Full text of DOI:10.1097/00005373-200210000-00026

CASE
REPORT
௡
The Journal of TRAUMA Injury, Infection, and Critical Care
Small Bowel Transplantation and Staged Abdominal Wall
Reconstruction after Shotgun Injury
Evangelos Tzoracoleftherakis, MD, Mimis Cohen, MD, Pierpaolo Sileri, MD, Luca Cicalese, MD, and
Enrico Benedetti, MD
J Trauma. 2002;53:770–776.
enetrating injuries to the abdomen from shotguns carry
significant mortality and morbidity.1 Such injuries can
result in loss of the abdominal wall and various injuries
infection and total dehiscence. The duodenal stump dehisced
as well, and high-output content was spilling into the abdom-
inal cavity.
The patient was transferred to our institution approxi-
mately 5 weeks after the injury and the initial exploration for
further care. On transfer, the patient was cachectic and suf-
fered from multiple organ failure. His weight was 45 kg. The
abdominal wound was open, the abdominal cavity, including
the left lobe of the liver, was entirely exposed, and the
duodenal stump was wide open.
P
to the viscera.
Complex injuries require early extensive evaluation, re-
suscitation, and management. Cooperation with other special-
ists, prioritization, and staging of procedures is extremely
important and highly recommended to reduce the number of
complications and achieve the best possible results and
rehabilitation.
We present the case of a victim of a shotgun wound to
the abdomen who was transferred to our institution with
multiple organ failure; partial loss of the abdominal wall;
dehiscence of a previous laparotomy incision; and loss of the
entire small bowel, the left kidney, and spleen (Fig. 1). He
was managed jointly by the transplant and the plastic surgery
services and underwent several procedures before total
rehabilitation.
After resuscitation and stabilization, the abdominal
wound was debrided to healthy tissue, a draining catheter was
placed in the duodenal stump, and total parenteral nutrition
CASE REPORT
A 28-year-old man sustained an extensive abdominal
injury from a close-range shotgun blast. He was admitted to
a local hospital with an entrance wound on the right upper
quadrant of the abdomen and significant soft tissue damage.
The patient was unstable and underwent urgent exploration
through a midline incision. Because of significant damage to
the superior mesenteric pedicle and the lack of perfusion to
the small bowel, a near-total resection of the small bowel was
performed. In addition, a left nephrectomy, distal pancreate-
ctomy, and splenectomy were performed at the same setting.
The duodenum was stapled 3 cm distally to the ampulla. A
gastrostomy was also placed and the abdominal wall was
closed in layers after debridement of all grossly devitalized
tissues. A few days later, the abdominal wound developed an
Submitted for publication December 10, 2001.
Accepted for publication January 30, 2002.
Copyright © 2002 by Lippincott Williams & Wilkins, Inc.
From the Divisions of Transplantation (E.T., P.S., L.C., E.B.) and
Plastic Surgery (M.C.), University of Illinois at Chicago, and Division of
Plastic Surgery, Cook County Hospital (M.C.), Chicago, Illinois.
Address for reprints: Mimis Cohen, MD, Division of Plastic Surgery,
University of Illinois at Chicago, 820 South Wood Street, Suite 515, CSN,
M/C 958, Chicago, IL 60612.
DOI: 10.1097/01.TA.0000032119.91608.DF
Fig. 1. Surgical events following the trauma.
770
October 2002

Abdominal Surgery after Shotgun Injury
Fig. 3. Meshed split-thickness skin graft directly applied to the
exposed liver.
Fig. 2. Open abdominal wound with draining catheter in the duo-
denal stump and exposed left lobe of the liver.
rectus abdominis muscle was mobilized, based on the supe-
rior epigastric pedicle and sutured around the anastomosis,
providing an additional layer of coverage with well-vascular-
ized tissue (Figs. 4–6). The abdominal wound was then
(TPN) was initiated. Two weeks after admission to our insti-
tution, the patient was stabilized. At this point, we were faced
with the following problems: an open abdominal wound 24
cm in length with partial loss of tissue in the right upper
quadrant and exposed liver (Fig. 2), high-output drainage
from the duodenal stump, and the complete absence of the
small bowel.
A staged approach was designed in collaboration be-
tween the transplant and the plastic surgery services. Because
of the patient’s general condition and poor nutritional status,
we first elected to perform a simple coverage of the liver with
a split-thickness skin graft (Fig. 3). This procedure was suc-
cessful, with complete acceptance of the graft and coverage
of the exposed liver. Three weeks later, we proceeded to
repair the duodenal stump and close the abdominal cavity.
After limited debridement of the stump, because of the prox-
imity to the ampulla, the stump was oversewn. Because
tissues were extremely friable and to prevent further compli-
cations with dehiscence of the stump, the suture line was
reinforced with a rectus abdominis muscle flap. The left
Fig. 4. Duodenal stump (arrow) after debridement and repair.
Volume 53 • Number 4
771

௡
The Journal of TRAUMA Injury, Infection, and Critical Care
Fig. 5. Superiorly based left rectus abdominis muscle mobilized to
completely cover the repair of the duodenum.
closed in layers using a modification of the components
separation technique2 (Fig. 7).
This procedure was successful, the duodenal stump
healed uneventfully, and the abdominal wound healed as
well. The patient’s general condition continued to improve.
However, the patient experienced multiple episodes of line
sepsis. Furthermore, he presented incipient cholestasis related
to TPN, with a total bilirubin up to 3.2 mg/dL but with no
evidence of cirrhosis by liver biopsy. The patient was eval-
uated as a candidate for intestinal transplantation. Six months
after the initial injury, he underwent a living-related trans-
plantation of 185 cm of ileus harvested from his father. The
father was 1 haplotype-matched and ABO compatible with
him. Both donor and recipient were found to have a negative
Fig. 7. Closure of abdominal wound after repair of the duodenal
stump fistula.
cytomegalovirus serology. The details of the surgical tech-
nique have already been described3 (Fig. 8). Briefly, the
segment of ileum was resected 15 cm from the ileocecal valve
after measurement of the total length of the donor ileum. The
donor ileum was primarily reanastomosed in end-to-end fash-
ion using 4-0 polyglyconate for the mucosal layer and 4-0
polypropylene for the seromuscular layer. The graft was
transplanted suturing the ileocolic vessels in an end-to-side
fashion to the aorta and inferior vena cava using 6-0 polypro-
pylene with a cold ischemia time of less than 10 minutes and
a warm ischemia time of 30 to 40 minutes. The intestinal
continuity was immediately reestablished proximally anasto-
mosing the graft to the recipient’s intestinal stump and dis-
tally to sigmoid colon, using similar suture materials as in the
donor. A temporary distal loop ileostomy was performed to
monitor graft output and to perform endoscopic surveillance
and biopsies.
Fig. 6. Coverage of duodenal stump. 1, Rectus abdominis; 2,
superior epigastric pedicle; 3, stomach; 4, duodenum; 5, area of
duodenal repair; 6, stump of right rectus abdominis; 7, stump of left
rectus abdominis.
To prevent closure of the abdominal wall with undue
tension, a 5 ϫ 6-cm area of the abdomen was left open,
covered only with an absorbable mesh. One week later, a
772
October 2002

Abdominal Surgery after Shotgun Injury
fluids were collected as part of routine surveillance or when
infections were clinically suspected and cultures were per-
formed using standard microbiologic techniques. Serial biop-
sies (weekly for the first postoperative month and then on the
basis of clinical indication) of the intestinal grafts were per-
formed to evaluate rejection or viral infections.
Tacrolimus target blood levels were 20 to 25 ng/mL the
first month, 15 to 20 ng/mL the second month, and 10 to 15
ng/mL thereafter.
His liver function tests normalized within 2 weeks; at
that time, bilirubin was down to 1 mg/dL and remained
normal thereafter. He experienced high stomal output in the
first month posttransplant that progressively decreased to the
point that the patient did not require any further assistance at
6 weeks posttransplant. The ileostomy was successfully taken
down 4 months after the transplant. From a nutritional stand-
point, a fat-soluble vitamin level and Schilling test were
already normal 1 month posttransplant. Fat absorption docu-
mented by fecal fat studies was normalized by 6 months.
The D-xylose absorption test was back to normal at 6
months. The serum albumin level increased steadily after the
transplant and has remained normal to date. Importantly, the
patient has been able to gain weight and maintain it at 61 kg
compared with 45 kg before the transplant. Thirty months
after the procedure, he has doing well and has a stable
abdomen (Fig. 9).
DISCUSSION
This case clearly demonstrates the need for close coop-
eration between specialists for the successful management of
the various problems associated with complex penetrating
injuries to the abdomen. Penetrating trauma is the leading
cause of duodenal injury in countries such as the United
States and South Africa, where 75% to 80% of the trauma is
caused by gunshot wounds. According to the 1997 report of
the International Registry for Intestinal Transplantation, 21%
of the adult transplants are caused by ischemia and 15% are
caused by trauma.4 Our patient presented with superior mes-
enteric vessel ischemia, and with soft tissue trauma.
Injuries to the superior mesenteric artery are primarily
because of penetrating trauma. Entrance wounds can be to the
anterior abdomen, the flank, or the back. The majority of
portal-superior mesenteric vein injuries are penetrating, and
many are associated with an adjacent arterial injury.5 Com-
bination of superior mesenteric artery and vein injury has a
significant mortality, reported to be 43%.6 In a recent report
of a multi-institutional experience with the management of
superior mesenteric artery injuries, injury to the Fullen’s
anatomic zone I (trunk proximal to first major branch, infe-
rior pancreaticoduodenal) corresponding to the maximal isch-
emic category affecting jejunum, ileum, and right colon pre-
sents the highest mortality (76.5%).7 The American
Association for the Surgery of Trauma Organ Injury Scale for
abdominal vascular injury grades the superior mesenteric
trunk injury as grade IV injury,8 with overall mortality of
Fig. 8. Small bowel transplant. Side-to-side anastomosis of the
proximal end of the small bowel to the duodenum and the distal end
to the sigmoid colon. The graft’s ileocolic vessels were anastomosed
end-to-side to the infrarenal aorta and vena cava. A temporary loop
ileosigmoidostomy was performed to monitor the output and to
perform biopsy of the bowel.
split-thickness skin graft was applied over the granulation
tissue on the mesh. The recipient had a smooth and uncom-
plicated recovery. He was able to tolerate a general diet by
the seventh postoperative day, allowing early discontinuation
of TPN. The patient did not experience any rejection episode
and he has had no infectious complications to date. Short-
course perioperative antibiotic prophylaxis with vancomycin
(1 g intravenously before surgery) and piperacillin (3 g in-
travenously six or eight times a day, depending on renal
function, for 3 days) was used, and antiviral prophylaxis was
achieved using perioperative ganciclovir (5 mg/kg intrave-
nously every 12 hours for 14 days) followed by high-dose
acyclovir (800 mg per os four times a day for 3 months).
Immunosuppression consisted of oral tacrolimus (Fuji-
sawa Healthcare, Inc., Deerfield, IL), prednisone, and intra-
venous induction with Atgam (antithymocyte globulin,
Kalamazoo, MI) until therapeutic blood levels of tacrolimus
were reached. Blood, stool, urine, sputum, and peritoneal
Volume 53 • Number 4
773

௡
The Journal of TRAUMA Injury, Infection, and Critical Care
fistula. Advantages of skin graft application include wound
stabilization, reduction of risk for secondary infection, rup-
ture of anastomosis when present, and fistula formation. The
loss of fluid and electrolytes from the open wound is reduced
as well.
Furthermore, the procedure is quick and safe and is
strongly indicated for debilitated patients, as in our case.
There was significant tissue loss over the right upper quad-
rant, and coverage with complex reconstructive procedures
and flaps was contraindicated at the time. It was our rationale
that after stabilization of the wound over the liver we could
close the abdominal defect without tension.
The use of muscle flaps to support friable anastomosis is
also a well-accepted principle. Most reported cases include
reinforcement of anastomotic areas of the bronchus, thoracic
esophagus, heart, great vessels, and the bladder. A few scat-
tered reports limited to a small number of patients have
appeared in the literature for management of difficult and
recalcitrant fistulas with muscle flaps.11–13 Muscle flaps pro-
vide coverage of the suture line with well-vascularized tissue.
Healthy tissue brought from areas outside the site of injury,
fibrosis, and scarring are extremely helpful in promoting
healing and reducing fistula recurrence.
The open duodenal stump in our patient was extremely
difficult to manage because of its proximity to the ampulla;
an ischemic and fibrotic stump would have resulted in an
injury or ligation around the area of the ampulla. A major
reconstructive procedure for biliary and pancreatic ducts was
certainly not an option at the time.
Fig. 9. Result 18 months after small bowel transplant with com-
pletely healed and stable abdominal wall. The bulge in the abdomen
(arrow) is not a hernia, but caused by the rectus abdominis flap.
In addition to the closure and coverage of the duodenal
stump, the abdominal wall was closed at the same setting.
There was very little tissue loss after the debridement of the
dehisced midline incision. However, because of the unop-
posed pull of the lateral abdominal muscles and the subse-
quent tissue fibrosis, direct layered closure was not possible.
We elected to use a modification of the components separa-
tion technique described by Ramirez et al.2 The plane be-
tween the rectus muscle and the posterior rectus sheath on the
right side was developed. A relaxing incision was then car-
ried on the fascia of the external oblique and a plane was
developed between this muscle and the internal oblique mus-
cle. With this maneuver, the right rectus, the internal oblique,
and the transversalis myofascial complex were mobilized
medially by approximately 3 cm.
The final reconstructive problem we were faced with was
the open abdominal wound after the small bowel transplant.
Because of the contracted size of the abdominal cavity, pri-
mary closure of the entire cavity after small bowel transplan-
tation was not feasible. Attempts for further closure would
have resulted in dehiscence and further loss of tissue and
could have placed the transplanted small bowel at risk from
undue tension. An absorbable mesh was used to temporarily
cover the remaining 5 ϫ 6-cm defect. One week after the
procedure, a split-thickness skin graft was applied over the
granulating bed. The graft healed uneventfully.
53.6%.7 The retroperitoneal location of the duodenum and its
close proximity to a number of other viscera and major
vascular structures means that isolated penetrating injury of
the duodenum itself is rare.9
In our case, there was a left kidney injury along with the
injury to the superior mesenteric artery and vein. This injury
resulted in left nephrectomy and near total enterectomy. A
duodenal fistula and an extensive abdominal wall defect were
present. A staged approach was elected in an effort to stabi-
lize the patient and manage each injury in a logical approach,
after prioritization of the patient’s needs and establishment of
a short- and long-term plan of action. Our initial attention was
directed toward stabilization of the patient, management of
the multiple organ failure, control of sepsis, improvement of
nutritional status with TPN, and wound care.
The application of split-thickness skin grafts for tempo-
rary or permanent coverage of abdominal viscera has been
well described and accepted.10 The success rate of these
grafts is extremely high, provided that there is no active
infection and a granulative tissue bed has formed over the
viscera. Even the presence of fistulas is not a contraindication
for grafting. The skin grafts take well in all areas around the
774
October 2002

Abdominal Surgery after Shotgun Injury
The patient’s general condition dramatically improved
after the transplant. He was managed according to existing
standard protocols, and at 18-month follow-up he presented
with significant weight gain, excellent bowel function, and a
stable abdomen. The progress in immunosuppressive therapy
and the continuous technical refinements in the past decade
have allowed a spectacular increase in the volume of clinical
bowel transplantation. The most recent report of the Interna-
tional Intestinal Transplant Registry documented a total of
over 270 cases of intestinal transplantation performed world-
wide until February 1997, with a linear rate of growth from
1994.4 The great majority of these cases are from cadaveric
donors, and patient and graft survival rates (69% and 55% at
1 year, respectively) are still poor even only considering
cases performed after 1995, according to the registry data.
Furthermore, only 77% of the patients with functioning graft
achieved the ultimate target of freedom from TPN. Technical
complications, infectious complications, and rejection ac-
count for most of the patient and graft losses.14 Segmental
bowel transplantation from living-related (LR) donors could
potentially improve the outcomes of the procedure in many
ways. To date, 21 cases of intestinal transplantation using LR
donors have been reported worldwide.15 Similar to the trans-
plant of other organs, intestinal living donation offers several
advantages, such as reduced preservation injury, better HLA
matching, and optimal donor and graft conditions. However,
this procedure cannot be performed from living donors using
the standardized techniques used with cadaver grafts, and a
series of transplants using LR donors has not been available
to unequivocally demonstrate such advantages. Moreover,
LR small bowel transplantation has not encountered initial
preference among the intestinal transplant surgeons because
bowel grafts are widely available from cadavers. A standard-
ized technique has been recently proposed for small bowel
transplantation by Gruessner et al.,3 who clearly demon-
strated that the use of a segmental ileal graft vascularized by
the ileocolic artery and vein, anastomosed to the infrarenal
aorta and vena cava, is a reliable technique.
clude the risks for the donor, technical problems in trans-
planting a segmental graft in the recipient, and the concern
regarding the functional ability of a relatively short graft to
provide adequate absorption. The surgical risk of an elec-
tive small bowel resection with primary anastomoses is
relatively low in experienced hands. The rationale to use
200 cm of bowel is dictated by the need for the longest
possible segment with minimum risk for the donor and no
long-term functional adaptation problems. The donor does
not suffer any long-term absorption problem with an ileal
resection limited to 200 cm, and the initial diarrhea can be
easily controlled with medical therapy without any signif-
icant malabsorption (in particular, vitamin B12).3,16 Con-
versely, there are data suggesting that, through a progres-
sive and relatively rapid functional adaptation, a 200-cm
ileal graft can fully support the nutritional requirements of
an active adult.16
In conclusion, several transplantation and reconstructive
principles were used to manage this patient. With close cooper-
ation between the transplant and the plastic surgeons, prioritiza-
tion of the patient’s needs, and short- and long-term planning,
we were able to achieve total functional rehabilitation.
REFERENCES
1. Mathes JS, Stone HH. Acute traumatic loss of abdominal wall
substance. J Trauma. 1975;15:386–390.
2. Ramirez OM, Ruas E, Dellon L. Components separation method for
closure of abdominal wall defects: an anatomic and clinical study.
Plast Reconstr Surg. 1990;86:519–523.
3. Gruessner RWG, Sharp HL. Living-related intestinal transplantation:
first report of a standardized surgical technique. Transplantation.
1997;64:1605–1607.
4. Grant D. On behalf of the Intestinal Transplant Registry: Intestinal
Transplantation—1997 Report of the International Registry.
Transplantation. 1999;67:1061–1064.
5. Mullins RJ, Huckfeld R, Trunkey DD. Abdominal vascular injuries.
Surg Clin North Am. 1996;76:813–832.
6. Asensio J, Chahwan S, Hanpeter D, et al. Operative management
and outcome of 302 abdominal vascular injuries. Am J Surg. 2000;
180:528–534.
Most likely, better HLA matching could help to de-
crease the unacceptably high rate of rejection currently
experienced in cadaveric small bowel transplants.14,16,17
The donor can be easily matched on the basis of cytomeg-
alovirus status. Once a suitable donor is identified, the
transplant can be carried out expeditiously in an elective
fashion, when both donor and recipient are in optimal
conditions. Appropriate bowel preparation can be given.
Cold ischemia time can be virtually eliminated and the
warm ischemia time minimized, potentially improving mu-
cosal preservation and decreasing bacterial translocation.
A segmental graft may be quite helpful for potential re-
cipients with a small abdominal cavity as a consequence of
abdominal trauma and/or massive bowel resection. Finally,
immunologic manipulations can be easily carried out in the
elective setting of a living-related transplant.16 The disad-
vantages of living-related small bowel transplantation in-
7. Asensio JA, Britt LD, Borzotta A, et al. Multi-institutional
experience with the management of superior mesenteric artery
injuries. J Am Coll Surg. 2001;193:354–366.
8. Moore EE, Cogbill TH, Jurkovice GJ, et al. Organ injury scaling:
III—chest wall, abdominal vascular, ureter, bladder and urethra.
J Trauma. 1992;33:337–339.
9. Degiannis E, Boffard K. Duodenal injuries. Br J Surg. 2000;
87:1473–1479.
10. Cohen M, Morales R, Fildes J, Barrett J. Staged reconstruction after
gunshot wounds to the abdomen. Plast Reconstr Surg. 2001;108:83–
92.
11. Cohen M, Greco GA, Ramasastry S. Staged abdominal wall
reconstruction: why? when? how? Plast Surg Forum. 2000;23:174–
177.
12. Lui R, Friedman R, Fleischer A. Management of postirradiation
recurrent enterocutaneous fistula by muscle flaps. Am Surg. 1989;
55:403–407.
13. Chang P, Chun JT, Bell JL. Complex enterocutaneous fistula closure
with rectus abdominis muscle flap. South Med J. 2000;93:599–602.
Volume 53 • Number 4
775

௡
The Journal of TRAUMA Injury, Infection, and Critical Care
14. Benedetti E, Baum C, Cicalese L, et al. Progressive functional
adaptation of segmental bowel graft from living-related donor.
Transplantation. 2001;71:569–571.
15. Cicalese L, Rastellini C, Sileri P, Abcarian H, Benedetti E. Living
related small bowel transplantation. Graft. 2001;4:526–534.
16. Reyes J, Bueno J, Kocoshis S, et al. Current status of intestinal
transplantation in children. J Pediatr Surg. 1998;33:243–249.
17. Cicalese L, Rastellini C, Sileri P, Abcarian H, Benedetti E.
Segmental living-related small bowel transplantation in adults.
J Gastrointest Surg. 2001;5:168–173.
776
October 2002