LPPH is a rare but fatal complication with an incidence of 1%–10%2 and a mortality rate of 14%–58%1, 5. In this study, the incidence of LPPH was 6.3%, which was consistent with the literature. LPPH is mainly associated with postoperative complications, including pancreatic fistula, biliary fistula, and abdominal infection, leading to peripancreatic vascular erosion, anastomosis ulceration, and pseudoaneurysm8, 9. The main complications are pancreatic fistula and abdominal infection, which impede successful treatment and result in recurrent bleeding and extremely poor prognosis. In this study, there were 11 cases of LPPH with comorbid pancreatic fistula, 5 cases of biliary fistula, 7 cases of abdominal infection, and 6 cases of arterial hemorrhage. LPPH can present as intraluminal hemorrhage, i.e., bloody fluid from the nasogastric tube, hematemesis, or melena. There were 4 cases of intraluminal hemorrhage. LPPH can also present as extraluminal hemorrhage, i.e., bloody fluid from the abdominal drainage tube, abdominal pain, and abdominal distension. There were 9 cases of extraluminal hemorrhage. One patient had extraluminal and intraluminal bloody fluid. After surgical exploration, GDA hemorrhage was considered, which entered the intestinal tract through the biliary-enteric anastomosis.
LPPH treatment is the focus and pain point of managing post-LPD complications. Currently, there is no golden standard for LPPH treatment. Observation should be strengthened after surgery. Once hemorrhage is suspected, emergency abdominal CT and routine blood examinations should be performed. Currently, there are three main treatment methods: (1) For patients with intraluminal hemorrhage, gastroscopic hemostasis could be considered first10. (2) In recent years, arterial embolization and covered stent placement, which are vascular intervention surgeries, have undergone rapid development and have demonstrated significant benefits in LPPH treatment11, 12. However, the mortality rate is still high. In the study by Zhou et al., 29 LPPH patients underwent endovascular treatment, but 8 (28%) patients had re-bleeding and 5 (17%) patients died11. This may be due to risk factors such as pancreatic fistula and abdominal infection, which were unresolved, resulting in recurrent bleeding, and are associated with liver abscess and liver failure. (3) Surgical exploration: Currently, suturing and hemostasis of bleeding points alone is mostly used but this has problems such as a high mortality rate and recurrent bleeding. The causes of death include recurrent hemorrhage, abdominal infection, and multiorgan failure13, 14. In the study by Liang et al., 7 (35%) of 20 PPH patients died after conservative treatment, endovascular treatment, and re-surgery. Five of these patients underwent reoperation, but three eventually died. They found that pancreatic fistula is an independent risk factor for LPPH15. Some researchers referred to late post-pancreatic surgery hemorrhage as “pancreatic fistula-related bleeding,”16 because a closed “pancreatic fistula–abdominal infection–bleeding” cycle occurs in most of these patients. Currently, there is no study on total pancreatectomy of the residual pancreas in the treatment of LPPH. In the present study, 8 patients underwent re-surgical exploration. Of these, 1 patient had grade B bleeding and underwent suturing and hemostasis; the other 7 patients, who were grade C bleeding and complicated with pancreatic fistula, abdominal infection or biliary fistula, underwent residual total pancreatectomy. All 14 LPPH patients recovered and were discharged. No patient died. The low mortality rate of our hospital may be due to LPPH patients undergoing total pancreatectomy of the residual pancreas, which disrupted the “pancreatic fistula–abdominal infection–bleeding” cycle, which ultimately prevented recurrent bleeding and abdominal infection.
During the early stage of surgical development, the loss of the pancreas will result in digestive impairment and dysregulated blood glucose levels in patients. Total pancreatectomy was popular for a period in the 1980s, before it was gradually considered as a surgical contraindication. However, due to improvements in endocrine treatment of blood glucose levels and the invention of digestive enzymes, loss of the pancreas no longer results in severe consequences and the value and significance of total pancreatectomy in the treatment of pancreas-related diseases have been re-recognized17. Pulvirenti et al. also believe that total pancreatectomy is an option18. In the present study, 7 total pancreatectomy patients received postoperative insulin treatment and glycemic control was stable; 5 patients had increased bowel movements but did not have diarrhea, and symptoms were alleviated after pancreatic enzymes were taken. One patient died due to tumor recurrence. Six patients survived until the present day, and the longest survival period was 106 months.
Based on the above cases and past practical experience, our team developed the following LPPH management method: (1) Once LPPH is considered, a multidisciplinary team discussion is immediately initiated to formulate the subsequent treatment regimen. (2) Endovascular treatment could be carried out for grade B bleeding and conservative treatment such as hemostatic drugs and blood transfusion are carried out. Patients are closely observed for changes. Exploratory laparotomy is performed if the effects of conservative treatment are poor. (3) For grade C bleeding, particularly patients with comorbid shock, surgical exploration should be first considered and endovascular treatment could be used together. For patients with pancreatic fistula who are critically ill, the preferred surgical procedure is total pancreatectomy.
In summary, we believe that total pancreatectomy of the residual pancreas is safe and feasible in grade C bleeding patients with comorbid pancreatic fistula or abdominal infection, and this can be combined with endovascular treatment. This could decrease the mortality rate of LPPH patients. However, this study has the following limitations: (1) small sample size; and (2) a retrospective nature lacking randomization. It is difficult from an ethical point of view to perform a randomized controlled study in life-threatening situations. In the future, more studies are required to validate our conclusions.