Abstract
Background
We have been in constant search of novel innovations to decrease the high morbidity after Pancreaticoduodenectomy (PD). Pancreaticojejunostomy (PJ) and pancreaticogastrostomy (PG) are the two different methods of reconstruction after PD. However, the existing data is ambiguous in supporting either of them as the preferred technique of reconstruction.
Methods
This was a single-center prospective observational study that included 64 patients who underwent PD over two years. We compared PG with PJ as a method of reconstruction after PD. The primary objective was to assess whether PG decreases the rate of postoperative pancreatic fistula (POPF) rates or not. Secondary objectives comprised analysis of perioperative outcomes, 30-day and 90-day mortality.
Results
Pancreatic fistula was significantly lower in PG as compared to the PJ group (24% vs. 47%) with a p-value of 0.027. The incidence of clinically pertinent (grade B) fistula was only 3% in the PG group and 32% in the PJ group. PG group had a higher incidence of post pancreatectomy hemorrhage (PPH) and delayed gastric emptying (DGE). No statistically significant difference was seen between either group need for blood transfusion, re-exploration, re-admissions, ICU stay, or length of hospital stay, and 30-day and 90-day mortality. Pancreatic texture and high BMI were independent predictors for pancreatic fistula.
Conclusion
PG when compared to PJ for reconstruction after PD, decreases the rate of POPF significantly; however, it is associated with an elevated risk of DGE and PPH. There was no difference in 30-day and 90-day mortality between both the treatment groups.
Keywords: Pancreaticogastrostomy, Pancreaticojejunostomy, Postoperative pancreatic fistula, Post-pancreatectomy hemorrhage, Delayed gastric emptying
Introduction
Walter Kausch has been credited with the first two stage Pancreaticoduodenectomy (PD) in 1935.1 Advances in medical and surgical care have caused a steep decline in the mortality rate after PD and are currently reported to be less than 3% for 30-day mortality.2 But, the associated morbidity remains high and varies from 30% to 45%.3,4 The major contributing factor leading to the increase in morbidity was POPF. Grave secondary complications can arise as a sequel of spillage of pancreatic juice from the disrupted pancreatic anastomosis.5 Diverse anastomotic techniques have been advocated to decrease postoperative pancreatic fistula (POPF) rates.6 Waugh and Clagett from Mayo Clinic were first to perform pancreaticogastrostomy (PG) in 1946.7 Various theoretical and technical advantages attributed to PG include (1) Pancreatic enzymes get inactivated by the low pH of the gastric lumen and the deficiency of enterokinase in the gastric mucosa; (2) Alkaline pancreatic secretions help protect the pancreaticogastric anastomosis against marginal ulceration; (3) Posterior gastric wall is in close proximity to the pancreatic remnant allowing us to do anastomosis without any undue tension (4) Stomach has thick walls and a robust blood supply helping in a secure anastomosis.8 When we look at multiple studies, they claim that PG has the upper hand over PJ in decreasing POPF rates, and at the same time, other complications too.9, 10, 11, 12 At the same time, the data obtained from other trials and meta-analyses do not suggest such advantages.13, 14, 15, 16, 17
Materials and methods
Study design, hypothesis, inclusion, and exclusion criteria
The study was a single-center prospective study that included 64 patients in whom PD was carried out over 2 years at a high-volume tertiary care center. We assumed a lesser rate of POPF to occur in the PG group. We included all patients undergoing PD for either pancreatic or periampullary carcinoma or benign conditions with high risk for malignant transformation and pancreatic lesions with suspicious malignancy in the study. We excluded patients with unresectable or locally advanced and metastatic cancer on exploration/diagnostic laparoscopy, patients who required resection of other organs other than standard PD, patients having dual malignancy, and those who were found unfit for surgery were excluded from the study. The research is based on 64 patients who underwent PD over the course of 2 years at our center (Fig. 1). Approval from the hospital ethical committee was sought before starting the study. Written informed consent was taken from the patients.
Fig. 1.
Study flow diagram.
Technique
We did classical “PD” and pylorus-preserving pancreaticoduodenectomy in distinctly prescribed steps facilitating the safe removal of the specimen at the same time ensuring negative margins. For all PJ, we practiced duct to mucosa anastomosis, while for all PG, we used the dunking method for anastomosis.
Endpoint definitions
Primary objective
The primary objective was to assess whether PG, as compared to PJ, when used for reconstruction after PD decreases the rate of POPF. The grades of POPF were defined per the latest International Study Group for Pancreatic Fistula (ISGPF) Consensus meeting 2016.18 Three distinct grades of POPF, BL (biochemical leak), B and C were followed.
Secondary objectives
Standard ISGPS classification was followed for grading complications like delayed gastric emptying (DGE) and post pancreatectomy hemorrhage (PPH).19,20 Other secondary endpoints included the need for blood transfusion, wound infection, leak other than pancreatic fistula, duration of surgery, length of hospital stay; further secondary endpoints were needed for re-operation or re-admission mortality occurring within 30 days and 90 days after the operation. Composite grading of postoperative complications was done using Clavien-Dindo grading.21
Statistical analyses
We have presented the data as numbers, percentages, and mean ± standard deviation (SD). Analyses of either demographic or perioperative continuous variables were done using the Student t-test and Mann–Whitney test; at the same time, categorical variables were subjected to a two-sided Chi-square test and Fischer’s exact test. We derived the p-value using univariate analysis, and a value of <0.05 was considered significant. We performed all statistical analyses using SPSS 24 software.
Results
We screened sixty-eight patients from April 2018 to March 2020. One patient withdrew consent, while one patient had metastatic disease on diagnostic laparoscopy. Two patients had unresectable tumors as it was completely encasing the SMA. Sixty-four patients underwent PD, out of which 34 patients underwent PJ, and 30 patients underwent PG for reconstruction (Fig. 1). The decision to do PG or PJ was made as per the surgeon’s expertise in the technique. Surgeries were performed by specialized gastrointestinal surgeons performing more than ten PD each year.
Patient baseline characteristics and operation
We analyzed the patient’s baseline character. PG and PJ groups were comparable in terms of demographics, underlying comorbid condition, symptomatology, need for preoperative biliary drainage, physical status as per American Society of Anesthesiologists (ASA) classification, and laboratory parameters except for the fact that patients with higher carcinoembryonic antigen (CEA) levels underwent PJ more frequently as compared to PG, which was an incidental finding and was discovered retrospectively on analyzing the data after completion of the study (Table 1). There was no statistically important difference between the PG and the PJ group based on the type of surgery (classical PD vs. PPPD) performed (Table 1).
Table 1.
Association between procedure and parameters.
| Parameters | Procedure |
p-value | |
|---|---|---|---|
| Pancreaticogastrostomy (n = 30) | Pancreaticojejunostomy (n = 34) | ||
|
Pre-operative variables | |||
| Age in years | 55.53 ± 12.91 | 55.47 ± 10.49 | 0.9831 |
| Sex | |||
| Male | 16 (53.3%) | 23 (67.6%) | 0.2422 |
| Female | 14 (46.7%) | 11 (32.4%) | |
| BMI inkg/m2 | 22.26 ± 2.16 | 22.68 ± 1.61 | 0.3911 |
| Jaundice | 24 (80.0%) | 32 (94.1%) | 0.1333 |
| Weight Loss | 6.17 ± 4.90 | 6.71 ± 3.94 | 0.3704 |
| Pain | 12 (40.0%) | 17 (50.0%) | 0.4232 |
| Itching | 23 (76.7%) | 31 (91.2%) | 0.1693 |
| H/O Alcohol | 8 (26.7%) | 16 (47.1%) | 0.0932 |
| H/O Smoking | 16 (53.3%) | 13 (38.2%) | 0.2262 |
| DM | 9 (30.0%) | 9 (26.5%) | 0.7542 |
| HTN | 9 (30.0%) | 12 (35.3%) | 0.6532 |
| CAD | 2 (6.7%) | 2 (5.9%) | 1.0003 |
| Cholangitis | 4 (13.3%) | 9 (26.5%) | 0.1922 |
| Haemoglobin (gm/dl) | 11.63 ± 1.76 | 11.68 ± 1.53 | 0.6814 |
| TLC (103/cumm) | 8.42 ± 3.01 | 8.75 ± 3.29 | 0.6234 |
| Albumin (g/dl) | 3.22 ± 0.59 | 3.33 ± 0.59 | 0.2964 |
| Bilirubin (mg/dl) | 12.02 ± 7.50 | 13.27 ± 7.64 | 0.5104 |
| INR | 1.16 ± 0.34 | 1.12 ± 0.29 | 0.6964 |
| ALP (IU/L) | 421.83 ± 225.10 | 418.85 ± 182.23 | 0.9541 |
| ASA Grade | |||
| I | 6 (20.0%) | 7 (20.6%) | 0.1692 |
| II | 20 (66.7%) | 16 (47.1%) | |
| III | 4 (13.3%) | 11 (32.4%) | |
| Pre op ERCP/PTBD | 4 (13.3%) | 7 (20.6%) | 0.4432 |
| Stent | 3 (10.0%) | 7 (20.6%) | 0.9532 |
| Tumor Size | |||
| <2 cm | 14 (46.7%) | 15 (44.1%) | 0.9413 |
| 2–5 cm | 12 (40.0%) | 15 (44.1%) | |
| >5 cm | 4 (13.3%) | 4 (11.8%) | |
| Lymph nodes | 10 (33.3%) | 17 (50.0%) | 0.1782 |
| Encasement of vessels | 1 (3.3%) | 1 (2.9%) | 1.0003 |
| CEA∗∗∗(ng/ml) | 4.12 ± 7.09 | 5.52 ± 6.33 | 0.0064 |
| CA 19.9(u/ml) |
234.82 ± 331.13 |
165.04 ± 261.74 |
0.8724 |
|
Intraoperative Variables | |||
| Blood loss | 338.33 ± 112.80 | 343.24 ± 111.29 | 0.5534 |
| Blood Transfusion (units) | 0.47 ± 0.63 | 0.65 ± 0.73 | 0.3384 |
| Duration of surgery (mins) | 313.67 ± 38.73 | 310.00 ± 35.51 | 0.8664 |
| Pancreatic texture Firm/Soft | |||
| Firm | 9 (30.0%) | 8 (23.5%) | 0.5592 |
| Soft | 21 (70.0%) | 26 (76.5%) | |
| Pancreatic duct diameter (mm) | 3.77 ± 1.01 | 3.79 ± 1.04 | 0.8834 |
| Intraoperative Lymph nodes | 9 (30.0%) | 16 (47.1%) | 0.1632 |
| Vessel encasement | 1 (3.3%) | 0 (0.0%) | 0.4693 |
| Classical Whipples | 28 (93.3%) | 33 (97.1%) | 0.5963 |
| PPPD |
0.07 ± 0.25 |
0.03 ± 0.17 |
0.4974 |
|
Postoperative Variable | |||
| Wound infection (Present) | 10 (33.3%) | 12 (35.3%) | 0.8692 |
| Biliary leak (Present) | 0 (0.0%) | 2 (5.9%) | 0.4943 |
| Gastric leak (Present) | 0 (0.0%) | 0 (0.0%) | 1.0002 |
| ISGPS DGE grading | |||
| A | 3 (38.5%) | 2 (5.8%) | 0.0383 |
| B | 6 (20.0%) | 1 (2.9%) | |
| C | 2 (6.0%) | 0 (0.0%) | |
| ISGPS PPH grading | |||
| A | 5 (16.6%) | 3 (8.82%) | 0.7943 |
| B | 6 (20%) | 1 (2.94%) | |
| C | 2 (6.6%) | 1 (2.94%) | |
| ISGPS PF grading∗∗∗ | |||
| BL | 7 (23.3%) | 5 (14.7%) | 0.0273 |
| B | 1 (3%) | 11 (32.3%) | |
| C | 0 (0.0%) | 0 (0.0%) | |
| Reexploration | 1 (3.3%) | 1 (2.9%) | 1.0003 |
| Readmission | 0 (0.0%) | 2 (5.9%) | 0.4943 |
| ICU stay | 1.77 ± 1.43 | 1.74 ± 2.47 | 0.3754 |
| Total Hospital stay | 16.73 ± 8.89 | 14.62 ± 6.56 | 0.1154 |
| Pancreatic Adenocarcinoma | 15 (50.0%) | 16 (47.1%) | 0.8142 |
| Duodenal Carcinoma | 2 (6.7%) | 4 (11.8%) | 0.6763 |
| Ampullary Carcinoma | 7 (23.3%) | 7 (20.6%) | 0.7912 |
| Cholangiocarcinoma | 4 (13.3%) | 2 (6.1%) | 0.4123 |
| Cystic neoplasm of pancreas | 1 (3.3%) | 1 (2.9%) | 1.0003 |
| Chronic Pancreatitis | 0 (0.0%) | 2 (5.9%) | 0.4943 |
| Neuroendocrine Tumour | 1 (3.3%) | 2 (5.9%) | 1.0003 |
| R0 RESECTION | 30 (100.0%) | 33 (97.1%) | 1.0003 |
| Mortality (within 1 month) | 1 (3.3%) | 0 (0.0%) | 0.4693 |
| 90 days mortality | 2 (6.6%) | 1 (2.9%) | 0.5963 |
| Clavien Dindo Grade | |||
| No Complication | 8 (26.6%) | 8 (23.5%) | 0.7733 |
| I(Basal atelectasis,SSI) | 3 (10.0%) | 5 (14.7%) | |
| II(DGE, Pneumonia, VT, POPF, PPH) | 5 (16.6%) | 10 (29.4%) | |
| IIIA (DGE,POPF,PPH) | 6 (20%) | 12 (35.2%) | |
| IIIB(PPH, Biliary leak) | 2 (6.6%) | 1 (2.9%) | |
| IV(AKI) | 0 (0) | 1 (2.9%) | |
| V (MI,MSOF) | 2 (6.6%) | 1 (2.9%) | |
∗∗∗Significant at p < 0.05, 1: t-test, 2: Chi-Squared Test, 3: Fisher’s Exact Test, 4: Wilcoxon-Mann-Whitney U Test,; CA19-9, Carbohydrate antigen19-9; CEA, Carcinoembryonic antigen; TLC, Total leukocyte count; ALP, Alkaline phosphatase; INR, International normalized ratio ASA; American Society of Anesthesiologists; BMI, body mass index; CAD, Coronary artery disease; COPD, Chronic obstructive pulmonary disease; PPPD, Pylorus preserving pancreaticoduodenectomy; ICU, Intensive care unit; NG, Nasogastric tube DGE, Delayed gastric emptying; PPH, Post pancreatectomy hemorrhage; MOSF, Multisystem organ failure.
Analyses of primary and secondary objectives
The pancreatic fistula was significantly lower in PG as compared to the PJ group (24% vs. 47%), with a p-value of 0.027. The incidence of clinically pertinent (grade B) fistula was only 3% in the PG group and 32% in the PJ group (Fig. 2). The relative risk for pancreatic fistula was in PG vs. PJ was 0.51 (95% CI 0.28–1.09). There was no patient having grade C type pancreatic fistula. Both PG and PJ groups were comparable in terms of duration of surgery and blood loss/intraoperative transfusion requirements, and none of them attained statistical significance. The overall incidence of delayed gastric emptying was higher in the PG group as compared to the PJ group and was statistically significant with a p-value of 0.038 (Fig. 3). Compared to PJ, there was a higher rate of PPH in the PG group; however, it did not reach statistical significance with a p-value of 0.794 (Fig. 4). Collectively complications were graded as per by Clavien, most commonly belonged to grade II and III. Patients in the PJ group had higher chances of complications as compared to the PG group but were not statistically significant with a p-value of 0.73 (Table 1). There was no significant difference between either group in need of postoperative blood transfusion, re-exploration, and admissions. Length of hospital stay, ICU stay, 30-day and 90-day mortality did not attain statistical significance while comparing PG vs. PJ.
Fig. 2.
Bar chart showing International study group of pancreatic surgery (ISGPS) grades of pancreatic fistula in PG and PJ group after PD along the X-axis and percentage of pancreatic fistula in the Y-axis.
Fig. 3.
Bar chart showing ISGPS grades of delayed gastric emptying in PG and PJ group after PD along the X-axis and its percentage among patients having DGE along the Y-axis.
Fig. 4.
Bar chart showing ISGPS grades of post pancreatectomy hemorrhage in PG and PJ group after PD along the X-axis and its percentage among patients having PPH along the Y-axis.
Discussion
The first Whipple procedure dates back to 1935 and is still evolving.1 We have achieved significant success in decreasing mortality, but high morbidity still poses a challenge, the most frequent morbidity being POPF. Researchers have constantly claimed about finding new techniques to decrease the incidence of POPF. A recently concluded retrospective case series by Krishna et al included 467 patients. All patients underwent PG as a method of reconstruction after PD, and it was reported that none of the patients developed clinically significant pancreatic fistula12 and concluded that PG is a safe and effective method for reconstruction after PD. Our study concluded that the rate of pancreatic fistula was significantly lower in the PG group as compared to the PJ group. We also concluded that the PG group had a majority of cases with a less severe type of pancreatic fistula (grade BL) as compared to the PJ group, where the majority of patients had a more severe type of pancreatic fistula (grade B). Similar results have been reported by Menahem et al in a meta-analysis of 1121 patients with 562 patients in the PG arm and 559 in the PJ arm. They found that the PG group had a lower rate of pancreatic fistula compared to the PJ group (11.2% vs. 18.7%).9 Zhou et al in the meta-analysis of six RCT, including 1005 patients, inferred that the PG group was associated with a significantly lower pancreatic fistula rate as compared with the PJ group with a p-value of 0.01.10 Guerrini et al too showed a similar benefit of PG. His meta-analysis of 8 RCTs, which included 1211 patients, came up with the inference that the PG group was associated with a significantly lower rate of pancreatic fistula with a p-value of 0.03 and odds ratio of 0.053.11
DGE is one of the commonest complications after pancreaticoduodenectomy. By and large, DGE is self-limiting; however, it adds to patient discomfort and also leads to an increase in the length of hospital stay, thereby increasing healthcare-related costs.21 The possible mechanisms behind DGE include ischemic insult to pylorus and antrum, denervation of the pylorus, and reduced level of motilin.22 In our study, patients in the PG group had a higher incidence of DGE as compared to the PJ group. In a study of 73 patients followed retrospectively by Hayama et al, a higher rate of DGE was associated with PG (40.5%) in comparison to the PJ group (17.3%) with a p-value of 0.019.23
Bleeding was more commonly associated with the PG group as compared to the PJ group (43% vs. 14.7%). The results are consistent with the findings of the systemic review by Clerveus et al.24 Jin et al in his meta-analysis of 1756 patients from 11 studies reported a higher incidence of PPH in the PG group when compared with the PJ group with an odds ratio of 1.47 (95%CI 1.05–2.06) with a p-value of 0.03.13 Similar findings were reported by Tobias et al in a multicenter German RCT of 320 patients.15 Lyu et al in meta-analyses of 1184 patients also pointed towards the increased risk of PPH in the PG group.25 Endoscopic intervention was needed in two patients who underwent PG as a method of reconstruction for UGI bleed. Both the patients needed re-exploration because of failed endoscopic intervention. One patient in the PJ group had intra-abdominal bleed following an iatrogenic injury during radiologically guided percutaneous drain placement for an intra-abdominal collection.
The duration of surgery was lower in the PG group; however, the same was not statistically significant. The same has been reported by Perivoliotis et al in a meta-analysis of ten studies, including 1629 patients.8 The study by Perivoliotis et al also concluded that both PG and PJ groups were similar in terms of intraoperative blood loss and SSI. In our study, we too could not find a statistically significant difference considering the intraoperative blood loss and wound infection rate between both the treatment groups. More number of patients underwent re-exploration in the PG group in contrast to the PJ group, but the association was not statistically significant. The overall re-exploration rate was 4.6%. Re-exploration was mainly done for severe UGI bleed after failed endoscopy, as already discussed. Krishna et al, in a large retrospective study, reported a re-operation rate of 4.7%, and the most common cause of re-exploration was bleeding from the pancreatic stump, similar to our finding.12 Qiu et al, in a retrospective study of 330 patients, reported the rate of re-exploration to be 6.67%, and the most common cause for re-exploration was PPH.26 Total hospital stay and ICU stay of both the groups was not statistically significant, PG vs. PJ group with a p-value of 0.115 and 0.375, respectively. The mean hospital stay was around 16 days, which is in concordance with the available literature.15 The main cause of prolonged hospital stay was either pancreatic fistula or delayed gastric emptying. We did not have any case of a gastric leak in our study; however, we had two patients with a biliary leak in the PJ group. One patient had bile coming out through the intraoperatively placed drain, which subsided on its own on the fourth postoperative day, while the other patient required radiologically guided percutaneous drain placement after an ultrasound showed collection near the hepaticojejunostomy site. But increased biliary leak associated with the PJ group did not acquire any statistical significance while comparing with the PG group. Similar was the finding by Lyu and Tobias et al.15,20 The 30-day mortality and 90-day mortality were also statistically insignificant between PG and PJ with a p-value of 0.469 and 0.596, respectively. A similar finding was reported by Tobias et al.21 We had in total three mortalities. There was 1 (1.5%) 30-day mortality because of a cardiac event in an elderly patient. There were two (3.15%) 90-day mortalities; both died of multisystem organ failure. The retrospective study by Narayanan et al consisting of 551 patients, showed a 1.1% 30-day mortality and 3.6% 90-day mortality.27 Our mortality rates were within the acceptable range.26,27
Our study is the only prospective study comparing PG vs. PJ as a method of reconstruction after PD group for reconstruction after PD, done after the revised ISGPS 2016 classification. To the best of our knowledge, the last prospective study on this topic was RECOPANC trial done from 2011 to 2012, which concluded that the rate of the clinically pertinent fistula was not different between the PG and PJ groups after PD.15 However the study was flawed in certain aspects, 50% of the PD was performed surgeons performing less than 10 PD/year whereas it has previously been seen that surgeon’s experience has a direct influence on both mortality and morbidity after PD.28 It was a multicentric study, so the training was not uniform. The median BMI for their study was 25 kg/m2 and at least 50% of the patients had soft pancreas, and both of them are proven risk factors for pancreatic fistula that could lead to confounding bias.24,25 We had multiple meta-analyses in the last decade, all quoting older studies. To the best of our knowledge latest available meta-analyses on this topic are by Jin et al, published in 2019, who quoted older studies.13 So our study is looking at an old issue with a fresh set of eyes in the milieu of recent advances with the availability of advanced energy sources, better stapling devices, improved anesthetic care, and postoperative care. We followed standard protocols for perioperative and postoperative care for all our patients.
We had a small sample size of 64 patients based on routine hospital admissions, which is a limitation of our study but based on our limited experience, we can emphasize the fact that the risk of POPF can be significantly decreased by doing PG for reconstruction in place of PJ; however, we further recommend fresh multicentric RCTs to validate the findings of our study as there has been tremendous improvisation of surgical techniques, better energy sources, improved intraoperative and postoperative care using latest facilities at our disposal. Patient who had higher CEA levels underwent PJ more frequently, which can be a limitation of our study, but this was an accidental finding after completion of the whole study and analysis of the data. Moreover, none of the available literature suggest a higher CEA as a risk factor for POPF. The surgeon performing the surgery according to his expertise in the technique of either PG or PJ can lead to bias and is another limitation of our study. We performed invaginating PG in all our cases, which could have led to a higher incidence of DGE in this group. Hayama et al had postulated a greater disturbance in peristaltic activity with invaginating PG leading to a higher incidence of DGE in the group.23 The same can also be considered as a limitation of our study.
Conclusion
Our hypothesis is that PG, when used as a method for reconstruction after PD, decreases the rate of pancreatic fistula as compared to PJ holds. We have a higher risk of having PPH with PG as a method of reconstruction. There was an increased risk of associated DGE in the PG group, but it was mostly self-limiting. Both procedures were comparable in terms of 30-day and 90-day mortality. So, the take-home message is that performing PG can help us in dealing with the most limiting complication of POPF, but we need to be very meticulous with the pancreatic stump to decrease the chances of PPH, especially in the early postoperative period.
Disclosure of competing interest
The authors have none to declare.
References
- 1.Are C., Dhir M., Ravipati L. History of pancreaticoduodenectomy: early misconceptions, initial milestones and the pioneers. HPB (Oxford) 2011 Jun;13(6):377–384. doi: 10.1111/j.1477-2574.2011.00305.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gleeson E.M., Shaikh M.F., Shewokis P.A., et al. Whipple-ABACUS, a simple, validated risk score for 30-day mortality after pancreaticoduodenectomy developed using the. ACS-NSQIP database. Surgery. 2016;160(5):12791. doi: 10.1016/j.surg.2016.06.040. [DOI] [PubMed] [Google Scholar]
- 3.Giuliano K., Ejaz A., He J. Technical aspects of pancreaticoduodenectomy and their outcomes. Chin Clin Oncol. 2017 Sep 26;6(6):8. doi: 10.21037/cco.2017.09.01. [DOI] [PubMed] [Google Scholar]
- 4.Reyna-Sepúlveda F., Muñoz-Maldonado G., Pérez-Rodríguez E., Hernández-Trejo F., Guevara-Charles A., Hernández-Guedea M.A. Prognostic factors for survival and surgical complications in Whipple's pancreatoduodenectomy during a 10-year experience. CIRU. 2019 Feb 5;87(2):1233. doi: 10.24875/CIRU.18000526. [DOI] [PubMed] [Google Scholar]
- 5.Nahm C.B., Connor S.J., Samra J.S., Mittal A. Postoperative pancreatic fistula: a review of traditional and emerging concepts. Clin Exp Gastroenterol. 2018 Mar 15;11:105–118. doi: 10.2147/CEG.S120217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kawaida H., Kono H., Hosomura N., et al. Surgical techniques and postoperative management to prevent postoperative pancreatic fistula after pancreatic surgery. World J Gastroenterol. 2019 Jul 28;25(28):3722–3737. doi: 10.3748/wjg.v25.i28.3722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Waugh J.M., Clagett O.T. Resection of the duodenum and head of the pancreas for carcinoma. Surgery. 1946;20:224–232. [PubMed] [Google Scholar]
- 8.Perivoliotis K., Sioka E., Tatsioni A., Stefanidis I., Zintzaras E., Zacharoulis D. Pancreatogastrostomy versus pancreatojejunostomy: an up-to-date meta-analysis of RCTs. Int J Surg Oncol. 2017;2017:7526494. doi: 10.1155/2017/7526494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Menahem B., Guittet L., Mulliri A., Alves A., Lubrano J. Pancreaticogastrostomy is superior to pancreaticojejunostomy for prevention of pancreatic fistula after pancreaticoduodenectomy: an updated meta-analysis of randomized controlled trials. Ann Surg. 2015 May;261(5):882–887. doi: 10.1097/SLA.0000000000000806. [DOI] [PubMed] [Google Scholar]
- 10.Zhou Y., Yu J., Wu L., Li B. Meta-analysis of pancreaticogastrostomy versus pancreaticojejunostomy on occurrences of postoperative pancreatic fistula after pancreaticoduodenectomy. Asian J Surg. 2015 Jul 1;38(3):155–160. doi: 10.1016/j.asjsur.2015.02.002. [DOI] [PubMed] [Google Scholar]
- 11.Guerrini G.P., Soliani P., D'Amico G., et al. Pancreaticojejunostomy versus pancreaticogastrostomy after pancreaticoduodenectomy: an up-to-date meta-analysis. J Invest Surg. 2016 Jun;29(3):175–184. doi: 10.3109/08941939.2015.1093047. [DOI] [PubMed] [Google Scholar]
- 12.Krishna A., Bansal V.K., Kumar S., Garg P.K., Kapoor B.M., Misra M.C. Pancreaticogastrostomy after Whipple's surgery avoids pancreatic fistula—a large case series analysis. Indian J Surg. 2019;82(3):415–420. [Google Scholar]
- 13.Jin Y., Feng Y.-Y., Qi X.-G., et al. Pancreatogastrostomy vs pancreatojejunostomy after pancreaticoduodenectomy: an updated meta-analysis of RCTs and our experience. World J Gastrointest Surg. 2019 Jul 27;11(7):322–332. doi: 10.4240/wjgs.v11.i7.322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hallet J., Zih F.S., Deobald R.G., et al. The impact of pancreaticojejunostomy versus pancreaticogastrostomy reconstruction on pancreatic fistula after pancreaticoduodenectomy: meta-analysis of randomized controlled trials. HPB (Oxford) 2015 Feb;17(2):113–122. doi: 10.1111/hpb.12299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Keck T., Wellner U.F., Bahra M., et al. Pancreatogastrostomy versus pancreatojejunostomy for RECOnstruction after PANCreatoduodenectomy (RECOPANC, DRKS 00000767): perioperative and long-term results of a multicenter randomized controlled trial. Ann Surg. 2016 Mar;263(3):440–449. doi: 10.1097/SLA.0000000000001240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dudeja V., Christein J.D., Jensen E.H., Vickers M.S. In: Sabiston Textbook of Surgery. 20th ed. Townsend J.C.M., Beauchamp R.D., editors. Elsevier; Philadelphia: 2016. Exocrine pancreas; pp. 1520–1555. [Google Scholar]
- 17.Bartoli F.G., Arnone G.B., Ravera G., Bachi V. Pancreatic fistula and relative mortality in malignant disease after pancreaticoduodenectomy. Review and statistical meta-analysis regarding 15 years of literature. Anticancer Res. 1991;11:1831–1848. [PubMed] [Google Scholar]
- 18.Bassi C., Marchegiani G., Dervenis C., et al. The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 years after. Surgery. 2017;161(3):584–591. doi: 10.1016/j.surg.2016.11.014. [DOI] [PubMed] [Google Scholar]
- 19.Wente M.N., Bassi C., Dervenis C., et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS) Surgery. 2007 Nov;142(5):761–768. doi: 10.1016/j.surg.2007.05.005. [DOI] [PubMed] [Google Scholar]
- 20.Wente M.N., Veit J.A., Bassi C., et al. Postpancreatectomy hemorrhage (PPH): an international study group of pancreatic surgery (ISGPS) definition. Surgery. 2007 Jul;142(1):20–25. doi: 10.1016/j.surg.2007.02.001. [DOI] [PubMed] [Google Scholar]
- 21.Dindo D., Demartines N., Clavien P.A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–213. doi: 10.1097/01.sla.0000133083.54934.ae. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Parmar A.D., Sheffield K.M., Vargas G.M., et al. Factors associated with delayed gastric emptying after pancreaticoduodenectomy. HPB (Oxford) 2013 Oct;15(10):763–772. doi: 10.1111/hpb.12129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hayama S., Senmaru N., Hirano S. Delayed gastric emptying after pancreatoduodenectomy: comparison between invaginated pancreatogastrostomy and pancreatojejunostomy. BMC Surg. 2020 Apr 3;20(1):60–63. doi: 10.1186/s12893-020-00707-w. 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Clerveus M., Morandeira-Rivas A., Picazo-Yeste J., Moreno-Sanz C. Pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy: a systematic review and meta-analysis of randomized controlled trials. J Gastrointest Surg. 2014 Sep;18(9):1693–1704. doi: 10.1007/s11605-014-2557-6. official journal of the Society for Surgery of the Alimentary Tract. [DOI] [PubMed] [Google Scholar]
- 25.Lyu Y., Li T., Cheng Y., Wang B., Chen L., Zhao S. Pancreaticojejunostomy versus pancreaticogastrostomy after pancreaticoduodenectomy: an up-to-date meta-analysis of RCTs applying the ISGPS (2016) criteria. Surg Laparosc Endosc Percutaneous Tech. 2018 Jun;28(3):139–146. doi: 10.1097/SLE.0000000000000530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Qiu H., Shan R.-F., Ai J.-H., Ye S.-P., Shi J. Risk factors for 30-day unplanned reoperation after pancreatoduodenectomy: a single-center experience. J Canc Res Therapeut. 2019;15(7):1530–1534. doi: 10.4103/jcrt.JCRT_137_19. [DOI] [PubMed] [Google Scholar]
- 27.Narayanan S., Martin A.N., Turrentine F.E., Bauer T.W., Adams R.B., Zaydfudim V.M. Mortality after pancreaticoduodenectomy: assessing early and late causes of patient death. J Surg Res. 2018;231:304–308. doi: 10.1016/j.jss.2018.05.075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Schmidt C.M., Turrini O., Parikh P., et al. Effect of hospital volume, surgeon experience, and surgeon volume on patient outcomes after pancreaticoduodenectomy: a single-institution experience. Arch Surg. 2010 Jul;145(7):634–640. doi: 10.1001/archsurg.2010.118. [DOI] [PubMed] [Google Scholar]