Jpn J Clin Oncol 2014;44(5)408 – 415 doi:10.1093/jjco/hyu031 Advance Access Publication 9 April 2014
Original Articles
Does Overweight Affect Outcomes in Patients Undergoing Gastrectomy for Cancer? A Meta-analysis of 25 Cohort Studies Lun Li1,2, Xiuxia Li1, Shengping Chu3, Jinhui Tian1, Juan Su4, Hongliang Tian1,2, Rao Sun1 and Kehu Yang1,2,* 1
*For reprints and all correspondence: Kehu Yang, Dong Gang West Road No. 199, Lanzhou, Gansu, China. E-mail: [email protected]; Lun Li, [email protected] Received April 17, 2013; accepted March 3, 2014
Objective: Overweight was regarded as one of the risk factors for poor outcome after gastrectomy, but its influence on the surgical and postoperative outcomes of gastrectomy was unclear. Methods: Comprehensive searches were conducted to include cohort studies which evaluated the influence of overweight on the surgical and postoperative outcomes of gastrectomy. Data was analyzed by RevMan 5.0. Results: Twenty-five cohort studies (18 518 patients) were included. Overweight patients were associated with longer operation time (mean difference 20.88, 95% confidence interval 14.07, 27.69), more intraoperative blood loss (mean difference 35.45, 95% confidence interval 9.24, 61.67), and less retrieved lymph nodes (mean difference 22.17, 95% confidence interval 23.51, 20.83) than normal patients undergoing laparoscopy-assisted gastrectomy. And overweight patients were associated with longer operation time (mean difference 26.31, 95% confidence interval 21.92, 30.70), more intraoperative blood loss (mean difference 130.02, 95% confidence interval 75.49, 184.55), less retrieved lymph nodes (mean difference 23.18, 95% confidence interval 24.74, 21.61), longer postoperative hospital stay (mean difference 2.37, 95% confidence interval 0.03, 4.70) and more postoperative complications (risk ratio 1.53, 95% confidence interval 1.29, 1.80) than normal patients in open gastrectomy. Conclusions: Overweight might affect the clinical results of both laparoscopy-assisted and open gastrectomy, especially for open gastrectomy. Key words: overweight – gastrectomy – meta-analysis
Adiposity is associated with increased morbidity and disability (1). Body mass index (BMI) (calculated as weight [kg]/ height [m2]) is commonly used to classify adiposity as overweight (BMI 25) and obesity (BMI 30) in adults (2). In 2008 an estimated 1.5 billion adults worldwide were overweight and 500 million were obese (1). A recent meta-analysis showed statistically significant associations for overweight with the incidence of 18 co-morbidities, including diabetes, cancers, cardiovascular diseases, asthma, gallbladder disease, osteoarthritis and chronic back pain (3). Each five units above the overweight category (BMI .25) are associated with 30% higher overall mortality and 40% higher for cardiovascular mortality (1,4). It is reported by World Health
Organization (WHO) that at least 2.8 million people die every year because of being overweight or obese worldwide, and an estimated 35.8 million (2.3%) of global Disability Adjusted Life Years are caused by overweight (2). In the context of surgical procedures, overweight itself is regarded as one risk factor for poor outcomes after complicated surgical procedures such as gastrectomy (5). It is reported that overweight was associated with unfavorable surgical outcomes, such as higher postoperative complication rates, more intraoperative blood loss, longer operation time and mortality and for both open and laparoscopy-assisted gastrectomy (6 – 11). However, the influence of overweight on the surgical and postoperative outcomes of gastrectomy were not
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou, The First Clinical College of Lanzhou University, Lanzhou, 3Department of Gastroenterology, First Affiliated Hospital of Dalian Medical University, Dalian and 4Shuocheng District People’s Hospital, Shuozhou, China 2
Jpn J Clin Oncol 2014;44(5)
clear, as studies (6 – 11) reported conflicted results for operation time and intraoperative blood loss. Meanwhile, the differences of overweight on the surgical and postoperative outcomes between open and laparoscopy-assisted gastrectomy were not clear, and no such a meta-analysis was conducted. As a result, we conducted this meta-analysis to comprehensively evaluate the influence of overweight on the surgical and postoperative outcomes of gastrectomy.
409
DATA ABSTRACTION Data were entered into an Excel database by two independent authors (L.L. and H.T.). The following fields were abstracted: country, patient characteristics, disease, follow-up duration and outcomes. Any disagreements were resolved by a third reviewer (K.Y.). DATA ANALYSIS AND SUBGROUP ANALYSIS
We conducted this systematic review in accordance with Guidelines for Meta-Analyses and Systematic Reviews of Observational Studies (MOOSE) (12). SEARCH STRATEGY PubMed, the Cochrane library, EMBASE, ISI Web of Knowledge, and Chinese Biomedical Database were searched using overweight, adiposity, BMI, gastrectomy and gastrectomies. Some studies used obese or obesity instead of overweight where the patients’ BMIs were ,30, but .25, so obese or obesity was also searched as a supplement. If possible, subject heading terms such as Medical Subject Headings terms were added. Reference lists of the metaanalysis, review articles about this topic and identified trials were hand-searched to identify further relevant citations. All the searches were conducted in August 2012 without language, date and publication status restrictions.
Data were analyzed by Review Manager Version 5.0. For dichotomous outcomes, results were expressed as risk ratio (RR) with 95% confidence interval (CI). If there were continuous scales of measurement, the mean difference (MD) was used to assess the effects of treatment. The percentage of variability across trials attributable to heterogeneity beyond chance was estimated with the I2 statistic, which was deemed significant when P was ,0.05 or I-square was .50%. Data were pooled using the fixed-effect model but the random-effects model was also considered in the case of significant heterogeneity. In our study, we conducted subgroup analyses of different kinds of gastrectomy (laparoscopy-assisted vs. open gastrectomy) and study types (prospective vs. retrospective cohort studies). Meanwhile, sensitive analysis was also conducted for those studies which applied WHO classification. We used a funnel plot based on operation time to evaluate publication bias. The small-study effect in terms of publication bias was also estimated using Egger’s linear regression test.
RESULT SEARCH RESULT
INCLUSION CRITERIA AND STUDY SELECTION Cohort studies which evaluated the influence of overweight on the surgical and postoperative outcomes of gastrectomy were included. The definition of overweight was not restricted by WHO definitions (BMI 25), but according to different countries’ condition. That is why we also conducted meta-analysis for overweight according to WHO definitions (BMI 25). The outcomes we evaluated were intraoperative outcomes (length of the operation, intraoperative blood loss, number of retrieved lymph nodes and lymph node dissection extent), postoperative short-term outcomes (postoperative mortality, time of first flatus, postoperative hospital stay and postoperative complications), and postoperative long-term outcomes (recurrence and 5-year survival rate). The postoperative complications we evaluated included anastomotic leak, abscess, pancreatic fistula, wound infection, postoperative bleeding and bowel obstruction. Studies that reported the following one or more outcomes were included; letters, comments, editorials, practice guidelines and trials published without the outcome measures of interest were excluded. Two reviewers (L.L and X.L.) independently assessed potentially relevant citations for inclusion, disagreements were resolved involved with a third reviewer (K.Y.).
After comprehensive searches, we found 1058 citations. We excluded 395 duplicates and 574 citations based on screening the titles and abstracts. After screening full texts, we excluded 64 citations (32 were not about gastrectomy, 21 were not about overweight and 11 were reviews). Finally, 25 cohort studies (5,7–11,13–31) (18 518 patients) were included (Fig. 1). CHARACTERISTICS OF INCLUDED TRIALS Of these 25 trials, they were from Japan (n ¼ 15), Korea (n ¼ 5), China (n ¼ 2), Germany (n ¼ 1), Poland (n ¼ 1) and UK (n ¼ 1). Of them, five were prospective cohort studies and two were randomized trials) and 20 were retrospective cohort studies. Eight of them were laparoscopy-assisted gastrectomy and 17 were open gastrectomy; all of them were for gastric cancer patients. The other characteristics of included trials were presented in Table 1. THE TOTAL RESULTS BASED ON META-ANALYSIS STUDIES
OF ALL INCLUDED
INTRAOPERATIVE OUTCOMES Overweight was associated with longer operation time (MD 24.71, 95% CI 21.02, 28.40), more intraoperative blood loss
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
PATIENTS AND METHODS
410
Overweight on outcomes of gastrectomy
postoperative complications including anastomotic leak, abscess, pancreatic fistula and longer postoperative hospital stay than normal patients in open gastrectomy, but not for laparoscopy-assisted gastrectomy (Table 3). SUBGROUP ANALYSIS OF PROSPECTIVE AND RETROSPECTIVE COHORT STUDIES
META-ANALYSIS OF THE STUDIES THAT DEFINED BMI >25 AS OVERWEIGHT ACCORDING TO WHO CRITERION
Figure 1. The flow chart.
(MD 105.50, 95% CI 66.76, 144.24), less retrieved lymph nodes (MD 22.87, 95% CI 23.96, 21.77) and more D1 lymph node dissection (RR 1.14, 95% CI 1.02, 1.27), but not with more D2 lymph node dissection (RR 0.98, 95% CI 0.91, 1.05) (Table 2). POSTOPERATIVE SHORT-TERM OUTCOMES Overweight was associated with more postoperative complications (RR 1.45, 95% CI 1.25, 1.69), but not with postoperative mortality (RR 1.22, 95% CI 0.80, 1.85), time of first flatus (MD 0.12, 95% CI 20.03, 0.27), and postoperative hospital stay (MD 0.20, 95% CI 20.99, 1.39) (Table 3). For postoperative complications, overweight was associated with more anastomotic leak (RR 1.68, 95% CI 1.15, 2.47), abscess (RR 2.31, 95% CI 1.73, 3.09) and pancreatic fistula (RR 2.18, 95% CI 1.67, 2.85), but not with more wound infection (RR 1.16, 95% CI 0.81, 1.66), postoperative bleeding (RR 1.21, 95% CI 0.77, 1.88), and bowel obstruction (RR 1.27, 95% CI 0. 71, 2.28) (Table 3).
Analyzing those studies which defined BMI .25 as overweight according to WHO criterion, the pooled results did not change and showed that overweight patients were associated with longer operation duration, more intraoperative blood loss, less retrieved lymph nodes and higher postoperative complication rates Supplementary Table 2. Subgroup analysis showed that overweight patients were associated with longer operation time, more intraoperative blood loss, and less retrieved lymph nodes for both laparoscopy-assisted and open gastrectomy. But overweight patients were associated with more postoperative complications in open gastrectomy. The details were presented in Supplementary Table 2 (online). PUBLICATION BIAS There seemed to be a significant publication bias based on funnel plot (Fig. 2). Egger’s test indicated that there was a possibility of publication bias for operation time (intercept 1.17, 95% CI 0.00 2.34, P ¼ 0.05).
DISCUSSION SUMMARY OF FINDINGS
POSTOPERATIVE LONG-TERM OUTCOMES Overweight was not associated with higher 5-year survival rates for Stages I, II and III gastric cancer and recurrence (including all recurrence, local recurrence, distant recurrence, both local and distant recurrence) (Table 4). SUBGROUP ANALYSIS OF LAPAROSCOPY-ASSISTED OPEN GASTRECTOMY
AND
Overweight patients were associated with longer operation time, more intraoperative blood loss, and less retrieved lymph nodes for laparoscopy-assisted and open gastrectomy (Table 2). But overweight was associated with more
Overweight patients were associated with longer operation time, more intraoperative blood loss and less retrieved lymph nodes than normal patients in both laparoscopy-assisted and open gastrectomy. Overweight affected the surgical results and postoperative outcomes via many operative factors, such as excessive adipose tissue (gastric, omental and perigastric fat), comorbidities, etc. (14,21,29). The biggest factor in overweight patients was adipose tissue, and this could result in longer operation time, more intraoperative blood loss and more postoperative complications because of technical difficulties accessing and dissecting lymph nodes deeply embedded in fatty tissues around the major abdominal vessels
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Overweight patients were associated with longer operation time, less retrieved lymph nodes and more postoperative complications based on both retrospective and prospective cohort studies. Overweight was associated with higher D1 lymph node dissection based on prospective studies and more intraoperative blood loss based on retrospective studies. The details were presented in Supplementary Table 1.
Table 1. Characteristics of included trials Study
Country
Study type
Overweight
Surgery
SZ E
Adachi et al. (1995) (13)
Follow-up C
Sex (M/F)
Age (years)
E
C
E
C
29–129 months
44/18
136/65
62.1
62
11–43% .Ideal
TG, DG
62
201
PC
.24.7 (M), .22.6 (F)
D2 G
38
46
36 (12–84) months
27/12
31/14
69 (44– 83)
71 (27– 86)
RC
25
TG, D2 G
81
82
–
51/30
41/41
61.3
57.4
695
^2 years
55/ 37
470/225
–
–
239
4.77 years
34/20
153/86
58.7
60.17
–
25/20
29/23
55 + 11
56 + 12
104 months
331/161
1033/467
59
60
Japan
RC
Barry et al. (2003) (14)
UK
Gretschel et al. (2003) (15)
Germany
Dhar et al. (2000) (29)
Japan
RC
.24.7 (M), .22.6 (F)
PG, TG
92
Inagawa et al. (2000) (17)
Japan
RC
25
D2 G
54
Kim et al. (2006) (18)
Korea
RC
23
LADG
Kulig et al. (2010) (10)
Poland
RC
25
G
Kunisaki et al. (2009) (31)
Japan
RC
25
LADG
29
123
–
–
–
–
–
Lee et al. (2009) (28)
Korea
PC
25
LADG
432
1053
–
279/153
653/400
59.4 + 10.8
56.9 + 12.5
394
153
–
269/125
88/65
59.78 + 12
60.8 + 12.56
99
545
–
77/22
371/174
62 + 11.3
62 + 10.5
50 (2– 95) months
70/29
210/101
–
–
–
19/6
19/18
56.42 + 11.50
52.30 + 10.91
Moriwaki et al. (2003) (20)
Japan
RC
18.5– 21
D2 G
Nobuoka et al. (2011) (9)
Japan
RC
25
TG
45
52
492
1500
Oh et al. (2009) (11)
Korea
RC
25
TG
99
311
Oh et al. (2012) (21)
Korea
PC
25
TG
24
37
Oki et al. (2012) (7)
Japan
RC
25
LADG
20
118
–
12/8
63/55
63.5 + 11.1
63.4 + 11.2
Tsujinaka et al. (2007) (5)
Japan
PC
25
G
77
446
–
57/20
301/145
–
–
27.29 + 2.4
21.53 + 2.2
Japan
RC
25
DG
China
RC
25
G
Kodera et al. (2004) (19)
Japan
RC
27
D2 G
Ojima et al. (2009) (23)
Japan
RC
25
DG, TG
60
188
45 (8– 118) months
46/14
134/54
103
468
5y
404/167
62.7 + 12.3
44
395
–
16/34
271/124
59.2
60.17
116
573
^5 years
92/24
405/168
63 + 12
64 + 12
19 (2– 39) months
68/15
325/123
60.7 + 9.7
59.7 + 11.3
–
37/29
23/27
55.8 + 11.6
55.0 + 10.5
Chen et al. (2011) (22)
China
RC
25
LADG
83
325
Shim et al. (2009) (24)
Korea
PC
23
LADG
60
56
Tokunaga et al. (2009) (30)
Japan
RC
25
TG, PG
1126
6799
–
825/301
4590/2290
59.1 + 11.6
58.5 + 11.8
Nozoe et al. (2012) (25)
Japan
RC
25
TG, PG
44
264
–
189/75
28/16
65.6 + 9.9
67.8 + 10.9
Yasuda et al. (2004) (27)
Japan
RC
25
LADG
16
83
48 months
8/8
48/35
66.4 + 10.9
66.7 + 10.0
Noshiro et al. (2003) (26)
Japan
RC
24.2
LADG
19
57
24 (1– 75) months
18/1
39/18
64 + 10
62 + 12
SZ, sample size; E, overweight group; C, control group; RC, Retrospective cohort; PC, prospective cohort; GC, gastric cancer; G, gastrectomy; LADG, laparoscopy-assisted distal gastrectomy; DG, distal gastrectomy; TG, total gastrectomy; PG, Partial gastrectomy; D2 G, gastrectomy with D2 lymphadenectomy.
Jpn J Clin Oncol 2014;44(5)
Yamada et al. (2008) (8) Gu and Zhang (2012) (16)
411
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
412
Overweight on outcomes of gastrectomy
Table 2. Subgroup analysis of laparoscopy-assisted versus open gastrectomy about intraoperative outcomes Outcomes
Studies
Sample size E
Operation time Laparoscopic
20 (7– 9,11,13–28) 7 (7,8,18,24,26–28)
Effect size (95% CI)
Heterogeneity I2 (%) (P)
C
1857
5276
MD 24.71 (21.02, 28.40)
34 (0.05)
621
1522
MD 20.88 (14.07, 27.69)
14 (0.32)
Open
14 (8,9,11,13– 17,19–23,25)
1236
3754
MD 26.31 (21.92, 30.70)
39 (0.05)
Intraoperative blood loss
14 (7– 9,13,15–17,19,20,22,23,25– 27)
1159
3721
MD 105.50 (66.76, 144.24
80 (0.00001)
4 (7,8,26,27) 11 (8,9,13,15– 17,19,20,22,23,25)
74
337
1085
3384
MD 35.45 (9.24, 61.67) MD 130.02 (75.49, 184.55)
0 (0. 71) 84 (0.00001)
D1 LN dissection
5 (7,10,18,24,25)
661
1990
RR 1.14 (1.02, 1.27)
0 (0.44)
Laparoscopic
3 (7,18,24)
125
226
RR 1.20 (0.94, 1.54)
51 (0.13)
Open
2 (10,25)
D2 LN dissection
6 (5,7,10,18,24,25)
536
1764
RR 1.12 (0.95, 1.31)
0 (0.46)
1170
3489
RR 0.98 (0.91, 1.05)
13 (0.33)
Laparoscopic
3 (7,18,24)
557
1279
RR 0.95 (0.87, 1.05)
45 (0.14)
Open
3 (5,10,25)
613
2210
RR 1.01 (0.91, 1.12)
0 (0.90)
E, overweight group; C, control group; LN, lymph node.
(2,11,13,14,28,29). For example, the N2 and N3 regional lymph nodes lie deep within the fatty mesentery around the major abdominal vessels which hampered complete removal in overweight patients (14,29). So extensive lymph node dissections on overweight patients have often been unsuccessful and the number of resected or retrieved lymph nodes (an indicator of adequacy of lymphadenectomy) was significantly smaller for the obese patient group (13,19,25,29). The presence of excessive subcutaneous fat predisposed the obese to impaired wound healing and thus wound infections and anastomotic leakage (32). Another factor that might be linked to higher postoperative complication rates is the co-morbidities such as hypertension, diabetes mellitus, coronary heart disease and respiratory dysfunction (11,13). However, the postoperative complications in laparoscopyassisted gastrectomy did not differ between overweight and normal patients. This suggested that laparoscopy-assisted gastrectomy for gastric cancer patients has advantages compared with a conventional open method. Recent meta-analyses showed that laparoscopy-assisted gastrectomy has the advantages of minimal invasion, faster recovery and fewer complications compared with open gastrectomy with the same short-term prognosis (33 –35). Overweight was hypothesized to be associated with a poorer prognosis due to associated co-morbidities and insufficient lymph node dissection (11,13,29). However, our meta-analysis failed to confirm it. Available evidence for 5-year survival rates for Stages I, II, III gastric cancer was based on two small studies (11,20). Their sample size was too small to detect the differences in the long-term survival between overweight and normal patients. Although another four studies (10,14,29,30) also reported the 5-year survival
rates, their included patients were in different stages. Three studies (10,14,29) of small sample size showed no differences between overweight and normal patients, while one large study (30) showed higher 5-year survival rate for overweight patients. All relevant characteristics were balanced in the three small studies (10,14,29), but not in the largest one (30). The number of patients with IA (UICC) was more in overweight group than that in normal, while the number of patients with IIIA, IIIB and IV was less in overweight group than that in normal. The unbalances were also seen in histology and surgical procedure. These unbalanced characteristics might lead to the unexpected results.
STRENGTH AND LIMITATIONS Our meta-analysis was the first one which evaluated the influence of overweight on the surgical and postoperative outcomes for both laparoscopy-assisted and open gastrectomy. However, our meta-analysis has several limitations. First, most the studies were retrospective studies (n ¼ 20) and from Asian countries (Japan, Korea and China; n ¼ 22). Credibility of meta-analysis data is defined by the type and quality of the trials included in the study. It is taken for granted that data and results from retrospective studies are heavily biased. Second, the definitions of overweight across different countries were different. Even though most studies used BMI .25 to judge whether a patient was overweight or not, but some other studies also used 23, 27 or other to define the overweight. This might be due to the nutritional condition and diet in their countries and could introduce clinical heterogeneity in our meta-analysis.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Laparoscopic Open
Jpn J Clin Oncol 2014;44(5)
413
Table 3. Subgroup analysis of laparoscopy-assisted versus open gastrectomy on postoperative short-term outcomes Outcomes
Studies
Sample size E
Effect size (95% CI)
Heterogeneity I2 (%) (P)
57 (0.01)
C
Length of hospital stay
9 (7,9,11,18,21,24,26–28)
891
2757
MD 0.20 (20.99, 1.39)
Laparoscopic
6 (7,18,24,26– 28)
586
1416
MD 20.64 (21.43, 0.15)
Open Postoperative mortality
2 (8,28) 10 (5,9– 11,13,15,16,19,23,29)
305
1341
MD 2.37 (0.03, 4.70)
59 (0.06)
2883
13 256
RR 1.22 (0.80, 1.85)
0 (0.81)
492
1241
RR 1.63 (0.49, 5.41)
0 (0.49)
2391
12 015
RR 1.17 (0.75, 1.83)
0 (0.73)
Time of first flatus
7 (11,18,22,24,26–28)
748
2057
MD 0.12 (20.03, 0.27)
48 (0.07)
Laparoscopic
5 (18,24,26–28)
566
1298
MD 0.15 (20.10, 0.40)
62 (0.03)
Open Postoperative complications Laparoscopic
2 (11,22) 23 (5,7,9 –11,13–15,17–31)
759 14 273
MD 0.13 (20.02, 0.27) RR 1.45 (1.25, 1.69)
0 (0.56) 66 (,0.00001)
615
1539
RR 1.16 (0.75, 1.79)
40 (0.13)
Open
16 (5,9– 11,13– 15,17,19– 23,25,29,30)
2925
12 734
RR 1.53 (1.29, 1.80)
72 (,0.00001)
Anastomotic leak
14 (5,8– 11,14– 16,18,20,23,25–28)
2084
5736
RR 1.68 (1.15, 2.47)
19 (0.25)
541
1348
RR 1.02 (0.44, 2.34)
0 (0.54)
Laparoscopic Open Abscess
7 (7,18,24,26– 28,31)
182 3540
5 (8,18,26–28) 9 (5,9– 11,14– 16,20,23,25) 12 (5,8,10,15–20,23,24,26)
Laparoscopic
4 (8,18,24,26)
Open
8 (5,10,15–17,19,20,23)
Wound infection Laparoscopic Open Postoperative bleeding
10 (7 –11,13,14,18,24,27) 5 (7,8,18,24,27) 5 (9 –11,13,14) 10 (7,10,11,15,17,18,20,24,26,28)
1543
4388
RR 1.89 (1.21, 2.96)
29 (0.18)
1514
4124
RR 2.31 (1.73, 3.09)
0 (0.78)
153
268
RR 0.93 (0.27, 3.17)
0 (0.93)
1361
3856
RR 2.47 (1.83, 3.34)
0 (0.72)
966
3018
RR 1.16 (0.81, 1.66)
0 (0.98)
176
415
RR 0.65 (0.23, 1.85)
0 (0.97)
790
2603
RR 1.27 (0.86, 1.87)
0 (0.95)
1690
3618
RR 1.21 (0.77, 1.88)
0 (0.96)
Laparoscopic
5 (7,18,24,26,28)
570
1333
RR 1.28 (0.73, 2.23)
0 (0.90)
Open
5 (10,11,15,17,20)
1120
2285
RR 1.10 (0.53, 2.28)
0 (0.73)
415
1765
RR 2.18 (1.67, 2.85)
55
224
RR 0.69 (0.09, 5.62)
0 (0.89)
Pancreatic fistula
6 (5,7– 9,15,16)
Laparoscopic
2 (7,8)
Open
4 (5,9,15,16)
Bowel obstruction
7 (7,11,13,15,20,24,28)
36 (,0.00001)
360
1541
RR 2.26 (1.72, 2.96)
55 (0.08)
1148
1974
RR 1.27 (0. 71, 2.28)
0 (0.69)
Laparoscopic
3 (7,24,28)
512
1227
RR 1.72 (0.84, 3.53)
0 (0.40)
Open
4 (11,13,15,20)
636
747
RR 0.72 (0.25, 2.12)
0 (0.93)
E, overweight group; C, control group.
IMPLICATIONS FOR RESEARCH AND PRACTICE As most studies were from Asian countries and retrospective studies, future prospective studies in other continental countries were needed. Some trials that used ‘obese’ or ‘obesity’ for those patients whose BMIs were .25 and ,30. So the definition of overweight and obesity should be well classified, and studies that assessed the effects of obesity for gastrectomy should be conducted. In the future, when conducting gastrectomy for overweight gastric cancer patients, surgeons should retrieve metastatic
lymph nodes as much as possible and be aware of the postoperative complications when conducting open gastrectomy. If possible, laparoscopy-assisted gastrectomy should be applied to overweight gastric cancer patients because of lower postoperative complication rates (safety), less intraoperative blood loss (safety), shorter operation time and postoperative hospital stay (more economic), and more retrieved lymph nodes (better survival) when compared with open gastrectomy. So for overweight gastric cancer patients, laparoscopy-assisted gastrectomy seems to be another better treatment.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Laparoscopic Open
3 (9,11,21) 12 (5,8– 11,13,15,16,19,23,28,29)
0 (0.52)
414
Overweight on outcomes of gastrectomy
Table 4. Subgroup analysis of laparoscopy-assisted versus open gastrectomy on postoperative long-term outcomes Outcomes
Studies
Effect size (95% CI) Heterogeneity I2 (%) (P)
Patient no. E
C
5-Year survival rate for Stage I gastric cancera
2 (11,20)
99
112
RR 0.99 (0.88, 1.12)
5-Year survival rate for Stage II gastric cancera
2 (11,20)
220
133
RR 1.05 (0.99, 1.10) 93 (0.0002)
5-Year survival rate for Stage III gastric cancera
2 (11,20)
174
219
RR 0.97 (0.78, 1.21)
All recurrenceb
3 (11,14,29)
200
777
RR 1.24 (0.78, 1.99) 69 (0.04)
Local recurrenceb
3 (11,14,29)
200
777
RR 1.60 (0.94, 2.74)
Distant recurrenceb
3 (11,14,29)
200
777
RR 0.82 (0.50, 1.34) 33 (0.22)
Both local and distant recurrenceb
3 (11,14,29)
115
550
RR 1.36 (0.63, 2.94) 10 (0.33)
0 (0.90)
0 (0.89)
a
Laparoscopy assisted. Open; E: overweight group; C: control group.
b
Figure 2. Publication bias.
CONCLUSION Overweight might affect the clinical results of gastrectomy, especially for open gastrectomy. Most studies were retrospective cohort studies and from Asian countries, so future prospective studies on other continental countries were needed.
Supplementary data Supplementary data are available at http://www.jjco.oxfordjournals.org.
Conflict of interest statement None declared.
References 1. Nguyen T, Lau DC. The obesity epidemic and its impact on hypertension. Can J Cardiol 2012;28:326– 33.
2. WHO: Global Health Observatory (GHO) http://wwwwhoint/gho/ncd/ risk_factors/overweight_text/en/indexhtml 2012. 3. Guh DP, Zhang W, Bansback N, et al. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009;9:88. 4. Whitlock G, Lewington S, Sherliker P, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet 2009;373:1083– 96. 5. Tsujinaka T, Sasako M, Yamamoto S, et al. Influence of overweight on surgical complications for gastric cancer: results from a randomized control trial comparing D2 and extended para-aortic D3 lymphadenectomy (JCOG9501). Ann Surg Oncol 2007;14:355–61. 6. Lee JH, Paik YH, Lee JS, et al. Abdominal shape of gastric cancer patients influences short-term surgical outcomes. Ann Surg Oncol 2007;14:1288– 94. 7. Oki E, Sakaguchi Y, Ohgaki K, et al. The impact of obesity on the use of a totally laparoscopic distal gastrectomy in patients with gastric cancer. J Gastric Cancer 2012;12:108–12. 8. Yamada H, Kojima K, Inokuchi M, et al. Effect of obesity on technical feasibility and postoperative outcomes of laparoscopy-assisted distal gastrectomy—comparison with open distal gastrectomy. J Gastrointest Surg 2008;12:997– 1004. 9. Nobuoka D, Gotohda N, Kato Y, et al. Influence of excess body weight on the surgical outcomes of total gastrectomy. Surg Today 2011;41:928–34. 10. Kulig J, Sierzega M, Kolodziejczyk P, et al. Implications of overweight in gastric cancer: a multicenter study in a Western patient population. Eur J Surg Oncol 2010;36:969– 76. 11. Oh SJ, Hyung WJ, Li C, et al. Effect of being overweight on postoperative morbidity and long-term surgical outcomes in proximal gastric carcinoma. J Gastroenterol Hepatol 2009;24:475–9. 12. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008– 12. 13. Adachi W, Kobayashi M, Koike S, et al. The influence of excess body weight on the surgical treatment of patients with gastric cancer. Surg Today 1995;25:939– 45. 14. Barry JD, Blackshaw GR, Edwards P, et al. Western body mass indices need not compromise outcomes after modified D2 gastrectomy for carcinoma. Gastric Cancer 2003;6:80–5. 15. Gretschel S, Christoph F, Bembenek A, et al. Body mass index does not affect systematic D2 lymph node dissection and postoperative morbidity in gastric cancer patients. Ann Surg Oncol 2003;10:363–8. 16. Gu JS, Zhang WH. Effect of overweight on patients undergoing gastrectomy for gastric cancer. J China Med Univ 2012;40:372– 5. 17. Inagawa S, Adachi S, Oda T, et al. Effect of fat volume on postoperative complications and survival rate after D2 dissection for gastric cancer. Gastric Cancer 2000;3:141– 4.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
0 (0.44)
Jpn J Clin Oncol 2014;44(5)
27. Yasuda K, Inomata M, Shiraishi N, et al. Laparoscopy-assisted distal gastrectomy for early gastric cancer in obese and nonobese patients. Surg Endosc 2004;18:1253 –6. 28. Lee HJ, Kim HH, Kim MC, et al. The impact of a high body mass index on laparoscopy assisted gastrectomy for gastric cancer. Surg Endosc 2009;23:2473– 9. 29. Dhar DK, Kubota H, Tachibana M, et al. Body mass index determines the success of lymph node dissection and predicts the outcome of gastric carcinoma patients. Oncology 2000;59:18 –23. 30. Tokunaga M, Hiki N, Fukunaga T, et al. Better 5-year survival rate following curative gastrectomy in overweight patients. Ann Surg Oncol 2009;16:3245 –51. 31. Kunisaki C, Makino H, Takagawa R, et al. Predictive factors for surgical complications of laparoscopy-assisted distal gastrectomy for gastric cancer. Surg Endosc 2009;23:2085–93. 32. Mullen JT, Davenport DL, Hutter MM, et al. Impact of body mass index on perioperative outcomes in patients undergoing major intra-abdominal cancer surgery. Ann Surg Oncol 2008;15:2164–72. 33. Ohtani H, Tamamori Y, Noguchi K, et al. Meta-analysis of laparoscopy-assisted and open distal gastrectomy for gastric cancer. J Surg Res 2011;171:479– 85. 34. Liang Y, Li G, Chen P, et al. Laparoscopic versus open gastrectomy for early distal gastric cancer: a meta-analysis. ANZ J Surg 2011;81:673–80. 35. Ding J, Liao GQ, Liu HL, et al. Meta-analysis of laparoscopy-assisted distal gastrectomy with D2 lymph node dissection for gastric cancer. J Surg Oncol 2012;105:297 –303.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
18. Kim KH, Kim MC, Jung GJ, et al. The impact of obesity on LADG for early gastric cancer. Gastric Cancer 2006;9:303–7. 19. Kodera Y, Ito S, Yamamura Y, et al. Obesity and outcome of distal gastrectomy with D2 lymphadenectomy for carcinoma. Hepatogastroenterology 2004;51:1225– 8. 20. Moriwaki Y, Kunisaki C, Kobayashi S, et al. Does body mass index (BMI) influence morbidity and long-term survival in gastric cancer patients after gastrectomy? Hepatogastroenterology 2003;50:284–8. 21. Oh CA, Kim DH, Oh SJ, et al. Impact of body mass index on surgical outcomes in radical total gastrectomy. Hepatogastroenterology 2012;59:934–7. 22. Chen JX, Huang CM, Zheng CH, et al. Impact of obesity on laparoscopic-assisted radical gastrectomy for gastric cancer. Zhonghua Wei Chang Wai Ke Za Zhi 2011;14:781– 4. 23. Ojima T, Iwahashi M, Nakamori M, et al. Influence of overweight on patients with gastric cancer after undergoing curative gastrectomy: an analysis of 689 consecutive cases managed by a single center. Arch Surg 2009;144:351– 8. discussion 358. 24. Shim JH, Song KY, Kim SN, et al. Laparoscopy-assisted distal gastrectomy for overweight patients in the Asian population. Surg Today 2009;39:481–6. 25. Nozoe T, Kohno M, Iguchi T, et al. Analysis of the impact of the body mass index in patients with gastric carcinoma. Surg Today 2012;42:945–9. 26. Noshiro H, Shimizu S, Nagai E, et al. Laparoscopy-assisted distal gastrectomy for early gastric cancer: is it beneficial for patients of heavier weight? Ann Surg 2003;238:680–5.
415
Original Articles
Does Overweight Affect Outcomes in Patients Undergoing Gastrectomy for Cancer? A Meta-analysis of 25 Cohort Studies Lun Li1,2, Xiuxia Li1, Shengping Chu3, Jinhui Tian1, Juan Su4, Hongliang Tian1,2, Rao Sun1 and Kehu Yang1,2,* 1
*For reprints and all correspondence: Kehu Yang, Dong Gang West Road No. 199, Lanzhou, Gansu, China. E-mail: [email protected]; Lun Li, [email protected] Received April 17, 2013; accepted March 3, 2014
Objective: Overweight was regarded as one of the risk factors for poor outcome after gastrectomy, but its influence on the surgical and postoperative outcomes of gastrectomy was unclear. Methods: Comprehensive searches were conducted to include cohort studies which evaluated the influence of overweight on the surgical and postoperative outcomes of gastrectomy. Data was analyzed by RevMan 5.0. Results: Twenty-five cohort studies (18 518 patients) were included. Overweight patients were associated with longer operation time (mean difference 20.88, 95% confidence interval 14.07, 27.69), more intraoperative blood loss (mean difference 35.45, 95% confidence interval 9.24, 61.67), and less retrieved lymph nodes (mean difference 22.17, 95% confidence interval 23.51, 20.83) than normal patients undergoing laparoscopy-assisted gastrectomy. And overweight patients were associated with longer operation time (mean difference 26.31, 95% confidence interval 21.92, 30.70), more intraoperative blood loss (mean difference 130.02, 95% confidence interval 75.49, 184.55), less retrieved lymph nodes (mean difference 23.18, 95% confidence interval 24.74, 21.61), longer postoperative hospital stay (mean difference 2.37, 95% confidence interval 0.03, 4.70) and more postoperative complications (risk ratio 1.53, 95% confidence interval 1.29, 1.80) than normal patients in open gastrectomy. Conclusions: Overweight might affect the clinical results of both laparoscopy-assisted and open gastrectomy, especially for open gastrectomy. Key words: overweight – gastrectomy – meta-analysis
Adiposity is associated with increased morbidity and disability (1). Body mass index (BMI) (calculated as weight [kg]/ height [m2]) is commonly used to classify adiposity as overweight (BMI 25) and obesity (BMI 30) in adults (2). In 2008 an estimated 1.5 billion adults worldwide were overweight and 500 million were obese (1). A recent meta-analysis showed statistically significant associations for overweight with the incidence of 18 co-morbidities, including diabetes, cancers, cardiovascular diseases, asthma, gallbladder disease, osteoarthritis and chronic back pain (3). Each five units above the overweight category (BMI .25) are associated with 30% higher overall mortality and 40% higher for cardiovascular mortality (1,4). It is reported by World Health
Organization (WHO) that at least 2.8 million people die every year because of being overweight or obese worldwide, and an estimated 35.8 million (2.3%) of global Disability Adjusted Life Years are caused by overweight (2). In the context of surgical procedures, overweight itself is regarded as one risk factor for poor outcomes after complicated surgical procedures such as gastrectomy (5). It is reported that overweight was associated with unfavorable surgical outcomes, such as higher postoperative complication rates, more intraoperative blood loss, longer operation time and mortality and for both open and laparoscopy-assisted gastrectomy (6 – 11). However, the influence of overweight on the surgical and postoperative outcomes of gastrectomy were not
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou, The First Clinical College of Lanzhou University, Lanzhou, 3Department of Gastroenterology, First Affiliated Hospital of Dalian Medical University, Dalian and 4Shuocheng District People’s Hospital, Shuozhou, China 2
Jpn J Clin Oncol 2014;44(5)
clear, as studies (6 – 11) reported conflicted results for operation time and intraoperative blood loss. Meanwhile, the differences of overweight on the surgical and postoperative outcomes between open and laparoscopy-assisted gastrectomy were not clear, and no such a meta-analysis was conducted. As a result, we conducted this meta-analysis to comprehensively evaluate the influence of overweight on the surgical and postoperative outcomes of gastrectomy.
409
DATA ABSTRACTION Data were entered into an Excel database by two independent authors (L.L. and H.T.). The following fields were abstracted: country, patient characteristics, disease, follow-up duration and outcomes. Any disagreements were resolved by a third reviewer (K.Y.). DATA ANALYSIS AND SUBGROUP ANALYSIS
We conducted this systematic review in accordance with Guidelines for Meta-Analyses and Systematic Reviews of Observational Studies (MOOSE) (12). SEARCH STRATEGY PubMed, the Cochrane library, EMBASE, ISI Web of Knowledge, and Chinese Biomedical Database were searched using overweight, adiposity, BMI, gastrectomy and gastrectomies. Some studies used obese or obesity instead of overweight where the patients’ BMIs were ,30, but .25, so obese or obesity was also searched as a supplement. If possible, subject heading terms such as Medical Subject Headings terms were added. Reference lists of the metaanalysis, review articles about this topic and identified trials were hand-searched to identify further relevant citations. All the searches were conducted in August 2012 without language, date and publication status restrictions.
Data were analyzed by Review Manager Version 5.0. For dichotomous outcomes, results were expressed as risk ratio (RR) with 95% confidence interval (CI). If there were continuous scales of measurement, the mean difference (MD) was used to assess the effects of treatment. The percentage of variability across trials attributable to heterogeneity beyond chance was estimated with the I2 statistic, which was deemed significant when P was ,0.05 or I-square was .50%. Data were pooled using the fixed-effect model but the random-effects model was also considered in the case of significant heterogeneity. In our study, we conducted subgroup analyses of different kinds of gastrectomy (laparoscopy-assisted vs. open gastrectomy) and study types (prospective vs. retrospective cohort studies). Meanwhile, sensitive analysis was also conducted for those studies which applied WHO classification. We used a funnel plot based on operation time to evaluate publication bias. The small-study effect in terms of publication bias was also estimated using Egger’s linear regression test.
RESULT SEARCH RESULT
INCLUSION CRITERIA AND STUDY SELECTION Cohort studies which evaluated the influence of overweight on the surgical and postoperative outcomes of gastrectomy were included. The definition of overweight was not restricted by WHO definitions (BMI 25), but according to different countries’ condition. That is why we also conducted meta-analysis for overweight according to WHO definitions (BMI 25). The outcomes we evaluated were intraoperative outcomes (length of the operation, intraoperative blood loss, number of retrieved lymph nodes and lymph node dissection extent), postoperative short-term outcomes (postoperative mortality, time of first flatus, postoperative hospital stay and postoperative complications), and postoperative long-term outcomes (recurrence and 5-year survival rate). The postoperative complications we evaluated included anastomotic leak, abscess, pancreatic fistula, wound infection, postoperative bleeding and bowel obstruction. Studies that reported the following one or more outcomes were included; letters, comments, editorials, practice guidelines and trials published without the outcome measures of interest were excluded. Two reviewers (L.L and X.L.) independently assessed potentially relevant citations for inclusion, disagreements were resolved involved with a third reviewer (K.Y.).
After comprehensive searches, we found 1058 citations. We excluded 395 duplicates and 574 citations based on screening the titles and abstracts. After screening full texts, we excluded 64 citations (32 were not about gastrectomy, 21 were not about overweight and 11 were reviews). Finally, 25 cohort studies (5,7–11,13–31) (18 518 patients) were included (Fig. 1). CHARACTERISTICS OF INCLUDED TRIALS Of these 25 trials, they were from Japan (n ¼ 15), Korea (n ¼ 5), China (n ¼ 2), Germany (n ¼ 1), Poland (n ¼ 1) and UK (n ¼ 1). Of them, five were prospective cohort studies and two were randomized trials) and 20 were retrospective cohort studies. Eight of them were laparoscopy-assisted gastrectomy and 17 were open gastrectomy; all of them were for gastric cancer patients. The other characteristics of included trials were presented in Table 1. THE TOTAL RESULTS BASED ON META-ANALYSIS STUDIES
OF ALL INCLUDED
INTRAOPERATIVE OUTCOMES Overweight was associated with longer operation time (MD 24.71, 95% CI 21.02, 28.40), more intraoperative blood loss
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
PATIENTS AND METHODS
410
Overweight on outcomes of gastrectomy
postoperative complications including anastomotic leak, abscess, pancreatic fistula and longer postoperative hospital stay than normal patients in open gastrectomy, but not for laparoscopy-assisted gastrectomy (Table 3). SUBGROUP ANALYSIS OF PROSPECTIVE AND RETROSPECTIVE COHORT STUDIES
META-ANALYSIS OF THE STUDIES THAT DEFINED BMI >25 AS OVERWEIGHT ACCORDING TO WHO CRITERION
Figure 1. The flow chart.
(MD 105.50, 95% CI 66.76, 144.24), less retrieved lymph nodes (MD 22.87, 95% CI 23.96, 21.77) and more D1 lymph node dissection (RR 1.14, 95% CI 1.02, 1.27), but not with more D2 lymph node dissection (RR 0.98, 95% CI 0.91, 1.05) (Table 2). POSTOPERATIVE SHORT-TERM OUTCOMES Overweight was associated with more postoperative complications (RR 1.45, 95% CI 1.25, 1.69), but not with postoperative mortality (RR 1.22, 95% CI 0.80, 1.85), time of first flatus (MD 0.12, 95% CI 20.03, 0.27), and postoperative hospital stay (MD 0.20, 95% CI 20.99, 1.39) (Table 3). For postoperative complications, overweight was associated with more anastomotic leak (RR 1.68, 95% CI 1.15, 2.47), abscess (RR 2.31, 95% CI 1.73, 3.09) and pancreatic fistula (RR 2.18, 95% CI 1.67, 2.85), but not with more wound infection (RR 1.16, 95% CI 0.81, 1.66), postoperative bleeding (RR 1.21, 95% CI 0.77, 1.88), and bowel obstruction (RR 1.27, 95% CI 0. 71, 2.28) (Table 3).
Analyzing those studies which defined BMI .25 as overweight according to WHO criterion, the pooled results did not change and showed that overweight patients were associated with longer operation duration, more intraoperative blood loss, less retrieved lymph nodes and higher postoperative complication rates Supplementary Table 2. Subgroup analysis showed that overweight patients were associated with longer operation time, more intraoperative blood loss, and less retrieved lymph nodes for both laparoscopy-assisted and open gastrectomy. But overweight patients were associated with more postoperative complications in open gastrectomy. The details were presented in Supplementary Table 2 (online). PUBLICATION BIAS There seemed to be a significant publication bias based on funnel plot (Fig. 2). Egger’s test indicated that there was a possibility of publication bias for operation time (intercept 1.17, 95% CI 0.00 2.34, P ¼ 0.05).
DISCUSSION SUMMARY OF FINDINGS
POSTOPERATIVE LONG-TERM OUTCOMES Overweight was not associated with higher 5-year survival rates for Stages I, II and III gastric cancer and recurrence (including all recurrence, local recurrence, distant recurrence, both local and distant recurrence) (Table 4). SUBGROUP ANALYSIS OF LAPAROSCOPY-ASSISTED OPEN GASTRECTOMY
AND
Overweight patients were associated with longer operation time, more intraoperative blood loss, and less retrieved lymph nodes for laparoscopy-assisted and open gastrectomy (Table 2). But overweight was associated with more
Overweight patients were associated with longer operation time, more intraoperative blood loss and less retrieved lymph nodes than normal patients in both laparoscopy-assisted and open gastrectomy. Overweight affected the surgical results and postoperative outcomes via many operative factors, such as excessive adipose tissue (gastric, omental and perigastric fat), comorbidities, etc. (14,21,29). The biggest factor in overweight patients was adipose tissue, and this could result in longer operation time, more intraoperative blood loss and more postoperative complications because of technical difficulties accessing and dissecting lymph nodes deeply embedded in fatty tissues around the major abdominal vessels
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Overweight patients were associated with longer operation time, less retrieved lymph nodes and more postoperative complications based on both retrospective and prospective cohort studies. Overweight was associated with higher D1 lymph node dissection based on prospective studies and more intraoperative blood loss based on retrospective studies. The details were presented in Supplementary Table 1.
Table 1. Characteristics of included trials Study
Country
Study type
Overweight
Surgery
SZ E
Adachi et al. (1995) (13)
Follow-up C
Sex (M/F)
Age (years)
E
C
E
C
29–129 months
44/18
136/65
62.1
62
11–43% .Ideal
TG, DG
62
201
PC
.24.7 (M), .22.6 (F)
D2 G
38
46
36 (12–84) months
27/12
31/14
69 (44– 83)
71 (27– 86)
RC
25
TG, D2 G
81
82
–
51/30
41/41
61.3
57.4
695
^2 years
55/ 37
470/225
–
–
239
4.77 years
34/20
153/86
58.7
60.17
–
25/20
29/23
55 + 11
56 + 12
104 months
331/161
1033/467
59
60
Japan
RC
Barry et al. (2003) (14)
UK
Gretschel et al. (2003) (15)
Germany
Dhar et al. (2000) (29)
Japan
RC
.24.7 (M), .22.6 (F)
PG, TG
92
Inagawa et al. (2000) (17)
Japan
RC
25
D2 G
54
Kim et al. (2006) (18)
Korea
RC
23
LADG
Kulig et al. (2010) (10)
Poland
RC
25
G
Kunisaki et al. (2009) (31)
Japan
RC
25
LADG
29
123
–
–
–
–
–
Lee et al. (2009) (28)
Korea
PC
25
LADG
432
1053
–
279/153
653/400
59.4 + 10.8
56.9 + 12.5
394
153
–
269/125
88/65
59.78 + 12
60.8 + 12.56
99
545
–
77/22
371/174
62 + 11.3
62 + 10.5
50 (2– 95) months
70/29
210/101
–
–
–
19/6
19/18
56.42 + 11.50
52.30 + 10.91
Moriwaki et al. (2003) (20)
Japan
RC
18.5– 21
D2 G
Nobuoka et al. (2011) (9)
Japan
RC
25
TG
45
52
492
1500
Oh et al. (2009) (11)
Korea
RC
25
TG
99
311
Oh et al. (2012) (21)
Korea
PC
25
TG
24
37
Oki et al. (2012) (7)
Japan
RC
25
LADG
20
118
–
12/8
63/55
63.5 + 11.1
63.4 + 11.2
Tsujinaka et al. (2007) (5)
Japan
PC
25
G
77
446
–
57/20
301/145
–
–
27.29 + 2.4
21.53 + 2.2
Japan
RC
25
DG
China
RC
25
G
Kodera et al. (2004) (19)
Japan
RC
27
D2 G
Ojima et al. (2009) (23)
Japan
RC
25
DG, TG
60
188
45 (8– 118) months
46/14
134/54
103
468
5y
404/167
62.7 + 12.3
44
395
–
16/34
271/124
59.2
60.17
116
573
^5 years
92/24
405/168
63 + 12
64 + 12
19 (2– 39) months
68/15
325/123
60.7 + 9.7
59.7 + 11.3
–
37/29
23/27
55.8 + 11.6
55.0 + 10.5
Chen et al. (2011) (22)
China
RC
25
LADG
83
325
Shim et al. (2009) (24)
Korea
PC
23
LADG
60
56
Tokunaga et al. (2009) (30)
Japan
RC
25
TG, PG
1126
6799
–
825/301
4590/2290
59.1 + 11.6
58.5 + 11.8
Nozoe et al. (2012) (25)
Japan
RC
25
TG, PG
44
264
–
189/75
28/16
65.6 + 9.9
67.8 + 10.9
Yasuda et al. (2004) (27)
Japan
RC
25
LADG
16
83
48 months
8/8
48/35
66.4 + 10.9
66.7 + 10.0
Noshiro et al. (2003) (26)
Japan
RC
24.2
LADG
19
57
24 (1– 75) months
18/1
39/18
64 + 10
62 + 12
SZ, sample size; E, overweight group; C, control group; RC, Retrospective cohort; PC, prospective cohort; GC, gastric cancer; G, gastrectomy; LADG, laparoscopy-assisted distal gastrectomy; DG, distal gastrectomy; TG, total gastrectomy; PG, Partial gastrectomy; D2 G, gastrectomy with D2 lymphadenectomy.
Jpn J Clin Oncol 2014;44(5)
Yamada et al. (2008) (8) Gu and Zhang (2012) (16)
411
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
412
Overweight on outcomes of gastrectomy
Table 2. Subgroup analysis of laparoscopy-assisted versus open gastrectomy about intraoperative outcomes Outcomes
Studies
Sample size E
Operation time Laparoscopic
20 (7– 9,11,13–28) 7 (7,8,18,24,26–28)
Effect size (95% CI)
Heterogeneity I2 (%) (P)
C
1857
5276
MD 24.71 (21.02, 28.40)
34 (0.05)
621
1522
MD 20.88 (14.07, 27.69)
14 (0.32)
Open
14 (8,9,11,13– 17,19–23,25)
1236
3754
MD 26.31 (21.92, 30.70)
39 (0.05)
Intraoperative blood loss
14 (7– 9,13,15–17,19,20,22,23,25– 27)
1159
3721
MD 105.50 (66.76, 144.24
80 (0.00001)
4 (7,8,26,27) 11 (8,9,13,15– 17,19,20,22,23,25)
74
337
1085
3384
MD 35.45 (9.24, 61.67) MD 130.02 (75.49, 184.55)
0 (0. 71) 84 (0.00001)
D1 LN dissection
5 (7,10,18,24,25)
661
1990
RR 1.14 (1.02, 1.27)
0 (0.44)
Laparoscopic
3 (7,18,24)
125
226
RR 1.20 (0.94, 1.54)
51 (0.13)
Open
2 (10,25)
D2 LN dissection
6 (5,7,10,18,24,25)
536
1764
RR 1.12 (0.95, 1.31)
0 (0.46)
1170
3489
RR 0.98 (0.91, 1.05)
13 (0.33)
Laparoscopic
3 (7,18,24)
557
1279
RR 0.95 (0.87, 1.05)
45 (0.14)
Open
3 (5,10,25)
613
2210
RR 1.01 (0.91, 1.12)
0 (0.90)
E, overweight group; C, control group; LN, lymph node.
(2,11,13,14,28,29). For example, the N2 and N3 regional lymph nodes lie deep within the fatty mesentery around the major abdominal vessels which hampered complete removal in overweight patients (14,29). So extensive lymph node dissections on overweight patients have often been unsuccessful and the number of resected or retrieved lymph nodes (an indicator of adequacy of lymphadenectomy) was significantly smaller for the obese patient group (13,19,25,29). The presence of excessive subcutaneous fat predisposed the obese to impaired wound healing and thus wound infections and anastomotic leakage (32). Another factor that might be linked to higher postoperative complication rates is the co-morbidities such as hypertension, diabetes mellitus, coronary heart disease and respiratory dysfunction (11,13). However, the postoperative complications in laparoscopyassisted gastrectomy did not differ between overweight and normal patients. This suggested that laparoscopy-assisted gastrectomy for gastric cancer patients has advantages compared with a conventional open method. Recent meta-analyses showed that laparoscopy-assisted gastrectomy has the advantages of minimal invasion, faster recovery and fewer complications compared with open gastrectomy with the same short-term prognosis (33 –35). Overweight was hypothesized to be associated with a poorer prognosis due to associated co-morbidities and insufficient lymph node dissection (11,13,29). However, our meta-analysis failed to confirm it. Available evidence for 5-year survival rates for Stages I, II, III gastric cancer was based on two small studies (11,20). Their sample size was too small to detect the differences in the long-term survival between overweight and normal patients. Although another four studies (10,14,29,30) also reported the 5-year survival
rates, their included patients were in different stages. Three studies (10,14,29) of small sample size showed no differences between overweight and normal patients, while one large study (30) showed higher 5-year survival rate for overweight patients. All relevant characteristics were balanced in the three small studies (10,14,29), but not in the largest one (30). The number of patients with IA (UICC) was more in overweight group than that in normal, while the number of patients with IIIA, IIIB and IV was less in overweight group than that in normal. The unbalances were also seen in histology and surgical procedure. These unbalanced characteristics might lead to the unexpected results.
STRENGTH AND LIMITATIONS Our meta-analysis was the first one which evaluated the influence of overweight on the surgical and postoperative outcomes for both laparoscopy-assisted and open gastrectomy. However, our meta-analysis has several limitations. First, most the studies were retrospective studies (n ¼ 20) and from Asian countries (Japan, Korea and China; n ¼ 22). Credibility of meta-analysis data is defined by the type and quality of the trials included in the study. It is taken for granted that data and results from retrospective studies are heavily biased. Second, the definitions of overweight across different countries were different. Even though most studies used BMI .25 to judge whether a patient was overweight or not, but some other studies also used 23, 27 or other to define the overweight. This might be due to the nutritional condition and diet in their countries and could introduce clinical heterogeneity in our meta-analysis.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Laparoscopic Open
Jpn J Clin Oncol 2014;44(5)
413
Table 3. Subgroup analysis of laparoscopy-assisted versus open gastrectomy on postoperative short-term outcomes Outcomes
Studies
Sample size E
Effect size (95% CI)
Heterogeneity I2 (%) (P)
57 (0.01)
C
Length of hospital stay
9 (7,9,11,18,21,24,26–28)
891
2757
MD 0.20 (20.99, 1.39)
Laparoscopic
6 (7,18,24,26– 28)
586
1416
MD 20.64 (21.43, 0.15)
Open Postoperative mortality
2 (8,28) 10 (5,9– 11,13,15,16,19,23,29)
305
1341
MD 2.37 (0.03, 4.70)
59 (0.06)
2883
13 256
RR 1.22 (0.80, 1.85)
0 (0.81)
492
1241
RR 1.63 (0.49, 5.41)
0 (0.49)
2391
12 015
RR 1.17 (0.75, 1.83)
0 (0.73)
Time of first flatus
7 (11,18,22,24,26–28)
748
2057
MD 0.12 (20.03, 0.27)
48 (0.07)
Laparoscopic
5 (18,24,26–28)
566
1298
MD 0.15 (20.10, 0.40)
62 (0.03)
Open Postoperative complications Laparoscopic
2 (11,22) 23 (5,7,9 –11,13–15,17–31)
759 14 273
MD 0.13 (20.02, 0.27) RR 1.45 (1.25, 1.69)
0 (0.56) 66 (,0.00001)
615
1539
RR 1.16 (0.75, 1.79)
40 (0.13)
Open
16 (5,9– 11,13– 15,17,19– 23,25,29,30)
2925
12 734
RR 1.53 (1.29, 1.80)
72 (,0.00001)
Anastomotic leak
14 (5,8– 11,14– 16,18,20,23,25–28)
2084
5736
RR 1.68 (1.15, 2.47)
19 (0.25)
541
1348
RR 1.02 (0.44, 2.34)
0 (0.54)
Laparoscopic Open Abscess
7 (7,18,24,26– 28,31)
182 3540
5 (8,18,26–28) 9 (5,9– 11,14– 16,20,23,25) 12 (5,8,10,15–20,23,24,26)
Laparoscopic
4 (8,18,24,26)
Open
8 (5,10,15–17,19,20,23)
Wound infection Laparoscopic Open Postoperative bleeding
10 (7 –11,13,14,18,24,27) 5 (7,8,18,24,27) 5 (9 –11,13,14) 10 (7,10,11,15,17,18,20,24,26,28)
1543
4388
RR 1.89 (1.21, 2.96)
29 (0.18)
1514
4124
RR 2.31 (1.73, 3.09)
0 (0.78)
153
268
RR 0.93 (0.27, 3.17)
0 (0.93)
1361
3856
RR 2.47 (1.83, 3.34)
0 (0.72)
966
3018
RR 1.16 (0.81, 1.66)
0 (0.98)
176
415
RR 0.65 (0.23, 1.85)
0 (0.97)
790
2603
RR 1.27 (0.86, 1.87)
0 (0.95)
1690
3618
RR 1.21 (0.77, 1.88)
0 (0.96)
Laparoscopic
5 (7,18,24,26,28)
570
1333
RR 1.28 (0.73, 2.23)
0 (0.90)
Open
5 (10,11,15,17,20)
1120
2285
RR 1.10 (0.53, 2.28)
0 (0.73)
415
1765
RR 2.18 (1.67, 2.85)
55
224
RR 0.69 (0.09, 5.62)
0 (0.89)
Pancreatic fistula
6 (5,7– 9,15,16)
Laparoscopic
2 (7,8)
Open
4 (5,9,15,16)
Bowel obstruction
7 (7,11,13,15,20,24,28)
36 (,0.00001)
360
1541
RR 2.26 (1.72, 2.96)
55 (0.08)
1148
1974
RR 1.27 (0. 71, 2.28)
0 (0.69)
Laparoscopic
3 (7,24,28)
512
1227
RR 1.72 (0.84, 3.53)
0 (0.40)
Open
4 (11,13,15,20)
636
747
RR 0.72 (0.25, 2.12)
0 (0.93)
E, overweight group; C, control group.
IMPLICATIONS FOR RESEARCH AND PRACTICE As most studies were from Asian countries and retrospective studies, future prospective studies in other continental countries were needed. Some trials that used ‘obese’ or ‘obesity’ for those patients whose BMIs were .25 and ,30. So the definition of overweight and obesity should be well classified, and studies that assessed the effects of obesity for gastrectomy should be conducted. In the future, when conducting gastrectomy for overweight gastric cancer patients, surgeons should retrieve metastatic
lymph nodes as much as possible and be aware of the postoperative complications when conducting open gastrectomy. If possible, laparoscopy-assisted gastrectomy should be applied to overweight gastric cancer patients because of lower postoperative complication rates (safety), less intraoperative blood loss (safety), shorter operation time and postoperative hospital stay (more economic), and more retrieved lymph nodes (better survival) when compared with open gastrectomy. So for overweight gastric cancer patients, laparoscopy-assisted gastrectomy seems to be another better treatment.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
Laparoscopic Open
3 (9,11,21) 12 (5,8– 11,13,15,16,19,23,28,29)
0 (0.52)
414
Overweight on outcomes of gastrectomy
Table 4. Subgroup analysis of laparoscopy-assisted versus open gastrectomy on postoperative long-term outcomes Outcomes
Studies
Effect size (95% CI) Heterogeneity I2 (%) (P)
Patient no. E
C
5-Year survival rate for Stage I gastric cancera
2 (11,20)
99
112
RR 0.99 (0.88, 1.12)
5-Year survival rate for Stage II gastric cancera
2 (11,20)
220
133
RR 1.05 (0.99, 1.10) 93 (0.0002)
5-Year survival rate for Stage III gastric cancera
2 (11,20)
174
219
RR 0.97 (0.78, 1.21)
All recurrenceb
3 (11,14,29)
200
777
RR 1.24 (0.78, 1.99) 69 (0.04)
Local recurrenceb
3 (11,14,29)
200
777
RR 1.60 (0.94, 2.74)
Distant recurrenceb
3 (11,14,29)
200
777
RR 0.82 (0.50, 1.34) 33 (0.22)
Both local and distant recurrenceb
3 (11,14,29)
115
550
RR 1.36 (0.63, 2.94) 10 (0.33)
0 (0.90)
0 (0.89)
a
Laparoscopy assisted. Open; E: overweight group; C: control group.
b
Figure 2. Publication bias.
CONCLUSION Overweight might affect the clinical results of gastrectomy, especially for open gastrectomy. Most studies were retrospective cohort studies and from Asian countries, so future prospective studies on other continental countries were needed.
Supplementary data Supplementary data are available at http://www.jjco.oxfordjournals.org.
Conflict of interest statement None declared.
References 1. Nguyen T, Lau DC. The obesity epidemic and its impact on hypertension. Can J Cardiol 2012;28:326– 33.
2. WHO: Global Health Observatory (GHO) http://wwwwhoint/gho/ncd/ risk_factors/overweight_text/en/indexhtml 2012. 3. Guh DP, Zhang W, Bansback N, et al. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009;9:88. 4. Whitlock G, Lewington S, Sherliker P, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet 2009;373:1083– 96. 5. Tsujinaka T, Sasako M, Yamamoto S, et al. Influence of overweight on surgical complications for gastric cancer: results from a randomized control trial comparing D2 and extended para-aortic D3 lymphadenectomy (JCOG9501). Ann Surg Oncol 2007;14:355–61. 6. Lee JH, Paik YH, Lee JS, et al. Abdominal shape of gastric cancer patients influences short-term surgical outcomes. Ann Surg Oncol 2007;14:1288– 94. 7. Oki E, Sakaguchi Y, Ohgaki K, et al. The impact of obesity on the use of a totally laparoscopic distal gastrectomy in patients with gastric cancer. J Gastric Cancer 2012;12:108–12. 8. Yamada H, Kojima K, Inokuchi M, et al. Effect of obesity on technical feasibility and postoperative outcomes of laparoscopy-assisted distal gastrectomy—comparison with open distal gastrectomy. J Gastrointest Surg 2008;12:997– 1004. 9. Nobuoka D, Gotohda N, Kato Y, et al. Influence of excess body weight on the surgical outcomes of total gastrectomy. Surg Today 2011;41:928–34. 10. Kulig J, Sierzega M, Kolodziejczyk P, et al. Implications of overweight in gastric cancer: a multicenter study in a Western patient population. Eur J Surg Oncol 2010;36:969– 76. 11. Oh SJ, Hyung WJ, Li C, et al. Effect of being overweight on postoperative morbidity and long-term surgical outcomes in proximal gastric carcinoma. J Gastroenterol Hepatol 2009;24:475–9. 12. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008– 12. 13. Adachi W, Kobayashi M, Koike S, et al. The influence of excess body weight on the surgical treatment of patients with gastric cancer. Surg Today 1995;25:939– 45. 14. Barry JD, Blackshaw GR, Edwards P, et al. Western body mass indices need not compromise outcomes after modified D2 gastrectomy for carcinoma. Gastric Cancer 2003;6:80–5. 15. Gretschel S, Christoph F, Bembenek A, et al. Body mass index does not affect systematic D2 lymph node dissection and postoperative morbidity in gastric cancer patients. Ann Surg Oncol 2003;10:363–8. 16. Gu JS, Zhang WH. Effect of overweight on patients undergoing gastrectomy for gastric cancer. J China Med Univ 2012;40:372– 5. 17. Inagawa S, Adachi S, Oda T, et al. Effect of fat volume on postoperative complications and survival rate after D2 dissection for gastric cancer. Gastric Cancer 2000;3:141– 4.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
0 (0.44)
Jpn J Clin Oncol 2014;44(5)
27. Yasuda K, Inomata M, Shiraishi N, et al. Laparoscopy-assisted distal gastrectomy for early gastric cancer in obese and nonobese patients. Surg Endosc 2004;18:1253 –6. 28. Lee HJ, Kim HH, Kim MC, et al. The impact of a high body mass index on laparoscopy assisted gastrectomy for gastric cancer. Surg Endosc 2009;23:2473– 9. 29. Dhar DK, Kubota H, Tachibana M, et al. Body mass index determines the success of lymph node dissection and predicts the outcome of gastric carcinoma patients. Oncology 2000;59:18 –23. 30. Tokunaga M, Hiki N, Fukunaga T, et al. Better 5-year survival rate following curative gastrectomy in overweight patients. Ann Surg Oncol 2009;16:3245 –51. 31. Kunisaki C, Makino H, Takagawa R, et al. Predictive factors for surgical complications of laparoscopy-assisted distal gastrectomy for gastric cancer. Surg Endosc 2009;23:2085–93. 32. Mullen JT, Davenport DL, Hutter MM, et al. Impact of body mass index on perioperative outcomes in patients undergoing major intra-abdominal cancer surgery. Ann Surg Oncol 2008;15:2164–72. 33. Ohtani H, Tamamori Y, Noguchi K, et al. Meta-analysis of laparoscopy-assisted and open distal gastrectomy for gastric cancer. J Surg Res 2011;171:479– 85. 34. Liang Y, Li G, Chen P, et al. Laparoscopic versus open gastrectomy for early distal gastric cancer: a meta-analysis. ANZ J Surg 2011;81:673–80. 35. Ding J, Liao GQ, Liu HL, et al. Meta-analysis of laparoscopy-assisted distal gastrectomy with D2 lymph node dissection for gastric cancer. J Surg Oncol 2012;105:297 –303.
Downloaded from http://jjco.oxfordjournals.org/ at Uniwersytet Warszawski Biblioteka Uniwersytecka on April 18, 2015
18. Kim KH, Kim MC, Jung GJ, et al. The impact of obesity on LADG for early gastric cancer. Gastric Cancer 2006;9:303–7. 19. Kodera Y, Ito S, Yamamura Y, et al. Obesity and outcome of distal gastrectomy with D2 lymphadenectomy for carcinoma. Hepatogastroenterology 2004;51:1225– 8. 20. Moriwaki Y, Kunisaki C, Kobayashi S, et al. Does body mass index (BMI) influence morbidity and long-term survival in gastric cancer patients after gastrectomy? Hepatogastroenterology 2003;50:284–8. 21. Oh CA, Kim DH, Oh SJ, et al. Impact of body mass index on surgical outcomes in radical total gastrectomy. Hepatogastroenterology 2012;59:934–7. 22. Chen JX, Huang CM, Zheng CH, et al. Impact of obesity on laparoscopic-assisted radical gastrectomy for gastric cancer. Zhonghua Wei Chang Wai Ke Za Zhi 2011;14:781– 4. 23. Ojima T, Iwahashi M, Nakamori M, et al. Influence of overweight on patients with gastric cancer after undergoing curative gastrectomy: an analysis of 689 consecutive cases managed by a single center. Arch Surg 2009;144:351– 8. discussion 358. 24. Shim JH, Song KY, Kim SN, et al. Laparoscopy-assisted distal gastrectomy for overweight patients in the Asian population. Surg Today 2009;39:481–6. 25. Nozoe T, Kohno M, Iguchi T, et al. Analysis of the impact of the body mass index in patients with gastric carcinoma. Surg Today 2012;42:945–9. 26. Noshiro H, Shimizu S, Nagai E, et al. Laparoscopy-assisted distal gastrectomy for early gastric cancer: is it beneficial for patients of heavier weight? Ann Surg 2003;238:680–5.
415
