Laparoscopic Surgery of Gastric Cancer with D2 Lymphadenectomy and Omentum Preservation: Our 10 Years Experience

Abstract

Introduction:

The debate is still open about laparoscopic treatment of gastric cancer. The aim of this retrospective study is to analyze our short-, medium-, and long-term surgical and oncological results in laparoscopic treatment of gastric cancer with D2 lymphadenectomy and omentum preservation.

Materials and Methods:

From January 2010 to June 2018, after >150 surgical procedures for gastric cancer performed by minimally invasive approach, we performed 100 laparoscopic subtotal gastrectomies and 38 total gastrectomies, both for early gastric cancer (EGC) and advanced gastric cancer (AGC). We always made a D2 lymphadenectomy or higher. As often as possible, we performed omentum-preserving technique. Primary outcomes analyzed included incidence of medical and surgical complications. Secondary outcomes analyzed were survival probability and incidence of relapse. Every patient read and signed informed consent before surgery.

Results:

Mean operative time: 2.4 ± 0.7 hours (range 1.2–4.7 hours). Rate of conversions: 14.5% (20/138); intraoperative complications: 1.4% (2/138) and positive resection margins: 6.5% (9/138). Overall incidence of duodenal fistula: 3.6% (5/138). Rate of reoperation was 7.3% (10/138). Postoperative complications according to Clavien–Dindo classification: I 3.6% (5/138); II 13.0% (18/138); III 5.8% (8/138); III B 0.7% (1/138); V 1.4% (2/138). Overall survival with 60 months follow-up was 58%. Overall 60 months incidence of relapse was 44%. Patients with omentum preservation had a lower incidence of relapse than patients with omentectomy (40% versus 57% P = .002).

Conclusions:

Laparoscopic treatment of gastric cancer with D2 lymphadenectomy and omentum preservation is safe and feasible, both for EGC and for AGC. Although this study has limitations, omentum-preserving technique was associated with a statistically lower recurrence rate.

Introduction

The number of laparoscopic procedures for gastric cancer for the past decade has grown quickly. Laparoscopic surgery provides considerable advantages compared with traditional surgery regarding postoperative course and rate of complications, allowing surgeons to perform the same operation with less physical stress and discomfort for the patient. For early gastric cancer (EGC), evidence supports the feasibility and oncological safety of laparoscopic surgery^1,2; however, for advanced gastric cancer (AGC) there is still open debate. In addition, there is also open discussion about the usefulness of carrying out a complete omentectomy in the course of gastrectomy for cancer, regardless of the stage of the tumor.^3,4

The history of laparoscopic gastric surgery began in 1992, when Peter Goh (Singapore) performed the first laparoscopic subtotal Billroth II gastrectomy in a patient with gastric ulcer.⁵ In the same year, other Asian authors proposed laparoscopic treatment of neoplastic disease localized to the gastric mucosa with low risk of metastatic spread (EGC), intending to minimize the extent of gastric resection and reduce postoperative complications compared with traditional gastrectomy. The first wedge resections were laparoscopic intragastric mucosal resections for EGC.^6,7 In June 1993, Azagra successfully performed the first laparoscopic gastrectomy for cancer.⁸ Since then, many surgeons have contributed to the study and progress of laparoscopic techniques for the treatment of gastric cancer. New instruments (stapler, ultrasonic dissector, etc.) and developments in laparoscopic surgical techniques have improved operational feasibility, patient safety, and oncological radicality.

Although laparoscopic surgery of the stomach was initially limited to EGC,⁹ today laparoscopic surgery is applied to AGC with similar results to open technique in terms of oncological radicality, plus all the benefits of minimally invasive surgery. Except for very early reports, the number of lymph nodes removed during laparoscopy gastrectomy is equivalent to that of open surgery.¹⁰ Two randomized controlled trials (total 87 patients) showed no differences in morbidity and mortality between laparoscopic and open surgery for patients with EGC^1,2 and AGC.² A randomized study in 2005 showed a survival and disease-free interval of 55.7% and 54.8% in patients with gastric cancer treated by open surgery, compared with 54.8% and 58.9% in patients undergoing laparoscopic approach.²

A multicenter randomized trial of 342 patients¹¹ showed a higher incidence, although not statistically significant, of short-term complications of open surgery (15.1% [27/161]) compared with laparoscopy (11.6% [20/179]). In general, in this study, the morbidity and mortality between the two techniques were not statistically different (P > .49). In a meta-analysis based only on randomized trials, surgical complications were less common after laparoscopic surgery compared with open surgery, although again without statistical significance.¹² A retrospective analysis of 1185 patients with gastric cancer who underwent laparoscopic-assisted surgery, with a mean follow-up of 36 months, found an estimated disease-free survival at 5 years of 99.8% for stage IA, 98.7% for stage IB, and 85.7% for stage II.¹³ A multicenter retrospective study of 1485 patients with gastric cancer having laparoscopic-assisted surgery showed a 5-year survival of 95.5% and a disease-free survival of 94.1%.¹⁴ In the same study, 1.6% of patients with EGC and 13.4% of patients with AGC developed tumor recurrence at 5 years.

A recent large-scale multicenter retrospective study¹⁵ showed that the number of retrieved lymph nodes after laparoscopic-assisted gastrectomy was oncologically acceptable, complication rates and mortality were also acceptable compared with previous reports, and long-term outcomes for AGC were comparable with those previously reported for open gastrectomy. In addition, some authors have highlighted the possibility of performing a partial omentectomy instead of a complete omentectomy. Theoretically, the residual omentum might fill up any anastomotic microleakages by adhesions to the inflamed bowel. Macrophages residing within the omentum might play an important role against minimal residual disease or free peritoneal tumor cells; alternatively, a complete omentectomy may impair antitumor immune response.^16,17

A randomized phase II trial is being conducted to evaluate the impact of omentectomy for AGC on patient survival.³ Results reported in the literature are encouraging: in one retrospective study⁴ omentum-preserving gastrectomy for AGC did not increase peritoneal relapse rate and did not affect survival of patients compared with conventional gastrectomy.

These results justify and encourage the use and development of laparoscopic surgery in the treatment of gastric cancer. The aim of this retrospective study is to present our short-, medium-, and long-term results after laparoscopic treatment of gastric cancer with omentum preservation.

Materials and Methods

We present the short-, medium-, and long-term results of our study, after >150 operations for gastric cancer performed with minimally invasive approach, from January 2010 to June 2018 at the San Marco Hospital—GSD (Bergamo, Italy). After the initial laparoscopic staging, we proceeded to laparoscopic gastric resection in 138 cases. The purpose of this analysis was to describe the population of patients who underwent a total or subtotal gastrectomy due to stomach cancer and to gather information on patient's outcomes in terms of incidence of complications, survival, and relapse. Outcomes were compared by type of complication, type of surgery, and cancer stage. Informed consent was obtained from each patient before surgery.

Surgical technique

The patient was positioned as follows: the legs are spread and the surgeon is positioned between the legs. The right arm is extended along the body, whereas the left arm is abducted. The patient is positioned in mild anti-Trendelenburg. The equipment for laparoscopy is placed behind the right shoulder of the patient. The cameraman is sitting at the right side of the patient, operating the camera with his right arm. The second surgeon is on the patient's left side. After pneumoperitoneum induction with umbilical Veress needle (or left subcostal in the case of obese patients), four trocars are introduced: subumbilical 10 mm, 12 mm left side, 5 mm right side, and 5 mm in the epigastrium. In cases of total gastrectomy, there may be value in an additional 5 mm trocar in the left lateral position.

We proceed to explore the peritoneal cavity with confirmation of staging and operability. This essential step avoids unnecessary laparotomy for inoperable gastric cancers and allows a final evaluation and confirmation of appropriateness of the intervention planned (total versus subtotal gastrectomy). With the aid of an ultrasonic dissector (ACE Ultracision™ 5 mm; Ethicon Endosurgery, Cincinnati, OH), or radiofrequency dissector (Ligasure Maryland™; Medtronic, Minneapolis, Minnesota) we open the gastrocolic ligament, starting about 4–5 cm from the gastroepiploic vessels but sparing the omentum on the side and performing a partial omentectomy, thus accessing the lesser sac. Then we continue skeletonization of the great curvature of the stomach, to Van Goethem's point for cases of subtotal gastrectomy, or completely for cases of total gastrectomy.

The right gastroepiploic vessels and the right gastric vessels are ligated at their origin. Duodenal division is performed by laparoscopic stapler (Echelon 60™, Ethicon Endosurgery; or EndoGIA 60™, Medtronic; or EndoGIA SIGNIA™, Medtronic). Reinforcement may be performed by running suture of PDS 3/0 (MIC e55™; Ethicon Endosurgery, Cincinnati, OH) or material reinforcements (Seaguard^®; Peristrip^®). The height of the staplers is always chosen in relation to the thickness of the bowel. Through dissection of the hepatoduodenal ligament, we proceed to lymphadenectomy along the hepatic artery. Then, we proceed to the section of the left gastric vessels.

Lymphadenectomy ends with a complete removal of the hepatogastric ligament to the right diaphragmatic pillar. At this point, we proceed to the proximal transection at the level of the upper third of the stomach if a partial gastrectomy is being performed. In cases of total gastrectomy, we further divide the short gastric vessels and freno-fundic ligament until Bertelli's ligament, and then divide proximally at the esophagus-gastric junction. The resection specimen is now completely liberated and is temporarily placed over the liver, to allow anastomoses, all of which are performed with intracorporeal laparoscopic technique.

Reconstruction is normally performed, in the case of subtotal gastrectomy, according to Roux-en-Y, except in selected cases, where we prefer to reconstruct Omega, for example, patients with important comorbidities or very elderly patients; in these cases we also made a reconstruction with a jejunojejunal anastomosis sec. Brown. All anastomoses in the case of total gastrectomy are performed according to the technique of Orringer, a mechanical isoperistaltic esophagojejunal anastomosis. Then we make mechanical jejunojejunal anastomosis. We always position a periduodenal drain and, in patients with a high risk or in the case of total gastrectomy, a further retroanastomotic drain. During hospitalization, 4–5 days after surgery, we performed routine upper esophageal-gastric contrast study in all total gastrectomy patients and selected subtotal gastrectomy patients.

Statistical analysis

Descriptive statistics were used to summarize patient baseline characteristics and clinical data. This includes mean and standard deviation, minimum, maximum, and median with the interquartile range (IQR) for continuous variables, and counts and percentages for categorical variables. Summary statistics were reported with maximum two decimals, as appropriate. The analysis for primary objective was performed using the logistic regression model. Bivariate odds ratios and 95% confidence intervals (95% CIs) were estimated for each potential predictor present in the database. The multivariate logistic regression model was performed using the stepwise selection procedure considering a selection based on the significant association (P ≤ .10) identified from the bivariate logistic analysis, to avoid overfitted and unstable model the correlation coefficient between them should be <0.20.

For the multivariate models the presence of event will be the dependent variable. The analyses of time-to-the-event were described by means of Kaplan–Meier curves and compared between the groups by means of the log-rank test. Cox bivariate models were fitted and hazard ratios with 95% CIs were computed and plotted. The exposure time (days) was estimated from the date of surgery to the date of the last available follow-up or date of event (death or relapse), as appropriate. The analysis was supported by graphical representation. The SAS software version 9.4 (SAS Institute, Inc., Cary, NC) was used to perform statistical analysis. Statistical tests were based on a two-sided significance level of 0.05.

Results

A total of 138 patients (Table 1) with endoscopic diagnosis of AGC/EGC were subjected to laparoscopic total or subtotal gastrectomy with D2 lymphadenectomy and attempted omentum preservation at the San Marco Hospital—GSD (Bergamo, Italy). Median age was 72 (IQR, 63.5–79) years; 57.2% of patients were male. Thirty-eight (27.5%) patients underwent a total gastrectomy and 100 (72.5%) a subtotal gastrectomy. The median surgery duration was 2.3 (IQR, 1.8–2.8) hours, and median length of hospital stay was 10 days (IQR, 9–12). Table 2 summarizes the main results related to surgery. Thirty-four (25%) patients had a complication after surgery, 19 (14%) had a medical complication, and 17 (12%) had a surgical complication. Additional information regarding complications is summarized in Table 3. The most common complication was anastomosis fistula (7/138; 5.1%). The most concerning complication was fistula at the level of the duodenal transection. The overall incidence of duodenal fistula across the study population was 3.6% (5/138). The rate of reoperation was 7.2% (10/138).

Table 1.

Patients' Demographic and Baseline Characteristics

Baseline characteristics	Summary statistics	Total (n = 138)
Age (years)	Median (IQR)	72.0 (63.5–79.0)
	Mean ± SD	70.7 ± 10.1
	Min–max	34.8–87.4
Gender (male)	%, n/pts	57.2 (79/138)
BMI	Median (IQR)	27.0 (24.0–29.0)
	Mean ± SD	26.6 ± 2.7
	Min–max	21.0–31.0
Preoperative tumor localization (endoscopy)
Antrum	%, n/pts	65.9 (91/138)
Cardias	%, n/pts	7.2 (10/138)
Body	%, n/pts	16.6 (23/138)
Fundus	%, n/pts	7.2 (10/138)
Greater curvature	%, n/pts	0.7 (1/138)
Lesser curvature	%, n/pts	2.1 (3/138)
TC: Liver metastasis	%, n/pts	5.8 (8/138)
TC: LN+	%, n/pts	29.7 (41/138)
Pre-EGC	%, n/pts	30.4 (42/138)
Pre-AGC	%, n/pts	69.6 (96/138)
ASA score
I	%, n/pts	10.6 (10/94)
II	%, n/pts	51.1 (48/94)
III	%, n/pts	37.2 (35/94)
IV	%, n/pts	1.1 (1/94)

AGC, advanced gastric cancer; ASA, American Society of Anesthesiologists; EGC, early gastric cancer; IQR, interquartile range; LN+, positive lymph nodes at preoperative TC scan; SD, standard deviation; TC, tomographic computed images.

Table 2.

Surgery Characteristics

Surgery characteristics	Summary statistics	Total (n = 138)
Conversion	%, n/pts	14.5 (20/138)
Adhesions	%, n/pts	37.0 (51/138)
Total gastrectomy	%, n/pts	27.5 (38/138)
Subtotal gastrectomy	%, n/pts	72.5 (100/138)
Associated surgery	%, n/pts	21.7 (30/138)
Omentum preserving	%, n/pts	83.3 (115/138)
Intraoperative tumor localization
Cardias	%, n/pts	8.0 (11/138)
Fundus	%, n/pts	8.7 (12/138)
Body	%, n/pts	18.1 (25/138)
Antrum	%, n/pts	65.2 (90/138)
Dissection and suture tools
EndoGIA™	%, n/pts	26.1 (36/138)
SIGNIA™	%, n/pts	14.5 (20/138)
Echelon™	%, n/pts	59.4 (82/138)
ACE™ dissector	%, n/pts	61.6 (85/138)
Maryland™ dissector	%, n/pts	38.4 (53/138)
Duodenum section
Echelon™ Green	%, n/pts	16.7 (23/138)
Echelon™ Gold	%, n/pts	36.2 (50/138)
Echelon™ Blue	%, n/pts	8.0 (11/138)
EndoGIA™/SIGNIA™ Black	%, n/pts	6.5 (9/138)
EndoGIA™/SIGNIA™ Purple	%, n/pts	23.2 (32/138)
EndoGIA™/SIGNIA™ Brown	%, n/pts	9.4 (13/138)
Reinforcement: Seamguard^®	%, n/pts	49.3 (68/138)
Reinforcement: Peristrips^®	%, n/pts	18.8 (26/138)
Reinforcement: MIC™	%, n/pts	26.1 (36/138)
Reinforcement: nothing	%, n/pts	5.8 (8/138)
Stomach section
Echelon™ Blue	%, n/pts	36.2 (50/138)
Echelon™ Green	%, n/pts	22.5 (31/138)
Echelon™ Gold	%, n/pts	23.9 (33/138)
EndoGIA/SIGNIA™ Black	%, n/pts	2.9 (4/138)
EndoGIA/SIGNIA™ Purple	%, n/pts	13.8 (19/138)
EndoGIA/SIGNIA™ Brown	%, n/pts	0.0 (0/138)
Reinforcement: MIC™	%, n/pts	41.3 (57/138)
Reinforcement: nothing	%, n/pts	58.7 (81/138)
Reconstruction
Reconstruction Roux	%, n/pts	50.0 (69/138)
Reconstruction Omega	%, n/pts	22.5 (31/138)
Reconstruction Orringer	%, n/pts	27.5 (38/138)
Echelon™ Blue	%, n/pts	36.2 (50/138)
Echelon™ Green	%, n/pts	43.5 (60/138)
Echelon™ Gold	%, n/pts	0.0 (0/138)
EndoGIA™/SIGNIA™ Purple	%, n/pts	6.5 (9/138)
EndoGIA™/SIGNIA™ Black	%, n/pts	12.3 (17/138)
EndoGIA™/SIGNIA™ Brown	%, n/pts	1.4 (2/138)
Intraoperative complications	%, n/pts	1.4 (2/138)
Positive margins	%, n/pts	6.5 (9/138)
Negative margins	%, n/pts	93.5 (129/138)
Anastomosis drainage	%, n/pts	67.4 (93/138)
Length of surgery (hours)	Median (IQR)	2.3 (1.8–2.8)
	Mean ± SD	2.4 ± 0.7
	Min-Max	1.2–4.7
Length of hospital stay (days)	Median (IQR)	10.0 (9.0–12.0)
	Mean ± SD	13.4 ± 12.8
	Min−max	6.6–115.6
Blood loss (mL)	Median (IQR)	150.0 (100.0–200.0)
	Mean ± SD	189.0 ± 124.2
	Min−max	50.0–500.0

ACE™: ultrasound dissector; EndoGIA™, SIGNIA™, Echelon™: laparoscopic staplers; Maryland™: radiofrequency dissector. MIC™: 3/0 self-absorbable suture; Seamguard^®, Peristrips^®: bioabsorbable staple line reinforcement.

Table 3.

Postoperative Results

Variable	Summary statistics	Total (n = 138)
Complication	%, n/pts	24.6 (34/138)
Medical complication	%, n/pts	13.8 (19/138)
Surgical complication	%, n/pts	12.3 (17/138)
Days for canalization	Median (IQR)	3.0 (2.0–3.0)
	Mean ± SD	2.9 ± 1.0
	Min–max	0.0–6.0
Oral food	Median (IQR)	5.0 (5.0–6.0)
	Mean ± SD	5.6 ± 1.7
	Min–max	4.0–17.0
Walking	Median (IQR)	2.0 2.0–3.0)
	Mean ± SD	2.5 ± 1.1
	Min–max	0.0–6.0
Complications
Fistula	%, n/pts	5.1 (7/138)
Dehiscence	%, n/pts	0.0 (0/138)
Stenosis	%, n/pts	1.4 (2/138)
Duodenum	%, n/pts	3.6 (5/138)
Wound infection	%, n/pts	2.2 (3/138)
Pulmonary	%, n/pts	4.3 (6/138)
Cardiac	%, n/pts	3.6 (5/138)
Pancreatitis	%, n/pts	1.4 (2/138)
Ileus	%, n/pts	0.7 (1/138)
Bleeding	%, n/pts	10.9 (15/138)
Transfusion	%, n/pts	12.3 (17/138)
Reintervention	%, n/pts	7.3 (10/137)
DPO reintervention	Median (IQR)	18.0 (10.0–20.0)
	Mean ± SD	15.6 ± 9.1
	Min–max	2.0–30.0
Clavien–Dindo
I	%, n/pts	3.6 (5/138)
II	%, n/pts	13.0 (18/138)
III	%, n/pts	5.8 (8/138)
III B	%, n/pts	0.7 (1/138)
V	%, n/pts	1.4 (2/138)

DPO, postoperative day.

Medical treatments and chemotherapy were performed based on grade, stage, histological type, and patient characteristics. We proceeded to clinical and instrumental follow-up. We summarize these data in Table 4 and the distribution by stage in Figure 1. Univariate logistic regression modeling was applied to investigate the association between baseline characteristics and complication occurrences. Results are shown in Table 5. Greater patient age, omentectomy, and positive margins were statistically significant predictors of any complication. Use of Echelon Gold for duodenal stapling was a statistically significant protective factor for complication occurrence. Median length of hospital stay was higher in patients with complications. The same univariate logistic regression analysis was applied to investigate the association between measured variables and medical complications. Conversion, surgery type, and positive margins were statistically significant predictors of medical complications (P < .05). Likewise, similar analysis applied to surgical complications found higher age and ASA score III–IV were statistically significant predictors of surgical complication, whereas the presence of lymph nodes (P < .046) and use of Echelon Gold for duodenal stapling were protective factors (P < .05).

FIG. 1.

Tumor stage distribution.

Table 4.

Type of Cancer and Staging

Type of cancer	Summary statistics	Total (n = 138)
EGC	%, n/pts	11.6 (16/138)
AGC	%, n/pts	86.2 (119/138)
Tubular	%, n/pts	71.0 (98/138)
Mucinous	%, n/pts	5.8 (8/138)
Signet ring	%, n/pts	22.5 (31/138)
Grading
Low	%, n/pts	11.6 (16/138)
Moderate	%, n/pts	42.8 (59/138)
High	%, n/pts	44.2 (61/138)
Stage
0	%, n/pts	5.8 (8/138)
I A	%, n/pts	14.5 (20/138)
I B	%, n/pts	15.2 (21/138)
II A	%, n/pts	12.3 (17/138)
II B	%, n/pts	8.7 (12/138)
III A	%, n/pts	12.3 (17/138)
III B	%, n/pts	8.0 (11/138)
IV	%, n/pts	13.0 (18/138)

AGC, advanced gastric cancer; EGC, early gastric cancer.

Table 5.

Univariate Logistic Regression: Complications

Clinical characteristics	Summary statistics	No complications (n = 104)	Complications (n = 34)	OR (95% CI)	P
Age (years)	Mean ± SD	69.7 ± 9.9	73.9 ± 10.5	1.05 (1.00–1.09)	.041
Gender (male)	%, n/pts	57.7% (60/104)	55.9% (19/34)	0.93 (0.43–2.03)	.853
BMI	Mean ± SD	26.5 ± 2.5	27.6 ± 3.6	1.20 (0.77–1.87)	.421
Liver metastasis	%, n/pts	4.9% (4/82)	13.8% (4/29)	3.12 (0.73–13.40)	.126
LN	%, n/pts	39.0% (32/82)	31.0% (9/29)	0.70 (0.28–1.74)	.445
ASA III–IV	%, n/pts	53.8% (56/104)	70.6% (24/34)	2.06 (0.89–4.73)	.089
Conversion	%, n/pts	12.5% (13/104)	20.6% (7/34)	1.81 (0.66–5.00)	.249
Adhesions	%, n/pts	36.5% (38/104)	38.2% (13/34)	1.08 (0.48–2.39)	.859
Total gastrectomy	%, n/pts	29.8% (31/104)	20.6% (7/34)	0.61 (0.24–1.55)	.299
Subtotal gastrectomy	%, n/pts	70.2% (73/104)	79.4% (27/34)	1.64 (0.65–4.16)	.299
Associated surgery	%, n/pts	18.3% (19/104)	32.4% (11/34)	2.14 (0.89–5.13)	.088
Omentectomy	%, n/pts	12.5% (13/104)	29.4% (10/34)	2.92 (1.14–7.46)	.025
Omentum preserving	%, n/pts	87.5% (91/104)	70.6% (24/34)	0.34 (0.13–0.88)	.025
EndoGIA™	%, n/pts	22.1% (23/104)	38.2% (13/34)	2.18 (0.95–5.01)	.066
SIGNIA™	%, n/pts	14.4% (15/104)	14.7% (5/34)	1.02 (0.34–3.06)	.967
Echelon™	%, n/pts	63.5% (66/104)	47.1% (16/34)	0.51 (0.23–1.12)	.094
ACE™ dissector	%, n/pts	65.4% (68/104)	50.0% (17/34)	0.53 (0.24–1.16)	.112
Maryland™ dissector	%, n/pts	34.6% (36/104)	50.0% (17/34)	1.89 (0.86–4.14)	.112
Duodenum Echelon™ Green	%, n/pts	17.3% (18/104)	14.7% (5/34)	0.82 (0.28–2.42)	.724
Duodenum Echelon™ Gold	%, n/pts	41.3% (43/104)	20.6% (7/34)	0.37 (0.15–0.92)	.033
Duodenum Echelon™ Blue	%, n/pts	6.7% (7/104)	11.8% (4/34)	1.85 (0.51–6.75)	.353
Duodenum EndoGIA™/SIGNIA™ Black	%, n/pts	7.7% (8/104)	2.9% (1/34)	0.36 (0.04–3.02)	.349
Duodenum EndoGIA™/SIGNA™ Purple	%, n/pts	19.2% (20/104)	35.3% (12/34)	2.29 (0.97–5.39)	.058
Duodenum EndoGIA™/SIGNIA™ Brown	%, n/pts	7.7% (8/104)	14.7% (5/34)	2.07 (0.63–6.81)	.232
Duodenum Reinforcement: Seamguard^®	%, n/pts	47.1% (49/104)	55.9% (19/34)	1.42 (0.65–3.10)	.376
Duodenum Reinforcement: Peristrips^®	%, n/pts	17.3% (18/104)	23.5% (8/34)	1.47 (0.57–3.77)	.422
Duodenum reinforcement: MIC™	%, n/pts	29.8% (31/104)	14.7% (5/34)	0.41 (0.14–1.15)	.089
Duodenum reinforcement: nothing	%, n/pts	5.8% (6/104)	5.9% (2/34)	1.02 (0.20–5.31)	.980
Reconstruction Roux	%, n/pts	47.1% (49/104)	58.8% (20/34)	1.60 (0.73–3.51)	.238
Reconstruction Omega	%, n/pts	23.1% (24/104)	20.6% (7/34)	0.86 (0.33–2.23)	.763
Reconstruction Orringer	%, n/pts	29.8% (31/104)	20.6% (7/34)	0.61 (0.24–1.55)	.299
Intraoperative complications	%, n/pts	% (/104)	5.9% (2/34)	>999.9	.985
Positive margins	%, n/pts	2.9% (3/104)	17.6% (6/34)	7.21 (1.70–30.67)	.007
Length of surgery (hours)	Mean ± SD	2.3 ± 0.7	2.5 ± 0.7	1.29 (0.74–2.25)	.362
Length of hospital stay (days)	Mean ± SD	10.1 ± 2.6	23.8 ± 23.1	1.32 (1.15–1.52)	<.001

Bold values reflect that statistical tests were based on a two-sided significance level of 0.05.

ACE™: ultrasound dissector; EndoGIA™, SIGNIA™, Echelon™: laparoscopic staplers; Maryland™: radiofrequency dissector; MIC™: 3/0 self-absorbable suture; Seamguard^®, Peristrips^®: bioabsorbable staple line reinforcement.

CI, confidence interval.

We analyzed the probability of survival, overall and by clinical characteristics and the incidence of relapse. The overall 60-month probability of survival was 58%. Figure 2 shows the overall survival Kaplan-Meier (KM) curve. Patients with complications had a significant lower probability of survival compared with those without complications (24% versus 68%, P = .001), as shown in Figure 3. No survival differences were found between patients with and without conversion (P = .47), or between patients having total versus subtotal gastrectomy (P = .42). A statistically significant survival difference (P < .001) was found by cancer stage (Fig. 4).

FIG. 2.

Overall survival KM curve. KM, Kaplan-Meier.

FIG. 3.

Overall survival KM curve—survival by complications.

FIG. 4.

Overall survival KM curve—survival by stage.

Finally, the incidence of relapse, overall and by clinical variables, was analyzed. The overall 60-month incidence of disease relapse was 44%. There were no differences in relapse between patients with or without complication. Patients with medical complications had a significant higher incidence of relapse than patients without medical complications (69% versus 41%, P = .030), as shown in Figure 5. There was no association between relapse and surgical complications, but patients with conversion had a statistically higher incidence of relapse (63% versus 42%, P = .026). Patients who had omentum preservation had a lower incidence of relapse than patients who had omentectomy (40% versus 57%, P = .002)—Figure 6. A statistically significant difference (P < .001) was found in terms of relapse by cancer stage (Fig. 7).

FIG. 5.

Incidence of relapse KM curve by medical complications.

FIG. 6.

Incidence of relapse KM curve by omentectomy.

FIG. 7.

Incidence of relapse KM curve by stage.

Discussion

Laparoscopic surgery for gastric cancer has rapidly developed given the early patient benefits associated with this minimally invasive technique. The treatment of EGC with laparoscopic approach is now accepted,^13,18 whereas the treatment of the AGC remains controversial, due to supposed technical challenges in performing an appropriate lymphadenectomy. Retrospective studies on EGC have also shown that there are no differences in terms of survival in patients operated by laparoscopic approach compared with the open technique. Although the long-term results of the Korean Laparoscopic Surgery Study (KLASS) are still to be published, EGC's laparoscopic treatment is now accepted. Regarding the treatment of AGC, few data are available; however, both randomized and retrospective studies have shown that the number of lymph nodes removed and the disease-free survival do not differ significantly from the open technique.^19–21

Laparoscopic surgery for gastric cancer has been shown to be feasible and safe when performed by surgeons experienced with minimally invasive method for the treatment of EGC, but the long-term efficacy of laparoscopy for the treatment of AGC is still debated. Our data on the laparoscopic treatment of gastric cancers (both EGC and AGC) are encouraging both for the immediate postoperative period and for long-term follow-up. Regarding the postoperative course and complications, higher patient age at surgery, omentectomy, and positive margins were found as statistically significant predictors of complications. Older age is certainly indicative of an increased overall risk to the elderly patient. Likewise, positive resection margins indicate an advanced oncological disease, therefore worse condition of the patient. It is interesting to note that omentectomy was associated with increased risk of postoperative complications in our analysis. Theoretically, the residual omentum after a partial omentectomy might fill up some anastomotic microleakages by adhesions to the inflamed bowel. These benefits might derive from the protective action of the omentum at an inflammatory level, as it could limit abdominal inflammatory events that could lead to systemic or surgical problems.

Conversion to open surgery and the presence of positive margins were found as statistically significant predictors of medical complications, whereas older age and higher ASA score (III–IV) were statistically significant predictors of surgical complications. The association of surgical complications with open conversion is indicative of the well-known benefits of laparoscopic technique. Laparotomy is an enormous physical stress for a patient, with increased medical complications such as pneumonia or heart disease. The increased pain deriving from laparotomy prevents rapid mobilization, lengthens hospitalization and, therefore, portends a worse postoperative course. Regarding positive margins and higher ASA score, correlation with medical complications is similar to what was discussed about advanced cancer and the age of the patient. All of these are indices of a patient's worse global condition, which directly affect the risk of postoperative medical complications.

The overall 60-month probability of survival was 58% and the overall relapse rate was 44% (Fig. 2). It must be remembered that these are raw data, which do not stratify the risk of mortality or recurrence by disease stage. In our population there was a considerable preponderance of AGC (86.2%); therefore, a global 60-month survival rate of almost 60% is in line with other similar series.²² In terms of survival, differences between patients with or without complications (any, medical and surgical) were found (Fig. 3) (P = .0012). The interpretation of these data is twofold: on the one hand, a patient suffering complications may have longer recovery and, therefore, a delay in initiating chemotherapy; on the other hand, we have seen that positive margins, higher ASA score, and older age are independent variables associated with complications. Therefore, it is more likely that a patient with complications is already a patient with a risk of shorter survival. A statistically significant difference (P < .001) was found in terms of survival by cancer stage (Fig. 4). Regarding survival, our data show that stages 0–I–II had 60-month survival in >70% of patients. Regarding risk of relapse, regardless of survival, patients with medical complications had a significantly higher likelihood of relapse than those without medical complication (Fig. 5), whereas those with omentum preservation had a lower incidence of relapse than patients who had omentectomy. This last finding is extremely interesting, because it represents one of the endpoints of this study. From our data, it emerges that the omentum-preserving technique may protect against the risk of recurrence (Fig. 6; P = .002). Macrophages residing within the omentum may play an important role in clearing minimal residual disease or free peritoneal tumor cells; therefore, a complete omentectomy may impair the antitumor immune response.^16,17 However, this interpretation has some limitations: most patients had omentum preservation, with lower relapse rate, but there was no valid control group since the patients who had omentectomy did so because omentum-preserving technique was not possible. Therefore, this result may be limited by a selection bias. It is also possible that the greater complications of omentectomy were the reason omentum sparing was associated with lower relapse. At least, incidence of relapse statistically increased according to the stage of cancer.

The analysis described in this report had two major limitations. First, as we depended upon retrospective observational research data, selection bias was possible. Second, the low sample size and the low incidence of events limited our ability to identify statistically significant differences or to perform multivariate data analysis to evaluate if the demographic and baseline characteristics of the patients and of the cancer affect the outcomes. The results for some comparisons were reported from a descriptive perspective, as some P values could not be reliable due to the low observed frequencies. In the KM analysis, differences within groups were not tested.

Conclusions

Laparoscopic treatment of gastric cancer with D2 lymphadenectomy and omentum preservation is safe and feasible, both for EGC and for AGC. Our retrospective analysis shows survival and recurrence rates that confirm existing data. In this uncontrolled study, the omentum-preserving technique was associated with statistically lower recurrence rates.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Kitano

, Shiraishi

, Fujii

, et al. A randomized controlled trial comparing open vs laparoscopy-assisted distal gastrectomy for the treatment of early gastric cancer: An interim report. Surgery, 2002; 131,(1 Suppl):S306–S311.

Huscher

CGS

, Mingoli

, Sgarzini

, Sansonetti

, Paola

, Recher

, et al. Laparoscopic versus open subtotal gastrectomy for distal gastric cancer: Five-year results of a randomized prospective trial. Ann Surg, 2005; 241:232–237.

Hasegawa

, Yamamoto

, Taguri

, Morita

, Sato

, Yamada

, et al. A randomized phase II trial of omentum-preservation gastrectomy for advanced gastric cancer. Jpn J Clin Oncol, 2013; 43:214–216.

Hasegawa

, Kunisaki

, Ono

, Oshima

, Fujii

, Taguri

, et al. Omentum-preservation gastrectomy for advanced gastric cancer: A propensity-matched retrospective cohort study. Gastr Cancer, 2013; 16:383–388.

Goh

PMY

, Tekant

, Isaac

, et al. The technique of laparoscopic Billroth II gastrectomy. Surg Endosc Laparosc, 1992; 2:258–260.

Ohgami

, Otani

, Kumai

, Kubota

, Kim

, Kitajima

. Curative laparoscopic surgery for early gastric cancer: 5 Years experience. World J Surg, 1999; 23:187–192.

Ohashi

. Laparoscopic intraluminal (intragastric) surgery for early gastric cancer. A new concept in laparoscopic surgery. Surg Endosc, 1995; 9:169–171.

Azagra

, Goergen

, De Simone

, Ibañez-Aguirre

. Minimally invasive surgery for gastric cancer. Surg Endosc, 1999; 13:351–357.

Croce

, Olmi

, Magnone

, Russo

. [Laparoscopic surgery of the stomach: State of the art]. Chir Ital, 2003; 55:811–820.

10.

Song

, Shim

. Laparoscopic surgery for gastric cancer. In: Chapter 5 in Management of Gastric Cancer. IntechOpen, 2011, pp. 73–83.

11.

Kim

, Hyung

, Cho

, et al. Morbidity and mortality of laparoscopic gastrectomy versus open gastrectomy for gastric cancer: An interim report—A phase III multicenter, prospective, randomized trial (KLASS Trial). Ann Surg, 2010; 251:417–420.

12.

Kodera

, Fujiwara

, Ohashi

, Nakayama

, Koike

, Morita

, et al. Laparoscopic surgery for gastric cancer: A collective review with meta-analysis of randomized trials. J Am Coll Surg, 2010; 211:677–686.

13.

Kitano

, Shiraishi

, Uyama

, et al. A multicenter study on oncologic outcome of laparoscopic gastrectomy for early cancer in Japan. Ann Surg, 2007; 245:68–72.

14.

Song

, Lee

, Cho

, et al. Recurrence following laparoscopy assisted gastrectomy for gastric cancer: A multicenter retrospective analysis of 1,417 patients. Ann Surg Oncol, 2010; 17:1777–1786.

15.

Park do

, Han

, Hyung

, Kim

, Ryu

, et al. Korean Laparoscopic Gastrointestinal Surgery Study (KLASS) Group. Long-term outcomes after laparoscopy-assisted gastrectomy for advanced gastric cancer: A large-scale multicenter retrospective study. Surg Endosc, 2012; 26:1548–1553.

16.

Oosterling

, van der Bij

, Bögels

, van der Sijp

, Beelen

, Meijer

, et al. Insufficient ability of omental milky spots to prevent peritoneal tumor outgrowth supports omentectomy in minimal residual disease. Cancer Immunol Immunother, 2006; 55:1043–1051.

17.

Bosch

, Guller

, Schnider

, Maurer

, Harder

, Metzger

, et al. Perioperative detection of disseminated tumour cells is an independent prognostic factor in patients with colorectal cancer. Br J Surg, 2003; 90:882–888.

18.

Kim

, Kim

, et al. Decreased morbidity of laparoscopic distal gastrectomy compared with open distal gastrectomy for stage I gastric cancer. Short-term outcomes from a multicenter randomized controlled trial (KLASS-01). Ann Surg, 2016; 263:28–35.

19.

Huscher

, Mingoli

, Sgarzini

, et al. Laparoscopic versus open subtotal gastrectomy for distal gastric cancer: Five-year results of a randomized prospective trial. Ann Surg, 2005; 241:232–237.

20.

Lee

, Lee

, Son

, Ahn

, Park

, Kim

. Laparoscopic versus open gastrectomy for gastric cancer: Long-term oncologic results. Surgery, 2014; 155:154–164.

21.

Lee

, Kim

, Son

, Shin

, Yoo

. Laparoscopic-assisted total gastrectomy versus open total gastrectomy for upper and middle gastric cancer in short-term and long-term outcomes. Surg Laparosc Endosc Percutan Tech, 2014; 24:277–282.

22.

Olmi

, Giorgi

, Cioffi

SPB

, Uccelli

, Villa

, Ciccarese

, et al. Total and subtotal laparoscopic gastrectomy for the treatment of advanced gastric cancer: Morbidity and oncological outcomes. J Laparoendosc Adv Surg Tech A, 2018; 28:278–285.