 (75)
Primary repair                   1 (8)       1 (25)
None (due to exsanguination)     1 (8)

Table 4. American Association for Surgery of Trauma-Organ Injury Scale
(AAST-OIS) for abdominal vascular injuries

                   Mortality rate
           n (%)       n (%)

Grade I    0 (0)       0 (0)
Grade II   0 (0)       0 (0)
Grade III  7 (54)      3 (43)
Grade IV   2 (15)      1 (50)
Grade V    4 (31)      4 (100)

Table 5. Celiac axis injuries reported in literature

               No. of    Mortality
Author   Year  patients  rate (%)

Patman   1964      1        NR
Perry    1971      2        NR
Mattox   1975      3        NR
Graham   1978     13        38
Kashuk   1982      6         0
Adkins   1985      1         0
Davis    2001      6        NR
Kavic    2001      1         0
Asensio  2002     13        50

NR, Not reported.

Accepted May 4, 2004.

References

1. Asensio JA, Forno W, Roldan G, et al. Visceral vascular injuries. Surg Clin North Am 2002;82:1-20.

2. Feliciano DV, Burch JM, Graham JM. Abdominal Vascular Injury. In Mattox KL, Feliciano DV, Moore EE (eds). Trauma. New York, NY, McGraw Hill, 1999, ed 4, pp 783-805.

3. Patman RD, Poulos E, Shires GT. The management of civilian arterial injuries. Surg Gyne Obs 1964;118:725-738.

4. Perry MO, Thal ER, Shires GT. Management of arterial injuries. Ann Surg 1971;173:403-408.

5. Mattox KL, McCollins WM, Beall AC, et al. Management of penetrating injuries of the suprarenal aorta. J Trauma 1975;15:808-815.

6. Graham JM, Mattox KL, Beall AC, et al. Injuries to the visceral arteries. Surgery 1978;84:835-839.

7. Kashuk JL, Moore EE, Millikan JS, et al. Major abdominal vascular trauma: a unified approach. J Trauma 1982;22:672-679.

8. Adkins RB, Bitseff EL, Meacham PW. Abdominal vascular injuries. South Med J 1985;78:1152-1160.

9. Davis TP, Feliciano DV, Rozycki GS, et al. Results with abdominal vascular trauma in the modern era. Am Surg 2001;67:565-571.

10. Kavic SM, Atweh N, Ivy ME, et al. Celiac axis ligation after gunshot wound to the abdomen: case report and literature review. J Trauma 2001;50:738-739.

11. Asensio JA, Chahwan S, Hanpeter D, et al. Operative management and outcome of 302 abdominal vascular injuries. Am J Surg 2000;180:528-534.

12. Asensio JA, Buckman RF Jr, Mauro L. Exsanguination from penetrating injuries. Trauma Q 1989;6:1-25.

13. Asensio JA, Ierardi R. Exsanguination. Emerg Care Q. 1991;7:59-75.

14. Moore EE, Cogbill TH, Jurkovich GJ, et al. Organ injury scaling, III: chest wall, abdominal vascular, ureter, bladder, and urethra. J Trauma 1992;33:337-339.

15. Asensio JA, Britt LD, Borzotta A, et al. Multi-institutional experience with the management of superior mesenteric artery injuries. J Am Coll Surg 2001;193:354-356.

16. Asensio JA, Berne JD, Chahwan S, et al. Traumatic injury to the superior mesenteric artery. Am J Surg 1999;178:235-239.

17. Makins GH. Gunshot Injuries to the Blood Vessels. Bristol, England, John Wright and Sons, Ltd, 1919.

18. DeBakey ME, Simeone FA. Battle injuries of the arteries in World War II: an analysis of 2,471 cases. Ann Surg 1946;123:534.

19. Hughes CW. Acute vascular trauma in Korean War casualties: an analysis of 180 cases. Surg Gyne Obst 1954;99:91-100.

20. Rich NM, Baugh JH, Hughes CW. Acute arterial injuries in Victnam: 1,000 cases. J Trauma 1970;10:359.

21. Feliciano DV, Bitondo CG, Mattox KL, et al. Civilian trauma in the 1980s: a 1-year experience with 456 vascular and cardiac injuries. Ann Surg 1984;199:717.

22. Mattox KL, Feliciano DV, Burch J, et al. Five thousand seven hundred sixty cardiovascular injuries in 4459 patients: epidemiologic evolution 1958 to 1987. Ann Surg 1989;209:698.

23. Tyburski JG, Wilson RF, Dente C, et al. Factors affecting mortality rates in patients with abdominal vascular injuries. J Trauma 2001;50:1020-1026.

24. Lucas AE, Richardson JD, Flint LM. Traumatic injury of the proximal superior mesenteric artery. Ann Surg 1981;193:30-34.

25. Accola KD, Feliciano DV, Mattox KL, et al. Traumatic injury of the proximal superior mesenteric artery. J Trauma 1986;26:313-319.

26. Graham JM, Mattox KL, Beall AC. Injuries to the visceral arteries. Surgery 1978;84:835-839.

27. Sirinek KR, Levine BA. Trauma injury to the proximal superior mesenteric vessels. Surgery 1985;98:831-835.

28. Strandness DE. Collateral circulation in clinical surgery. Philadelphia, WB Saunders, 1969, pp 508-536.

29. Bongard FS, Wilson SE, Perry MO. Vascular injuries in surgical practice. Norwalk, Appleton & Lange, 1991, pp 165-184.

30. Flannigan DP. Civilian vascular injuries. Philadelphia, PA, Lea & Febiger, 1992, pp 176-190.

31. Rotondo MF, Schwab CW, McGonigal MD, et al. "Damage control": an approach for improved survival in exsanguinating penetrating abdominal injury. J Trauma 1993;35:375.

32. Asensio JA, McDuffie L, Petrone P, et al. Reliable variables in the exsanguinated patient which indicate damage control and predict outcome. Am J Surg 2001;182:743-751.

33. Friedman SG. A history of vascular surgery. Mt Kisco, NY, Futura Publishing Co, 1989.

34. Moore WS. Vascular Surgery: A Comprehensive Series. New York, NY, Grune & Stratton, 1983, pp 465-470.

RELATED ARTICLE: Key Points

* Injuries to the celiac axis incur high mortality rates.

* Exsanguinating hemorrhage is the most common cause of early death.

* Associated vascular injuries increase mortality rates.

* Operative treatment can be achieved with ligation or repair.

Juan A. Asensio, MD, FACS, Patrizio Petrone, MD, Brian Kimbrell, MD, and Eric Kuncir, MD, FACS

From the Division of Trauma and Surgical Critical Care, Department of Surgery, University of Southern California Keck School of Medicine, Los Angeles County--University of Southern California Medical Center, Los Angeles, CA.

Reprint requests to Juan A. Asensio, MD, FACS, Division of Trauma and Critical Care, Department of Surgery, University of Southern California Keck School of Medicine, LAC + USC Medical Center, 1200 North State Street, Room 10-750, Los Angeles, CA 90033-4525. Email: [e-mail]