MedComm Oncology - 2023 - Chen

Received: 28 March 2023 | Revised: 24 July 2023 | Accepted: 24 July 2023
DOI: 10.1002/mog2.48
ORIGINAL ARTICLE
Comparison of Billroth II with Braun and Roux‐en‐Y

reconstructions after distal gastrectomy for gastric
cancer: A meta‐analysis
Defei Chen | Chenglin Tang | Fan He | Fuyu Yang | Saed Woraikat |

Kun Qian
Department of Gastrointestinal Surgery,

The First Affiliated Hospital of Abstract
Chongqing Medical University, Roux‐en‐Y (RY) and Billroth II with Braun (BB) reconstruction are two similar
Chongqing, China
methods of reconstruction after distal gastrectomy (DG). Currently, it is unclear
Correspondence which method is superior. This meta‐analysis was performed to compare the safety
Kun Qian, Department of and efficacy of BB and RY reconstruction after DG for gastric cancer. A literature
Gastrointestinal Surgery, the First
Affiliated Hospital of Chongqing Medical search of Pubmed, Embase, Web of Science, and Cochrane Library was conducted
University, Chongqing 400016, China. to identify studies comparing BB with RY after DG for gastric cancer until the
Email: [email protected]
end of October 2022. Main outcomes assessed were perioperative outcomes,
Funding information postoperative complications, functional findings, and nutritional status. Meta‐
Chongqing Medical Scientific Research analyses were performed using RevMan 5.4 software. Finally, eight studies with a
Project (Joint project of Chongqing
total of 910 patients were considered for the meta‐analysis. The meta‐analysis
Health Commission and Science and
Technology Bureau), results revealed that operation time, anastomosis time, incidence of total
Grant/Award Number: 2021MSXM096 complications, and delayed gastric emptying were reduced, and the incidence of
bile reflux was increased in patients undergoing BB compared to RY reconstruction.
In conclusion, BB has the advantage of reducing operative time, anastomotic time,
intraoperative blood loss, overall postoperative complications, and delayed gastric
emptying. RY has the advantage of preventing bile reflux and gastritis after surgery.
KEYWORDS
Billroth II Braun, distal gastrectomy, gastric cancer, meta‐analysis, reconstruction,
Roux‐en‐Y
1 | INTRODUCTION respectively.1 Radiotherapy as well as chemotherapy and

immunotherapy may improve the outcomes, but surgical
Globally, gastric cancer is one of the most prevalent resection with proper lymph node dissection remains the
malignant tumors of the digestive tract. Among malignant primary option for curable gastric cancer.2 Distal gastrec-
tumors, its morbidity and mortality rank fifth and fourth, tomy (DG) with sufficient resection margin is the indicated
Defei Chen, Chenglin Tang, and Fan He contributed equally to this study.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided
the original work is properly cited.
© 2023 The Authors. MedComm – Oncology published by John Wiley & Sons Australia, Ltd on behalf of Sichuan International Medical Exchange & Promotion Association
(SCIMEA).
MedComm – Oncology. 2023;2:e48. wileyonlinelibrary.com/journal/mog2 | 1 of 13

https://doi.org/10.1002/mog2.48
27696448, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mog2.48 by Cochrane Saudi Arabia, Wiley Online Library on [08/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 of 13 | CHEN ET AL.
treatment for cancer located in the distal stomach. search. After reviewing the publications, the full text
Reconstruction of the digestive tract after gastrectomy is of 11 studies was further assessed. Of these, three
one of the most important aspects of surgery. The goal of papers did not meet the criteria and were excluded.
optimal reconstruction is maintenance of satisfactory Finally, one randomized controlled trial (RCT) 14 and
nutritional status and quality of life, with the simplest seven nonrandomized observational clinical studies
techniques available while reducing postoperative morbidity (OCS) 15–21 with a total of 910 patients were consid-
as much as possible. ered for the meta‐analysis (Figure 1). The population
The main reconstruction methods for DG include characteristics of each study are shown in Table 1. A
Billroth I (BI) reconstruction,3 Billroth II (BII) reconstruc- total of 518 and 392 patients were included in the BB
tion,4 Billroth II Braun (BB) reconstruction,5 Roux‐en‐Y (RY) and RY groups, respectively. Studies were conducted
reconstruction,6 and uncut RY (URY) reconstruction,7 BI in South Korea (four studies)14–17 and China (four
reconstruction has specific requirements in terms of gastric studies). 18–21 Quality scores of each included article
cancer location while the other reconstructions have broader were between 6 and 9, indicating sufficient study
indications. BI and BII retain intestinal continuity and are quality (Table 1).
easier to perform while other reconstructions are technically
challenging and alter the intestinal anatomy, however, the
chronic bile reflux into the stomach may cause remnant 2.2 | Comparison of perioperative
gastritis with a potential risk for gastric metaplasia.8,9 BB outcomes between BB and RY
reconstruction was designed specifically to reduce the flow of
bile into the stomach and URY is a modification of the BB To compare the feasibility of BB and RY, perioperative
anastomosis in which a jejunal occlusion is fashioned.5,7,10 outcomes including operation time, anastomosis
RY reconstruction addresses the problem of alkaline time, intraoperative blood loss, harvested lymph
bile reflux in BI and BII reconstruction but it leaves the nodes, length of hospital stay, and time to first flatus
patient prone to RSS.11,12 URY reconstruction can effectively or defecation were evaluated (Table 2). Eight
prevent reflux gastritis and is superior to RY in preventing studies 14–21 reported operation time and the meta‐
RSS after DG,10,13 however, it is more complex and expen- analysis showed operation time was lower in the
sive compared with BB and RY. Compared with BII BB group (weighted mean difference [WMD],
reconstruction, BB reconstruction has a supplementary −21.06 min; 95% confidence interval [CI], −31.50
Braun anastomosis, through which the bile can flow into to −10.63; p < 0.0001; Figure 2A). Among the two
distal jejunum directly, so it can reduce the bile reflux studies reporting anastomosis time18,20 anastomosis
theoretically.4,5 Everything considered, BB and RY seem to time was lower in the BB group (WMD, −8.65 min;
be the preferred reconstruction method after DG. 95% CI, −9.20 to −8.10; p < 0.00001; Figure 2B).
BB and RY reconstruction are similar anatomically, both Eight studies 14–21 reported intraoperative blood
of them have an anastomosis of stomach and jejunum plus loss and the meta‐analysis revealed the value was
an anastomosis of jejunum and jejunum, but which one is similar in the two groups (WMD, −13.44 mL;
superior remains controversial. Therefore, we performed this 95% CI, −28.67 to 1.78; p = 0.08; Figure 2C). The
meta‐analysis to compare the safety and efficacy of BB and number of harvested lymph nodes was reported in
RY reconstruction after DG for gastric cancer. Finally, eight seven studies 15–21 and was similar between the two
studies with a total of 910 patients were considered for the groups (WMD, 0.50; 95% CI, −1.71 to 2.71; p = 0.66;
meta‐analysis and we found that BB has the advantage of Figure 2D). No significant difference was observed
reducing operative time, anastomotic time, intraoperative between the two groups in terms of hospital stay
blood loss, overall postoperative complications, and delayed (WMD, −0.02; 95% CI, −0.19 to 0.14; p = 0.77;
gastric emptying. RY has the advantage of preventing bile Figure 2E) and time to first flatus or defecation
reflux and gastritis after surgery. (WMD, −0.09; 95% CI, −0.27 to 0.09; p = 0.35;
Supporting Information: Figure S1A).
2 | RESUL TS
2.3 | Comparison of postoperative
2.1 | Studies with sufficient quality complications between BB and RY
were selected for meta‐analysis
To compare the safety of BB and RY, postoperative
To search for all eligible studies for the meta‐analysis, complications were evaluated (Table 2). The overall
125 publications were selected during the initial postoperative complications were reported in eight
CHEN ET AL. | 3 of 13
F I G U R E 1 PRISMA flow diagram of

studies selected for meta‐analysis; PRISMA:
Preferred Reporting Items for Systematic
Reviews and Meta‐Analyses.
studies.14–21 There was no significant difference in the p = 0.74; Supporting Information: Figure S2F) was
risk of total complications (odds ratio [OR], 0.68; 95% similar between the BB and RY groups.
CI, 0.46 to 1.00; p = 0.05; Figure 3A). Four stud-
ies 15,17,20,21 reported postoperative complications ac-
cording to the Clavien–Dindo classification22 and risk 2.4 | Comparison of postoperative
of total complications (OR, 0.63; 95% CI, 0.40 to 0.99; functional findings and nutritional status
p = 0.04; Supporting Information: Figure S2A) was a between BB and RY
lower in the BB group. However, subgroups of grade
Grades I–II (OR, 0.66; 95% CI, 0.40 to 1.09; p = 0.11; To compare the efficacy of BB and RY, postoperative
Supporting Information: Figure S2B) and Grades functional findings and nutritional status were
III–IV (OR, 0.61; 95% CI, 0.24 to 1.58; p = 0.31; assessed. Three studies 15,16,20 reported postoperative
Supporting Information: Figure S2C) did not differ functional findings including residual food, gastritis,
between groups. Delayed gastric emptying was re- and bile reflux; the results of the meta‐analysis are
ported in five studies 16,17,19–21 and the BB group had a shown in Table 2. No difference was found in the risk
lower risk (OR, 0.40; 95% CI, 0.18 to 0.89; p = 0.02; of residual food (OR, 0.68; 95% CI, 0.39 to 1.16;
Figure 3B). The risk of other complications, including p = 0.16; Figure 4A) and gastritis (OR, 4.63; 95% CI,
anastomotic leakage (OR, 0.48; 95% CI, 0.09 to 2.58; 0.92 to 23.29; p = 0.06; Figure 4B) between BB and RY
p = 0.39; Figure 3C), ileus (OR, 0.58; 95% CI, 0.19 to groups. Risk of bile reflux was significantly lower in
1.84; p = 0.36; Figure 3D), postoperative pancreatitis the RY group (OR, 4.77; 95% CI, 1.23 to 18.53;
(OR, 4.33; 95% CI, 0.96 to 19.49; p = 0.06; Figure 3E), p = 0.02; Figure 4C). Postoperative nutritional status
duodenal stump leakage (OR, 0.96; 95% CI, 0.25 to 1 year after surgery was reported in two studies
3.76; p = 0.95; Supporting Information: Figure S2D), (Table 2).15,19 No significant between‐group differ-
intra‐abdominal abscess (OR, 1.06; 95% CI, 0.24 to ences were observed in body weight (WMD, 1.64; 95%
4.59; p = 0.94; Supporting Information: Figure S2E), CI, −0.52 to 3.80; p = 0.14; Figure 4D), hemoglobin
wound problem (OR, 1.18; 95% CI, 0.45 to 3.12; (WMD, 2.01; 95% CI, −1.29 to 5.31; p = 0.23;
| 4 of 13
TABLE 1 Characteristics of the included studies.

Quality Study Operation Surgical Sample Sex BMI Mean ASA score TNM stage
Author Country Design score period type group size male/female (kg/m2) age (yr) (I/II/III) (I/II/III/IV)
Lee et al.14 South RCT 6 2006–2007 LA/open BB 52 42/10 NA 59.7 ± 10.9 NA NA
Korea RY 47 28/19 NA 58.5 ± 10.7 NA NA
15
Park and Kim South Retro 7 2005–2013 LA/TL BB 76 N/A NA NA NA NA
Korea RY 55 N/A NA NA NA NA
16
Choi et al. South Retro 7 2010–2012 LA BB 26 18/8 23.4 ± 2.0 59.7 ± 9.1 4/22/0 22/3/1/0
Korea RY 40 28/12 23.7 ± 2.4 57.2 ± 10.7 18/20/2 38/2/0/0
17
Cui et al. South Retro 7 2013–2015 TL BB 26 15/11 NA 60.1 ± 13.3 13/13/0 12/7/7/0
Korea RY 30 22/8 NA 57.6 ± 12.6 18/12/0 23/5/2/0
18
Chi et al. China Retro 7 2019–2020 TL BB 54 39/15 21.8 ± 2.4 65.3 ± 9.8 24/24/6 7/19/28/0
RY 51 31/20 22.0 ± 2.6 67.3 ± 7.9 25/21/5 8/18/25/0
Jun et al.19 China Retro 7 2017–2019 LA/Open BB 36 18/18 22.47 ± 2.61 65.64 ± 8.59 9/20/7 0/14/19/3
RY 36 22/14 21.00 ± 3.39 63.86 ± 9.22 8/23/5 0/11/23/2
20
Yalikun et al. China Retro 7 2016–2019 TL BB 145 97/48 23.4 ± 5.8 62 61/72/12 44/35/62/4
RY 102 72/30 22.8 ± 6.2 63 45/42/15 21/22/56/3
21
Yan et al. China Retro 7 2015–2020 LA/TL BB 103 75/28 23.3 ± 3.5 68.0 ± 9.3 29/68/6 24/26/53/0
RY 31 20/11 22.5 ± 2.9 62.4 ± 10.6 11/19/1 10/9/12/0
Note: Data are presented as mean ± standard deviation, median or n corresponding to groups.
Abbreviations: ASA, American Society of Anesthesiologists; BB, Billroth II with Braun; BMI, body mass index; LA, laparoscopy‐assisted; NA, not available; RCT, randomized controlled trial; RY, Roux‐en‐Y;
TL, totally laparoscopic.
CHEN
ET AL.
TABLE 2 Meta‐analysis of outcomes of interest.

Outcome of interest Studies Patients (n) OR/WMD 95% CI I² (%) p‐value
Perioperative outcomes
Operation time 8 910 −21.06 −31.50, −10.63 75 <0.0001
Anastomosis time 2 352 −8.65 −9.20, −8.10 89 <0.00001
Intraoperative blood loss 8 910 −13.44 −28.67, 1.78 73 0.08
Harvested lymph nodes 7 811 0.50 −1.71, 2.71 66 0.66
Hospital stay 8 910 −0.02 −0.19, 0.14 0 0.77
Time to first flatus or defecation 4 410 −0.09 −0.27, 0.09 43 0.35
Postoperative complications
Total complications 8 910 0.68 0.46, 1.00 0 0.05
Clavien‐Dindo classification 4 568 0.63 0.40, 0.99 0 0.04
Grades I–II 4 568 0.66 0.40, 1.09 1 0.11
Grades III–IV 4 568 0.61 0.24, 1.58 0 0.31
Anastomotic leakage 4 536 0.48 0.09, 2.58 0 0.39
Delayed gastric emptying 5 650 0.40 0.18, 0.89 20 0.02
Ileus 5 579 0.58 0.19, 1.84 0 0.36
Postoperative pancreatitis 3 276 4.33 0.96, 19.49 0 0.06
Duodenal stump leakage 2 381 0.96 0.25, 3.76 0 0.95
Intra‐abdominal abscess 3 270 1.06 0.24, 4.59 21 0.94
Wound problem 6 713 1.18 0.45, 3.12 0 0.74
RGB classification
Residual food 3 424 0.68 0.39, 1.16 0 0.16
Gastritis 3 424 4.63 0.92, 23.29 86 0.06
Bile reflux 3 424 4.77 1.23, 18.53 82 0.02
Postoperative nutritional status
Body weight 2 187 1.64 −0.52, 3.80 0 0.14
Hemoglobin 2 187 2.01 −1.29, 5.31 46 0.23
Albumin 2 187 0.27 −0.68, 1.23 36 0.58
Abbreviations: CI, confidence interval; OR, odds ratio; RGB, residue, gastritis, bile; WMD, weighted mean difference.
Figure 4E), or albumin (WMD, 0.27; 95% CI, −0.68 to p = 0.06) and bile reflux (I² = 82%; p = 0.02). Sensitivity
1.23; p = 0.58; Figure 4F). analysis of intraoperative blood loss showed heterogene-
ity was reduced to 44% after excluding the study by
Yalikun et al.,20 but the difference was not significant
2.5 | Sensitivity analyses of outcomes (p = 0.58) (Supporting Information: Table S1). Sensitivity
with high heterogeneity analysis of incidence of gastritis and bile reflux indicated
the heterogeneity was decreased to 8% and 0% respec-
To explore the source of heterogeneity, sensitivity tively after excluding the study by Yalikun et al.,20
analyses were performed. The results of the meta‐ moreover, a significantly lower risk of incidence of
analysis revealed significant heterogeneity in some gastritis (p < 0.00001) and bile reflux (p < 0.00001) was
outcomes, such as operative time (I² = 75%; p < 0.0001), found in BB group after excluding the aforementioned
anastomosis time (I2 = 89%; p < 0.00001), intraoperative study (Supporting Information: Table S1). However, a
blood loss (I² = 73%; p = 0.08), harvested lymph nodes thorough review of this study revealed no reason for
(I2 = 66%; p = 0.66), and incidence of gastritis (I² = 86%; exclusion.
F I G U R E 2 Meta‐analysis of surgical outcomes. (A) Pooled data on operation time, (B) pooled data on anastomosis time, (C) pooled data
on intraoperative blood loss, (D) pooled data on number of harvested lymph nodes, (E) pooled data on length of hospital stay. BB, Billroth II
with Braun; CI, confidence interval; RY, Roux‐en‐Y; SD, standard deviation.
F I G U R E 3 Meta‐analysis of postoperative complications. (A) Pooled data on total postoperative complications, (B) pooled data on
delayed gastric emptying, (C) pooled data on anastomotic leakage, (D) pooled data on postoperative ileus, (E) pooled data on postoperative
pancreatitis. BB, Billroth II with Braun; CI, confidence interval; RY, Roux‐en‐Y; SD, standard deviation.
F I G U R E 4 Meta‐analysis of postoperative functional findings and nutritional status. (A) Pooled data on residual food, (B) pooled data
on gastritis, (C) pooled data on bile reflux, (D) pooled data on postoperative body weight, (E) pooled data on postoperative hemoglobin,
(F) pooled data on postoperative albumin. BB, Billroth II with Braun; CI, confidence interval; RY, Roux‐en‐Y; SD, standard deviation.
presence of anastomotic tension. One of the primary

limitations associated with BI and BII anastomoses is the
persistent occurrence of chronic bile reflux into the
remnant, leading to subsequent alterations in the physiolog-
ical gastric acid environment,24 which may lead to
esophagitis, Barrett's esophagus, and remnant gastritis.25
The BB anastomosis has been proposed as a potential
method to decrease the occurrence of bile reflux into
the remnant. The RY and URY techniques, which are
predominantly implemented in Western countries, have
been introduced with the intention of mitigating the
potential for bile reflux. However, they are more compli-
F I G U R E 5 Funnel plot of total complications rate in all cated to perform and some patients may develop functional
included studies. OR, odds ratio; SE, standard error. RSS with a potential risk for internal hernia (Petersen's
hernia).26 BB seems to be a promising method for
2.6 | Subgroups analyses of studies gastrointestinal reconstruction after distal gastrectomy but
using totally laparoscopic technique controversy remains regarding which is the optimal method
for reconstruction after distal gastrectomy between BB and
We noticed that a totally laparoscopic technique was RY. Therefore, this meta‐analysis was performed to evaluate
used for all patients in three studies.17,18,20 To identified and compare the safety and efficacy of BB reconstruction
the influence of totally laparoscopic technique to the and RY reconstruction after distal gastrectomy.
results, subgroup analyses were undertaken for relevant The meta‐analysis revealed that the operative time and
outcome measures by including studies with totally anastomotic time were reduced in patients who under-
laparoscopic technique only (Supporting Information: went BB reconstruction compared to RY reconstruction.
Table S2). Interestingly, the subgroup analyses showed Despite a lack of significance, BB displayed a trend toward
that more harvested lymph nodes in the BB group reducing intraoperative blood loss. Furthermore, by
(WMD, 2.75; 95% CI, 1.98 to 3.52; p < 0.00001; Support- excluding the study by Chi et al.,18 a significant difference
ing Information: Figure S1B), which differed from the (p = 0.04) was found between the two groups with
results when all studies were included. Further, signifi- moderate heterogeneity (I2 = 62%). Therefore, the results
cant differences were observed in outcomes, including may be influenced by the related heterogeneity and
operation time (WMD, −20.98; 95% CI, −27.06 to −14.90; smaller sample size of the included studies. Thus, it is
p < 0.00001), total complications (OR, 0.50; 95% CI, 0.26 possible that BB was related to reduced intraoperative
to 0.96; p = 0.04) and Clavien–Dindo classification grades blood loss. Hence, BB has an advantage in shortening the
I–II (OR, 0.41; 95% CI, 0.17 to 0.95; p = 0.04), which were operative time, anastomotic time, and intraoperative blood
similar to the results when all studies were included. loss. This is explained by the additional procedures
required in RY, including cutting off the bowel, separating
the part of the mesentery, and handling the mesenteric
2.7 | Evaluation of publication bias vessels, which require more time and cause more trauma
than BB.
To identify potential publishing bias, a funnel plot of the The number of harvested lymph nodes was similar
overall postoperative complications was created in which between the two groups when all studies were included.
no asymmetry was observed, indicating no evidence of However, in the subgroup of studies using the totally
publication bias in the study (Figure 5). laparoscopic technique, BB was associated with more
harvested lymph nodes (Supporting Information:
Figure S2B), and interestingly, when a fixed model was
3 | DISCUS SION used and all studies were included, the groups differed
significantly (p < 0.00001) with 66% heterogeneity. Con-
The optimal choice for reconstruction after DG remains sidering the heterogeneity of the included studies and that
controversial. BI reconstruction, as it maintains the number of harvested lymph nodes was influenced by the
physiological passage of food into the duodenum, is usually different dissection type, increased use of stapling and
performed after DG when technically feasible.23 Likewise, energy devices, operating surgeon experience, and hospital
the utilization of BII reconstruction is frequently employed volume, it is difficult to determine which technique is
in cases where BI is deemed technically unviable due to the associated with more harvested lymph nodes.
The incidence of postoperative complications is distance between the gastrojejunostomy and Braun
often used to assess the safety of a surgical procedure. anastomosis was increased by 10 cm compared to
In this meta‐analysis, despite the lack of statistical conventional BB reconstruction, and for this reason,
significance (p = 0.05), BB reconstruction was associ- bile reflux and gastritis were reduced in their BB group.
ated with reduced overall postoperative complications. Therefore, BB is inferior to RY in terms of preventing
Furthermore, the groups differed significantly (p = 0.04, bile reflux and gastritis, but hopefully, by modifying
Supporting Information: Figure S1A) when only the BB reconstruction, this drawback can be remedied.
studies categorizing complications according to the Only two studies reported changes in nutritional
Clavien–Dindo classification were included. Thus, BB status at 1 year of follow‐up after surgery. This meta‐
is associated with lower risk of overall postoperative analysis revealed no significant difference in body
complications, possibly because of the need to cut off weight, hemoglobin, or albumin levels between the two
the bowel and separate part of the mesentery in RY, groups, however, body weight was higher in the BB
which disrupts intestinal continuity and physiology group than in the RY group. Therefore, postoperative
leading to more trauma than BB. nutritional status could not be fully assessed due to lack
The incidence of delayed gastric emptying (DGE) was of sufficient data. Consequently, further studies with
2.3% and 5.7% in the BB and RY groups, respectively and larger samples, including more nutritional indicators and
risk was significantly lower in the BB group (p = 0.02, longer follow‐up are needed to address this question.
Figure 3B), consistent with a previously published This meta‐analysis has several advantages. First, it is the
analysis27 that reported a higher incidence of DGE after first study comparing BB with RY anastomosis. Second,
RY reconstruction (known as RSS). RSS is characterized unlike the comparison of the procedures in previous reports,
by abdominal pain, vomiting, and nausea after oral our study focused on a comparison of BB and RY and
intake of food. RSS appears to be associated with provides relevant evidence‐based findings. Third, all the
functional rather than mechanical obstruction of the extracted data were cross‐checked and sensitivity and
“Y” limb. The possible mechanisms of this syndrome subgroup analyses were performed to improve the credibility
include the altered electrical stimulation and retrograde of our results. However, there are several limitations to this
peristalsis of the “Y” limb caused by the altered intestinal study. First, all the included studies were conducted in Asian
continuity and intestinal innervation. Although these countries, probably because the incidence of gastric cancer is
factors may lead to delayed gastric emptying, the higher in East Asia than in most Western countries,1,30
incidence and cause of RSS are not yet clearly elucidated. making the results more generalizable to East Asia than
The incidence of anastomotic leakage was similar other populations. Second, there was potential selection bias,
between the two groups, consistent with previous findings.28 given that the included studies are mostly observational.
No significant difference was found in postoperative Third, although we included all relevant studies, the sample
complications including anastomotic leakage, ileus, pancrea- size may not be sufficient. Hence, additional RCTs and large‐
titis, duodenal stump leakage, intra‐abdominal abscess, or scale studies are needed. Nonetheless, our study provides
wound issues between the two groups. This may be because clinical evidence of the optimal reconstruction for surgeons,
of the small difference between the two groups, and another which we hope is useful for surgeons worldwide.
reason to explain this is that postoperative complications
such as surgical site infection have been reduced with the
development of surgical technique and the using of 4 | CONCLUSIONS
antimicrobial prophylaxis these years,29 which make it
difficult to observed the difference between the two groups BB has the advantage in shortening the operative time,
with the small sample size of the included studies. anastomotic time, and intraoperative blood loss, and
Three studies15,16,20 reported postoperative functional reducing overall postoperative complication and delayed
findings using RGB classification. A significant difference gastric emptying. RY has the advantage of preventing bile
was observed only for bile reflux (p = 0.02) with high reflux and gastritis after surgery.
heterogeneity (I2 = 82%) when including all three studies
(Figure 4C), however, when excluding the study by
Yalikun et al,20 a significantly lower risk of bile reflux 5 | MATERIALS AND METHODS
(p < 0.00001) and gastritis (p < 0.00001) was observed in
the RY group with low heterogeneity (Supporting The study protocol was registered on PROSPERO
Information: Table S1), similar to previous reports.15,16 (Registration Number: CRD42022331178). The study
Interestingly, we noticed that Yalikun et al.20 used a was performed according to assessing the methodological
modified BB reconstruction technique, in which the quality of systematic reviews (AMSTAR) guidelines and
the Preferred Reporting Items for Systematic Reviews using the same or similar terms. Residual food, gastritis,
and Meta‐Analyses (PRISMA).31,32 and bile reflux as evaluated by the postoperative endo-
scopic findings 1 year after surgery and based on food
residue, gastritis, and bile reflux classification (RGB score),
5.1 | Literature search strategy score ≥ grade 1 were regarded as positive findings.33 Body
weight, hemoglobin, and albumin were defined as the
The PubMed, Embase, Web of Science, and Cochrane data measured 1 year after surgery reported in the study.
Library databases were searched for primary studies
published between 2000 and October 2022. Search terms
included the following MeSH terms (Medical Subject 5.4 | Quality assessment
Headings) and their combinations: “Billroth” (tiab) AND
“Braun” (tiab) AND “Roux en Y” (tiab) with no language The quality of the selected studies was assessed by two
restriction. To search for additional potentially eligible authors independently. The nonrandomized OCS was
studies, all references of the retrieved studies were evaluated using the Newcastle‐Ottawa Quality Assess-
reviewed. Both RCT and OCS were included in the study ment Scale and the RCT was assessed using the Jadad
and all titles and abstracts were extracted and evaluated. scoring system.34,35
The eligible studies were assessed and research data were
extracted by two reviewers independently. In there was
any disagreement, consensus was reached after discuss- 5.5 | Outcomes of interest
ing with a third researcher.
Perioperative outcomes, postoperative complications,
postoperative functional findings, and nutritional status
5.2 | Inclusion and exclusion criteria were evaluated. The main perioperative outcomes were
operation time, anastomosis time, intraoperative blood
All studies that met all of the following criteria were loss, harvested lymph nodes, hospital stay, and time to
included: (1) patients with gastric cancer who underwent first flatus or defecation. Postoperative complications
DG; (2) study compared BB and RY reconstruction; (3) study included anastomotic leakage, ileus, postoperative pan-
included at least one of the surgical outcomes mentioned; (4) creatitis, duodenal stump leakage, DGE, wound problem,
original research with ≥ 10 patients; and (5) study was and intra‐abdominal abscess. Postoperative functional
published in English. If two or more articles were published findings included residual food, gastritis, and bile reflux
by the same study group, institution, or used the same data which were endoscopically proven 1 year after surgery
set, the article with the largest sample size or the longest using RGB score. Postoperative nutritional status
follow‐up was selected. Any study that met one of the included serum hemoglobin, total protein, albumin
following criteria was excluded: (1) article type was abstract, levels, and body weight.
case reports, review articles, expert opinions, basic research,
animal experiments, or letters. (2) study without available
data or full text; (3) study including patients with benign 5.6 | Data extraction
disease; (4) sample size <10; or (5) study published
before 2000. Two authors collected the data from each of the selected
research papers independently, and disagreements were
discussed before a final decision was made. Outcomes of
5.3 | Definitions interest as well as population characteristics of each
study were collected carefully. The population character-
Perioperative outcomes included operation time, anasto- istics included author name, study period, geographical
mosis time, intraoperative blood loss, harvested lymph region, operation type, sample size, mean age, sex, tumor
nodes, hospital stay, and time to first flatus or defecation stage, body mass index, and American Society of
as defined as reported in the studies using the same or Anesthesiologists (ASA) score.
similar terms. Total complications were defined as all the
postoperative complications reported or categorized ac-
cording to the Clavien–Dindo classification.22 Anasto- 5.7 | Statistical analysis
motic leakage, ileus, postoperative pancreatitis, duodenal
stump leakage, DGE, wound problems, and intra‐ Review Manager Version 5.4 software (Nordic Cochrane
abdominal abscess were defined as reported in the studies Centre; Denmark) was used to conduct meta‐analyses. The
OR with a 95% CI was used to examine the dichotomous REFER ENCES

variables. The WMD with 95% CI was used to evaluate the 1. Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020:
continuous variables. Statistical significance was defined as GLOBOCAN estimates of incidence and mortality worldwide
a p‐value < 0.05. Before performing a meta‐analysis, the I2 for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):
209‐249.
value was calculated using a chi‐square test to evaluate the
2. Mocan L. Surgical management of gastric cancer: a systematic
homogeneity of effect sizes. If there was a significant
review. J Clin Med. 2021;10(12):11.
heterogeneity (I2 > 50%), a random‐effects model was
3. Kanaya S, Gomi T, Momoi H, et al. Delta‐shaped anastomosis
applied. If the heterogeneity was not statistically significant in totally laparoscopic Billroth I gastrectomy: new technique
(I2 < 50%), a fixed effects statistical model was applied. of intraabdominal gastroduodenostomy. J Am Coll Surg.
Subgroup and sensitivity analyses were carried out to 2002;195(2):284‐287.
identify the cause of the heterogeneity if it was high. To 4. Lee WJ, Wang W, Chen TC, Chen JC, Ser KH. Totally
identify potential publishing bias, a funnel plot was created. laparoscopic radical BII gastrectomy for the treatment of
The mean and standard deviation figures were not gastric cancer: a comparison with open surgery. Surg Laparosc
accessible in some publications, the procedures utilized to Endosc Percutan Tech. 2008;18(4):369‐374.
5. Vogel SB, Drane WE, Woodward ER. Clinical and radio-
calculate these values using the available median and range
nuclide evaluation of bile diversion by Braun enteroenter-
data have been reported previously.36
ostomy: prevention and treatment of alkaline reflux gastritis.
An alternative to Roux‐en‐Y diversion. Ann Surg. 1994;219(5):
AUTHOR CONTRIBUTIONS 458‐466.
Defei Chen: Conceptualization (equal); methodology 6. Takaori K, Nomura E, Mabuchi H, et al. A secure technique of
(equal); writing—original draft (equal). Chenglin Tang: intracorporeal Roux‐Y reconstruction after laparoscopic distal
Data curation (equal); formal analysis (equal). Fan He: gastrectomy. Am J Surg. 2005;189(2):178‐183.
Data curation (equal); formal analysis (equal). Fuyu 7. Uyama I, Sakurai Y, Komori Y, et al. Laparoscopy‐assisted
uncut Roux‐en‐Y operation after distal gastrectomy for gastric
Yang: Writing—review and editing (equal). Saed Wor-
cancer. Gastric Cancer. 2005;8(4):253‐257.
aikat: Writing—review and editing (equal). Kun Qian: 8. Fein M. Duodenoesophageal reflux induces esophageal adeno-
Conceptualization (equal); funding acquisition (equal); carcinoma without exogenous carcinogen. J Gastrointest Surg.
investigation (equal). All authors have read and approved 1998;2(3):260‐268.
the final manuscript. 9. Sato T, Miwa K, Sahara H, Segawa M, Hattori T. The
sequential model of Barrett's esophagus and adenocarci-
ACKNOWLEDGMENTS noma induced by duodeno‐esophageal reflux without
The authors are grateful to all their colleagues who helped exogenous carcinogens. Anticancer Res. 2002;22(1a):
39‐44.
prepare this article and Editage (www.editage.com) for
10. Li Y, Wang Q, Yang KL, Wang J, Jiang KW, Ye YJ. Uncut
English language editing. This study was funded by the Roux‐en‐Y might reduce the rate of reflux gastritis after
Chongqing Medical Scientific Research Project (Joint project radical distal gastrectomy: an evidence mapping from a
of Chongqing Health Commission and Science and systematic review. Int J Surg. 2022;97:11.
Technology Bureau) General Program under Grant (number 11. Xiong JJ. Roux‐en‐Y versus Billroth I reconstruction after
2021MSXM096). The funder was not involved in the study distal gastrectomy for gastric cancer: a meta‐analysis. World
design, collection, analysis, interpretation of data, writing of J Gastroenterol. 2013;19(7):1124‐1134.
12. Zong L, Chen P. Billroth I vs. Billroth II vs. Roux‐en‐Y
this article, or decision to submit it for publication.
following distal gastrectomy: a meta‐analysis based on 15
studies. Hepatogastroenterology. 2011;58(109):1413‐1424.
CONFLI CT OF I NTER EST STATEMENT 13. Sun MM, Fan YY, Dang SC. Comparison between uncut Roux‐
The authors declare no conflict of interest. en‐Y and Roux‐en‐Y reconstruction after distal gastrectomy
for gastric cancer: a meta‐analysis. World J Gastroenterol.
DATA A VAILABILITY S TATEMENT 2018;24(24):2628‐2639.
All data generated or analyzed during this study are 14. Lee MS, Ahn SH, Lee JH, et al. What is the best reconstruction
included in this published article. No additional method after distal gastrectomy for gastric cancer? Surg
unpublished data are available. Endosc. 2012;26(6):1539‐1547.
15. Park JY, Kim YJ. Uncut Roux‐en‐Y reconstruction after
laparoscopic distal gastrectomy can be a favorable method in
ETHI CS STATEMENT
terms of gastritis, bile reflux, and gastric residue. J Gastric
Our study is a systematic review and meta‐analysis; Cancer. 2014;14(4):229‐237.
hence, ethical approval was not required. 16. Choi C, Baek DH, Lee SH, et al. Comparison between Billroth‐
II with Braun and Roux‐en‐Y reconstruction after laparo-
ORCID scopic distal gastrectomy. J Gastrointest Surg. 2016;20(6):
Defei Chen http://orcid.org/0000-0002-2579-3866 1083‐1090.
17. Cui LH, Son SY, Shin HJ, et al. Billroth II with Braun 29. Marano L, Carbone L, Poto GE, et al. Antimicrobial
enteroenterostomy is a good alternative reconstruction to prophylaxis reduces the rate of surgical site infection in upper
Roux‐en‐Y gastrojejunostomy in laparoscopic distal gastrec- gastrointestinal surgery: a systematic review. Antibiotics‐Basel.
tomy. Gastroenterol Res Pract. 2017;2017:6. 2022;11(2):15.
18. Chi F, Lan Y, Bi T, Zhou S. Billroth‐II with Braun versus Roux‐ 30. Zhang T, Chen H, Yin X, et al. Changing trends of disease
en‐Y reconstruction in totally laparoscopic distal gastrectomy for burden of gastric cancer in China from 1990 to 2019 and its
gastric cancer. Wideochir Inne Tech Maloinwazyjne. 2021;16(4): predictions: findings from global burden of disease study. Chin
664‐668. J Cancer Res. 2021;33(1):11‐26.
19. Jun B, Nian L, Shan H, et al. Effects of different gastro- 31. Pieper D, Buechter RB, Li L, Prediger B, Eikermann M. Systematic
intestinal reconstruction techniques on nutrition, anemia, and review found AMSTAR, but not R(evised)‐AMSTAR, to have good
quality of life in laparoscopic distal gastrectomy for gastric measurement properties. JCE. 2015;68(5):574‐583.
cancer. Acta Cirúrgica Brasileira. 2022;37(4):e370408. 32. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement
20. Yalikun A, Aikemu B, Li S, et al. A modified Billroth‐II with for reporting systematic reviews and meta‐analyses of studies that
Braun anastomosis in totally laparoscopic distal gastrectomy: evaluate healthcare interventions: explanation and elaboration.
initial experience compared with Roux‐en‐Y anastomosis. Ann British Medical Journal. 2009;339:37.
Surg Oncol. 2022;29(4):2359‐2367. 33. Kubo M, Sasako M, Gotoda T, et al. Endoscopic evaluation of
21. Yan Y, Wang D, Liu Y, et al. Optimal reconstruction after the remnant stomach after gastrectomy: proposal for a new
laparoscopic distal gastrectomy: a single‐center retrospective classification. Gastric Cancer. 2002;5(2):83‐89.
study. Cancer Control. 2022;29:107327482210870. 34. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of
22. Dindo D, Demartines N, Clavien PA. Classification of surgical reports of randomized clinical trials: is blinding necessary?
complications: a new proposal with evaluation in a cohort of Control Clin Trials. 1996;17(1):1‐12.
6336 patients and results of a survey. Ann Surg. 2004;240(2): 35. Stang A. Critical evaluation of the Newcastle‐Ottawa scale for
205‐213. the assessment of the quality of nonrandomized studies in
23. Yoshino K. History of gastric cancer surgery. Nihon Geka meta‐analyses. Eur J Epidemiol. 2010;25(9):603‐605.
Gakkai zasshi. 2000;101(12):855‐860. 36. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and
24. Fukuhara K, Osugi H, Takada N, Takemura M, Higashino M, variance from the median, range, and the size of a sample.
Kinoshita H. Reconstructive procedure after distal gastrec- BMC Med Res Methodol. 2005;5:13.
tomy for gastric cancer that best prevents duodenogastroeso-
phageal reflux. World J Surg. 2002;26(12):1452‐1457. SU PPOR T ING INF OR MATI ON
25. Piessen G, Triboulet JP, Mariette C. Reconstruction after
Additional supporting information can be found online
gastrectomy: which technique is best? J Visceral Surg.
in the Supporting Information section at the end of this
2010;147(5):e273‐e283.
26. Kojima K, Inokuchi M, Kato K, Motoyama K, Sugihara K.
article.
Petersen's hernia after laparoscopic distal gastrectomy with
Roux‐en‐Y reconstruction for gastric cancer. Gastric Cancer.
2014;17(1):146‐151. How to cite this article: Chen D, Tang C, He F,
27. Ma Y, Li F, Zhou X, et al. Four reconstruction methods after Yang F, Woraikat S, Qian K. Comparison of
laparoscopic distal gastrectomy: a systematic review and
Billroth II with Braun and Roux‐en‐Y
network meta‐analysis. Medicine. 2019;98(51):e18381.
reconstructions after distal gastrectomy for gastric
28. Lombardo F, Aiolfi A, Cavalli M, et al. Techniques for
reconstruction after distal gastrectomy for cancer: updated
cancer: a meta‐analysis. MedComm – Oncology.
network meta‐analysis of randomized controlled trials. 2023;2:e48. doi:10.1002/mog2.48
Langenbecks Arch Surg. 2022;407(1):75‐86.

MedComm Oncology - 2023 - Chen

Uploaded by

Copyright:

Available Formats

MedComm Oncology - 2023 - Chen

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MedComm Oncology - 2023 - Chen

Uploaded by

Copyright:

Available Formats

Received: 28 March 2023 | Revised: 24 July 2023 | Accepted: 24 July 2023

Comparison of Billroth II with Braun and Roux‐en‐Y

Defei Chen | Chenglin Tang | Fan He | Fuyu Yang | Saed Woraikat |

Department of Gastrointestinal Surgery,

1 | INTRODUCTION respectively.1 Radiotherapy as well as chemotherapy and

MedComm – Oncology. 2023;2:e48. wileyonlinelibrary.com/journal/mog2 | 1 of 13

F I G U R E 1 PRISMA flow diagram of

TABLE 1 Characteristics of the included studies.

TABLE 2 Meta‐analysis of outcomes of interest.

presence of anastomotic tension. One of the primary

OR with a 95% CI was used to examine the dichotomous REFER ENCES

You might also like