Gynecologic Oncology 121 (2011) 258–263

Contents lists available at ScienceDirect

Gynecologic Oncology
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y g y n o

Celiac lymph node resection and porta hepatis disease resection in advanced or
recurrent epithelial ovarian, fallopian tube, and primary peritoneal cancer
A. Martinez a,⁎, C. Pomel b, E. Mery c, D. Querleu a, L. Gladieff d, G. Ferron a
a
Claudius Regaud Comprehensive Cancer Center, Department of Surgical Oncology, Toulouse, France
b
Jean Perrin Comprehensive Cancer Center, Department of Surgical Oncology, Clermont-Ferrand, France
c
Claudius Regaud Comprehensive Cancer Center, Department of Pathology, Toulouse, France
d
Claudius Regaud Comprehensive Cancer Center, Department of Medical Oncology, Toulouse, France

a r t i c l e i n f o a b s t r a c t

Article history: Introduction. Prognostic value of complete macroscopic resection of primary disease has been reported and
Received 18 August 2010 confirmed in several publications. Published data indicate that extensive upper abdominal disease involving
Available online 3 February 2011 the hepatic pedicle and celiac trunk is associated with an abortion of the surgical procedure or with suboptimal
residual disease.
Keywords: Methods. All patients who had disease at the porta hepatis or celiac lymph node resection as part of
Ovarian cancer
cytoreductive surgery were included. Medical and operative records with particular emphasis on extent and
Celiac lymph nodes
Cytoreductive surgery
distribution of disease spread, number of peritonectomy procedures, visceral resections, and lymphadenectomy
Porta hepatis procedures were examined.
Results. A total of 28 patients who underwent some kind of celiac lymph node resection or resection of
metastatic involvement of the porta hepatis were included. Median preoperative serum Ca-125 level was
78 U/ml (range, 30–2950 U/ml), and median ascites volume was 1900 ml (range, 0–10,000 ml). Of the 28
patients, 23 underwent supra-radical surgery for diffuse peritoneal carcinomatosis. Median operative time
was 252 minutes (range, 100–540 minutes). Complete cytoreduction to CCO was achieved in all except one
case, who was cytoreduced to millimetric residue. Fifteen patients had positive celiac nodes and nineteen
patients had peritoneal disease in the porta hepatis region.
Discussion. Resection of enlarged nodes and metastatic disease to the porta hepatis is feasible with an
acceptable morbidity. The decision to undergo an aggressive cytoreductive surgery is based on appropriate
patient selection depending on the extension of surgical procedure, on medical comorbidities, and on the
potential to tolerate an extensive procedure, rather than on specific anatomic locations.
© 2010 Elsevier Inc. All rights reserved.

Introduction
The surgical approach to advanced ovarian cancer involving the
upper abdomen has changed in the last decades, and upper abdominal
surgical procedures have been incorporated to achieve technically
complete yet safe cytoreduction. Treating upper abdominal disease
increases the rate of optimal cytoreduction from 50% to 76% [1]. The
addition of these procedures is both feasible and associated with
acceptable morbidity in selected patients. However, celiac lymph
node (CLN) metastases, usually due to retroperitoneal spread from
upper abdominal disease, constitute a major obstacle to achieve
complete cytoreduction. The potential benefit of pelvic and paraaortic
lymphadenectomy [2,3] may reconsider the role of lymphadenectomy
in other abdominal areas, especially when bulky suspect nodes are
encountered. Lymphadenectomies of the porta hepatis and of the
⁎ Corresponding author. Claudius Regaud Comprehensive Cancer Center, 20-24, Rue
Pont-Saint-Pierre, 31052 Toulouse, France. Fax: +33 561424117.
E-mail address: martinez.alejandra@claudiusregaud.fr (A. Martinez).
celiac trunk are performed by surgical oncologists for staging some
digestive cancers and germ cell tumors.
In an attempt to improve optimal cytoreduction rates, we
expanded the surgical strategy to resection of metastatic involvement
of the porta hepatis and suspect CLNs in patients with epithelial
ovarian, fallopian, and primary peritoneal cancer.
Material and methods
Anatomy
The celiac artery is a short, thick trunk about 1.25 cm in length,
arising from the front of the aorta, just below the aortic hiatus of the
diaphragm. At the upper border of the pancreas, it divides into three
large branches: the left gastric, the splenic, and the common hepatic
arteries. The hepatic artery is directed to the upper margin of the
superior part of the duodenum, forming the lower boundary of the
epiploic foramen. It crosses the portal vein anteriorly and ascends
between the layers of the lesser omentum, where it gives rise to the

0090-8258/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.ygyno.2010.12.328
 A. Martinez et al. / Gynecologic Oncology 121 (2011) 258–263
right gastric artery above the pyloric end of the stomach, and the
gastroduodenal artery. The gastroduodenal artery divides at the lower
border of the duodenum into the superior pancreaticoduodenal artery
and the right gastroepiploic artery. The superior pancreaticoduodenal
artery descends between the duodenum and pancreas. It anastomoses
with the inferior pancreaticoduodenal artery, a branch of the superior
mesenteric artery, and with pancreatic branches of the splenic artery.
Celiac and hepatic artery anatomic variations are encountered
in approximately 35% of patients. Bifurcation of the celiac trunk is
observed in 12% of cases [4], and an inferior phrenic artery originating
from the celiac trunk is reported in around 8% of patients [4]. Variant
left and right hepatic arteries arising occur in 11% and 5%, respectively.
A common hepatosplenic trunk with a left gastric artery arising
directly from the aorta is identified in approximately 5.5% of patients,
and a common hepatic artery arising from the superior mesenteric
artery is present in approximately 3.5% of patients [5].
The celiac nodes consist on two or three nodes located at the celiac
trunk, which drain the supramesocolic part of the peritoneal cavity.
The hepatic nodes form a chain of three to six nodes situated along the
course of the hepatic artery. The inferior nodes are located at the
superior border of the pancreas and continue to follow the anterior
surface of the portal vein to finish on the hepatic hilum, surrounding
the left and right division of the hepatic artery.
Procedure
Exposure
The abdominal approach to the upper abdomen in the setting of
cytoreductive surgery for ovarian cancer entails a long midline
incision from the pubis to the symphysis. Sometimes, the incision
can be extended to the sternum to the side of the xiphoid process and,
less frequently, resection of the xiphoid is included. A versatile self-
retaining retractor is used for generous abdominal exposure and to
free the assistant's hands. Before dissection is started, the extent of
disease spread and of surgical procedures required must be evaluated
to ensure complete cytoreduction with acceptable morbidity.
Hepatic pedicle, splenic vein, and celiac trunk
Celiac lymphadenectomy comprises all fatty tissue and lymph
nodes between the hepatic artery and the root of the left gastric
artery. In the present series, resection of CLNs including hepatic, celiac
trunk, splenic and/or gastric lymph node (LN) groups was limited only
to suspect ones.
LNs along the hepatic artery, around the celiac trunk, and along the
splenic artery were sequentially separated and dissected when
suspect. The stomach was retracted cephalad, the bilateral peritone-
um of the gastropancreatic peritoneal fold was, and the pancreas were
gently retracted downwards. While tracting the stomach upwards,
the proper hepatic artery was identified and exposed from the gas-
troduodenal artery to the hepatic hilum. The dissection was started by
separating the lymphofatty tissue around the common bile duct and
proper hepatic artery. The hepatic artery was retracted to the right
side of the abdomen to dissect the nodal tissue lying behind it and the
nodal tissue surrounding the surface of the portal vein. The hepatic
pedicle was rotated on itself when posterior LNs wanted to be
resected. Approach to the LN on the right side of the hepatic pedicle
and LNs lying on the anterior region of the hepatoduodenal ligament
often required a cholecystectomy to improve exposure and facilitate
dissection. The dissection was continued by removing the lympho-
fatty tissue surrounding the common hepatic artery and the splenic
vein, along the superior border of the pancreas, and around the
anterior surface of the celiac trunk and the beginning of the left gastric
artery (Fig. 1). Finally, when suspect suprarenal LNs were encoun-
tered, the dissection proceeded in a caudal direction to remove nodal
tissue with adjacent peritoneum on the anterior surface of the aorta to
the level of the superior mesenteric artery. A Kocher's procedure was
259
performed until the inferior vena cava, left renal vein, and abdominal
aorta were visualized. When metastatic deposits overlying the hepatic
pedicle were found with or without palpation of suspect lymph nodes,
resection of porta hepatis peritoneal disease was performed.
A computer-generated search of the institutional patient database
was performed to retrospectively identify all patients who had disease
at the porta hepatis or celiac lymph node resection as part of
cytoreductive surgery for epithelial ovarian, fallopian, or primary
peritoneal cancer between January 2009 and June 2010 at Claudius
Regaud Comprehensive Cancer Center (Toulouse) and at Jean Perrin
Comprehensive Cancer Center (Clermont Ferrand), France.
Preoperative image study evaluation included in all cases a CT of
the chest, abdomen, and pelvis, and in selected cases, a positron
emission tomography.
All surgical procedures were performed by the same senior
surgical oncologists (GF, CP). Extent and disease distribution
throughout 13 abdominopelvic regions and surgical outcome were
evaluated by the peritoneal cancer index (PCI) and completeness of
cytoreduction score (CC) [6]. The PCI is a clinical integration of
peritoneal implant size and lesion distribution within the abdominal
cavity. Tumor volume was quantified in 9 abdominopelvic regions
defined by two horizontal lines (through the lowest aspect of the rib
cage and through the iliac crest) and two vertical lines (equally
dividing the abdomen into three sections),and on four small bowel
areas (upper jejunum, lower jejunum, upper ileum, and lower ileum).
The size of the largest lesion in each of the 13 abdominopelvic regions
was scored from 0 to 3 (0, no tumor; 1, nodules smaller than 0.5 mm;
2, tumor size from 0.5 mm to 5 cm; 3, tumor larger than 5 cm or
confluent disease) and added together to give a PCI between 0 and 39
(3 × 13). Residual disease after surgery was classified according to CC:
CC-0, no residual disease; CC-1, minimal residual disease of 0–
2.5 mm; CC-2, residual disease of 2.5 mm–2.5 cm; CC-3, residual
disease of more than 2.5 cm. Surgical technique for cytoreductive
surgery including peritonectomy procedures and visceral resections
was performed as described by Sugarbaker [7]. Supra-radical
cytoreductive surgery was defined as a debulking operation requiring
extensive peritonectomies associated to at least two multivisceral
resections including extensive peritonectomy with at least two
visceral resections: combined large and small bowel resection,
splenectomy with or without distal pancreatectomy, liver resection,
diaphragm resection, and/or pelvic exenteration.
Indication for hepatic and/or celiac lymph node resection was
based on intraoperative findings of suspect LNs, which included those
measuring more than 1 cm and/or indurated at palpation (Fig. 2).
The study was approved by the institutional review board from our
institutions. Medical records were examined and patient demograph-
ic data with particular emphasis on operative records to detail extent
and distribution of disease spread, number of peritonectomy
procedures, visceral resections, and lymphadenectomy procedures
were included. Pathological data were also analyzed. Adjuvant
chemotherapy was administered within 2 months of surgery, when
feasible, at the discretion of the treating oncologist. Postoperative
complications and death were defined as any adverse events related
to surgery occurring within 30 days of cytoreduction and were graded
according to the NCI scale [8]. For patients who had more than one
complication, the highest grade complication was used in the analysis.
Results
A total of 28 patients who underwent some kind of celiac lymph
node resection (CLNR) or resection of metastatic involvement of the
porta hepatis region for epithelial ovarian, fallopian tube, or primary
peritoneal cancer were identified. Median age of the patients included
in the study group was 58.5 years (range, 22–72 years). Median age in
the primary cytoreduction group was 58 years (range, 22–72 years)
and in the secondary cytoreduction group was 61 years (range, 47–
260 A. Martinez et al. / Gynecologic Oncology 121 (2011) 258–263

Fig. 1. Celiac lymphadenectomy. The left triangular hepatic ligament has been divided to retract the left lobe of the liver to the right. The lesser omentum has been resected.
Lymphofatty tissue around the common bile duct, proper hepatic artery, portal vein, and splenic vein, lying along the superior border of the pancreas, has been removed. LLL
indicates left lobe of the liver; SV, splenic vein; Spa, superior aspect of the pancreas; LGC, left gastric curvature.

72 years). Patient characteristics, stage, and type of tumor are
depicted on Table 1. CLNR occurred in the context of primary or late
recurrent disease, and only one patient had one prior recurrence.
Median time to recurrence was 9 months (range, 8–60 months).
Eleven patients underwent cytoreduction followed by adjuvant
platinum-based chemotherapy, and the remaining seventeen patients
received neoadjuvant chemotherapy before debulking surgery. From
the 8 patients who had received 6 cycles of chemotherapy prior to
cytoreductive surgery, 3 received 3 additional cycles and another one
received a single cycle. Patients who had received 3 cycles prior to
surgery completed the 6 cycles of taxol carboplatin after cytoreduc-
tion. Although neoadjuvant chemotherapy decreased tumor burden
in some patients, lymphadenectomy after prior chemotherapy was
more challenging as sclerose made dissection more challenging.
The median preoperative serum Ca-125 level was 78 U/ml (range,
30–2950 U/ml), and the median ascites volume was 1900 ml (range,
0–10,000 ml). Of the 28 patients, 23 included in the study underwent
supra-radical surgery for diffuse peritoneal carcinomatosis. The 5
remaining cases were an isolated splenic recurrence in one case, a
hepatodiaphragmatic recurrence in the second one, and an isolate
lymph node recurrence to the aortic and celiac lymph nodes in the
third one. The fourth patient had a previous incomplete surgery and
came to our institution after the completion of chemotherapy. The
fifth patient underwent interval surgery after prior chemotherapy for
stage IV ovarian cancer due to a mediastinal adenopathy and did not
undergo aortic lymphadenectomy in the context of a randomized trial
(see Table 2). Median peroperative volume loss (estimated blood
loss and ascites) was 1700 ml (range, 97–4697 ml), and median
operative time was 252 minutes (range, 100–540 minutes). Complete
cytoreduction to CCO was achieved in all except one case, who was
cytoreduced to millimetric residue (CC-1). Details of associated
surgical procedures are shown in Table 2.

Table 1
Patient characteristics undergoing HCLN.

N %

Operation
Primary 20 71.5
Recurrent 8 28.5
Neoadjuvant chemotherapy 17 60.7
Stage
III 22 78.5
IV 6 21.5
Histologic subtype
Serous ovarian cancer 19 68
Clear cell ovarian cancer 1 3.5
Endometrioid ovarian cancer 1 3.5
Fig. 2. Bulky nodes at the hepatic pedicle. The round ligament of the liver has been
Mixed serous/mucinous adenocarcinoma and carcinosarcoma 1 3.5
tracted cranially to expose the porta hepatis. BB indicates biliary bladder; HP, hepatic
Primary fallopian tube cancer 2 7
pedicle; BB, bulky nodes; RL, round ligament; LLL, left lobe of the liver; LO, lesser
Primary peritoneum cancer 4 14.5
omentum; LGC, lesser gastric curvature.
 A. Martinez et al. / Gynecologic Oncology 121 (2011) 258–263 261

Table 2

P Age PCI NAR NPP NVR Associated surgical procedures N positive

LN/total

1 47 34 13 6 5 Glissonectomy + 20 × 10 cm full-thickness right diaphragm resection + CH + SCP + partial gastrectomy + OM + RLS + triple 0/1
bowel resection (rectosigmoid, ileocecal, and resection of the duodenojejunal angle) + AL
2 72 28 13 6 2 SCP + OM + RLS + left hemicolectomy including rectosigmoid resection + lateral and umbilical port-site metastasis resection PHD
3 65 17 11 6 2 CH + SCP + OM + RLS + appendicectomy + posterior pelvic exenteration + PPLND 5/5
4 59 22 10 6 2 Full-thickness right diaphragm resection + OM + RLS + double bowel resection (ileocecal and resection of the duodenojejunal 14/15
angle) + superior colpectomy + AL
5 57 22 13 6 4 8 × 2 cm full-thickness right diaphragm resection + CH + SP + OM + RLS + triple bowel resection (rectosigmoid, right 6/7
hemicolectomy, and jejunal resection)
6 52 21 13 6 1 Transdiaphragmatic resection of mediastinal lymph node + glissonectomy + CH + SCP + OM + RLS + pelvic mass resection 3/3
7 39 7 5 4 1 5 × 2 cm full-thickness right diaphragm resection + OM + appendicectomy + PPLND + TAH + BSO 0/5
8 71 13 9 4 1 CH + SCP + RLS + lateral port-site metastasis resection 6/7
9 52 21 10 6 4 SP + OM + RLS + double bowel resection (ileal and transverse and left hemicolectomy with a posterior exenteration) + 1/6
PPLND
10 72 0 0 0 2 Appendicectomy + double small bowel resection (secondary to close adhesions) + AL 14/14
11 65 0 0 0 1 SP + OM + appendicectomy + AL 0/3
12 66 9 3 3 2 10 × 5 cm full-thickness right diaphragm + wedge hepatic resection + CH + OM + rectosigmoid resection + AL 0/1
13 22 31 12 6 3 Glissonectomy + CH + SCP + OM + RLS + double bowel resection (ileocecal and rectosigmoid) + PPLND PHD
14 63 33 12 6 3 8 × 3 cm full-thickness right diaphragm + glissonectomy + CH + SP + OM + RLS + posterior exenteration with double bowel 0/3
resection (ileocecal and transverse and left hemicolectomy) + PPLND
15 72 19 10 5 2 Glissonectomy + CH + SP + OM + posterior pelvic exenteration + PPLND 2/4
16 53 3 1 1 0 OM + PPLND 0/3
17 61 17 9 5 2 OM + RLS + ileocecal resection + coagulation of superficial mesenteric implants + TAH/BSO + PPLND 0/2
18 70 15 8 6 2 SCP + OM + RLS + HT + PPLND 0/2
19 53 22 9 6 3 CH + SP + OM + RLS + left hemicolectomy + appendicectomy + HT/BSO + PPLND 0/2
20 57 24 11 6 3 SCP + OM + RLS + double bowel resection (ileocecal and posterior exenteration) + PPLND 1/1
21 39 31 12 6 2 CH + SCP + OM + RLS + total colectomy (ileorectal anastomose) + coagulation of mesenteric superficial implants + PPLND 3/8
22 76 23 13 6 2 Glissonectomy + CH + SCP + OM + RLS + total colectomy (ileorectal anastomose) + coagulation of mesenteric superficial 1/2
implants + PPLND
23 53 20 11 4 1 OM + TAH/BSO 1/1
24 60 14 6 4 2 Full-thickness right diaphragmatic resection + SP + OM + resection of a parietal metastasis to the abdominal wall + posterior 3/3
exenteration + PPLND
25 58 11 6 2 1 Full-thickness right diaphragmatic resection + OM + TAH/BSO 0/1
26 58 11 4 2 1 Full-thickness right diaphragmatic resection + CH + OM + RLS + posterior exenteration + douglassectomy + PPLND 0/6
27 67 2 1 1 1 OM + TAH/BSO 1/1
28 53 3 1 1 1 Full-thickness right diaphragmatic resection + wedge hepatic resection + glissonectomy 1/1

P indicates patients; NAR, number of involved anatomic regions; NPP, number of peritonectomy procedures; NVR, number of visceral resections; CH, cholecystectomy; RLS, resection
of the lesser sac; SCP, splenectomy and caudal pancreatectomy; OM, infragastric omentectomy; AL, paraaortic lymphadenectomy; PPLND, pelvic and paraaortic lymph node
dissection; TAH, total abdominal hysterectomy; BSO, bilateral salpingo-oophorectomy; LN, lymph node; PHD, porta hepatis disease with no LN identified in pathology exam.

In this series, there was one postoperative death due to septic
complications at the fourth postoperative day. Major morbidity
(grade 3–5) occurred in 10 (35.7%) patients. Four patients underwent
reoperation due to bowel complications: an ileostomy was performed
in a patient with an ileocecal perforation, a protective colostomy was
performed in two patients with rectal anastomotic leaks, and a
transverse colectomy was required in a patient with a necrosis of the
transverse colon. Two patients were reoperated for abdominal
hemorrhage, one patient had an intraabdominal abscess requiring
radiologic drainage, one patient developed pneumonia requiring
intubation, one patient had a congestive heart failure which
responded to medical treatment, and another patient was dead of
septic complications. Peroperative complications directly related to
CLN resection or resection of disease at the porta hepatis were
identified in one patient. A lateral common bile duct injury occurred
during resection of tumor at the porta hepatis, and primary suture
with drainage followed by intraoperative cholangiography to ensure
correct drain placement was performed. The drain was removed
4 months after surgery without any complications.
Pathologic evaluation of the resected hepatic and celiac lymph
nodes confirmed LN resection in 26 patients. The 2 remaining patients
presented with bulky disease at the porta hepatis, but no LN was
identified on pathology exam. A total of 107 celiac LNs were removed,
with a median of 3 LNs per patient (range, 1–14). Metastatic
involvement of CLNs was identified in 15 patients; 7 of the 13
remaining patients without celiac lymph node disease had peritoneal
metastatic involvement of the porta hepatis region. Almost all
patients with metastatic celiac LNs presented with some form of
upper abdominal disease, 12 had diffuse peritoneal carcinomatosis
with a median PCI of 21 (range, 17–31), 1 patient had a mediastinal
positive node and underwent surgery after neoadjuvant chemother-
apy, another had an isolated paraaortic and celiac LN recurrence, and
the last one had a hepatodiaphragmatic recurrence. Pathology exam
confirmed metastatic involvement of the porta hepatic peritoneum in
19 patients. Disease involving the small bowel and/or the right and
transverse colon was confirmed in 15.4% of patients without
metastatic disease to the celiac LN and in 40% of patients with
positive celiac LNs. Excluding 5 patients whose paraaortic LN status
data were missing, paraaortic LN metastases were present in 41.6% of
patients with negative celiac LNs and in 81.1% of patients with celiac
LN involvement.
Discussion
Ovarian cancer dissemination pattern most commonly occurs
through the intraperitoneal route, followed by lymphatic invasion.
Incidence of retroperitoneal LN metastases in advanced ovarian
cancer ranges from 62% to 75% in reported studies, being paraaortic
involvement the most frequent [9–11]. However, frequency of
metastatic involvement of the porta hepatis region and CLN
involvement is not well detailed on the medical literature.
Prognostic value of complete macroscopic resection of primary
disease has been reported and confirmed in several publications [12–14].
The largest meta-analysis performed by Bristow et al. [12], which
included 81 studies accounting for 6885 patients with stage III–IV
ovarian cancer, demonstrated a statistically significant positive correla-
tion between percent of maximal cytoreduction and median survival
time. Patient selection for secondary cytoreduction remains
262 A. Martinez et al. / Gynecologic Oncology 121 (2011) 258–263
controversial. Although several authors report a survival benefit after
surgical cytoreduction [15–17], the heterogeneity of patients included in
the published studies precludes to clearly define surgical candidates.
A survey of SGO (Society of Gynecologic Oncologists) members
showed that tumor involving the diaphragm, bowel, mesentery, and
portal triad precluded optimal cytoreduction, being technically
unresectable metastatic disease the most important reason why
optimal cytoreduction was not achieved [18]. Consequently, extensive
upper abdominal disease involving the hepatic pedicle and celiac
trunk are associated with an abortion of the surgical procedure or
with suboptimal residual disease. Is this specific anatomic marker of
unresectable disease an adequate predictor of suboptimal surgical
outcome? Variable surgical effort among different gynecologic
oncologists makes it difficult to generalize and define criteria for
unresectability based on anatomical sites.
Patients with positive nodes at the celiac axis or with involvement of
the porta hepatis presented with extensive upper abdominal tumor
burden in our series, requiring ultra-radical procedures. Disease to these
specific locations was completely resected in all cases. Memorial Sloan–
Kettering Cancer Center (MSKCC) series reported resection of tumor on
the porta hepatis resection in 5% of patients with advanced ovarian
cancer operated from 2001 to 2004 [19]. Other series have reported on
optimal resection of the porta hepatis disease in 9 of the 12 patients with
metastatic involvement in this region [20]. Extensive surgical approach
to the upper abdomen improves optimal cytoreduction rate but
associates to increased blood loss and higher peroperative complica-
tions [19]. Possible complications related to the addition of resection of
the porta hepatis tumor and morbidity directly related to celiac LN
resection mainly include vascular and bile duct injury. The single
complication in this series was a bile duct injury repaired during the
procedure. The most frequent problem encountered during dissection
was anatomic variations of the celiac axis. Extreme care not to injure the
left gastric vessels is necessary, especially when a previous infragastric
omentectomy and splenectomy with transection of the right and left
gastroepiploic vessels and splenic vessels have been performed as it
remains the only blood supply to the stomach. An abnormal left hepatic
artery arising from the celiac trunk or from the left gastric artery can also
be found at the pars flaccida of the lesser omentum (Fig. 3). Left hepatic
artery must also be preserved as its injury can lead to left hepatic
necrosis.
Based on the rationale that cytoreduction to no macroscopic tumor
nodule is critical to the efficacy of chemotherapy and to long-term
survival, grossly abnormal enlarged nodes should be removed when
discovered during the procedure. Recent publications demonstrate
beneficial therapeutic effect of pelvic and paraaortic lymphadenect-
omy in epithelial ovarian cancer, regardless of the stage of the disease
[3]. This therapeutic effect is based on the negative impact of
understaging and on insufficient extensive surgery when not
performed. The combination of both residual disease and clinical LN
status shows a significant impact of lymphadenectomy in patients
with no residual disease and clinically suspect nodes [2]. Resection of
suspect LNs at the celiac axis participates in debulking by removal of
macroscopic tumor based on previous reports indicating that more
than 90% of patients with clinically suspect LNs had histological
positive LNs [2]. Other series demonstrates low sensitivity and
specificity of nodal size as an indicator of ovarian cancer metastasis,
even when done by experienced gynecologist oncologists [21,22]. In
this series, 57.7% of patients with at least one celiac LN identified at
microscopic exam had metastatic disease. When considering patients
with diffuse upper abdominal disease and/or more than 4 metastatic
aortic LNs, positivity of celiac LNs increased to 75%. Bulky metastatic
disease at the porta hepatis peritoneum can also be confused with
node disease at the hepatic pedicle, as identified in 2 patients of this
series.
The location of suspect lymph nodes at the celiac axis per se should
not strictly be a formal contraindication for surgery, except when
there is diffuse lymph node dissemination to mesenteric LN or when a
complex surgical procedure with unacceptable morbidity is required.
As described in this series, it is feasible to achieve an optimal
cytoreductive outcome when metastatic disease involves celiac nodes
or porta hepatis region. The question is whether patients with bulky
upper abdominal disease requiring extremely aggressive procedures
obtain a survival benefit with an acceptable quality of life. Reports
from the MSKCC have demonstrated that patients requiring extensive
upper abdominal procedures to achieve optimal cytoreduction have
similar initial response, progression-free survival, and overall survival
to patients optimally cytoreduced by standard surgery [23]. Other
authors have also shown that the completeness of cytoreduction has a
more significant influence on survival than the extent of metastatic
disease before surgery [24]. However, metastases to the celiac lymph
nodes associate to an increased risk of hepatic metastases and
progression to mediastinal LN. In this series, 20% of patients with
metastatic celiac lymph nodes had also a suspect mediastinal LN at
image study, before prior chemotherapy. One of the three cases
underwent resection of the mediastinal LN by a diaphragmatic direct
approach. In the other two cases, image study after neoadjuvant
chemotherapy demonstrated a complete response at this level.
Peroperative morbidity associated to extensive upper abdominal
surgery, as well as quality of life concerns, must be considered and
balanced with oncological benefit of surgery. Multivisceral resections,
multiple anastomoses, patient medical status, and prior chemother-
apy each increase the potential for postoperative complications. It is
also well known that patients who undergo suboptimal cytoreduction
sustain the morbidity of a cytoreductive attempt without an
associated survival benefit. Therefore, preoperative risk-assessment
and prudent surgical judgment to proceed are required on initial
exploration of the abdominal cavity as preoperative imaging often
underestimates surgical findings. Preoperative CT and PET scan
detected only 2 (13.3%) of the 15 patients with CLN metastases. The
decision to undergo an aggressive cytoreductive surgery is based on
appropriate patient selection depending on the extension of surgical
procedure, on medical comorbidities, and on the potential to tolerate
an extensive procedure, rather than on specific anatomic locations.
Fig. 3. Left hepatic artery arising from left gastric artery. The left triangular hepatic
ligament has been divided to retract the left lobe of the liver to the right and completely
expose the hepatogastric ligament. The lesser omentum has been resected. An
accessory left hepatic artery is viewed running from the left gastric artery. LHA
indicates left hepatic artery; Pa, pancreas; LLL, left lobe of the liver; LGC, lesser gastric
curvature.
 A. Martinez et al. / Gynecologic Oncology 121 (2011) 258–263
Conclusions
Metastatic involvement of the porta hepatis and celiac axis should
not be considered strict criteria for abandoning an up-front
cytoreductive effort. Resection of enlarged nodes and metastatic
disease to the porta hepatis is feasible with an acceptable morbidity.
The decision to proceed with surgery depends not on specific
anatomic locations but on the sum of the procedures required to
achieve complete eradication of the disease.
Conflicts of interest statement
The authors declare no conflicts of interest.
Acknowledgments
Alejandra Martínez works at ICR and is supported by a grant from
the Fundación Alfonso Martín Escudero, Madrid, Spain.
References
[1] Chi DS, Franklin CC, Levine DA, et al. Improved optimal cytoreduction rates for
stages IIIC and IV epithelial ovarian, fallopian tube, and primary peritoneal cancer:
a change in surgical approach. Gynecol Oncol 2004;94:650–4.
[2] du BA, Reuss A, Harter P, et al. Potential role of lymphadenectomy in advanced
ovarian cancer: a combined exploratory analysis of three prospectively random-
ized phase III multicenter trials. J Clin Oncol 2010;28:1733–9.
[3] Rouzier R, Bergzoll C, Brun JL, et al. The role of lymph node resection in ovarian
cancer: analysis of the surveillance, epidemiology, and end results (SEER)
database.; 2010.
[4] Cicekcibasi AE, Uysal II, Seker M, et al. A rare variation of the coeliac trunk. Ann
Anat 2005;187:387–91.
[5] Chen H, Yano R, Emura S, Shoumura S. Anatomic variation of the celiac trunk with
special reference to hepatic artery patterns. Ann Anat 2009;191:399–407.
[6] Sugarbaker PH. Peritoneal carcinomatosis. Principles of management; 1996.
[7] Sugarbaker PH. Peritonectomy procedures. Ann Surg 1995;221:29–42.
[8] DTCD NCI DHHS. Common terminology criteria for adverse events v4.0 (CTAE).
Cancer therapy evolution program; 2009. Ref Type: Report.
[9] Benedetti-Panici P, Greggi S, Maneschi F, et al. Anatomical and pathological study of
retroperitoneal nodes in epithelial ovarian cancer. Gynecol Oncol 1993;51:150–4.
263
[10] Morice P, Joulie F, Camatte S, et al. Lymph node involvement in epithelial ovarian
cancer: analysis of 276 pelvic and paraaortic lymphadenectomies and surgical
implications. J Am Coll Surg 2003;197:198–205.
[11] Onda T, Yoshikawa H, Yokota H, et al. Assessment of metastases to aortic and
pelvic lymph nodes in epithelial ovarian carcinoma. A proposal for essential sites
for lymph node biopsy. Cancer 1996;78:803–8.
[12] Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal
cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a
meta-analysis. J Clin Oncol 2002;20:1248–59.
[13] Reuss A, Pujade-Lauraine E, et al. Role of surgical outcome as prognostic factor in
advanced epithelial ovarian cancer: a combined exploratory analysis of 3
prospectively randomized phase 3 multicenter trials: by the Arbeitsgemeinschaft
Gynaekologische Onkologie Studiengruppe Ovarialkarzinom (AGO-OVAR) and
the Groupe d'Investigateurs Nationaux Pour les Etudes des Cancers de l'Ovaire
(GINECO). Cancer 2009;115:1234–44.
[14] Winter III WE, Maxwell GL, Tian C, et al. Prognostic factors for stage III epithelial
ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol 2007;25:
3621–7.
[15] Eisenkop SM, Friedman RL, Spirtos NM. The role of secondary cytoreductive
surgery in the treatment of patients with recurrent epithelial ovarian carcinoma.
Cancer 2000;88:144–53.
[16] Janicke F, Holscher M, Kuhn W, et al. Radical surgical procedure improves survival
time in patients with recurrent ovarian cancer. Cancer 1992;70:2129–36.
[17] Tay EH, Grant PT, Gebski V, Hacker NF. Secondary cytoreductive surgery for
recurrent epithelial ovarian cancer. Obstet Gynecol 2002;99:1008–13.
[18] Eisenkop SM, Spirtos NM. What are the current surgical objectives, strategies, and
technical capabilities of gynecologic oncologists treating advanced epithelial
ovarian cancer? Gynecol Oncol 2001;82:489–97.
[19] Chi DS, Eisenhauer EL, Zivanovic O, et al. Improved progression-free and overall
survival in advanced ovarian cancer as a result of a change in surgical paradigm.
Gynecol Oncol 2009;114:26–31.
[20] Salani R, Axtell A, Gerardi M, et al. Limited utility of conventional criteria for
predicting unresectable disease in patients with advanced stage epithelial ovarian
cancer. Gynecol Oncol 2008;108:271–5.
[21] Arango HA, Hoffman MS, Roberts WS, et al. Accuracy of lymph node palpation to
determine need for lymphadenectomy in gynecologic malignancies. Obstet
Gynecol 2000;95:553–6.
[22] Tangjitgamol S, Manusirivithaya S, Sheanakul C, et al. Can we rely on the size of the
lymph node in determining nodal metastasis in ovarian carcinoma? Int J Gynecol
Cancer 2003;13:297–302.
[23] Eisenhauer EL, Abu-Rustum NR, Sonoda Y, et al. The addition of extensive upper
abdominal surgery to achieve optimal cytoreduction improves survival in patients
with stages IIIC-IV epithelial ovarian cancer. Gynecol Oncol 2006;103:1083–90.
[24] Eisenkop SM, Spirtos NM, Friedman RL, et al. Relative influences of tumor volume
before surgery and the cytoreductive outcome on survival for patients with
advanced ovarian cancer: a prospective study. Gynecol Oncol 2003;90:390–6.