External versus internal pancreatic duct drainage for early efficacy after laparoscopic pancreaticoduodenectomy in the early stages of the low-flow center learning curve: a retrospective comparative study

Introduction

Laparoscopic pancreaticoduodenectomy (LPD) is one of the most complex procedures in general surgery, and its postoperative complication rate can be as high as 20% to 50% (1,2). And the incidence of pancreatic fistulae, as the most important postoperative complication, can reach 10% to 30% (3,4). Despite these, LPD is becoming more common and sophisticated to be performed in high-flow centers. However, it is still difficult to perform LPD in low-flow centers due to the characteristics of extensive resection and difficult laparoscopic reconstruction. It has been demonstrated that the application of intraoperative pancreatic duct drainage in pancreaticoduodenectomy (PD) can reduce the incidence of postoperative pancreatic fistulae (5). However, there is no uniform conclusion as to whether internal or external drainage of the supporting duct is more clinically effective, especially in LPD which has not yet been reported in clinical studies.

To this end, we retrospectively analyzed the clinical outcomes of 73 patients who underwent LPD treatment in Jingzhou Hospital Affiliated to Yangtze University from January 2022 to June 2024, comparing the effects of external pancreatic duct drainage and internal pancreatic duct drainage on pancreatic fistulae after LPD and early clinical outcomes. Further, the safety and feasibility of the two drainage modalities are explored to clarify which drainage modality is more helpful in reducing the incidence of postoperative pancreatic fistulae and more conducive to the operator’s safe passage through the early stages of the LPD learning curve. We present this article in accordance with the STROBE reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-24-316/rc).

Methods

Patients and design

Data from patients who attended Jingzhou Hospital Affiliated to Yangtze University from January 2022 to June 2024 were retrospectively analyzed. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013), and the study protocol was approved by the Institutional Review Board of Jingzhou Hospital Affiliated to Yangtze University (No. 2024-080-01). All patients signed an informed consent form.

Between January 2022 and June 2024, 73 consecutive patients, 41 males and 32 females, were treated with LPD at Jingzhou Hospital Affiliated to Yangtze University. The age of the patients ranged from 40 to 82 years old. Forty-eight cases completed by doctors from Jingzhou Hospital Affiliated to Yangtze University were set as not having passed the early stages of the learning curve, and 25 cases completed by the invited team with rich LPD experience were set as having passed the early stages of the learning curve. According to whether they had passed the early stages of the learning curve and the method of pancreatic duct drainage used, the 73 patients were categorized into the external drainage group (EDG) that did not pass the early stages of the learning curve (n=24), the internal drainage group (IDG) that did not pass the early stages of the learning curve (n=24), and the IDG that passed the early stages of the learning curve (n=25).

Surgery procedure

All patients participated in routine preoperative consultations and preoperative design of surgical procedures. About 3–5 cm from the blind end of the jejunum, a hole was made in the opposite mesenteric margin of the jejunum corresponding to the main pancreatic duct. Separation forceps were probed into the intestinal loop more than 5 cm through this hole, and a hole was made in the contralateral intestinal wall using the embedded tip as a guide, and the external pancreatic drainage tube was directed through this hole down the intestinal collaterals to the pancreatic cross-section. The external pancreatic duct drain was inserted into the main pancreatic duct for 3 to 5 cm, and the pancreatic fluid drain was fixed with a 4-0 absorbable suture. Where the external pancreatic drainage tube leads out of the jejunal wall, the plasma muscular layer of the jejunum was closed with purse-string sutures and the drainage tube was fixed, and at the end of the procedure, the drainage tube was threaded through the greater omental tissue and covered, and drained out of the body by the abdominal wall and fixed, which is called external drainage (6). A suitable drain is placed in the opening of the pancreatic duct at the severed end of the pancreas and fixed to the pancreatic tissue, 3–5 cm in the pancreatic duct and about 10 cm in the jejunum, which is called internal drainage. After performing pancreaticojejunostomy, the hepaticojejunostomy anastomotic stoma is selected 10 cm away from the pancreaticojejunal anastomotic stoma. The bile duct jejunal anastomosis is performed using an end-to-side technique. The clamped stump of the bile duct is excised. If the diameter of the bile duct is greater than 8 mm, a continuous suture using 4-0 barbed threads is used to sew the anterior and posterior walls of the bile duct jejunal anastomosis. If the diameter of the bile duct is less than 8 mm, a continuous suture using 5-0 polydioxanone sutures is used to sew the anterior and posterior walls of the bile duct jejunal anastomosis, and a T-tube is placed for support and drainage. After completing the bile intestinal anastomosis, on the right side of the anastomosis, a one-stitch figure-eight suture is performed using 4-0 absorbable sutures to secure the bridge flap jejunal seromuscular layer with the reserved connective tissue of the gallbladder bed, ensuring that the bile intestinal anastomosis is tension-free in the normal position. Finally, a gastrojejunostomy is performed approximately 60 cm below the hepaticojejunostomy anastomotic stoma, with a diameter of approximately 5 cm.

Perioperative management

Total parenteral nutritional support, rehydration, broad-spectrum antibiotics, proton pump inhibitor (PPI), and growth inhibitors were given within 72 hours after surgery, and water, electrolyte, and acid-base balance were maintained. Postoperative blood routine, blood biochemistry and ascites and serum amylase were regularly reviewed, and the nature and total amount of pancreatic fluid in the drainage bag of the EDG were recorded. Enteral nutrition therapy was performed after 72 hours depending on the recovery of bowel sounds and gas expulsion. On the 7th postoperative day, all patients underwent abdominal computed tomography (CT) scanning to check for the presence of ascites in the abdominal cavity. If the patients did not have complications such as pancreatic fistulae or abdominal bleeding, the abdominal drainage tubes were gradually removed at 1 week postoperatively. Patients in the EDG were discharged with a pancreatic duct stent and returned to the hospital for review and removal of the drain at 1 month postoperatively. If pancreatic fistulae combined with hemorrhage occurred, angiography and other targeted interventions were performed according to the situation, and emergency reoperation was performed if necessary.

Data collection

Clinical data on patients’ basic conditions, concomitant diseases, disease characteristics, intraoperative conditions, and postoperative complications were retrospectively collected. The incidence of postoperative pancreatic fistulae was the main observation index. According to the International Study Group of Pancreatic Surgery (ISGPS) 2016 version of the postoperative pancreatic fistulae definition and grading system (7), pancreatic fistulae was categorized into three different grades of A, B, and C. Other observation records included the incidence of postoperative related complications, intraoperative bleeding, operative time, postoperative hospitalization time, perioperative mortality, and reoperation rate.

Statistical analysis

The experimental data were statistically analyzed using SPSS 25.0 software. Discrete data were expressed as frequencies (percentages) and differences between groups were compared using the χ² test. If the discrete data set did not meet the assumptions of the χ² test, Fisher’s exact test was used. Comparison of continuous data between two independent samples was performed using the Wilcoxon rank sum test. Differences were considered statistically significant when P<0.05.

Results

The general clinical characteristics and preoperative baseline data of all patients are listed in Table 1. There were no statistically significant differences between the EDG and the two IDGs in terms of sex, age, body mass index, total bilirubin, American Society of Anesthesiologists Physical Status Classification (ASA PS Classification), cardiopulmonary disease, diabetes mellitus, history of abdominal surgery, and biliary drainage.

The patients’ intraoperative performance and histopathological diagnosis are shown in Table 2. There was no statistical difference between the EDG and the two IDGs in terms of lesion site, pathological diagnosis, pancreas texture, diameter of pancreatic duct, and resection of portal or superior mesenteric vein. The IDG that passed the early part of the learning curve had a shorter total duration of surgery (300 vs. 570 min, P<0.001), fewer intraoperative bleeding (200 vs. 275 mL, P=0.03), and lower proportion of intraoperative blood transfusion (16.0% vs. 41.7%, P=0.04) compared to EDG that did not pass the early part of the learning curve.

Postoperative complications during the perioperative period are a major indicator of surgical safety. Except for the incidence of biliary fistulae in the EDG that did not pass through the early part of the learning curve (0.0%), which was lower than that in the IDG that did not pass through the early part of the learning curve (25.0%) (P=0.02), there were no statistically significant differences in the complications such as Clavien-Dindo classification ≥ III, abdominal hemorrhage (intra-abdominal hemorrhage, gastrointestinal hemorrhage), delayed gastric emptying (DGE), abdominal infection, lung infection, urinary tract infection, cardiac complication, wound infection, other serious medical complications, and length of stay (days). Differences in complications were not statistically significant [see reoperation and mortality (all cases within 90 days postoperatively) for details in Table 3].

Postoperative pancreatic fistulae is one of the most common and dangerous complications, the total pancreatic fistulae incidence rate of the whole group was 16.4% (12/73), and the incidence of pancreatic fistulae was significantly lower in the EDG that did not pass the early stages of the learning curve (8.3%) than in the IDG that did not pass the early stages of the learning curve (33.3%), with statistically significant difference (P=0.03); the incidence of pancreatic fistulae was slightly higher in the EDG that did not pass the early stages of the learning curve (8.3%) than in the IDG that passed the early stages of the learning curve (8.0%), with statistically not significant difference (P>0.99) (see Table 3).

Discussion

Like open pancreaticoduodenectomy (OPD), pancreatic fistulae is one of the most common complications after LPD. Nowadays, pancreatic fistulae are generally categorized into three grades; grade A pancreatic fistulae that do not require any special treatment, grade B pancreatic fistulae that can be effectively intervened by intervention and puncture, and grade C pancreatic fistulae that are mostly accompanied by abdominal bleeding that requires surgical intervention (8). Therefore, to safely perform LPD surgery within the learning curve, the focus is on effective intraoperative management to prevent grade B and C pancreatic fistulae.

The surgical debate on how to reduce pancreatic fistulae has been categorized into two main areas: first, effective pancreaticoenteric anastomosis technique, i.e., building a strong “dam” (9). Although the surgeons that did not pass the beginning of the learning curve may have prior experience in laparoscopic liver resection and laparoscopic bile duct surgery, their laparoscopic suturing skills are insufficient to ensure the quality of the pancreaticoenteric anastomosis (10). Moreover, successful laparoscopic surgery depends on proficient teamwork, which is particularly crucial during pancreaticoenteric anastomosis (11). The surgical teams that did not pass the beginning of the learning curve in the early stages of LPD are unable to coordinate effectively, leading to lower suture quality at the anastomosis site and an increased risk of postoperative pancreatic fistulae. So, in order to master the pancreaticoenteric anastomosis technique, it is essential to get through the learning curve. In the literature, the number of surgical cases that need to be performed for an operator to pass the learning curve fluctuates between 20 and 100 cases (12-16). Speicher et al. demonstrated that when the LPD is >50 cases, the completion time of the surgery will be significantly shorter and intraoperative bleeding will be drastically reduced (12). Prof. Hong (14) concluded that 50–60 LPDs are required to overcome the learning curve, while Kim et al. (15) analyzed the surgical history of surgeons and found that 84 LPDs is a “watershed” for the reduction of complications. In summary, a low-flow PD center (<20 cases/year) may need 100 cases to complete the LPD learning curve period; however, in a high-flow PD center only 20 cases may be needed (8). Thus, surgeons in low-flow centers need more procedures to get through the LPD learning curve and a longer period of time to learn to build the “dam”. Furthermore, the initial safety of LPD in low-flow centers is much lower than that in high-volume centers. However, LPD in low-flow centers should not be performed at the expense of high complication and mortality rates, so we can only look for more effective intraoperative management to make up for the lack of skills and experience, and thus reduce the complication rate.

At this time, the reasonable placement of catheter drainage, that is, the smooth flow of the “channel” has become a breakthrough. The fundamental reason why fistulae is more likely to occur in pancreaticoenteric anastomosis than in gastrointestinal and enteroenteric anastomosis is that the anastomosis healing in pancreaticoenteric anastomosis mainly relies on the slow adhesive healing between the pancreatic segment and the jejunal plasma muscle layer (8). And this adhesive healing takes longer than the growth healing of the pancreatic duct and jejunal mucosa. Inserting a pancreatic fluid drain into the main pancreatic duct to drain the pancreatic fluid secreted by the pancreas across the pancreaticoenteric anastomosis allows sufficient time for the growth healing of the pancreatic duct and jejunal mucosa to wait for the pancreatic parenchyma and the jejunal plasma musculature to complete the adhesive healing first (17), which then reduces the occurrence of pancreatic-enteric anastomotic fistulae at the initial stage of the learning curve from the root cause. There have been a number of studies that during the slow adhesive healing between the pancreatic segment and the jejunal plasma muscle layer, patients who undergo internal drainage will have their early postoperative pancreatic fluid directly discharged into the intestine, where they come into contact with intestinal fluids, bile, and digestive enzymes (18-20). This premature activation of digestive enzymes and bile increases the corrosiveness of the digestive fluids and also increases the chance of contact with the anastomosis site, leading to a higher incidence of pancreatic fistulae. Instead, external pancreatic duct drainage theoretically allows all pancreatic fluid to flow out of the body, avoiding the corrosive effects of pancreatic fluid on the pancreatic-intestinal anastomosis and reducing the occurrence of pancreatic fistulae. And the results of this study also suggest that the incidence of pancreatic fistulae in the EDG that did not pass the early stages of the learning curve was lower than that in the IDG that did not pass the early stages of the learning curve, with statistically significant difference. Thus, placing an external pancreatic fluid drain in the pancreatic duct provides a way for physicians in low-flow centers to safely navigate the learning curve.

It has been continuously shown that leaving a drain in the pancreatic duct can effectively reduce the incidence of pancreatic fistulae (9,21,22). And in recent years, the use of pancreatic duct drains in OPD has been promoted by more and more pancreatic surgeons. However, no consensus has been reached on whether to choose internal or external drainage for pancreatic duct placement and drainage in OPD (23). Especially for the specific period at the beginning of the learning curve period and the specific setting of low-flow centers, few studies have reported on whether to choose internal or external drainage of pancreatic duct support tubes in LPD surgery.

The team has been independently performing LPD since 2022 and is now at the beginning of the LPD learning curve, making it a reference implementation team for LPD in low-traffic centers with a total of 24 cases of LPD in the EDG that did not pass the early stages of the learning curve, 24 cases of LPD in the IDG that did not pass the early stages of the learning curve, and 25 cases of LPD in the IDG that passed the early stages of the learning curve. Both of the average number of cases performed by surgeons in the EDG and IDG who had not yet passed the early learning curve are 24 cases. All 24 operations in each group were performed by the same group of surgeons. The overall pancreatic fistulae incidence in the three groups was 16.4%, which is consistent with literature reports such as Hackert et al. (24) and Wang et al. (4). In the results of this study, there was no statistically significant reduction in the incidence of pancreatic fistulae in the EDG that did not pass the early stages of the learning curve compared to the IDG that passed the early stages of the learning curve (P>0.99). However, the surgeons in the IDG that passed the early stages of the learning curve had a had a shorter total duration of surgery, fewer intraoperative bleeding, and lower proportion of intraoperative blood transfusion compared to the surgeons in the EDG that did not pass the early stages of the learning curve. This result indicates that the surgeons in the IDG that passed the early stages of the learning curve had a proven pancreatic-intestinal anastomosis technique in performing LPD, whereas the incidence of pancreatic fistula was extremely similar between the two groups. This implies that operators in low-flow centers can prevent pancreatic fistulae by using external pancreatic duct drainage at the beginning of the learning curve, making the safety of the procedure close to that of high-flow centers that have passed the beginning of the learning curve. In the results of this study compared with internal pancreatic duct drainage, the use of external pancreatic duct drainage by surgeons who had not yet passed the early learning curve in low-flow centers was more effective in reducing the incidence of pancreatic fistulae (P=0.03). Moreover, two cases of pancreatic fistulae in the EDG that did not pass the early stages of the learning curve were cured by symptomatic treatment, and no patients died as a direct result of pancreatic fistulae. As for the 8 cases of pancreatic fistulae in the IDG that did not pass the early stages of the learning curve, there were 2 cases of anastomotic hemorrhage and reoperation, of which 1 case was successfully cured by external pancreatic drainage during reoperation, while the other one died of systemic multiorgan failure caused by pancreatic fistulae.

Conclusions

In summary, our study demonstrated that external pancreatic duct drainage at the beginning of the learning curve can effectively reduce the incidence of pancreatic fistulae after LPD. This surgical method is safe, effective, easy to operate, and has great clinical application potential. It has certain clinical promotion value in low volume centers for LPD. Of course, due to the relatively limited case data of LPD carried out in Jingzhou Hospital Affiliated to Yangtze University so far, further confirmation by subsequent large-sample data or prospective randomized controlled trial studies is still needed.

Acknowledgments

We would like to thank all our research team members for their hard work.

Funding: This work was supported by a grant from the Natural Science Foundation of Hubei Province (No. 2022CFB346).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-24-316/rc

Data Sharing Statement: Available at https://gs.amegroups.com/article/view/10.21037/gs-24-316/dss

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-24-316/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-24-316/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013), and the study protocol was approved by the Institutional Review Board of Jingzhou Hospital Affiliated to Yangtze University (No. 2024-080-01). All patients signed an informed consent form.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Guo C, Xie B, Guo D. Does pancreatic duct stent placement lead to decreased postoperative pancreatic fistula rates after pancreaticoduodenectomy? A meta-analysis. Int J Surg 2022;103:106707. [Crossref] [PubMed]
2. Sofue K, Ueshima E, Masuda A, et al. Estimation of pancreatic fibrosis and prediction of postoperative pancreatic fistula using extracellular volume fraction in multiphasic contrast-enhanced CT. Eur Radiol 2022;32:1770-80. [Crossref] [PubMed]
3. Perri G, Marchegiani G, Partelli S, et al. Preoperative risk stratification of postoperative pancreatic fistula: A risk-tree predictive model for pancreatoduodenectomy. Surgery 2021;170:1596-601. [Crossref] [PubMed]
4. Wang GQ, Yadav DK, Jiang W, et al. Risk Factors for Clinically Relevant Postoperative Pancreatic Fistula (CR-POPF) after Distal Pancreatectomy: A Single Center Retrospective Study. Can J Gastroenterol Hepatol 2021;2021:8874504. [Crossref] [PubMed]
5. Zhang M, Meng L, Zhang Z, et al. Meta-analysis of the efficacy of pancreatico-enteric anastomotic internal drainage compared with external drainage after pancreaticoduodenectomy. Journal of Hepatobiliary and Pancreatic Surgery 2019;31:746-52+760.
6. Liang Y, Gong S, Li S, et al. Clinical application of external pancreatic duct drainage in laparoscopic pancreaticoduodenectomy. Journal of Clinical Surgery 2024;32:396-9.
7. 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;161:584-91. [Crossref] [PubMed]
8. Hong D. Hot issues of laparoscopic pancreaticoduodenectomy reconsidered from the Dutch study. Journal of Hepatobiliary Surgery 2021;29:309-11+317.
9. Feng Y, Pan C, Zhou J, et al. Analysis of the efficacy of internal and external drainage of the pancreatic duct during pancreaticoduodenectomy. Chinese Journal of Pancreatic Disease 2020;20:142-4.
10. Yang Z. Some experiences of safely and quickly passing laparoscopic pancreaticoduodenectomy. Abdominal Surgery 2021;34:347-50.
11. Liang M, Li X, Liang Z, et al. Experience of laparoscopic pancreaticoduodenectomy during the learning curve period. Journal of Hepatobiliary and Pancreatic Surgery 2022;34:93-7+103.
12. Speicher PJ, Nussbaum DP, White RR, et al. Defining the learning curve for team-based laparoscopic pancreaticoduodenectomy. Ann Surg Oncol 2014;21:4014-9. [Crossref] [PubMed]
13. Wang M, Peng B, Liu J, et al. Practice Patterns and Perioperative Outcomes of Laparoscopic Pancreaticoduodenectomy in China: A Retrospective Multicenter Analysis of 1029 Patients. Ann Surg 2021;273:145-53. [Crossref] [PubMed]
14. Hong D. Experiences and technological innovations in routinely performing laparoscopic pancreaticoduodenectomy. Journal of Hepatobiliary and Pancreatic Surgery 2017;29:89-92.
15. Kim S, Yoon YS, Han HS, et al. Evaluation of a single surgeon's learning curve of laparoscopic pancreaticoduodenectomy: risk-adjusted cumulative summation analysis. Surg Endosc 2021;35:2870-8. [Crossref] [PubMed]
16. Nagakawa Y, Nakamura Y, Honda G, et al. Learning curve and surgical factors influencing the surgical outcomes during the initial experience with laparoscopic pancreaticoduodenectomy. J Hepatobiliary Pancreat Sci 2018;25:498-507. [Crossref] [PubMed]
17. Hong D. My opinion on the choice of open, laparoscopic, and da Vinci robotic pancreaticoduodenectomy. Journal of Hepatobiliary Surgery 2019;27:314-6.
18. Tu J, Huang C, Xu W, et al. Application of split pancreatic duct stent in laparoscopic pancreaticoduodenectomy. Medicine (Baltimore) 2023;102:e34049. [Crossref] [PubMed]
19. Wang Q, He XR, Tian JH, et al. Pancreatic duct stents at pancreaticoduodenectomy: a meta-analysis. Dig Surg 2013;30:415-24. [Crossref] [PubMed]
20. Wang G, Li L, Ma Y, et al. External Versus Internal Pancreatic Duct Drainage for the Early Efficacy After Pancreaticoduodenectomy: A Retrospectively Comparative Study. J Invest Surg 2016;29:226-33. [Crossref] [PubMed]
21. Iguchi T, Motomura T, Uchiyama H, et al. Impact of a 7.5-Fr Pancreatic Stent for Preventing Pancreatic Fistula after Pancreaticoduodenectomy. Dig Surg 2021;38:361-7. [Crossref] [PubMed]
22. Jiang Y, Chen Q, Wang Z, et al. The Prognostic Value of External vs Internal Pancreatic Duct Stents in CR-POPF after Pancreaticoduodenectomy: A Systematic Review and Meta-analysis. J Invest Surg 2021;34:738-46. [Crossref] [PubMed]
23. Wu Q, Han R, Chen L, et al. Prospective comparative study of the effect of internal and external pancreatic duct drainage on pancreatic fistula after pancreaticoduodenectomy. Chinese Journal of Practical Surgery 2023;43:1272-8.
24. Hackert T, Klaiber U, Pausch T, et al. Fifty Years of Surgery for Pancreatic Cancer. Pancreas 2020;49:1005-13. [Crossref] [PubMed]