If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Systematic review and meta-analysis comparing proximal gastrectomy with double-tract-reconstruction and total gastrectomy in gastric and gastroesophageal junction cancer patients: Still no sufficient evidence for clinical decision-making
To compare proximal gastrectomy with double-tract reconstruction and total gastrectomy in patients with gastroesophageal junction (AEG II–III) and gastric cancer.
Methods
We conducted systematic searches in Medline, Web of Science, and Cochrane Library until December 20, 2021 (PROSPERO registration number: CRD42021291500). Risk of bias was assessed using the revised Cochrane risk of bias tool and the ROBINS-I tool, as applicable. Evidence was rated by the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach.
Results
One randomized controlled trial (RCT) and 13 non-RCTs with 1,317 patients (715 patients with total gastrectomy and 602 patients with proximal gastrectomy with double-tract reconstruction) were included. Patients treated by total gastrectomy had a significantly higher proportion of advanced cancer stages International Union Against Cancer IB–III (odds ratio: 0.68, 95% confidence interval: 0.51–0.91, P = .01). This heterogeneity biases the observed improved overall survival of patients after proximal gastrectomy with double-tract reconstruction (odds ratio: 0.67, 95% confidence interval: 0.44–1.01, P = .05). Both procedures were comparably efficient regarding perioperative parameters. Postoperative/preoperative bodyweight ratio (mean difference: 3.56, 95% confidence interval: 1.32–5.79, P = .002), postoperative/preoperative serum-hemoglobin ratio (mean difference 3.73, 95% confidence interval: 1.59–5.88, P < .001), and postoperative serum vitamin B12 levels (mean difference 42.46, 95% confidence interval: 6.37–78.55, P = .02) were superior after proximal gastrectomy with double-tract reconstruction, while postoperative/preoperative serum-albumin ratio (mean difference 1.24, 95% confidence interval: -4.76 to 7.24, P = .69) and postoperative/preoperative serum total protein ratio (mean difference 1.12, 95% confidence interval: -2.77 to 5.00, P = .57) were not different. Health-related quality of life data were reported in only 2 studies, which found no significant advantages for proximal gastrectomy with double-tract reconstruction.
Conclusion
Proximal gastrectomy with double-tract reconstruction offers advantages in postoperative nutritional parameters compared to total gastrectomy (GRADE: moderate quality of evidence). Oncological effectiveness of proximal gastrectomy with double-tract reconstruction cannot be assessed (GRADE: very low quality of evidence). Further thoroughly planned randomized controlled trials in Western patient cohorts are necessary to improve treatment for gastric cancer patients.
Introduction
Upper gastrointestinal tract cancers originating in the gastroesophageal junction (AEG) or the stomach are one of the most frequent causes of cancer-related death worldwide. In Germany, AEG and gastric cancers are the fifth (men) and sixth (women) most commonly diagnosed types of cancer. AEG cancers are showing a dramatic increase in incidence, whereas gastric cancers of the distal part of the stomach are becoming less common in Western countries.
For early gastric cancer without risk factors (T1a/b sm1), endoscopic resection with endoscopic submucosal dissection (ESD) or endoscopic mucosa resection (EMR) is an option. The standard of surgical treatment for all other patients with resectable gastric cancer in the upper third of the stomach and AEG cancer type Siewert II and III
is total gastrectomy (TG) or transhiatal-extended gastrectomy. Surgical treatment is combined with perioperative chemotherapy in locally advanced cases.
Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial.
Total gastrectomy significantly impairs the long-term health-related quality of life (HRQoL) of the patients. Compared to subtotal-distal gastrectomy, HRQoL of patients with TG is substantially impaired for physical and role functioning, appetite loss, and eating restrictions.
The nutritional status and the HRQoL of patients undergoing subtotal gastrectomy has been shown to be superior to TG, while both procedures offer equal overall survival in 1 randomized controlled trial (RCT), provided that the proximal margin of resection falls in healthy tissue.
Subtotal versus total gastrectomy for gastric cancer: five-year survival rates in a multicenter randomized Italian trial. Italian Gastrointestinal Tumor Study Group.
As the results of subtotal-distal gastrectomy are superior to TG, efforts have been made to develop an organ-preserving approach for proximal gastric cancer and AEG cancers as well. Proximal gastrectomy offers survival similar to TG in retrospective studies, while HRQoL appeared to be improved.
The proximal gastrectomy procedure includes D1 and D2 lymphadenectomy and resection of the upper half to two-thirds of the stomach and the distal esophagus, followed in the past by esophagogastrostomy with gastric tube reconstruction or Merendino reconstruction with jejunal interposition.
Quality of life after surgical treatment of early Barrett’s cancer: a prospective comparison of the Ivor-Lewis resection versus the modified Merendino resection.
Total vs proximal gastrectomy for adenocarcinoma of the upper third of the stomach: a propensity-score-matched analysis of a multicenter Western experience (on behalf of the Italian Research Group for Gastric Cancer-GIRCG).
Comparison of three digestive tract reconstruction methods for the treatment of Siewert II and III adenocarcinoma of esophagogastric junction: a prospective, randomized controlled study.
PG-DTR is a proximal gastrectomy with preservation of the distal stomach. Double-tract reconstruction is performed after standardized Roux-en-Y reconstruction with an additional side-to-side anastomosis of the distal stomach to the alimentary limb of the Roux-en-Y reconstruction (Figure 1). According to the literature, this procedure has a lower rate of postoperative reflux esophagitis and other beneficial long-term effects compared to TG, including reduced postoperative loss of bodyweight and improved serum hemoglobin, albumin, and vitamin B12 levels
As there seems to be a complete lack of RCTs comparing PG-DTR and TG in patients with resectable nonmetastatic gastric cancer and AEG carcinoma type II and III in a prospective, randomized trial with oncologic endpoints (overall survival, disease-free/local recurrence-free survival) and HRQoL as endpoint, we aim to perform this RCT to assess the comparative effectiveness of PG-DTR according to the IDEAL framework for surgical innovation.
In the project development phase for this trial, a systematic review and meta-analysis were necessary to finalize the definition and adapt the research question including the design and methodology of the planned RCT taking the findings into account.
The review will address the following questions:
1.
Is PG-DTR noninferior compared to TG with regard to overall survival according to the current study pool?
2.
Is PG-DTR noninferior compared to TG with regard to disease-free survival/local recurrence-free survival according to the current study pool?
3.
Is PG-DTR superior to TG with regard to HRQoL?
4.
In what populations/settings has PG-DTR been evaluated?
5.
What patients are feasible for randomization between PG-DTR and TG?
a.
With regard to TNM-stage/neoadjuvant treatment.
b.
With regard to tumor localization in the stomach (minimal distance to the pylorus).
6.
What is the necessary size/volume of the remnant stomach for improvement of postoperative HRQoL compared to TG? Is TG with double-tract reconstruction feasible and beneficial?
7.
What are the chosen distances between esophagojejunostomy and jejunogastrostomy and Roux-en-Y jejunojejunostomy in the current study pool? Is HRQoL impaired by the distances between the anastomoses?
Methods
This systematic review is written with reference to the preferred reporting items for systematic reviews and meta-analyses statement 2020 (PRISMA 2020 statement).
Approval by the Ethics Committee of the University of Freiburg was not needed for this systematic review as this study is based exclusively on published literature and was therefore not attained.
Eligibility criteria
Participants/population
We focused on studies including patients with resectable nonmetastatic gastric cancer and AEG carcinoma type II and III. Studies including patients with and without multimodal treatment/(neo-)adjuvant treatment were eligible.
Intervention and comparator treatment
We considered proximal gastrectomy with double-tract reconstruction (PG-DTR) as eligible intervention. Total gastrectomy (TG) was the comparator treatment. We considered open surgical and laparoscopic/robotic approaches as eligible treatments.
Outcomes
The following outcomes were extracted:
•
Overall survival
•
Disease-free survival
•
Local recurrence-free survival
•
Health-related quality of life (HRQoL)
•
Postoperative/preoperative bodyweight ratio
•
Postoperative/preoperative serum hemoglobin ratio, postoperative/preoperative serum iron ratio, postoperative/preoperative serum vitamin B12 ratio, postoperative/preoperative serum albumin ratio, and postoperative/preoperative serum total protein ratio
•
Incidence of postoperative vitamin B12 substitution
Study types
RCTs, nonrandomized controlled studies (NRS), and observational studies (with control group) were eligible for the systematic review and meta-analysis. We did not consider single-arm studies because of the missing control group in that study design. Furthermore, review articles, clinical guidelines, and work that has not been peer-reviewed were excluded. We did not apply any exclusion criteria regarding study duration and setting. We only considered studies written in English or German.
Information sources
The searches for this systematic review were performed on December 20, 2021. The searches were conducted following the recommendations of PRESS (Peer Review of Electronic Search Strategies).
Titles and abstracts of records identified by the searches were screened, and the full text of all potentially relevant articles was obtained. Two independent reviewers checked full texts for eligibility, and reasons for exclusion were documented during full-text screening.
Extraction of study data
For quality assurance purposes, 2 independent reviewers extracted the study data. A third reviewer resolved discrepancies between the 2 reviewers.
•
Study characteristics: title, author, year of publication, journal, language, setting (geographical), trial duration, trial design, total number of patients, number of treatment groups
•
Patient characteristics: age, sex, disease, tumor stage (International Union Against Cancer [UICC] stages), multimodal treatment/(neo-)adjuvant treatment
•
Outcome parameters: as previously mentioned.
Risk of bias assessment
The risk of bias (RoB) was assessed by 2 independent reviewers using the revised Cochrane risk-of-bias tool for randomized trials (RoB 2)
as applicable. The RoB 2 includes 5 standard domains of bias: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcome, and bias in selection of the reported result. Each domain was judged to be “low risk of bias,” “some concerns,” or “high risk of bias.” The ROBINS-I tool covers 7 domains through which bias might be introduced into a nonrandomized study: bias due to confounding, bias in selection of participants into the study, bias in classification of interventions, bias due to deviations from intended interventions, bias due to missing data, bias in measurement of outcomes, and bias in selection of the reported result. The response options for each domain level are: “low risk of bias,” “moderate risk of bias,” “serious risk of bias,” “critical risk of bias,” and “no information.” Rating of quality of evidence was performed for each outcome variable using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach.
and SPSS version 28.0.1.0 (IBM Corp, Armonk, NY). Relevant outcome parameters from the included trials were assessed for estimation of treatment effects. Odds ratios (OR) and associated 95% confidence intervals were calculated for dichotomous data by Mantel-Haenszel models. Weighted mean differences (MD) and associated 95% confidence intervals were calculated using inverse-variance models. When necessary, time-to-event data were estimated with indirect methods.
Statistical heterogeneity among trials was evaluated using I2 statistics. In the case of statistical heterogeneity (I2 > 50%), a random effects model instead of a fixed effects model was used for meta-analysis. To investigate the risk of publication bias, funnel plots were generated for meta-analyses and tested for asymmetry with the Harbord test.
A total of 1,189 articles were screened for eligibility. Of these, 90 were assessed in full-text analysis. A total of 76 publications were excluded during full-text analysis because of studies concerning reconstructions other than PG-DTR after PG as experimental intervention (n = 28), studies not using TG as comparator treatment (n = 7), studies without comparator treatment (n = 1), studies with missing details on reconstruction technique after PG (n = 13), publications as letters to the editor (n = 2), study protocol publications (n = 12), and studies with mixed reconstruction techniques after PG in which PG-DTR data were not dissoluble from other reconstructions after PG (n = 6) (Figure 2). Therefore, 14 studies with a total of 1,317 patients (715 patients with TG and 602 patients with PG-DTR; 113 patients from 2 studies with jejunal interposition not considered as eligible intervention for further analysis
Comparison of three digestive tract reconstruction methods for the treatment of Siewert II and III adenocarcinoma of esophagogastric junction: a prospective, randomized controlled study.
Functional evaluations comparing the double-tract method and the jejunal interposition method following laparoscopic proximal gastrectomy for gastric cancer: an investigation including laparoscopic total gastrectomy.
Table 1 presents the key characteristics of the identified 1 RCT, and 13 non-RCT studies (1 prospective and 12 retrospective cohort studies). Inclusion criteria were patients with AEG II–III and/or proximal gastric cancer. Most studies included patients with early gastric cancer; only 5 studies included more advanced tumor stages with T1-3/4a tumors and N+ patients.
Comparison of three digestive tract reconstruction methods for the treatment of Siewert II and III adenocarcinoma of esophagogastric junction: a prospective, randomized controlled study.
Functional evaluations comparing the double-tract method and the jejunal interposition method following laparoscopic proximal gastrectomy for gastric cancer: an investigation including laparoscopic total gastrectomy.
Clinical comparison of antrum-preserving double tract reconstruction vs Roux-en-Y reconstruction after gastrectomy for Siewert types II and III adenocarcinoma of the esophagogastric junction.
Mean patient age was 62.4 years (TG: 62.0 years; PG-DTR: 63.0 years), and 78.2% of patients were male (TG: 74.5%; PG-DTR: 78.2%). Only 1 study considered patients after neoadjuvant treatment, and even in that 1 study, only a small minority (6.6%) of patients received neoadjuvant treatment before the operation.
Comparison of three digestive tract reconstruction methods for the treatment of Siewert II and III adenocarcinoma of esophagogastric junction: a prospective, randomized controlled study.
Functional evaluations comparing the double-tract method and the jejunal interposition method following laparoscopic proximal gastrectomy for gastric cancer: an investigation including laparoscopic total gastrectomy.
Functional outcomes according to the size of the gastric remnant and type of reconstruction following open and laparoscopic proximal gastrectomy for gastric cancer.
Clinical comparison of antrum-preserving double tract reconstruction vs Roux-en-Y reconstruction after gastrectomy for Siewert types II and III adenocarcinoma of the esophagogastric junction.
Comparison of laparoscopic proximal gastrectomy with double-tract reconstruction and laparoscopic total gastrectomy in terms of nutritional status or quality of life in early gastric cancer patients.
Similar hematologic and nutritional outcomes after proximal gastrectomy with double-tract reconstruction in comparison to total gastrectomy for early upper gastric cancer.
Short-term surgical outcomes of laparoscopic proximal gastrectomy with double-tract reconstruction versus laparoscopic total gastrectomy for adenocarcinoma of esophagogastric junction: a matched-cohort study.
Short and long-term outcomes after proximal gastrectomy with double tract reconstruction for Siewert type III adenocarcinoma of the esophagogastric junction: a propensity score matching study from a 10-year experience in a high-volume hospital.
Initial results of laparoscopic proximal gastrectomy with double-tract reconstruction using oblique jejunogastrostomy method on the long-term outcome of postoperative nutritional status: a propensity score-matched study.
AEG, gastroesophageal junction; BMI, body mass index; C, control; DTR, double-tract reconstruction; I, intervention; JG, jejunogastrostomy ; NA, not available; PG, proximal gastrectomy; TG, total gastrectomy; UICC, International Union Against Cancer.
Eleven studies included only laparoscopic/robotic procedures, 1 study performed open procedures, and 2 studies combined laparoscopic and open surgical access.
Besides the available study pool of completed studies, our screening of trial registries yielded 4 ongoing/nonpublished RCTs and 1 NRS comparing PG-DTR and TG. All ongoing studies are located in Asian countries (Supplementary Table S2).
Risk of bias assessment
The RCT reported adequate measures of randomization, and the trial was of low risk of bias due to missing outcome data and selective reporting of results. The effect of assignment and the measurement of outcome were of some concern. The overall RoB assessment was “some concerns” (Supplementary Table S3).
The prospective cohort study and 7 retrospective cohort studies were considered to be of serious risk of bias, and 6 studies were considered to be of moderate risk of bias due to confounding. Bias due to selection of participants into the study was considered serious in 5 retrospective cohort studies, moderate in the prospective cohort study and 6 retrospective cohort studies, and low in 1 retrospective cohort study. Bias in classification of intervention, due to deviation from intended intervention and due to measurement of outcome, was considered low in all studies. Bias due to missing data was moderate in the prospective cohort study and 4 retrospective cohort studies and low in 8 retrospective cohort studies. Bias in selection of the reported results was moderate in the prospective cohort study and 11 retrospective cohort studies due to the missing prospective study protocols/registration in all studies and serious in 1 study.
Overall RoB assessment was moderate risk of bias in 6 retrospective cohort studies and serious risk of bias in the prospective cohort study and 6 retrospective cohort studies (Supplementary Table S4).
Quantitative analyses
A summary of all quantitative results is shown in Table 2 and Supplementary Figures S1 and S2. No differences were observed concerning patients’ age and sex. A statistically significant and clinically relevant imbalance was identified in the tumor stage of the treated patients. A significantly higher proportion of advanced tumor stages was found in patients undergoing TG compared to PG-DTR procedures (OR 0.68, 95% CI: 0.51–0.91, P = .010; Figure 3, A ).
Table IIQuantitative analyses comparing PG-DTR and TG
Postoperative/preoperative bodyweight ratio (12 mo)
11 (1,055)
MD: 3.56
1.32, 5.79
.002
88%
<.001
Random
Favors PG-DTR
Postoperative/preoperative serum hemoglobin ratio (12 mo)
8 (741)
MD: 3.73
1.59, 5.88
<.001
65%
.005
Random
Favors PG-DTR
Serum iron (12 mo)
3 (141)
MD: 13.85
4.13, 23.58
.005
72%
.03
Random
Favors PG-DTR
Serum vitamin B12 (12 mo)
4 (343)
MD: 42.46
6.37, 78.55
.02
80%
.002
Random
Favors PG-DTR
Vitamin B12 substitution (24 mo)
3 (408)
OR: 0.06
0.00, 1.02
.05
93%
<.001
Random
Favors PG-DTR
Postoperative/preoperative serum Albumin ratio (12 mo)
7 (493)
MD: 1.24
-4.76, 7.24
.69
93%
<.001
Random
Favors-PG-DTR
Postoperative/preoperative serum total protein ratio (12 mo)
6 (463)
MD: 1.12
-2.77, 5.00
.57
91%
<.001
Random
Favors PG-DTR
CI, confidence interval; DTR, double-tract reconstruction; MD, mean difference; OR, odds ratio; PG, proximal gastrectomy; TG, total gastrectomy; UICC, International Union Against Cancer.
Figure 3Forest plots with the comparison of (A) the proportion of patients with more advanced UICC stages IB–III in the current study pool comparing PG-DTR and TG and (B) overall survival after resection of gastric cancer (all patients/tumor stages). A bias toward higher UICC stages in patients treated by TG can be observed, severely limiting the validity of the overall survival analysis.
Operation time and postoperative hospital stay showed no significant differences between the 2 procedures. Meta-analysis showed nonsignificant differences between PG-DTR and TG in terms of overall postoperative Clavien-Dindo classification grade III–V complications and incidence of postoperative anastomotic leakage. In terms of long-term clinical outcomes, there were no significant differences between the 2 procedures in the incidence of reflux esophagitis and anastomotic strictures. Overall survival as the primary oncological outcome was significantly improved in the PG-DTR group. This result has to be interpreted very carefully, as statistically significant differences in the tumor grades of the treated patients were present that explain the worse survival probability in the TG group (Figure 3, A and B). Only very limited survival data from 2 studies were available for more advanced tumor stages UICC stage II–III.
Functional outcomes according to the size of the gastric remnant and type of reconstruction following open and laparoscopic proximal gastrectomy for gastric cancer.
Short and long-term outcomes after proximal gastrectomy with double tract reconstruction for Siewert type III adenocarcinoma of the esophagogastric junction: a propensity score matching study from a 10-year experience in a high-volume hospital.
There was no significant difference in overall survival in these patients. Data concerning disease-free and recurrence-free survival were reported in 3 retrospective cohort studies.
Functional outcomes according to the size of the gastric remnant and type of reconstruction following open and laparoscopic proximal gastrectomy for gastric cancer.
Similar hematologic and nutritional outcomes after proximal gastrectomy with double-tract reconstruction in comparison to total gastrectomy for early upper gastric cancer.
Neither local nor distant recurrences were observed in these 227 patients; therefore, quantitative analysis was not possible for these outcomes (follow-up interval: 12, 24, and 22.7 months).
Significant differences were observed in the 12-month postoperative/preoperative bodyweight ratio, as less postoperative loss of bodyweight was observed in patients after PG-DTR compared to TG. Furthermore, patients showed statistically significantly less postoperative reduction of serum hemoglobin, serum iron, and serum vitamin B12 levels after PG-DTR (Figure 4). The rate of patients requiring vitamin B12 substitution was lower in PG-DTR patients. No significant differences were observed in postoperative serum protein and albumin levels.
Figure 4Comparison of (A) postoperative/preoperative bodyweight ratio and (B) postoperative/preoperative serum hemoglobin ratio. PG-DTR offers statistically significant better nutritional long-term result compared to TG. This leads to 3.6% less change in postoperative/preoperative bodyweight ratio and 3.7% less decrease of postoperative/preoperative serum hemoglobin ratio.
Comparison of laparoscopic proximal gastrectomy with double-tract reconstruction and laparoscopic total gastrectomy in terms of nutritional status or quality of life in early gastric cancer patients.
The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
Characteristics and clinical relevance of postgastrectomy syndrome assessment scale (PGSAS)-45: newly developed integrated questionnaires for assessment of living status and quality of life in postgastrectomy patients.
to evaluate HRQoL. Park et al observed no significant difference in the HRQoL between TG and PG-DTR. Both groups experienced a similar decline in symptom scales related to gastrointestinal symptoms in the early postoperative period, which partially improved after 1 year. Sato et al reported slight but nonsignificant advantages concerning quality of ingestion and amount of ingestion, whereas no clinically relevant advantages for PG-DTR were observed concerning clinical symptoms such as reflux, diarrhea, and dumping syndrome.
The surgical details regarding the distance between the anastomoses and the size of the gastric remnant were not quantifiable. Supplementary Table S5 provides a summary of the findings. Functional analyses regarding the size of the gastric remnant or the positions of the anastomoses were not available. One study found that a larger anastomosis between the gastric remnant and the alimentary limb seemed to offer functional benefits after PG-DTR.
Functional evaluations comparing the double-tract method and the jejunal interposition method following laparoscopic proximal gastrectomy for gastric cancer: an investigation including laparoscopic total gastrectomy.
Sensitivity analyses were performed to assess the effect of studies with serious risk of bias. No relevant differences regarding overall effects were observed when studies at a “serious risk of bias” were excluded from the meta-analyses (Supplementary Table S6). Furthermore, sensitivity analyses were performed to assess whether oncologic adequacy of PG-DTR was different between the indications gastric cancer and AEG cancer. No differences were found in these analyses (Supplementary Figure S3).
Publication bias
Funnel plots were created and analyzed visually and statistically for quantitative analysis of publication bias for tumor stage >UICC stage IA, perioperative complications, overall survival, the long-term outcomes reflux esophagitis and anastomotic stricture, and nutritional surrogate parameters bodyweight change, serum hemoglobin change, and vitamin B12 levels. No publication bias could be observed in the available study pool despite a small-study bias for the incidence of reflux esophagitis 12 months after surgery (Supplementary Figure S4).
Discussion
This systematic review investigated evidence comparing PG-DTR and TG in patients with AEG II–III or proximal gastric cancers. The quality of evidence was very low to moderate (Table III), which indicates a great demand for high-quality RCTs comparing the 2 procedures. The identified study pool consisted of Asian study cohorts only. The transferability of the results to a Western/European cohort may therefore be limited.
Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial.
It has to be acknowledged that the majority of included patients did not receive neoadjuvant treatment, although multimodal treatment of gastric and esophageal cancer is the standard of care for patients with advanced stages in Western countries, which represent most diagnosed cases in Western cohorts.
Selection bias present with serious effect on outcome
5-y overall mortality rate - only UICC stage II–III
462 per 1,000
361 per 1,000
0.66 [0.37, 1.17]
2 (111)
+ - - -Very low
Selection bias present with serious effect on outcome
Risk of major surgical complications (Clavien-Dindo classification grade III–V)
81 per 1,000
64 per 1,000
0.77 [0.43, 1.38]
8 (694)
+ + + -Moderate
Postoperative hospital stay
11.2 d
0.4 d less in intervention group
-0.44 d [-1.10, 0.21]
13 (1,274)
+ + - -Low
Reflux esophagitis
75 per 1,000
64 per 1,000
0.85 [0.51, 1.41]
10 (997)
+ + - -Low
Anastomotic stricture
24 per 1,000
21 per 1,000
0.88 [0.36, 2.15]
8 (727)
+ + - -Low
Postoperative/preoperative bodyweight ratio (12 mo)
84% of preoperative bodyweight 1 y after surgery
88% of preoperative bodyweight 1 y after surgery
3.56 [1.32,5.79]
11 (1055)
+ + + -Moderate
Postoperative/preoperative serum hemoglobin ratio (12 mo)
91% of preoperative serum hemoglobin value 1 y after surgery
95% of preoperative serum hemoglobin value 1 y after surgery
3.73 [1.59, 5.88]
8 (741)
+ + + -Moderate
Serum vitamin B12 (12 mo)
222 pg/mL
42 pg/mL higher serum vitamin B12
42.46 [6.37, 78.55]
4 (343)
+ + - -Low
DTR, double-tract reconstruction; GRADE, Grading of Recommendations Assessment, Development, and Evaluation; MD, mean difference; PG, proximal gastrectomy; TG, total gastrectomy; UICC, International Union Against Cancer.
GRADE Working Group grades of evidence: high quality, further research is very unlikely to change our confidence in the estimate of effect; moderate quality, further research is likely to have an important effect on our confidence in the estimate of effect and may change the estimate; low quality, further research is very likely to have an important effect on our confidence in the estimate of effect and is likely to change the estimate; very low quality, any estimate of effect is very uncertain.
The objective of this systematic review was to answer predefined research questions. The 2 main outcomes to be considered were overall survival probability as the main oncologic outcome and HRQoL as a patient-reported outcome. Especially the available data on overall survival have to be interpreted very carefully. The only available RCT had very few patients with tumor recurrences despite a substantial fraction of patients with advanced stages, and the other studies showed a significant misbalance of tumor stages with significantly higher UICC stages in the TG groups. The observation that overall survival after PG-DTR was improved compared to TG has to be assessed in knowledge of this clinical heterogeneity, which itself causes bias. Data concerning disease-free and local recurrence-free survival in patients after PG-DTR and TG were too limited to be evaluated. This research question therefore could not be answered. Concerning the extent of oncologic resection, we found that PG-DTR was associated with significantly fewer resected lymph nodes. This observation had to be expected as the extent of resection is greater in TG than in PG. Furthermore, resection of all lymph nodes of the lesser curvature (station 3 a/b) and partially of the lymph nodes at station 4d is possible with preservation of the distal stomach; therefore, lymph node yield could be improved in our opinion by a technically more radical lymphadenectomy even in PG. Moreover, the incidence of lymph node metastases is low at these aboral lymph node stations in AEG II and III cancers.
Incidence of lymph node metastasis at each station in Siewert types Ⅱ/Ⅲ adenocarcinoma of the esophagogastric junction: a systematic review and meta-analysis.
The number of resected lymph nodes is only a surrogate parameter for oncologic adequacy, and its effect on oncological survival data remains unclear. Whether PG-DTR offers noninferiority with regard to oncologic outcomes could therefore not be definitively answered. Other studies comparing PG (without double-tract reconstruction) with TG demonstrated that proximal gastric resection via PG seemed possible without compromising overall survival, even in patients with advanced proximal gastric cancer.
Especially for AEG II carcinoma, the optimal choice of resection (TG versus partial esophagectomy with gastric tube reconstruction) is still a matter of debate.
The CARDIA-trial protocol: a multinational, prospective, randomized, clinical trial comparing transthoracic esophagectomy with transhiatal extended gastrectomy in adenocarcinoma of the gastroesophageal junction (GEJ) type II.
Comparative data from randomized trials comparing PG and TG do not exist; further randomized trials investigating the potential of partially organ-preserving resection in patients with advanced proximal gastric cancer and AEG II/III and multimodal treatment are therefore necessary.
The second key outcome is HRQoL as a patient-reported outcome. The available HRQoL data were very sparse. Only 2 retrospective studies reported HRQoL data; the other studies and the RCT reported only surrogate parameters (eg, nutritional outcomes) for HRQoL. Both studies reporting HRQoL data found no measurable or statistically significant advantages in HRQoL for PG-DTR compared to TG. However, these HRQoL observations were made in 2 rather small cohorts of 80 and 49 patients, limiting the generalizability of the findings. The nutritional outcomes as surrogate parameters for better nutritional status and possibly a better general health status were generally found to be less affected by PG-DTR than by TG. TG leads to a significant and long-lasting impairment of postoperative HRQoL and nutritional status.
The meta-analyses in this systematic review showed a superiority of PG-DTR over TG in terms of prevention of profound postoperative loss of bodyweight and in the postoperative preservation of nutritional serologic parameters such as serum hemoglobin and serum vitamin B12 levels.
In relation to perioperative outcomes, PG-DTR was comparably efficient and safe as TG in terms of overall postoperative complication rate, incidence of anastomotic leakages, and length of hospital stay. PG-DTR was noninferior in terms of long-term functional outcomes anastomotic stricture and the incidence of reflux esophagitis. Anastomotic strictures and postoperative reflux esophagitis have been relevant problems with PG and reconstruction by esophagogastrostomy compared to TG, but this functional disadvantage to TG seems to be resolved by PG-DTR.
The current study pool did not reveal definitive answers regarding the necessary size of the gastric remnant and the exact positions of the anastomoses in the PG-DTR reconstruction procedure. One study found a correlation between the size of the gastric remnant, the diameter of the jejunogastrostomy, and functional advantages.
Functional outcomes according to the size of the gastric remnant and type of reconstruction following open and laparoscopic proximal gastrectomy for gastric cancer.
Double-tract reconstruction after TG was not beneficial in a randomized controlled trial; therefore, a gastric remnant and preservation of the pylorus seem to be necessary to facilitate PG-DTR’s functional advantages.
The minimal size of the gastric remnant for functional advantages, however, remains unclear, even if Japanese Gastric Cancer Guidelines recommend the preservation of at least 50% of the stomach.
A further detail with possible interference with postoperative functional results is the distance between the anastomoses in double-tract reconstruction after PG. In the current study pool, the distance between esophagojejunostomy and jejunogastrostomy ranged between 12–15 and 20 cm, and the distance between esophagojejunostomy and Roux-en-Y anastomosis ranged between 30 and 40–45 cm. The distance between the duodenum (Lig. of Treitz) and the Roux-en-Y anastomosis was consistently chosen to be 20 to 25 cm. Influences of these distances on functional outcomes have not yet been assessed. Differences between open and laparoscopic/robotic approaches have not been evaluated in the current study pool. Another study found that laparoscopic PG-DTR was a safe and feasible surgical method with less bleeding, similar postoperative complication rates, and reduced inflammatory response to the surgical trauma, but with an increased operation time.
Laparoscopic-assisted versus open proximal gastrectomy with double-tract reconstruction for Siewert type II-III adenocarcinomas of esophago-gastric junction: a retrospective observational study of short-term outcomes.
The main limitation of the current systematic review is the quality of the available studies. With only 1 RCT with a risk-of-bias assessment of “some concerns” and 6 cohort studies with “moderate risk of bias” and 7 studies with “serious risk of bias,” the quality of evidence is severely impaired, although sensitivity analyses revealed no relevant differences regarding the observed effects when studies with “serious risk of bias” were excluded (Table 3 and Supplementary Table S6). Another considerable limitation is the very sparse data concerning HRQoL, which make generalization of these results unfeasible. The review process was conducted strictly according to the PRISMA protocol. Another limitation was that 12 retrospective publications had to be excluded during the screening process, as data from different PG reconstruction techniques could not be distinguished, which limited the available study pool for PG-DTR. The overall quality of evidence and the conclusions of this systematic review would, however, most probably not have been changed by inclusion of subpopulations of these retrospective studies.
Although the available data suggest advantages regarding nutritional outcomes for patients with proximal gastric cancer without compromising oncological and surgical safety by PG-DTR compared to TG, the available evidence is insufficient. This leads to an ethical, clinical, and scientific need to further investigate PG-DTR as a surgical procedure and demonstrate its oncological noninferiority and possible beneficial aspects for patients’ postoperative HRQoL by means of a randomized controlled trial comparing PG-DTR with the standard treatment TG. Our aim is to conduct this trial comparing PG-DTR and TG in combination with modern multimodal treatment in a Western cohort of gastric cancer patients. To support this planned clinical trial, this systematic review was performed to analyze the current state of evidence concerning the comparison of TG and PG-DTR. We believe that with the results of our planned RCT, substantial progress can be achieved by optimization of the surgical treatment with prospective and randomized validation of PG-DTR for AEG and gastric cancer patients.
Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial.
Subtotal versus total gastrectomy for gastric cancer: five-year survival rates in a multicenter randomized Italian trial. Italian Gastrointestinal Tumor Study Group.
Quality of life after surgical treatment of early Barrett’s cancer: a prospective comparison of the Ivor-Lewis resection versus the modified Merendino resection.
Total vs proximal gastrectomy for adenocarcinoma of the upper third of the stomach: a propensity-score-matched analysis of a multicenter Western experience (on behalf of the Italian Research Group for Gastric Cancer-GIRCG).
Comparison of three digestive tract reconstruction methods for the treatment of Siewert II and III adenocarcinoma of esophagogastric junction: a prospective, randomized controlled study.
Functional evaluations comparing the double-tract method and the jejunal interposition method following laparoscopic proximal gastrectomy for gastric cancer: an investigation including laparoscopic total gastrectomy.
Clinical comparison of antrum-preserving double tract reconstruction vs Roux-en-Y reconstruction after gastrectomy for Siewert types II and III adenocarcinoma of the esophagogastric junction.
Functional outcomes according to the size of the gastric remnant and type of reconstruction following open and laparoscopic proximal gastrectomy for gastric cancer.
Short and long-term outcomes after proximal gastrectomy with double tract reconstruction for Siewert type III adenocarcinoma of the esophagogastric junction: a propensity score matching study from a 10-year experience in a high-volume hospital.
Similar hematologic and nutritional outcomes after proximal gastrectomy with double-tract reconstruction in comparison to total gastrectomy for early upper gastric cancer.
Comparison of laparoscopic proximal gastrectomy with double-tract reconstruction and laparoscopic total gastrectomy in terms of nutritional status or quality of life in early gastric cancer patients.
The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
Characteristics and clinical relevance of postgastrectomy syndrome assessment scale (PGSAS)-45: newly developed integrated questionnaires for assessment of living status and quality of life in postgastrectomy patients.
Incidence of lymph node metastasis at each station in Siewert types Ⅱ/Ⅲ adenocarcinoma of the esophagogastric junction: a systematic review and meta-analysis.
The CARDIA-trial protocol: a multinational, prospective, randomized, clinical trial comparing transthoracic esophagectomy with transhiatal extended gastrectomy in adenocarcinoma of the gastroesophageal junction (GEJ) type II.
Laparoscopic-assisted versus open proximal gastrectomy with double-tract reconstruction for Siewert type II-III adenocarcinomas of esophago-gastric junction: a retrospective observational study of short-term outcomes.
Short-term surgical outcomes of laparoscopic proximal gastrectomy with double-tract reconstruction versus laparoscopic total gastrectomy for adenocarcinoma of esophagogastric junction: a matched-cohort study.
Initial results of laparoscopic proximal gastrectomy with double-tract reconstruction using oblique jejunogastrostomy method on the long-term outcome of postoperative nutritional status: a propensity score-matched study.
00990-4&id=gr1.jpg){kind=link}
00990-4&id=gr2.jpg){kind=link}
00990-4&id=gr3.jpg){kind=link}
00990-4&id=gr4.jpg){kind=link}