CASE REPORT

Intestinal Autotransplantation for Adenocarcinoma

of Pancreas Involving the Mesenteric Root
Our Experience and Literature Review
Cristiano Quintini, MD,* Fabrizio Di Benedetto, MD, PhD,* Teresa Diago, MD,* Augusto Lauro, MD,Þ
Nicola Cautero, MD,* Nicola De Ruvo, MD,* Antonio Romano, MD,* Stefano Di Sandro, MD,*
Giovanni Ramacciato, MD,* and Antonio D. Pinna, MDÞ

subgroup of patients can potentially benefit from curative

Abstract: Ductal adenocarcinoma of pancreas represents one of the
surgery. So far, only the involvement of resection margins has
most aggressive tumor as demonstrated by 3- and 5-year survival
been shown to be strongly associated with better long-term
rates. Involvement of mesenteric pedicle affects both the possibility
survival rates, with no univocal consensus on other prog-
to perform a tumor-free margin resection and accounts for most
nostic factors such as tumor size, nodal status, histology, and
exploratory laparotomy for locally advanced disease. The ex vivo
vascular involvement.1,2 It is well established that the most
resection of the tumor (autotransplantation) after total exenteratio
common obstacle to performing clean surgical margins
and perfusion of the intestine might have a role to overcome some
during pancreatoduodenectomy is represented by tumor
technical obstacles. So far, only 5 patients have been reported to
involvement of superior mesenteric artery and vein.1 Based
have undergone small-bowel autotransplantation for tumor involv-
on this concept, any improvement in resection rate can
ing the mesenteric root. We describe 2 cases of adenocarcinoma of
potentially lead to an increase in survival rate. So far, only 2
pancreas involving mesenteric root treated by small-bowel auto-
reports on 5 patients3,4 describe the ex vivo resection and
transplantation. Both patients survived from the procedure and were
intestinal autotransplantation for tumor involving the mesen-
discharged home on postoperative days 16 and 29, respectively. The
teric root, with only 1 case reported in literature for pancreatic
tumor was resected with free surgical margins, and both patients
adenocarcinoma. In this last case, the resection did not meet
underwent adjuvant treatment. Intestinal autotransplantation can
oncological criteria late intraoperative finding of a small liver
represent a significant technical advance for increasing the
metastasis. In this article, we describe our experience in
resectability rate and, ultimately, the survival rate for advanced
intestinal autotransplantation with intention to cure per-
adenocarcinoma of the pancreas in highly selected patients.
formed on 2 patients affected by ductal adenocarcinoma of
Key Words: adenocarcinoma of pancreas, intestinal pancreas involving the mesenteric root.
autotransplantation, mesenteric root involvement, abdominal
exenteratio, technical advance
Abbreviations: SMA - superior mesenteric artery, SMV - superior CASE REPORTS
mesenteric vein, PV - portal vein Two patients underwent intestinal autotransplantation at our
Institute.
(Pancreas 2007;34:266Y268) Patient 1 is a 43-year-old man with a 6-week history of
generalized abdominal discomfort, nausea, and progressive jaundice.
A liver ultrasonography showed intrahepatic bile duct dilatation and
distended gallbladder. The patient also experienced a 6-kg weight

D uctal adenocarcinoma of pancreas represents the fourth

and fifth most common cancer in men and women,
respectively, with the lowest 5-year survival rate of any
cancer. Surgical resection associated with adjuvant therapy is
the only chance for cure. Many prognostic factors have been
extensively investigated in the attempt to establish which
Received for publication May 14, 2006; accepted November 3, 2006.
From the *Department of Surgery, Liver and Multivisceral Transplant Center,
University of Modena and Reggio Emilia, Policlinico di Modena, Modena,
Italy; and †Department of Surgery, Liver and Multiorgan Transplant Centre,
University of Bologna, Policlinico S. Orsola, Bologna, Italy.
Reprints: Fabrizio Di Benedetto MD, PhD, Centro Trapianti di Fegato e
Multiviscerale, Policlinico di Modena, Universita’ di Modena e Reggio
Emilia, Via del Pozzo 71. Modena 41100, Italy (e-mail: f.diben@virgilio.it).
Copyright * 2007 by Lippincott Williams & Wilkins
loss in the 4 months before jaundice presentation. At admission, the
biochemistry panel showed as follows: glutamic oxalacetic transa-
minase, 59 U/L; glutamic pyruvic transaminase, 69 U/L; alkaline
phosphatase, 456 U/L; and total/direct bilirubin level, 6.9/6.5 g/dL.
Patient 2 is a 51-year-old man presenting to the emergency
department with a 3-day history of right upper quadrant pain, nausea,
vomiting, and jaundice. The patient referred that a 5-kg weight loss
occurred within a few months before symptom presentation. At
admission, the biochemistry panel showed as follows: glutamic
oxalacetic transaminase, 68 U/L; glutamic pyruvic transaminase,
79 U/L; alkaline phosphatase, 532 U/L; and total/direct bilirubin
level, 5.9/5.6 g/dL.
Both patients presented in excellent general condition, well
nourished despite the weight loss and extremely motivated to an
aggressive surgical approach to the malignancy. No other comor-
bidity factors contraindicated the operation. At preoperative work up
a CT scan showed in both patients a mass at the level of pancreatic

266 Pancreas & Volume 34, Number 2, March 2007

Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pancreas & Volume 34, Number 2, March 2007
head with signs of duodenal and transverse colon infiltration.
Reconstruction of the CT scan demonstrated in both patient the
involvement of the Superior Mesenteric Vessels by the tumor. Both
patient underwent ERCP, which confirmed the obstructive jaundice
and allowed to decompress the biliary tree by stent placement.
Surgery occurred as soon the bilirubin level normalized.
Operative technique
In both cases, after excluding liver and peritoneal metastases,
an attempt to perform a standard resection of the tumor was done. In
the first case, at laparotomy, the tumor seemed to infiltrate the
superior mesenteric artery (SMA) and superior mesenteric vein
(SMV), portal vein (PV), main bile duct, duodenum, transverse
colon, and gastroduodenal artery near its origin from the hepatic
artery. Because of the massive mesenteric root invasion, we decided
to perform the dissection of tumor ex vivo to facilitate the maneuver,
minimize blood loss, and small-bowel ischemia-reperfusion injury
deriving from prolonged clamping of mesenteric pedicle. The
multivisceral block, composed of the distal two thirds of stomach,
duodenum, jejunum, ileum, transverse colon, pancreas, and spleen,
was removed and arranged to the back table. The SMA and the PV
were sectioned at the origin of the aorta and just below the
confluence of left gastric vein, respectively (Fig. 1). The graft was
perfused with 2 L of cold Celsior preservation solution through the
SMA before starting any dissection. At the back table, the stomach,
duodenum, pancreas, spleen, and ascending and transverse colons
were resected together with the neoplastic mass including mesenteric
vessels, carefully trying to save the ileocecal valve and its relative
vascularization. The SMV was sectioned before the confluence of the
first jejunal vein with the ileocolic vein as far as tumor-free tissue was
obtained. The SMA was also resected as far as the origin of ileocolic
artery. Venous outflow was ensured anastomosing at the back table a
preserved iliac venous allograft obtained from an isogroup cadaveric
donor; the internal iliac vein was anastomosed to the first jejunal vein,
and the external iliac vein was anastomosed to the ileocolic vein. The
graft was therefore brought back to the operating table, and the
common iliac vein was anastomosed end-to-end to the remnant PV
(Fig. 2). The ileocolic artery was sutured end-to-end to the aortic
stump of SMA. The graft was reperfused after 1 h 54 min of cold
ischemia and soon appeared to be pink, moist, with a good peripheral
pulse, and without signs of edema. Peristalsis started a few minutes
after declamping. The biliary drainage was achieved by an
FIGURE 1. Operative field after resection of the multivisceral
block. The PV and the left gastric vein are clamped. * indicates
SMA; 9, HA.
* 2007 Lippincott Williams & Wilkins
Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Intestinal Autotransplantation for Pancreatic Adenocarcinoma
FIGURE 2. The outflow reconstruction was obtained
anastomosing at the back table an isogroup cadaveric iliac
graft distally to the FJB and to the SMV. The common iliac vein
was proximally anastomosed to the MPV. FJB indicates first
jejunal vein; MPV, main portal vein.
hepaticojunostomy, and the intestinal continuity was achieved by a
gastrojejunostomy and a side-to-side anisoperistaltic cecocolostomy.
The length of residual intestine from the gastric stump to the ileocecal
valve was 130 cm.
In the second patient, the extent and relationship of the tumor
with the mesenteric root allowed to spare distally the SMA before its
branching and the PV at its base so that the inflow and outflow were
ensured by an end-to-end anastomosis between the stumps of SMA
and SMV, respectively. Therefore, the cold ischemia time was
limited to 55 min, with prompt recovering of intestinal perfusion and
motility after reperfusion. The intestinal reconstruction consisted of
an hepaticojejunostomy, a gastrojejunostomy, and a cecotransver-
susanastomosis. A Doppler of vasa recta of small bowel confirmed
optimal perfusion excluding outflow abnormality. The length of
residual intestine from the gastric stump to the ileocecal valve was
140 cm. Neither patient underwent intraoperative radiotherapy.
The duration of procedure was 11.35 and 9.25 h. Neither
patient required preoperative blood transfusion. Both were admitted
in the intensive care unit for 1 day. Continuous insulin infusion with
strict glucose level control was introduced soon after operation to
prevent severe hypoglycemia. The postoperative course was
uneventful for the first patient treated. The second one underwent
relaparotomy for hemoperitoneum in postoperative day (POD) 5. In
the early postoperative period, both patients were supported by total
parenteral nutrition. Progressive enteral nutrition was started
gradually after PODs 3 and 9. The first patient was discharged in
POD 16, whereas the second patient was discharged in POD 29.
Because of interruption of intestinal innervation, both patients
experienced increased peristalsis taken under control with low doses
of loperamide. Histological examination showed moderately dif-
ferentiated ductal adenocarcinoma in the first case, whereas a well-
differentiated adenocarcinoma was diagnosed in the second case.
The neural retropancreatic tissue was infiltrated by neoplastic tissue
with free surgical margins in both cases. Gastric and transverse colic
walls were confirmed to be infiltrated by the tumor in the first case
but not in the second one. Four nodes of 10 analyzed were metastatic
in 1 patient, whereas all the 11 examined nodes in the second patient
were reported as reactive. Both patients were referred to the
Oncology Service for adjuvant therapy; a gemcitabine-5-fluorouracil
regimen was started 1 month after surgery and was well tolerated
267
Quintini et al
in both cases. No major hypoglycemic crisis was reported, and a
good glucose level control was achieved in both patients. Periodic
close follow-up visits with serial CT scans, ultrasonography, and
laboratory data showed no evidence of tumor recurrence until after
16 and 9 months. One patient experienced PV thrombosis 15 months
after operation, with signs of peritoneal carcinosis. He died 19
months after operation. The other patient returned to work 40 days
after operation and experienced optimal clinical conditions until
massive liver metastatic involvement, peritoneal carcinosis, and
death occurred 10 months after the operation.
DISCUSSION
It is widely accepted that surgical excision of tumor
tissue associated with adjuvant therapy represents the only
chance of cure for patients with ductal adenocarcinoma of
pancreas. The analysis of major series clearly establishes that
the only prognostic factor predicting long-term survival rate
is the invasion of resected margins by the tumor. There is no
accordance on whether infiltration of PV/SMV should be
considered a negative prognostic factor and then a contra-
indication to curative resection.1,2 Whether some authors
believe that vascular invasion should be considered as a
synonym of remote spread of the tumor, others support the
concept proposed by Fuhrman et al1 that venous involvement
is a reflection of tumor size and location rather than an
indicator of aggressive tumor biology. Moreover, others have
reported high rates of overestimation in intraoperative vein
invasion assessment, demonstrating that in most specimens,
the tumor abutted the PV/SMV without infiltrating the venous
wall.6 As proposed by Van Geenen et al,2 the macroscopic
appearance of tumor infiltration can be explained by the
desmoplastic stromal reaction of the tumor, which cannot
macroscopically be distinguished from tumor infiltration.
According to these concepts, any technical advance able to
increase the curative resection rate can potentially reflect an
increase in survival rate, especially in those patients in whom
curative resection is denied because of the involvement of the
mesenteric root. Many reports in literature describe the ex
vivo resection of the tumor in heart, liver, and kidney
surgeries. According to this technique, the organ is removed
from the surgical field with its vascular pedicle, perfused on
the back table with cold preservation solution, and then
subjected to parenchymal dissection of the tumor in a
bloodless safe condition. Only 2 reports in literature describe
the extension of this procedure to gastrointestinal field.
Lai et al4 first described a total pancreatectomy with
small-bowel autotransplantation for a locally advanced islet-
cell carcinoma. Tzakis et al3 described the remnant 4 patients
reported in literature that underwent a partial abdominal
exenteratio; the indications derived in 1 case from a vascular
malformation and in the remnant cases from the mesenteric
root encasement by a pancreas head fibroma, a pancreas body
desmoid tumor, and a pancreatic cancer. The latter represents,
to our knowledge, the only case in literature of a patient who
has undergone intestinal autotransplantation for exocrine
268
Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pancreas & Volume 34, Number 2, March 2007
carcinoma of the pancreas. In our report, we describe 2 cases
of ductal adenocarcinoma of pancreas treated with such a
technique. The decision to undertake this option derived from
the major invasion of mesenteric root by the tumor, from the
absence of liver and peritoneal metastasis, the excellent
clinical status, and the young age of the patients at the time of
diagnosis, and, finally, from a well-motivated intention of the
patients to undergo a high-risk critical but potentially curative
procedure, although remote. As described by Tzakis et al,3 the
ex sito perfusion and dissection can help provide an adequate
tumor-free margin in 2 crucial areas such as the mesenteric
pedicle, where isolation is conducted as far as safe tissue is
encountered, and the retroperitoneal aspect of pancreatic bed,
which is frequently invaded by tumoral tissue even in case of
small tumors. From a technical point of view, due to its role in
the transit time regulation, to prevent short-bowel syndrome
and its clinical-functional consequences, the ileocecal valve
should be preserved, and although the minimum length of
intestine necessary to maintain nutritional status has not been
established, a 100-cm-long part of the small intestine is
thought to be enough to maintain sufficient function. In our
2 cases, a good quality of life was ensured for a reasonable
time before recurrence, as deducted comparing our follow-up
period with median survival time reported in literature for
stage IV pancreatic cancer.5 Furthermore, the analysis of 7
cases performed so far shows that no perioperative mortality
occurred, ensuring that in highly selected patients and centers
with intestinal transplant experience, this procedure can be
performed with acceptable rates of mortality and morbidity.
Because of the paucity of intestinal autotransplantation
performed and to the biologic aggressiveness of this tumor, it
is very difficult, as in many pancreatic resections, to settle the
border between palliation and cure. Despite this, we believe
that a young individual in excellent clinical status with a local
invasion of the mesenteric pedicle (in the absence of
peritoneal/liver metastasis and aware of the risk that such a
procedure implies) still has the right to pursue a cure even if
the chance is infinitesimal.
REFERENCES
1. Fuhrman GM, Leach SD, Staley CA, et al. Rationale for en bloc vein
resection in the treatment of pancreatic adenocarcinoma adherent to the
superior mesenteric-portal vein confluence. Ann Surg. 1996:154Y162.
2. Van Geenen RC, ten Kate FJ, de Wit LT, et al. Segmental resection and
wedge excision of the portal or superior mesenteric vein during
pancreatoduodenectomy. Surg. 2001;129:158Y163.
3. Tzakis AG, Tryphonopoulos P, De Faria W, et al. Partial abdominal
evisceration, ex vivo resection, and intestinal autotransplantation for the
treatment of pathologic lesions of the root of the mesentery. J Am Coll
Surg. 2003;197:770Y776.
4. Lai DT, Chu KM, Thompson JF, et al. Islet cell carcinoma treated by
induction regional chemotherapy and radical total pancreatectomy with
liver revascularization and small bowel autotransplantation. Surg.
1996;119:112Y114.
5. Kelsen DP, Portenoy R, Thaler H, et al. Pain as predictor of outcome in
patients with operable pancreatic carcinoma. Surg. 1997;122:53Y59.
6. Nakagohri T, Kinoshita T, Konishi M, et al. Survival benefits of portal
vein resection for pancreatic cancer. Am J Surg. 2003;186:149Y153.
* 2007 Lippincott Williams & Wilkins