Effects of Fasting and Administration of Octreotide Acetate and Ulinastatin on Clinical Outcomes of Pancreatic Fistula After Pancreatoduodenectomy

Pancreatoduodenectomy (PD) is a surgical procedure used as treatment for both benign and malignant diseases of the pancreatic head region, and it is one of the most technically demanding gastrointestinal surgeries to perform.¹ Postoperative pancreatic fistula (POPF) is the most critical complication of PD, with a relatively high incidence rate of 2% to 28%,^1–4 and carries the risk of intraperitoneal bleeding and postoperative mortality.⁵ In the presence of POPF, trypsinogen, a protein fusion enzyme existing as an inactive substance in pancreatic fluid, transforms into trypsin, its active form, in the presence of intestinal fluids or bile, including activated enterokinase. The activated trypsin in turn activates other protein fusion enzymes, inducing damage to adjacent tissues via autophagia. As a result, intraperitoneal bleeding due to ruptured pseudoaneurysm, multiorgan failure due to intraperitoneal abscess, or sepsis induced by infectious pancreatic fluid may occur. The incidence of intraperitoneal bleeding after PD ranges between 1% and 8% worldwide,^5–11 with the associated mortality rate of as high as 11% to 58%.^7–15

Although various drug therapies and surgical techniques have been used for the treatment of POPF, a decisive treatment for POPF has not yet been defined. Fasting is one of the possible medical interventions for the treatment of POPF. The rationale for fasting as an intervention is that oral intake induces the secretion of secretin or cholecystokinin, which promotes pancreatic exocrine function, leading to an exacerbation of POPF.¹⁶ Fujii et al¹⁷ evaluated the effects of fasting by randomizing 59 patients who developed POPF 5 days after PD to an oral intake group (n = 30) and fasting group (n = 29), and they compared the following factors between the 2 groups: duration of drain placement, grade of pancreatic fistula, presence or absence of intraperitoneal bleeding, length of hospitalization, and mortality rate during hospitalization. Fujii et al¹⁷ concluded that oral intake did not prolong the period of drain placement nor did it exacerbate the pancreatic fistula. Drug therapies, as well as the use of fibrin glue applied to the pancreatic-digestive anastomotic region, and intra- and postoperative administration of octreotide acetate (OC) and ulinastatin (UTI), also have been used as treatment options for POPF, but with limited reported evidence of success.^18–26 Therefore, the purpose of this study was to confirm the effects of fasting (no dietary intake [NDI]) and the administration of OC and UTI for the treatment of POPF.

Materials and Methods

Results

Relevant background characteristics of the 30 patients who developed POPF are summarized in Table 1. There was no significant difference between NDI and DI groups regarding age, sex, disease, diagnostic criteria of pancreatic fistula, and period from the surgery to the time of POPF diagnosis. The length of hospitalization was significantly longer in the NDI group than in the DI group (P = 0.002), with a higher incidence of intraperitoneal bleeding in the NDI group than in the DI group (P = 0.024, Table 2). Among the 30 patients who developed a POPF, a POPF grade B or C was identified in 26 patients (17 in the NDI group and 9 in the DI group). On subgroup analysis of such patients, length of hospitalization was again longer in the NDI group than in the DI group (P = 0.003), with a higher WBC at 7 days after diagnosis in the NDI group than in the DI group (P = 0.028, Table 3).

Table 1 Patient characteristics (n = 30)

View Fullscreen

Table 2 Results of NDI (n = 30)

View Fullscreen

Table 3 Results of NDI (POPF grade B or C only) (n = 26)

View Fullscreen

Regarding OC, 8 patients were in the NDI group and the other 22 in the DI group. The WBC was significantly higher at 7 days after the diagnosis of POPF in the OC group than in the non-OC group (22 patients; P = 0.021; Table 4). The effects of UTI administration are shown in Table 5. The length of hospitalization was significantly longer in the UTI than in the non-UTI group (P = 0.025).

Table 4 Results of octreotide (n = 30)

View Fullscreen

Table 5 Results of ulinastatin (n = 30)

View Fullscreen

Overall, NDI, OC, and UTI did not result in any significant favorable clinical outcome compared with DI.

Discussion

POPF is a critical complication of PD that increases the medical intervention required, prolongs hospitalization, increases the risk for morbidity and mortality, and decreases patient-reported quality of life. It is common for patients with a POPF to be on a prolonged period of fasting, requiring intravenous hyperalimentation for nutritional management, as well as the use of drug therapy to suppress the production of pancreatic fluid. Treatment of the POPF itself and treatments targeting biochemical and drug-specific effects have been evaluated. In the current study, we focused on evaluating the effects of NDI, OC, and UTI, and demonstrated no positive effect of any of them for POPF.

The effect of OC on the pancreatic fistula has been evaluated in numerous randomized studies, with no evidence of a positive effect identified.^20–24,27 However, 2 studies have reported a reduced incidence of POPF among patients treated with OC,^28,29 including a significantly shorter time to closure of the POPF.³⁰ Therefore, it has been suggested that OC may reduce the risk of POPF, with a possible advantage of prophylactic administration of OC.^31,32 Contrary to these findings, the current study demonstrated that OC administration was associated with a significantly higher WBC at 7 days after POPF diagnosis, with no favorable effect of OC on the measured outcomes noted.

UTI has been used clinically to alleviate acute circulatory disorders induced by acute or chronic pancreatitis, with evidence of effectiveness for the prevention of pancreatitis after endoscopic retrograde cholangiopancreatography.³³ However, 2 recent studies indicated that the incidence of postoperative pancreatitis was higher than expected after PD, with an associated increased risk for POPF.^34,35 To address this issue, Zhang et al³⁶ administered UTI intraoperatively and postoperatively in patients who underwent PD, and reported a decrease in the incidence rate of POPF (grade B or higher) compared with the placebo group. Uemura et al³⁷ reported that the group with UTI administered during the perioperative period was associated with a significant decrease in the level of amylase on postoperative days 1 and 3 compared with the placebo group; however, this prophylactic effect of UTI was not supported by a recent study evaluating the incidence rate of pancreatitis associated with endoscopic procedures.³⁸ In the current study, UTI administration was associated with a significantly longer length of hospitalization, with no indication of a favorable effect of UTI on clinical outcomes. It is important to note, however, that there are only a few reports currently available regarding the effects of UTI administration on POPF incidence. Therefore, the effects of UTI administration currently remain controversial.

DI induces the secretion of secretin and cholecystokinin, as well as promotes pancreatic exocrine function, which leads to exacerbation of the POPF. As previously described, Fujii et al¹⁷ did not identify a therapeutic benefit of fasting (NDI) over DI, concluding that fasting may not be required after PD surgery to reduce the risk of POPF. Klek et al,³⁹ however, did provide evidence in their randomized trial that nasogastric tube feeding of patients with POPF after PD (where PD included distal pancreatectomy, necrosectomy, and gastrectomy) significantly shortened the length of hospitalization compared with intravenous hyperalimentation by achieving pancreatic fistula resolution. This finding, however, should be cautiously considered for practice considering the findings of a similar study that reported an exacerbation of pancreatic juice leakage with nasogastric feeding under similar conditions.^40–43 Moreover, a multi-institutional, randomized controlled trial, which included 204 patients after PD, comparing the clinical course of patients provided with nasojejunal early enteral nutrition with those receiving total parenteral nutrition, reported that nasojejunal early enteral nutrition was associated with a higher rate of POPF (48.1% versus 27.7%, respectively; P = 0.012) and grade B or C of POPF (29.4% versus 13.9%, respectively; P = 0.007).⁴⁴ In the present study, the length of hospitalization was longer in the NDI group than in the DI group, with a higher incidence of intraperitoneal bleeding in the NDI group. According to a subanalysis of patients with grade B or C POPF, NDI was associated with a longer length of hospitalization and higher WBC; therefore, we did not identify any clinical advantage of NDI among patients who developed POPF after PD.

The primary rationale for fasting after PD was to lower the risk of POPF. Specifically, it has been postulated that oral intake may be associated with a lower secretion of secretin and cholecystokinin in the duodenum and upper jejunum in patients who have undergone PD than in healthy individuals. However, although some studies have confirmed a decrease in secretin and cholecystokinin after PD, others did not find a statistically significant difference.^45–48 Therefore, there remains no strong evidence to support the use of NDI after PD. However, there is general agreement that enteral nutrition has a more favorable impact on recovery than parenteral nutrition, promoting better wound healing and decreasing exuberant granulation tissue.⁴⁹ Moreover, related studies have suggested that DI may improve a patient's nutritional condition, as well as increase intestinal immunity. DI further promotes and improves self-reported satisfaction with daily life and quality of life.⁵⁰

Based on the evidence available, suppression of the production of digestive fluids may not improve the clinical course of POPF after PD. Moreover, fasting may indeed have a negative impact on intestinal immunity and nutritional status. In addition, long-term fasting may induce a decline in the function of the intestinal mucosa and pancreatic excretion, resulting in metabolic complications, such as morphological degeneration in the intestinal mucosa.⁵¹

Evaluation of clinical outcomes of interventions for POPF is important because unnecessary medical intervention can increase health care costs. Moreover, fasting, OC, and UTI typically require placement of a central venous catheter, which increases the risk for infection, as well as complications at the site of insertion.

The most prominent limitation of the current study was the fact that allocation of patients to the NDI, OC, and UTI groups was based on the subjective preference of the attending physician rather than by randomization. Thus, the groups were not balanced for comparison, which may have introduced significant bias and affected our results. Furthermore, the sample size was small, and all the patients were from one hospital. Thus, for future investigation, a multicenter randomized controlled study involving a larger number of patients is warranted to provide stronger evidence supporting the findings of the present study.

Conclusions

In conclusion, our data indicate that NDI, OC, and UTI did not confer a therapeutic benefit to the clinical course of patients with POPF after PD. Therefore, the most appropriate course of care for a POPF may be to secure anastomosis of the pancreas and digestive tract, followed with appropriate placement of a drain. Further, the management of DI may not adversely influence the outcomes of a POPF. Moreover, OC and UTI administration also have no notable effects on the patients' clinical course.

Data Availability

All data generated or analyzed during this study are included in this published article and are available from the corresponding author on reasonable request.