pISSN : 2093-582X, eISSN : 2093-5641 J Gastric Cancer 2016;16(2):63-71  http://dx.doi.org/10.5230/jgc.2016.16.2.63

Review Article

Pylorus-Preserving Gastrectomy for Gastric Cancer Seung-Young Oh1, Hyuk-Joon Lee1,2, and Han-Kwang Yang1,2 1

Department of Surgery and 2Cancer Research Institute, Seoul National University College of Medicine, Seoul, Korea

Pylorus-preserving gastrectomy (PPG) is a function-preserving surgery for the treatment of early gastric cancer (EGC), aiming to decrease the complication rate and improve postoperative quality of life. According to the Japanese gastric cancer treatment guidelines, PPG can be performed for cT1N0M0 gastric cancer located in the middle-third of the stomach, at least 4.0 cm away from the pylorus. Although the length of the antral cuff gradually increased, from 1.5 cm during the initial use of the procedure to 3.0 cm currently, its optimal length still remains unclear. Standard procedures for the preservation of pyloric function, infra-pyloric vessels, and hepatic branch of the vagus nerve, make PPG technically more difficult and raise concerns about incomplete lymph node dissection. The short- and longterm oncological and survival outcomes of PPG were comparable to those for distal gastrectomy, but with several advantages such as a lower incidence of dumping syndrome, bile reflux, and gallstone formation, and improved nutritional status. Gastric stasis, a typical complication of PPG, can be effectively treated by balloon dilatation and stent insertion. Robot-assisted pylorus-preserving gastrectomy is feasible for EGC in the middle-third of the stomach in terms of the short-term clinical outcome. However, any benefits over laparoscopyassisted PPG (LAPPG) from the patient’s perspective have not yet been proven. An ongoing Korean multicenter randomized controlled trial (KLASS-04), which compares LAPPG and laparoscopy-assisted distal gastrectomy for EGC in the middle-third of the stomach, may provide more clear evidence about the advantages and oncologic safety of PPG. Key Words: Pylorus-preserving gastrectomy; Stomach neoplasms; Review

Introduction

Laparoscopic gastrectomy is widely used to manage gastric cancer because of the benefits of the minimally invasive approach,

Due to the initiation of health screening programs in East

including less postoperative pain, better cosmetic results, early

Asian countries, including Korea and Japan, the proportion of

recovery of bowel function, and a rapid return to normal activ-

1

early gastric cancer (EGC) has been increasing. With the excel-

ity.2,4,5 The oncologic outcomes of laparoscopic gastrectomy for

lent outcomes obtained after early treatment of gastric cancer,

EGC have been found to be comparable.6 The Korean multi-

surgeons are now recognizing postoperative quality of life (QOL)

center randomized controlled trial (RCT) (KLASS-01 study)

to be as important to consider as survival for these patients.2,3

recently reported that the surgical morbidity rate, particularly

Two surgical approaches are widely used for the treatment

the wound complication rate, had decreased in cases undergoing

of EGC: laparoscopic surgery and function-preserving surgery.

laparoscopic gastrectomy compared with those undergoing open surgery.7 Thus, laparoscopic gastrectomy is considered as one of the standard procedures for EGC.

Correspondence to: Hyuk-Joon Lee Department of Surgery and Cancer Research Institute, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea Tel: +82-2-2072-1957, Fax: +82-2-766-3975 E-mail: [email protected] Received May 19, 2016 Accepted May 23, 2016

In function-preserving surgery, there are several methods for reducing the surgical extent to improve postoperative functional outcomes without compromising oncologic safety, such as pylorus-preserving gastrectomy (PPG), proximal gastrectomy, sentinel node navigation surgery, and vagus nerve-preserving

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Copyrights © 2016 by The Korean Gastric Cancer Association

www.jgc-online.org

64 Oh SY, et al.

surgery.8 Among these techniques, PPG was initially introduced

Indications and Surgical Techniques

9

by Maki et al. for the treatment of peptic ulcers, and was subsequently applied in gastric cancer in Japan and Korea. Although

1. Indications

several retrospective case-control studies have described the

The indications for PPG in several centers are EGCs located

functional benefits of PPG over distal gastrectomy (DG), a mul-

in the middle-third of the stomach with no evidence of regional

ticenter RCT has not yet been conducted to provide high quality

lymph node (LN) metastasis. According to the Japanese gastric

evidence supporting PPG.10-12

cancer treatment guidelines, PPG is indicated for the treatment

In the present review, we describe the current status of PPG, the technical information, and advantages and limitations. We

of cT1N0M0 gastric cancers in the middle-third of the stomach, at least 4.0 cm away from the pylorus.13

also briefly introduce our recent multicenter RCT that compares laparoscopic PPG and laparoscopic DG (KLASS-04 study).

2. Length of the antral cuff

The distance from the lesion to the pylorus needs to be care-

Methods

fully considered as a short antral cuff length may lead to postoperative gastric stasis, a typical complication of PPG. When

A PubMed search was conducted using the keywords ‘pylo-

PPG was initially performed in the treatment of gastric cancer,

rus-preserving gastrectomy’ AND ‘gastric cancer’ for all articles

surgeons usually maintained an antral cuff length of 1.5 cm.

published up to February 2016; only articles written in English

With this antral cuff length, incidence of immediate postopera-

were considered. For the analysis, meta-analyses and RCTs

tive delayed gastric emptying (DGE) was reported to range be-

were preferentially reviewed. Prospective cohort studies and ret-

tween 23% and 40%.14-16 The relationship between the length of

rospective case-control studies were also reviewed.

the antral segment and the incidence of DGE was investigated by Nakane et al.17 in 2002. In that study, the authors found that the incidence of DGE was 35.0% (7/20) in patients with an antral cuff length of 1.5 cm and only 10.0% (1/10) in patients with an antral cuff length of 2.5 cm, at 1 year after surgery. Nunobe

Table 1. Early experiences of pylorus-preserving gastrectomy No. of cases

Year

Pyloric branch of the vagus nerve & RGA

Length of the antral segment (cm)

Akita University

35

1989~1991

Preserved

1.5

Zhang et al.

University of Tokyo

15

1993~1995

Divided

1.5

Imada et al.15

Yokohama University

20

1992~1996

Preserved

1.5

Nihon University

10

1993~1996

Divided

1.5

Nakane et al.

Kansai University

30

1993~1999

Preserved

1.5 versus 2.5

Hotta et al.19

Wakayama University

19

1995~1998

Preserved

1.5

Gumma University

13

1995~1998

Preserved

2.0

Author

Institute

Kodama et al.14 16

26

Tomita et al.

17

24

Ohya et al.

10

Nunobe et al.

National Cancer Center

194

1993~1999

Preserved

2.5~6.0

Nagano et al.22

Fukui University

72

1991~2000

Preserved

-

Osaka University

12

1997~2000

Preserved

1.5

Urushihara et al.

Yoshida General Hospital

26

1998~2002

Preserved

3.0

Park et al.11

Seoul National University

22

1999~2003

Preserved

3.0

611

1995~2004

Preserved

2.0

23

Nishikawa et al.

27

21

National Cancer Center

Morita et al.

25

Tomikawa et al.

Fukuoka City Hospital

9

2004~2007

Preserved

3.0

Lee et al.20

Osaka Medical College

12

2000~2009

NA

≥4.0

RGA = right gastric artery; NA = not available.

65 Pylorus-Preserving Gastrectomy

et al.18 reported an incidence of DGE of 6% to 8% among 90

rior pancreatoduodenal artery (distal type, 64.2% of cases), the

patients after PPG in whom vagus innervation and blood flow to

right gastroepiploic artery (caudal type, 23.1% of cases), or the

the pylorus were preserved and the antral cuff length was main-

gastroduodenal artery (proximal type, 12.7% of cases). During

tained at 3 cm. In subsequent studies, the length of the antral cuff

dissection of LN station 6, the right gastroepiploic artery is li-

has tended to be longer than that used during the initial period

gated at its root in the distal or proximal types. For cases with a

10,11,14-17,19-27

However, a Japanese group did not identify

caudal type, the right gastroepiploic artery is ligated at a location

antral cuff length as a key factor of the PPG technique, report-

distal to the origin of the infra-pyloric artery.11,29,30,32 The hepatic

ing comparable postoperative outcomes among a group of pa-

branch of the vagus nerve that innervates the pylorus usually

tients with an antral cuff length ≤3 cm and a group of patients

follows the course of the supra-pyloric LNs (LN station 5) and

(Table 1).

with an antral cuff length >3 cm. Considering a sufficient distal

should be preserved to maintain the motility of the pylorus. In

resection margin of >1 cm for EGC in addition to the length of

the early years of PPG, surgeons commonly attempted to com-

the antral cuff, the distance from the lesion to the pylorus should

pletely dissect the supra-pyloric LNs.33 However, today, most

be maintained at >4.0 cm. Although guidelines suggest that the

surgeons prefer to focus on preservation of the vagus nerve,

minimum distance from the lesion to the pylorus should be 4.0

rather than on supra-pyloric LN dissection during PPG.12,15,18,23

cm, the optimal length for the antral cuff remains unclear yet.

These important procedures to preserve pyloric function make

28

PPG technically more difficult, when compared with DG.32 3. Lymph node metastasis around the pylorus

An important factor that should be considered prior to per-

5. Laparoscopic pylorus-preserving gastrectomy

forming a PPG is the likelihood of metastasis to LN station 5.

As most patients who undergo PPG are usually diagnosed

This is particularly important as the LN dissection of station 5

with EGC, laparoscopy-assisted pylorus-preserving gastrectomy

is usually omitted during PPG in order to preserve the hepatic

(LAPPG) is commonly used. Although the operation time is

branch of the vagus nerve. A review of PPGs performed at 144

longer in LAPPG than in conventional PPG, LAPPG provides

institutions in Japan indicated that dissection of LN station 5

several benefits over PPG, including reduced intraoperative

was not performed in 53 institutions (36.8%) and was partially

blood loss and postoperative pain, as well as a faster recov-

29

performed in 81 institutions (56.2%). At our institution, which

ery.18,33,34 Moreover, because LAPPG serves as combination of

is one of the institutions that actively performs PPG, dissection

minimally invasive surgery and function-preserving surgery,

of LN station 5 was performed in only 50% of cases of PPG

LAPPG may appear as an attractive treatment option for pa-

30

between 2003 and 2008. In addition to LN station 5, there is

tients.35

also a likelihood of incomplete LN dissection of station 6 dur-

Both extra-corporeal and intra-corporeal methods can be

ing skeletonization of the infra-pyloric artery. For these reasons,

used for anastomosis in LAPPG. For the extra-corporeal meth-

the presence or absence of LN metastasis should be carefully

od, a hand-sewn anastomosis is usually used, which generally

evaluated preoperatively using endoscopic ultrasonography and

involves an approximately 5.0 cm midline incision after mobili-

computed tomography (CT). The depth of invasion should also

zation of the stomach with LN dissection. The distal part of the

be evaluated, as the probability of LN metastasis increases as

stomach is retracted through the incision and resected first. After

the depth of the lesion increases18,30 Hence, PPG should only be

the resection of the proximal part of the stomach, a hand-sewn

considered only for patients with a cT1N0M0 gastric cancer.

gastro-gastrostomy is performed.33-35 Intra-corporeal anastomosis methods using linear staplers have only recently been

4. Techniques for preservation of the pylorus

introduced. For intra-corporeal anastomosis, transection of the

Although there are minor differences in the surgical tech-

stomach in the sagittal direction (i.e., posterior to anterior direc-

niques according to specific surgeons, the standard technique for

tion), rather than in the transverse direction (i.e., greater curva-

PPG includes preservation of the infra-pyloric vessels and the

ture to lesser curvature direction), can facilitate the alignment of

hepatic branch of the vagus nerve for structural and functional

the linear staplers.20,36 After resection of the distal and proximal

preservation of the pylorus.29 According to a study by Haruta et

parts of the stomach, one arm of a 60 mm linear stapler is in-

31

al., the infra-pyloric artery originates from the anterior supe-

serted into each gastric remnant through the gastrostomy on the

66 Oh SY, et al.

greater curvature side corner. The stapler has to be fired between the posterior walls on either side, and then the remaining gas-

2. Oncologic safety Preservation of the vessels and nerves in order to maintain pyloric function may result in insufficient LN dissection at LN

trostomy can be closed using further staplers.

stations 5, 6, and 12a, which could consequently compromise

Clinical Outcomes

the radicality of the curative gastrectomy for gastric cancer. According to the Japanese gastric cancer treatment guidelines (ver.

1. Complications

3), D1+ lymphadenectomy should be performed for patients

With regard to the short-term outcomes of PPG, Shibata et

with cT1N0.13 LN dissection of station 6 with infra-pyloric ar-

al.37 compared PPG and DG and our group compared LAPPG

tery preservation is a relatively easy technique, and LN station

12

and laparoscopy-assisted DG (LADG). Both studies indicated

12a is considered to be beyond the D1+ level in patients with

that the postoperative hospital stay, postoperative complications,

cT1N0M0. However, LN station 5 is considered to be D1 level.

and mortality did not differ between patients undergoing PPG

In PPG, dissection of LN station 5 is omitted to preserve the

and DG, regardless of the approach.

hepatic branch of the vagus nerve and preserve pyloric function.

In a study performed with 307 patients who underwent LAPPG by Jiang et al.,38 the overall complication rate was 17.3% (53/307) including a major complication rate (grade>IIIa, 39

This could lead to incomplete D1 LN dissection, which is associated with concerns regarding oncologic safety. In a study about a new index evaluating the therapeutic value

Clavien-Dindo classification) of only 1.3% (4/307). In an-

of LN dissection for gastric cancer, Sasako et al.40 reported that the

other study of complications (again, according to the Clavien-

index (estimated via multiplication of the incidence of metastasis

Dindo classification) of 116 patients who underwent LAPPG,

and the 5-year survival rate of patients with metastasis to LN sta-

the overall complication rate was 14.7% (17/116) and major

tion 5) was only 0.8 in patients with cancer of the middle-third of

12

complications, grade>IIIa, were found in 10 patients (8.6%). In

the stomach. In particular, a few studies have also focused on the

both studies, the most common complication was associated with

probability of metastasis to LN station 5 from EGC of middle-

postoperative impairment in pyloric function; gastric stasis was

third of the stomach. Kodera et al.41 reported that the metastasis

present in 6.2% in the former study and DGE in 7.8% in the latter.

rate to LN station 5 was