Obesity is rapidly increasing on a global scale and is defined as excess fat accumulation, which can lead to health problems. The criteria for obesity is a body mass index (BMI) ≥30 kg/m² in Western countries and ≥25 kg/m² in Asian countries.¹ Globally, 39% and 13% of adults aged ≥18 years were overweight and obese, respectively, in 2014. According to the Korea National Health and Nutrition Examination Survey, the obesity rate in south Korea is increasing, from 25.8% in 1998 to 32.8% in 2012.² As obesity in the population increases, the proportion of obese patients with gastric cancer is also increasing.

Generally, gastric cancer surgeries in obese patients are technically demanding, showing limited oncologic safety. Large amounts of intra-abdominal fat can obscure vascular structures and normal surgical planes during surgery. Preoperative BMI is a risk factor for postoperative complications.³ Moreover, complete lymph node dissection is challenging during surgery in obese patients. These issues are equally challenging during laparoscopic surgery. In the past, laparoscopic surgery was not attempted in obese patients due to the above-mentioned problems. However, technical advancements have overcome these limitations. At this time, laparoscopic gastrectomy has advantages over open gastrectomy regarding postoperative outcomes and quality of life.^4-6 In the recently published KLASS-01 trial morbidity and mortality report, laparoscopic gastrectomy showed advantages in terms of estimated blood loss, length of hospital stay, and overall complication rate, especially for wound complication.⁷ However, laparoscopic gastrectomy requires a longer operation time than that of open gastrectomy. Recently, the discrepancy in operation time has decreased due to technical improvements.⁸ However, the advantages of laparoscopic gastrectomy in obese patients remain controversial. In this study, we explored the safety of laparoscopic gastrectomy in terms of early surgical outcomes in obese gastric cancer patients.

From January 2000 to December 2014, 4,272 patients with gastric adenocarcinoma underwent gastrectomy at the Department of Surgery, Seoul, St. Mary’s Hospital. Eligible criteria were curative resection, distal gastrectomy, pathologically confirmed early gastric cancer, and range of lymph node dissection D1+ or greater. Patients who underwent palliative or non-curative resection or combined resection were excluded. The data were collected retrospectively from a prospectively collected database. Obesity and morbid obesity were defined as BMI ≥25 and BMI ≥30, respectively, according to Asian-Pacific guidelines. Among the 4,274 patients, two cohorts were defined: patients with BMI ≥25 and those with BMI ≥30. A total of 462 patients were included in the former cohort and 41 patients in the latter cohort. Those cohorts were further divided into the laparoscopic distal gastrectomy (LDG) and open distal gastrectomy (ODG) groups. Clinicopathologic characteristics and surgical results included the range of lymph node dissection, early surgical results including postoperative complications, and length of hospital stay. Comparisons of continuous variables were performed using Student’s t-test and categorical variables were assessed using Pearson’s χ² test or a linear-by-linear association χ² test. Postoperative complications were graded based on the Clavien-Dindo classification.⁹ In addition, univariate and multivariate analyses were performed to identify risk factors for the development of postoperative complications. This study followed the 7th UICC TNM staging system (according to the American Joint Committee on Cancer [AJCC] 7th edition).¹⁰

Clinicopathological characteristics

In the BMI ≥25 cohort, no significant differences in patient characteristics including age, sex, ECOG, number of comorbidities, social history, or previous operation history were observed between the LDG and ODG groups. Tumor location, degree of differentiation, and N stage also showed no significant differences between the two groups. In the BMI ≥30 cohort, similar results were observed between the two groups (excluding mean tumor size). Tumor size in the LDG group was 2.6±1.5 cm, which was smaller than that (3.9±2.4 cm) in the ODG group (Table 1).

Table 1 . Clinicopathologic characteristics of the patients.

Variables	BMI ≥25 (kg/m2) (n=426)			BMI ≥30 (kg/m2) (n=41)
	LDG (%) (n=260)	ODG (%) (n=166)	p value†	LDG (%) (n=24)	ODG (%) (n=17)	p value†
Age (year)	58.9±11.2*	59.8±9.9*	0.422	57.8±10.1*	61.2±9.9*	0.299
Sex			0.662			1.000
Male	182 (70.0)	120 (72.3)		16 (66.7)	11 (64.7)
Female	78 (30.0)	46 (27.7)		8 (33.3)	6 (35.3)
BMI (kg/m2)	27.3±2.2*	27.3±1.9*	0.918	32.7±2.6*	31.5±1.6*	0.093
ECOG PS			0.736			0.115
0	184 (70.8)	120 (72.3)		19 (79.2)	9 (52.9)
1	68 (26.2)	39 (23.5)		5 (20.8)	7 (41.2)
2	7 (2.7)	7 (4.2)		0 (0)	1 (5.9)
3	1 (0.4)	0 (0)		0 (0)	0 (0)
Number of comorbidities			0.470			0.780
0	107 (41.2)	64 (38.6)		10 (41.7)	4 (23.5)
1	111 (42.7)	76 (45.8)		10 (41.7)	9 (52.9)
2	30 (11.5)	14 (8.4)		1 (4.2)	1 (5.9)
≥3	12 (4.6)	12 (7.2)		3 (12.5)	3 (17.6)
Smoking history			0.713			0.233
Non-smoker	150 (57.7)	94 (56.6)		13 (54.2)	8 (47.1)
Ex-smoker	60 (23.1)	35 (21.1)		6 (25.0)	8 (47.1)
Current smoker	50 (19.2)	37 (22.3)		5 (20.8)	1 (5.9)
Alcohol history			0.545			0.176
Non-drinker	145 (55.8)	91 (54.8)		13 (54.2)	7 (41.2)
Social drinker	92 (35.4)	55 (33.1)		8 (33.3)	10 (58.8)
Heavy drinker	23 (8.8)	20 (12.0)		3 (12.5)	0 (0)
Previous abdominal operation history			0.803			0.499
Absent	210 (80.8)	132 (79.5)		15 (62.5)	13 (76.5)
Present	50 (19.2)	34 (20.5)		9 (37.5)	4 (23.5)
Tumor location			0.116			0.285
Upper	5 (2.0)	6 (3.6)		0 (0)	2 (11.8)
Middle	30 (11.5)	29 (17.5)		3 (12.5)	2 (11.8)
Lower	225 (86.5)	131 (78.9)		21 (87.5)	13 (76.5)
Tumor size (mm)	2.6±1.5*	3.9±2.4*	< 0.001	2.6±1.7*	4.8±3.2*	0.005
Differentiation			0.058			0.760
Differentiated	153 (58.8)	82 (49.4)		14 (58.3)	6 (52.9)
Undifferentiated	107 (41.2)	84 (50.6)		10 (41.7)	8 (47.1)
N stage			0.648			0.247
N0	229 (88.1)	142 (85.5)		21 (87.4)	12 (70.6)
N1	20 (7.7)	13 (7.8)		1 (4.2)	4 (23.5)
N2	6 (2.3)	8 (4.8)		1 (4.2)	1 (5.8)
N3	5 (1.9)	3 (1.8)		1 (4.2)	0 (0)

BMI = Body Mass Index; ECOG = Eastern Cooperative Oncology Group..

^*Standard deviation.

^†Indicates χ² test.

Postoperative and surgical results

In terms of surgical outcomes, there were differences in the type of reconstruction and extent of lymph node dissection between the groups in the BMI ≥25 cohort and in the extent of lymph node dissection in the BMI ≥30 cohort. Billroth-II anastomosis was performed more commonly in the LDG than in the ODG group. In addition, more D2 lymph node dissections were performed in the ODG than in the LDG (Table 2).

Table 2 . Postoperative and surgical outcomes.

Variables	BMI ≥25 (kg/m2) (n=426)			BMI ≥30 (kg/m2) (n=41)
	LDG (%) (n=260)	ODG (%) (n=166)	p value†	LDG (%) (n=24)	ODG (%) (n=17)	p value†
Reconstruction type			< 0.001			0.110
B-I	45 (17.3)	15 (9.0)		3 (12.5)	4 (23.5)
B-II	166 (63.8)	147 (88.6)		13 (54.2)	12 (70.6)
R-Y	49 (18.8)	4 (2.4)		8 (33.3)	1 (5.9)
LND			< 0.001			0.012
D1+β	173 (66.5)	55 (33.1)		17 (70.8)	5 (29.4)
≥D2	87 (33.5)	111 (66.9)		7 (29.2)	12 (70.6)
Retrieved LNs	39.5±15.2*	39.8±15.2*	0.834	41.1±14.5*	39.7±15.2*	0.754
Combine resection			1.000			1.000
Yes	4 (1.5)	2 (1.2)		23 (95.8)	17 (100)
No	256 (98.5)	164 (98.8)		1 (4.2)	0 (0)
Operation time (min)	190.7±49.7*	200.5±41.1*	0.033	190.0±64.4*	206.8±50.1*	0.375
EBL (ml)	155.2±124.9*	192.4±162.1*	0.008	151.9±100.2*	187.9±119.3*	0.300
First flatus (day)	3.2±1.7*	3.1±1.0*	0.549	3.1±0.6*	3.4±0.5*	0.212
Postoperative hospital stay (days)	8.2±4.7*	8.0±2.8*	0.629	9.6±5.8*	7.9±1.4*	0.231

LND = Lymph Node Dissection; EBL = Excessive Blood Loss..

^*Standard deviation.

^†Indicates χ² test.

In the LDG group, 87 patients (33.5%) underwent >D2 lymph node dissection, and 173 patients (66.5%) underwent D1+ lymph node dissection. In the ODG group, 111 patients (66.9%) underwent >D2 lymph node dissection and 55 patients (33.1%) D1+ lymph node dissection. Nevertheless, the number of retrieved lymph nodes showed no significant difference between the two groups (LDG vs. ODG: 39.5±15.2 vs. 39.8±15.2). An extended operation time was required in the ODG group compared with the LDG group in the BMI ≥25 cohort. However, the difference was minimal, and there was no significant difference in the BMI ≥30 cohort. Less blood loss was observed in the LDG group (155.2±124.9 ml) than in the ODG group (192.4±162.1 ml) in the BMI ≥25 cohort. The postoperative clinical course showed no significant differences between the two groups in terms of flatus day or length of hospital stay.

Postoperative complications

There were more overall complications in the ODG than in the LDG group in both cohorts. Among the complications, significant differences were observed in the rates of Clavien-Dindo I and II complications (Table 3), but not Clavien-Dindo III and IV complications, between the two groups in both cohorts.

Table 3 . Postoperative morbidities within 30 postoperative days.

Variables	BMI ≥25 (kg/m2) (n=426)			BMI ≥30 (kg/m2) (n=41)
	LDG (%) (n=260)	ODG (%) (n=166)	p value†	LDG (%) (n=24)	ODG (%) (n=17)	p value†
Total number of postoperative morbidity	49 (18.8)	60 (36.1)	<0.001	5 (20.8)	11 (64.7)	0.009
Grade I, II	39 (15)	50 (30.1)		4 (16.7)	11 (64.7)
Grade III, IV	10 (3.8)	10 (6.0)		1 (4.2)	0 (0)
Grade I	14 (5.4)	21 (12.7)	0.011	0 (0)	4 (23.5)	0.024
Grade II	25 (9.6)	29 (17.5)	0.024	4 (16.7)	7 (41.2)	0.151
Grade IIIa	8 (3.1)	10 (6.0)	0.216	1 (4.2)	0 (0)	1.000
Grade IIIb	1 (0.4)	0 (0)	1.000	0 (0)	0 (0)	1.000
Grade IVa	1 (0.4)	0 (0)	1.000	0 (0)	0 (0)	1.000
Medical
Pulmonary	7 (2.7)	17 (10.2)		0 (0)	4 (23.5)
Cardiovascular	0 (0)	2 (1.2)		0 (0)	0 (0)
Hepatic	1 (0.4)	2 (1.2)		0 (0)	1 (5.8)
Urinary	3 (1.2)	1 (0.6)		0 (0)	0 (0)
Intra-abdominal
Abscess	0 (0)	1 (0.6)		0 (0)	0 (0)
Anastomosis bleeding	1 (0.4)	0 (0)		0 (0)	0 (0)
Anastomosis leakage	5 (1.9)	3 (1.8)		1 (4.2)	1 (5.8)
Anastomosis stenosis	3 (1.2)	1 (0.6)		0 (0)	0 (0)
Bleeding	8 (3.1)	2 (1.2)		0 (0)	0 (0)
Ileus	14 (5.4)	10 (6.0)		2 (8.3)	0 (0)
Pancreatitis	1 (0.4)	3 (0.8)		1 (4.2)	0 (0)
Wound complications	3 (1.2)	18 (10.8)		0 (0)	5 (29.4)
Others	3 (1.2)	0 (0)		1 (4.2)	0 (0)

^†Indicates χ² test.

The most common complications observed in the LDG group were ileus and bleeding, pulmonary complications, and anastomosis leakage. Pulmonary and wound complications were more common in the ODG group than in the LDG group and represented the main differences in terms of complications between the two groups. Based on risk factor evaluation, open distal gastrectomy, age >60 years, and BMI ≥30 were independent risk factors for postoperative morbidity among obese patients (Table 4).

Table 4 . Risk factors for postoperative morbidity.

Variables	Univariate analysis			Multivariate analysis
	Number of patients (n=426)	Morbidities (%)	p value‡	Odds ratio	p value‡
Operative approach			< 0.001		< 0.001
Open	166	60 (36.1)		1
Laparoscopy	260	49 (18.8)		0.431 (0.273~0.683)
Age (years)			0.004		0.004
<60	213	41 (19.2)		1
≥60	213	68 (31.9)		1.987 (1.252~3.153)
Sex			0.038		0.057
Male	302	86 (28.5)		1
Female	124	12 (18.5)		0.592 (0.344~1.020)
BMI			0.041		0.039
<30	385	93 (24.2		1
≥30	41	16 (39.0)		2.106 (1.039~4.269)
ECOG PS			0.547
0, 1	411	104 (25.3)
2,3	15	5 (33.3)
Comorbidities			1.000
Absent	420	170 (25.5)
Present	6	2 (33.3)
History of abdominal surgery			0.488
Absent	342	85 (24.9)
Present	84	24 (28.6)
Reconstruction type			0.100
B-I	60	18 (30.0)
B-II	313	72 (23.0)
R-Y	53	19 (35.8)
LND			1.000
D1+β	228	58 (25.4)
D2	198	51 (25.8)
Operation time (min)			0.045		0.158
<200	228	49 (21.5)		1
≥200	198	60 (30.3)		1.396 (0.879~2.216)
EBL (ml)			0.094
<150	195	42 (21.5)
≥150	231	67 (29.0)
pN stage			1.000
N0	371	95 (25.6)
N1~3	55	14 (25.5)

Significant values are indicated in bold. BMI = Body Mass Index; ECOG = Eastern Cooperative Oncology Group..

^‡Binary logistic regression.

Major advancements in laparoscopic surgery for gastric cancer have been achieved since Kitano et al. first introduced laparoscopy-assisted distal gastrectomy in 1994.¹¹ Subsequently, several retrospective and prospective studies have explored the surgical and oncologic safety of this technique. A meta-analysis of seven randomized controlled trials concluded that laparoscopic assisted gastrectomy significantly reduced blood loss and postoperative complications compared with open distal gastrectomy.¹²

Obese patients undergoing surgery have been examined in several reports. The postoperative complication rate varies among studies. Some studies have suggested that obesity does not affect postoperative morbidity or mortality.^13,14 However, a longer operation time has been noted in obese patients.⁷

This study evaluated the postoperative complication rate after laparoscopic distal gastrectomy and the safety of laparoscopic distal gastrectomy in patients with a high BMI. The overall complication rate was higher in the ODG than in the LDG group in both cohorts. Specifically, the major complication rates did not differ between the two groups. However, minor complications, especially pulmonary and wound complications, were more common in the ODG group. Although more patients underwent lymph node dissection of >D2 in the ODG group compared with the LDG group, the number of retrieved lymph nodes did not differ between the two groups. In the BMI ≥25 cohort, the ODG group required a longer operation time. However, in the BMI ≥30 cohort, no significant difference in operation time was observed between the groups. Estimated blood loss was lower in the LDG group, but the postoperative clinical course showed no significant difference.

Kim W et al. reported that the wound complication rate of the LDG group was significantly lower than that in the ODG group.⁷ In our study, wound complications were more common in the ODG group. In obese patients, the risk of wound infections is likely to increase due to abundant subcutaneous fat.

In this study, pulmonary complications were significantly less common in the laparoscopic surgery group. Although objective measurements of postoperative pain were not performed, less pain in the LDG group allowed for early ambulation and prevented pulmonary complications.

In general, surgery in obese patients is associated with technical difficulties and a high complication rate. Especially in surgeries for cancer, in which sufficient lymph node dissection is important for curative resection, a large amount of intra-abdominal fat may obscure anatomical structures and hamper lymph node dissection. Laparoscopic gastrectomy for stomach cancer in patients with a high BMI is technically challenging; thus, the safety of this approach should be explored.

Based on our results, laparoscopic surgery can be more difficult during lymph node dissection for obese patients than for non-obese patients. However, laparoscopic surgery is not inferior to open surgery in obese patients.

This study has several limitations. First, it was designed as a single-institutional retrospective study with a focus on early gastric cancer. Further randomized controlled trials conducted in large cohorts are required to confirm the results of this study. Second, we enrolled patients from 2000 to 2014. In early 2000, the majority of surgeons preferred open gastrectomy over laparoscopic gastrectomy. Over time, as laparoscopic instruments became more developed and surgeons’ experiences increased, the preference shifted to laparoscopy. Patients who underwent laparoscopic gastrectomy at an earlier date may have had higher complication rates. In addition, the surgeon’s preference may lead to selection bias. Third, to ensure the safety of laparoscopic gastrectomy in obese patients, a comparison between obese patients and normal- or low-weight patients is required. In addition, future studies should explore the long-term safety, including cancer recurrence and disease-free survival rates, of laparoscopic gastrectomy in obese patients with gastric cancer.

In conclusion, LDG may improve surgical outcomes in obese patients undergoing surgery for early gastric cancer in terms of less excessive blood loss, shorter operation times, and lower complication rates. The conventional open gastrectomy approach, older age, and higher BMI are associated with postoperative morbidity in obese patients with early gastric cancer.