Continuous or Interrupted Suture for Hepaticojejunostomy in Pancreaticoduodenectomy (The HEKTIK Trial)

Findings of a randomized, controlled, single-center superiority trial

Background: Over 6000 pancreaticoduodenectomies are performed each year in Germany, and hepaticojejunostomy is a crucial step of the procedure. An anastomotic leak of hepaticojejunostomy can cause major postoperative complications. The aim of this trial was to compare the morbidity and efficiency of continuous versus interrupted suturing for hepaticojejunostomy in pancreaticoduodenectomy.

Methods: In a randomized, controlled, single-center trial (German Clinical Trials Register No. DRKS00024395), patients scheduled for elective open partial pancreaticoduodenectomy with hepaticojejunostomy between January 2020 and May 2023 were randomly assigned in a 1:1 ratio to suturing of the hepaticojejunostomy with either a continuous or an interrupted technique. The primary endpoint was anastomotic leakage from the hepaticojejunostomy in the first three days after surgery. Further perioperative parameters were secondary endpoints, including later leakage, other complications, the duration of hepaticojejunostomy, and the cost of hepaticojejunostomy.

Results: The 100 patients in the trial consisted of 50 in each group. The rate of early anastomotic leakage was 2% in both groups (95% confidence interval for the difference, [-5.5%; 5.5%]; p = 1.000). As for the secondary endpoints, there were no relevant intergroup differences in any other short-term or long-term morbidity parameters. Continuous suturing of the hepaticojejunostomy was, however, 31% faster and 68% cheaper in material costs.

Conclusion: These data imply that continuous and interrupted suturing techniques yield equally good clinical outcomes in hepaticojejunostomies of hepatic ducts with diameter 5 mm or more. Continuous suturing is, however, both faster and cheaper.

Hepaticojejunostomy is one of the three anastomoses required in pylorus-preserving pancreaticoduodenectomy (PPPD) and one of the four anastomoses in the classical Whipple procedure. Although the risk of complications is much lower for hepaticojejunostomy than for pancreatic anastomosis, any complications that occur can have serious consequences for the patients affected, including prolonged hospital stays and the need for interventions such as drain insertion or even relaparotomy. These procedures are associated with considerable morbidity and mortality, even at high-volume centers (1, 2, 3, 4).

The most important complications of hepaticojejunostomies include leakage, which is reported in the literature to occur in 1.2% to 12.5% of patients (1, 2, 5, 6, 7), and stenosis, found in 0.0% to 13.0% of cases (5, 6, 8, 9]. Both of these morbidities are potentially influenced by the suture technique used. The two most commonly used techniques are interrupted suturing and continuous suturing. A recent nationwide survey in Germany showed that 56% of the hospitals involved employ both techniques, depending on the intraoperative circumstances, while 40% consistently use one of the two techniques (10). This diversity illustrates the ongoing debate between the proponents of each technique. Advocates of the interrupted suture technique (IST) argue for its universal use (even for small bile ducts) and envisage potentially lower stenosis rates, while supporters of the continuous suture technique (CST) point to better sealing of the anastomosis and time saving.

Due to a lack of randomized studies comparing the two surgical techniques, we initiated this randomized controlled superiority trial comparing the interrupted and the continuous suture technique for hepaticojejunostomy.

Methods

Study design

The HEKTIK trial was designed as a randomized, controlled, patient-blinded superiority trial. In one arm of the trial the hepaticojejunostomy during pancreaticoduodenectomy was performed using the IST, in the other arm, the CST (Figure 1). The study commenced in January 2020 with the enrollment of the first patients and concluded in May 2024 with the last 1-year follow-up visit. The detailed study protocol has been published and is available online (11). Additional information about the participants, surgical technique, blinding, randomization, quality, and safety is presented in the eMethods. The trial was conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice (GCP) and was approved by the ethical committee of Friedrich Alexander University Erlangen–Nuremberg (reference number 167_19 B, approval date 22 May 2019). The trial was registered in the German Clinical Trials Register (DRKS00024395) and was reported in accordance with the CONSORT statement.

Figure 1

Flow chart of the HEKTIK study

Endpoints

The primary endpoint of this trial was the prevalence of early anastomotic leakage of the hepaticojejunostomy, defined as occurrence within the first 3 postoperative days (72 hours after the end of surgery), according to the criteria of the International Study Group of Liver Surgery (ISGLS) from 2011 (13). These define hepaticojejunostomy leakage as a bilirubin concentration in the drainage fluid at least 3 times as high as the serum bilirubin concentration on or after postoperative day 3 or the need for radiological or surgical intervention due to an intra-abdominal bile collection or biliary peritonitis.

The secondary endpoints included total operative time, the postoperative course of cholestasis parameters (γ-GT, alkaline phosphatase, bilirubin), reoperation rate, reintervention rate, morbidity (during hospital stay and at 3 months after operation, assessed by means of the Clavien–Dindo classification [14]), mortality (during hospital stay and at 3 months after operation), duration of hepaticojejunostomy, material costs of hepaticojejunostomy, and stenosis rate at 3 and 12 months after operation. Additionally, the overall anastomotic leakage rate (including leakage occurring after the first 3 postoperative days) was assessed in this analysis as an ad-hoc endpoint.

Statistics

Due to a lack of data regarding reliable sample size calculation at the time of study preparation, an adaptive sample size design was chosen. The intention was initial inclusion of a total of 100 patients (50 patients per arm) with the option of adjusting the sample size on the basis of these first 100 patients if necessary. A non-blinded interim analysis was planned after the inclusion of the first 100 patients to perform a final sample size calculation. The criteria for continuation of the trial were:

• An absolute difference of at least 1% in the primary endpoint, to justify the clinical relevance of the trial
• A calculated sample size per group of no more than 1000 patients, to ensure feasibility of the trial

Statistical analysis was carried out using SPSS (version 28.0). An intention-to-treat analysis was performed. Metric and ordinal data were analyzed using Student’s t-test or the Mann–Whitney U test, categorical data using the chi-square test. To determine mortality at 3 and 12 months after operation, a survival analysis was conducted using the Kaplan–Meier method and the log-rank test. The statistical significance for the primary endpoint was set at p < 0.05. Although not initially foreseen in the study protocol, and although the trial was designed as a superiority study, a post-hoc calculation of the 95% confidence interval for the primary endpoint was conducted, as conventionally done in a non-inferiority study.

End of trial

After conducting the planned interim analysis, the study was terminated as it failed to meet the continuation criteria: the effect size was zero, rendering the required sample size infinite and thus unattainable.

Results

Patient characteristics

A total of 100 patients were enrolled and randomized during the study period. Fifty participants received hepaticojejunostomy using the IST and 50 underwent hepaticojejunostomy with the CST. The patients’ characteristics are shown in Table 1. Except for age, which was higher in the IST group (70 versus 64 years, p = 0.017), the patients’ data were comparable between the two groups (Table 1). The preoperative liver-specific blood test results, presented in Figure 2, did not differ between the two groups.

Figure 2

Postoperative course of selected liver and cholestasis parameters, stratified by suture technique used

Table 1

Patient characteristics of the study population

Operative characteristics

The most common indication for surgery was pancreatic adenocarcinoma (49%), followed by other malignant pancreatic head tumors (29%), chronic pancreatitis (12%), and cystic pancreatic neoplasms (10%). Seventy-two percent of the patients received PPPD, 28% a classical Whipple procedure. Venous vascular resection was required in 15% of patients overall. The mean hepatic duct diameter as measured intraoperatively was 12 mm, and in 62% of cases the hepatic duct wall was thick. Operative details are summarized in eTable 1 and showed no relevant differences between the two groups.

Primary endpoint

Early anastomotic leakage of the hepaticojejunostomy occurred in one patient from each group, resulting in a prevalence of 2% [–5.5%; 5.5%] in both groups with no significant difference between them (p = 1.000) (Table 2). The two leakages occurred during operations performed by two different surgeons. Both patients underwent relaparotomy for repeated hepaticojejunostomy in the suture technique not performed in the first instance (freely chosen by the revising surgeon, who in each case was the person who had carried out the initial operation). Both newly performed hepaticojejunostomies healed without further complications.

Table 2

Endpoint analysis

Secondary endpoints

Reoperation and reintervention were necessary in 4% and 15% of patients respectively, while in-hospital morbidity and mortality were 51% and 1% respectively. There were no relevant differences between the two groups regarding the postoperative complication parameters (Table 2). Postoperative measurement of liver-specific blood values (AST, ALT, AP, γ-GT, and bilirubin) showed no relevant difference during the postoperative course (postoperative days 1 and 3) (Figure 2). Also with regard to the surgical parameters, there was no difference in total operative time between the two groups (303 minutes [CST] versus 301 minutes [IST], p = 0.898); however, the duration of the hepaticojejunostomy was shorter (700 versus 1047 seconds) and the material costs were lower (100% versus 317%) in the CST group (each p < 0.001) (Table 2).

There was again no relevant difference of morbidity and overall survival between the two groups at the 3-month and 12-month follow-up visits. At neither time were there any signs of stenoses of the hepaticojejunostomies (Table 2).

Ad-hoc endpoint

No late anastomotic leakage occurred, resulting in a total prevalence of hepaticojejunostomy leakage of 2% in each group, equivalent to the early leakage rate (p = 1.000) (Table 2).

Discussion

The significant consequences of hepaticojejunostomy leakage, the potential impact of the suture technique used, the diversity of suture techniques employed in Germany, and the lack of clear evidence-based data in the literature underscore the need for a randomized controlled trial on this topic.

The results of our clinical randomized controlled trial show that there is no significant difference in the occurrence of early anastomotic leakage or other postoperative complications in the short or long term between the interrupted suture technique and the continuous suture technique in fashioning a hepaticojejunostomy during pancreaticoduodenectomy. However, the continuous suture technique showed a shorter suturing time (−31%) and lower material costs (−68%). These findings are consistent with the sparse existing data from three other studies that compared the interrupted and continuous suture techniques for creation of a hepaticojejunostomy (5, 6, 7): Seifert et al. conducted a randomized controlled trial in 82 patients in whom hepaticojejunostomy was carried out and also found no difference in hepaticojejunostomy-specific or general postoperative complications. However, their study also demonstrated a shorter suturing time for hepaticojejunostomy with continuous suture technique (−46%) (6). It should nevertheless be noted that their data also included hepaticojejunostomies performed in the context of liver operations and isolated biliodigestive anastomoses, which limits comparability. Tatsuguchi et al. examined 228 patients undergoing pancreaticoduodenectomy or total pancreatectomy. Their prospective data also revealed no differences in short-term or long-term outcomes, including leakage rate. However, superiority of the continuous suture technique in terms of speed of hepaticojejunostomy (−40%) and cost efficiency (−92%) was reported (5). Yadav et al. came to similar conclusions (comparable morbidity, shorter anastomosis time [−58%], and lower costs [−75%]) in their comparison of the suture techniques for hepaticojejunostomy on the basis of prospective data (7). However, their study included only 34 patients and the hepaticojejunostomies were carried out in a wide variety of different procedures, which limits the evidential relevance of the findings.

Further data comparing the suture techniques are limited to patient cohorts in the setting of liver transplantation, which is probably not directly comparable with pancreatic surgery (14).

In our cohort, the hepaticojejunostomy leakage and stenosis rates were 2% and 0%, respectively. These outcomes fall within the lower range of previous study results, which exhibit a considerable variance in rates of leakage and stenosis. The leakage rate fluctuates between 1.2% and 12.5% (1,2, 5, 6, 7), while the stenosis rate varies from 0.0% to 13.0% (5, 6, 8, 9). This variability is most likely explained by the inclusion of hepaticojejunostomies performed in different surgical contexts, as well as inconsistent definitions of the outcome parameters.

In contrast to this variability our study offers high homogeneity, enhancing the validity of our results. First, hepaticojejunostomy leakage was assessed using the clearly defined and standardized criteria of the International Study Group of Liver Surgery (13). Second, the comparison of suture techniques was conducted exclusively in the context of pancreaticoduodenectomy and at a single center, ensuring a consistent approach in terms of both the scope of the operation and the technique used. Additionally, all surgical procedures were performed by experienced visceral surgeons specialized in pancreatic surgery, indicating a high level of expertise and procedural consistency.

Another notable strength of this study is that it stands as the largest randomized controlled trial to date on this topic. Nevertheless, there are also some limitations. Foremost among them is limited statistical power, with a relatively small sample of 100 participants. This sample size was chosen because of a lack of data on case number planning at the time of initial study design, with the option of adjustment after the first inclusion phase. However, the low incidence of leakage in our study and the missing effect size between the two suture techniques would have led to infinite or very high case numbers, respectively, rendering feasibility of the study questionable. Second, the single-center study design could entail bias. Third, patients with a bile duct diameter of less than 5 mm were excluded, which restricts the generalizability of the data. However, the hepatic duct is enlarged in the majority of patients who require pancreatic head resection, and this was indeed the case in our study. Fourth, the outcome assessor and data analyst in our study were not blinded due to the robust and well-defined outcome parameters, and this may be a source of bias. Finally, the follow-up period may be too short to permit reliable conclusions regarding long-term outcomes, especially concerning the stenosis rate of hepaticojejunostomies. Tatsuguchi et al. reported a median time of 13.5 months for the development of anastomotic stenoses (5), while Reid-Lombardo et al. showed reliable stenosis rates only after follow-up for 5 or 10 years (8). However, collection of long-term data is challenging due to the frequently limited survival of patients with pancreatic cancer.

Conclusion

Our findings suggest that neither the interrupted nor the continuous suture technique for hepaticojejunostomy with a hepatic duct diameter ≥ 5 mm is superior regarding clinical outcomes. However, the continuous suture technique offers advantages in terms of suturing time and costs.

Ethical approval and consent to participate

The clinical trial is conducted in accordance with with the Declaration of Helsinki. The study protocol and all trial documents were approved by the local ethics committee of Friedrich Alexander University Erlangen–Nuremberg, Germany (reference number 167_19 B, approval date 22 May 2019). All randomized patients were informed in detail about the study and the possible risks and gave their written informed consent.

 

Trial registration
German Clinical Trials Register, DRKS00024395. Registered on 1 February 2021, https://drks.de/search/en/trial/DRKS00024395

 

Funding
This trial was financed by internal hospital funds of the Department of General and Visceral Surgery, University Hospital Erlangen and by the Friends of the Tumor Center of Erlangen–Nuremberg University (“Verein zur Förderung des Tumorzentrums der Universität Erlangen-Nürnberg e.V.”). There was no external funding of this trial outside University Hospital Erlangen.

 

Data sharing
Individual patient data (anonymized) can be requested by scientists submitting a methodologically sound analysis proposal. Requests should be sent to: [e-mail].

Conflict of interest statement
The authors declare that no conflict of interest exists.

Manuscript received on 25 May 2024, revised version accepted on 21 August 2024.

Corresponding author
PD Dr. med. Maximilian Brunner, M.Sc.
Uniklinikum Erlangen, Chirurgie
Krankenhausstr. 12
91054 Erlangen, Germany

Cite this as:
Brunner M, Golcher H, Krautz C, Kersting S, Weber GF, Grützmann R: Continuous or interrupted suture for hepaticojejunostomy in pancreaticoduodenectomy (the HEKTIK trial): Findings of a randomized, controlled, single-center superiority trial. Dtsch Arztebl Int 2024; 121: 696–702. DOI: 10.3238/arztebl.m2024.0178

References

1. Antolovic D, Koch M, Galindo L, et al.: Hepaticojejunostomy—analysis of risk factors for postoperative bile leaks and surgical complications. J Gastrointest Surg 2007; 11: 555–61

2. de Castro SM, Kuhlmann KF, Busch OR, et al.: Incidence and management of biliary leakage after hepaticojejunostomy. J Gastrointest Surg 2005; 9: 1163–73

3. Akamatsu N, Sugawara Y, Hashimoto D: Biliary reconstruction, its complications and management of biliary complications after adult liver transplantation: a systematic review of the incidence, risk factors and outcome. Transpl Int 2011; 24: 379–92

4. Chok KS, Ng KK, Poon RT, Lo CM, Fan ST: Impact of postoperative complications on long-term outcome of curative resection for hepatocellular carcinoma. Br J Surg 2009; 96: 81–7

5. Tatsuguchi T, Takahashi H, Akita H, et al.: Short- and long-term outcomes of choledochojejunostomy during pancreaticoduodenectomy and total pancreatectomy: interrupted suture versus continuous suture. Langenbecks Arch Surg 2018; 403: 959–66

6. Seifert L, von Renesse J, Seifert AM, et al.: Interrupted versus continuous suture technique for biliary-enteric anastomosis: randomized clinical trial. BJS Open 2023; 7: zrac163

7. Yadav TN, Pandit N, Deo KB, Awale L, Neupane D, Adhikary S: Continuous versus interrupted anastomotic technique for the hepaticojejunostomy: a prospective cohort study. Ann Med Surg (Lond) 2024; 86: 1950–5

8. Reid-Lombardo KM, Ramos-De la Medina A, Thomsen K, Harmsen WS, Farnell MB: Long-term anastomotic complications after pancreaticoduodenectomy for benign diseases. J Gastrointest Surg 2007; 11: 1704–11

9. House MG, Cameron JL, Schulick RD, et al.: Incidence and outcome of biliary strictures after pancreaticoduodenectomy. Ann Surg 2006; 243: 571–8

10. Brunner M, Stockheim J, Krautz C, et al.: Continuous or interrupted suture technique for hepaticojejunostomy? A national survey. BMC Surg 2018; 18: 84

11. Brunner M, Golcher H, Krautz C, Kersting S, Weber GF, Grützmann R.: Continuous or interrupted suture technique for hepaticojejunostomy uring pancreatoduodenectomy (HEKTIK trial): study protocol of a randomized controlled multicenter trial. Trials 2022; 23: 467

12. Probst P, Zaschke S, Heger P, et al.: Evidence-based recommendations for blinding in surgical trials. Langenbecks Arch Surg 2019; 404: 273–84

13. Koch M, Garden OJ, Padbury R, et al.: Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery 2011; 149: 680–8

14. Dindo D, Demartines N, Clavien PA: Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205–13

15. Kasahara M, Egawa H, Takada Y, et al.: Biliary reconstruction in right lobe living-donor liver transplantation: Comparison of different techniques in 321 recipients. Ann Surg 2006; 243: 559–66