Pancreatic segmental portal hypertension (PSPH) is a rare, localized, regional, elevated pressure in the left portal vein system due to non-hepatic causes (1). Unlike generalized portal hypertension (GPH), PSPH is not caused by systemic diseases such as cirrhosis and portal vein thrombosis, but stems from local factors related to the pancreas. Because the splenic vein is close to the dorsal side of the tail of the pancreatic body, this anatomical basis makes it possible for various pancreatic lesions to involve the splenic vein, resulting in the splenic vein being susceptible to thrombosis or being affected by exogenous compression, and eventually leading to splenic vein occlusion (2). After splenic vein obstruction, varicose blood vessels may form in all of these vessels. According to the literature, 45% ~ 72% of patients with PSPH develop upper gastrointestinal bleeding (UGIB) due to variceal rupture, and severe UGIB is often at fatal risk (3). The most common causes of PSPH and UGIB include acute and chronic pancreatitis and pancreatic cancer, and other less common causes include pancreatic tuberculosis, pancreatic abscess, pancreatic trauma, and pancreatic surgery (3, 4). Currently, splenectomy is the most common treatment for patients with PSPH who develop UGIB, However, for GPH, comprehensive treatment for the primary liver disease is often required. For patients with mild UGIB, refusal of surgery, or poor general condition, endoscopic therapy and splenic artery embolization (SAE) are the best options. However, endoscopic treatment carries the risk of rebleeding, and “post-infarct” syndromes such as abdominal pain, splenic abscess, sepsis, hematoma, and ectopic embolization are prone to occur after SAE (5). To date, there are no guidelines or consensus on the treatment of PSPH with UGIB. Therefore, we report six cases of PSPH complicated with UGIB caused by different pancreatic etiologies to share clinical experience and summarize its treatment in combination with relevant literature.

We included six patients treated in our department between July 2022 and August 2024 who were considered to have a diagnosis of PSPH with UGIB. At present, no unified diagnostic criteria for PSPH have been established at home and abroad. In this study, the diagnosis of PSPH with gastrointestinal bleeding is mainly based on the patient’s clinical manifestations, CT, gastroscopy and other auxiliary examinations: 1. The patient has pancreatitis, pancreatic tumor and other pancreatic diseases, and has gastrointestinal bleeding manifestations such as hematemesis and hematochezia. 2. Pancreatic lesions, portal vein thickening, varicosities of the fundic veins and splenic vein thrombosis are seen on CT, while the liver is morphologically, size-wise, and texture-wise normal. 3. Isolated gastric varices diagnosed by gastroscopy, ultrasonography and gastroscopy. 4. Exclude portal hypertension due to cirrhosis.

Overall, six patients with PSPH and UGIB were enrolled. There were four males and two females. The average age was 44 ± 14.05 years (minimum 20 years, maximum 61 years). The six patients had different etiologies, one had chronic fibrous pancreatitis, one had type II autoimmune pancreatitis, one was diagnosed as pancreatic neuroendocrine tumor with multiple metastases, two had Pancreas Surgery, and one of them had concomitant pancreatic pseudocyst. Most patients had hematemesis and/or melena as first symptom, only one patient complained of fatigue, and two patients were accompanied by abdominal pain. Each patient presented with isolated GVs on EGD. All patients underwent CT, and had no liver diseases such as liver cirrhosis. One patient demonstrated a presence of tumor thrombus in the portal vein and splenic vein.

The average hemoglobin and albumin levels of all patients were low, with average values of 66 g/L and 33 g/L, respectively. There was no significant increase in liver function and indicators such as amylase and lipase.

One patient had visible bleeding under endoscopy and underwent gastric vein embolization and injection of Cyanoacrylate Glue. The remaining patients did not show active bleeding under endoscopy, so no endoscopic treatment was performed, and all were given conservative treatment mainly to inhibit acid and pancreatic enzyme secretion. Three patients were later transferred to the general surgery department for further treatment, one of whom underwent splenic vein stent implantation, and two underwent laparoscopic splenectomy and pericardial vascularization. Five patients did not experience UGIB during the follow-up period. The demographic and clinical characteristics of the patients are summarized in Table 1.

PSPH, also known as Left-Sided Portal Hypertension (LSPH), is a type of extrahepatic portal hypertension, accounting for 5 percent of extrahepatic portal hypertension (6), which is caused by the spleen vein being compressed by a mass or cyst, or blocked by thrombosis, making it difficult for blood from the spleen to flow through the short gastric vein to the stomach and eventually to the portal vein, but instead to the collateral circulation of the short gastric vein, resulting in GVs (3, 7). Isolated GVs is a clinical feature of PSPH, but GVs of PSPH may also be secondary to esophageal variceal rupture or portal hypertension gastropathy (8). Although the clinical incidence of PSPH complicated by UGIB is low, the associated GVs still have a high risk of mortality from ruptured bleeding.

When UGIB was caused by PSPH, hemodynamics should be optimized by injection of crystals or colloids and blood transfusions, in addition, vasoconstrictors (e.g., somatostatin, vasopressin), acid-suppressing drugs (e.g., PPIs, H2 receptor antagonists), and three-lumen two-cyst tubes should be used to stop the bleeding urgently. When the patient’s vital signs are stabilized, it is also necessary to address the root cause of hypertension on the splenic side of the portal venous circulation and eliminate the risk of recurrent bleeding combined with the patient’s primary disease (8, 9).

Unlike PSPH, GPH refers to a pathologic condition in which pressure within the portal system is generally elevated, usually caused by systemic disease inside and outside the liver, and involves the entire portal system, rather than being confined to a small segment. For this reason, its diagnosis usually requires evaluation of the entire portal venous system by ultrasonography, CT, endoscopy, and other multifaceted examinations. In terms of treatment, both need to be treated for the cause, elective transjugular intrahepatic portosystemic shunt (TIPS), and liver transplantation are often used for GPH (10). Taken together, GPH has a wider range of effects, which requires a comprehensive assessment of portal venous system involvement and comprehensive treatment measures for the underlying disease, while PSPH focuses more on the assessment and management of local pancreatic diseases.

Splenectomy is often recommended for patients with PSPH and UGIB (11, 12), it can not only reduce gastric venous congestion due to blood flow into the collateral circulation of the short gastric veins by removing the entire spleen to terminate the blood supply to the splenic artery, but also allows surgeons to treat possible pancreatic disease concurrently (13). However, in recent years, the disadvantages of splenectomy have also been frequently criticized. First, splenectomy, as an open procedure requiring general anesthesia, carries certain anesthetic and surgical risks, and a mortality rate of up to 8% has been reported with splenectomy (14). Second, pancreatic diseases such as pancreatitis and pancreatic tumors are the most common causes of PSPH (3, 15). However, surgery is often difficult in difficult settings due to inflammatory adhesion due to pancreatitis or other complications, particularly during acute exacerbations (16). At the same time, splenectomy results in a loss of spleen immunity, which may lead to overwhelming post-splenectomy infection (OPSI), also known as post splenectomy syndrome, with a mortality rate of up to 50% (17). Despite these risks, splenectomy is used as the mainstay in treatment of PSPH with UGIB because it minimizes blood flow to gastric and esophageal varices. In a meta-analysis, Liu et al. (18) concluded that splenectomy was more effective in reducing the incidence and mortality of UGIB than other treatment strategies.

In addition, prophylactic splenectomy is generally not recommended for patients with PSPH who have not yet experienced UGIB due to the insignificant differences in overall survival (19), the presence of biliary digestive anastomotic vein appendages (20), and avoidable complete loss of spleen function.

Based on the risks associated with splenectomy, SAE may be an alternative treatment. It works the same as splenectomy and can prevent bleeding by reducing blood flow from the spleen to the collateral veins (3), thereby decompressing the associated varices. The most important advantages of SAE over splenectomy are its minimally invasive nature and preservation of spleen function, and it has been shown that SAE is an effective method for UGIB caused by PSPH (21), but it may still cause some complications, including splenic abscess, sepsis, and splenic rupture (5). In general, SAE is the embolization of 50 to 70% of the splenic parenchyma (22), and when embolization of too much of the spleen parenchyma, more and more serious complications may arise (23).

The two-step embolization method and modified endoscopic ultrasound (EUS)-guided selective N-butyl-2-cyanoacrylate (NBC) injection have shown greater advantages in minimizing complications associated with SAE. When a large portion of the spleen is completely embolized in a single operation, postembolic syndrome may occur, presenting with fever, abdominal pain, etc. (24). Therefore, some patients may require a two-step embolization method. Wang et al. (25) suggested that partial splenic embolization can be performed to achieve infarction of 60 to 70% of the splenic volume in patients with PSPH and UGIB after their vital signs have stabilized, and a second embolization treatment was performed a month later. Patients treated with the two-step embolization method experienced a reduction in symptoms of postoperative fever and abdominal pain compared with those treated with one-step embolization, and in addition, the risk of pseudoaneurysms was reduced with this approach (26). Modified EUS-guided selective NBC injection therapy has also been widely used due to its outstanding advantages, and compared with traditional endoscopic injection, modified EUS-guided NBC injection has performed better in terms of reduced NBC dose, fewer complications, less risk of additional radiation exposure, and lower cost of endoscopic procedures (27–29). However, the disadvantage is that it is time-consuming and requires more technical skills from the operator (29). In addition, SAE can also be used as a preoperative management for splenectomy to reduce intraoperative blood loss and conversion to open surgery (30, 31). Liu (18) and Fujitani RM et al. (32) suggest that SAE-assisted splenectomy is more appropriate in high-risk patients and can significantly reduce intraoperative blood loss and operative time.

Unlike the two methods described above, SVS relieves pressure on collateral circulation by placing a stent in the splenic vein to restore blood flow to the collateral circulation (33), thus restoring normal blood circulation while completely preserving the spleen, thereby avoiding complications associated with loss of spleen function (34). SVS has been reported to be effective in the treatment of UGIB caused by PSPH in many previous case studies (30, 35–37). In a retrospective study of the effects of splenectomy and SVS on treatment of PSPH, Liu et al. (38) found that SVS was associated with a lower incidence of surgical complications such as bleeding and infection, a shorter hospital stay, and a lower overall cost of care compared with splenectomy. In another retrospective study comparing SAEs and SVS, Bo Wei et al. (33) noted that patients who received SVS (7.1%) had a significantly lower rate of rebleeding than those who received SAE (47.8%), and that SVS had a more significant advantage in preventing PSPH rebleeding. However, according to previous studies, the success rate of SVS is only 53.8% (33), and its use may be limited by a low success rate, and stent passage is difficult in some sites of severe splenic vein occlusion, particularly long splenic vein occlusion due to chronic pancreatitis, which is a major cause of SVS failure (33). In cases where it is difficult to open a stenosis segment of the splenic vein by portal vein approach, the addition of a splenic venous puncture method may increase the probability of reopening of the splenic vein occlusion segment (37).

In conclusion, the treatment of PSPH with gastrointestinal bleeding is divided into treatment of the primary disease of the pancreas and management of gastrointestinal bleeding. Splenectomy, splenic artery embolization, splenic vein stenting are all effective for gastrointestinal bleeding due to PSPH. Among these, splenectomy is most widely used because it minimizes blood flow to the stomach and simultaneously treats the primary disease of the pancreas. When the patient has recurrent pancreatitis causing tissue adhesion and it is difficult to perform surgery, splenic artery embolization can be used as an alternative, however its application is limited due to its possible complications such as splenic rupture and septicemia. Splenic vein stent implantation has the advantages of minimal invasion, preservation of spleen and fewer postoperative complications, but has certain technical requirements for the operator. When the patient is suffering from splenic vein occlusion due to chronic pancreatitis, the operation is difficult to carry out. A total of 3 patients were treated surgically in this study. Splenic vein, portal vein stent implantation in one patient for obstruction of portal, splenic, and superior mesenteric venous junctions. One patient had a pancreatic space-occupying lesion involving the spleen, and was selected for pancreatic lesion resection and splenectomy. Another one patient underwent splenectomy for massive splenosis and severe anemia. It can be seen that these patients have their own characteristics, and blind selection of splenectomy is not advisable, individualized treatment is particularly important.

In this study, six patients with pancreatic disease and UGIB were collected. The mean hemoglobin and albumin levels in six patients were low, which may be related to long-term chronic gastrointestinal blood loss. Five patients complained of melena and hematemesis, and were able to receive timely treatment. The patient in case 6 started with fatigue, and white cells, red blood cells and platelets were low. Complete abdominal CT showed pancreatic pseudocyst and portal hypertension. Gastroscopy shows dark red blood in the fundus and several varicose veins. After finding no contraindications, we performed ultrasound gastroscopic venous embolization combined with injection of Cyanoacrylate Glue, and the patient recovered well after surgery. This case is enough to remind us that for patients without typical UGIB manifestations, we should pay attention to routine blood count, abdominal CT and gastroscopic findings.

The etiology and treatment methods of the six patients in this study are abundant, but due to the small sample size, this study is only descriptive statistics, and a large sample study is needed to summarize the clinical characteristics of such patients in the future, so as to make timely diagnosis and treatment.

Since the initiating factors of PSPH are various types of primary pancreatic diseases, it is necessary to pay more attention to the primary pancreatic diseases while solving the problem of UGIB, and take corresponding treatment to achieve the purpose of effectively controlling the symptoms and alleviating the progression of the disease.

In conclusion, for patients with PSPH and UGIB, treatment should be individualized for their primary pancreatic disease. In this study, all six patients with PSPH and UGIB were treated with SVS, splenectomy, and treatment for primary pancreatic disease according to their actual conditions. In this study, we summarized the patients diagnosed with PSPH complicated with UGIB in our department and summarized the current mainstream clinical methods for the treatment of PSPH with UGIB, in order to provide reference opinions for the diagnosis, individualized treatment and prevention of PSPH complicated with UGIB.