Feeding Jejunostomy Following Pancreatoduodenectomy: A Double-edged Sword?

SUMMARY Pancreaticoduodenectomy (PD) still remains a complex procedure because it is technically difficult and is often followed by post-operative complications. Although the mortality has been reduced, in high-volume centres, to 1%–5%, morbidity levels remain at 30% to 60%. These include delayed gastric emptying (DGE), pancreatic anastomotic leaks and surgical site infection.[1] Early enteral feeding is thought to play a protective role in reducing post-operative complications and subsequent hospital stay.[2] However, because of DGE, feeding is not possible by mouth or via a Ryle’s tube. This is why most surgeons install a feeding jejunostomy (FJ) distal to the stomach because the small bowel regains its motility earlier and enteral feeding becomes possible. However, various studies have found that FJ is associated with increased morbidity in terms of DGE and catheter-related complications.[3-5] The PrEDICT-DGE scoring system, which was designed to predict the risk of DGE in patients undergoing PD, also included FJ as a risk factor for developing DGE.[6] However, the available studies in the literature are mostly retrospective. Therefore, the present authors planned this prospective observational study to analyse the early surgical outcomes among patients with FJ and without FJ following PD over a period of 2 years (August 2022–April 2024). The use of FJ was at the surgeon’s discretion. FJ was done by the Witzel technique using a 12-Fr feeding tube. DGE, post-operative pancreatic fistulae, hospital stay and FJ tube-related complications were then assessed as primary outcomes, whereas surgical site infections, haemorrhage, need for total parenteral nutrition and mortality were included in the secondary outcomes. Appropriate statistical tests were applied, and P < 0.05 was considered statistically significant. A total of 56 patients underwent PD during this time period, with 28 patients in each group. There were no significant differences between both the groups in relation to age, body mass index, diabetes mellitus, pancreatic duct size, texture of pancreas and pathological stage. Most of the post-operative parameters, such as mortality, haemorrhage, re-operation and parenteral nutrition use, were similar between the groups. The length of hospital stay was longer in the FJ group (13.6 ± 6.7 vs. 11.6 ± 6.3 days, P = 0.1), but without any statistical significance. However, the incidence of clinically significant DGE (grade B/C) was significantly higher in the FJ group (60.7% vs. 21.4%, P = 0.008). FJ tube-related morbidities were seen in three patients, with two of them requiring reoperations. Multivariate analysis showed patients without FJ (odds ratio [OR]: 9.1, 95% confidence interval [CI]: 0.2–0.4, P = 0.002) and without peri-operative blood transfusion (OR: 0.03, 95% CI: 0.03–0.3, P = 0.003) had a significantly lower incidence of DGE. The authors advocated selective rather than routine use of FJ following PD with the guidance of standardised criteria and patient-related factors. COMMENTARY Pancreatoduodenectomy is a complex procedure which is often associated with complications. DGE, which is one of the frequent and worrisome complication, although not life threatening, can lead to a delayed intake of solid food, electrolyte imbalances, prolonged hospital stay and a delayed start of adjuvant therapy. The incidence of DGE following PD may be as high as 60%.[1] It is thought to be caused by an imbalance or incoordination between the myogenic, hormonal and neuronal factors responsible for gastric motility. Its mechanism is poorly understood, with plausible explanations being reduced motilin receptors, damage to the vagus nerve, ischaemic damage to the stomach, gastroduodenal dysrhythmia due to anastomotic angulation or torsion and infection caused by anastomotic leakage.[7] The present study was planned in view of some evidence which showed an increased incidence of DGE after adding an FJ following a PD.[3-5] The hypothesis was that there was a disruption of coordinated peristalsis by the FJ.[4] The other problems of the FJ are the catheter-related complications such as torsion at the site of the FJ, leakage from the insertion site or obstruction. In this study, they found a two times higher rate of DGE in the FJ group in comparison to the non-FJ group. Similarly, the rate of clinically significant DGE was also higher in the FJ group. These findings were corroborated by another similar study by Waliye et al.[4] Despite an increased incidence of DGE, there was no significant difference in hospital stay between both the groups. About 10.7% of patients (three patients) suffered from FJ-related morbidity, out of which two patients were operated upon for FJ-related intestinal obstruction. The re-operation rate did not differ significantly between both the groups. PD is a complex procedure, and a successful outcome is affected by various factors, starting from simple surgical site infection to haemorrhage. An FJ allows early nutrition while the patient is still recovering from surgery. The inherent increased risk of DGE following PD usually leads to delayed tolerance to the oral diet, and it can be mitigated by an FJ. Similarly, severe forms of DGE from any cause usually take a long time to recover, and during this time period, there are two options to maintain enteral nutrition – either via a nasojejunal or nasogastric tube or via an FJ. The disadvantages of nasal tube feeding are discomfort to the patient, repeated displacement requiring reposition and difficulty in maintenance over the long term. Various factors have been attributed to the development of DGE following PD. The probable reasons for DGE after PD are intestinal dysmotility and obstruction at the FJ site. The first reason is hypothetical and without any concrete evidence to support it, while obstruction can be avoided by taking thin serosal sutures while creating the tunnel during the Witzel technique. Kinking at the FJ insertion site can be minimised by creating a wide-based fixation at multiple points rather than at a single point fixation. One of the limitations of this study was its small sample size, which could have been responsible for few of the parameters not achieving statistical significance. Furthermore, leaving FJ creation to the surgeon’s discretion could have led to the introduction of potential biases. Thus, although FJ has its own inherent complications, its benefits still outweigh the disadvantages following a pancreatoduodenectomy. Hence, a selective approach for inserting an FJ should be followed rather than a blanket omission. A further well-designed study may help to identify these patients. Article information Jebakumar GS, Tasgaonkar S, Muthiah J, Chinappa G, Anand KS, Jameel JK, et al. Feeding jejunostomy after pancreaticoduodenectomy: Benefit or burden? Ann Hepatobiliary Pancreat Surg 2025;29:317-22. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.