Non-cirrhotic portal hypertension is characterized by raised portal venous wedge pressure with normal hepatic venous wedge pressure and preserved liver functions. Extrahepatic portal vein obstruction (EHPVO) is a significant cause of portal hypertension in children and adults with normal liver.
There is a wide span of etiological factors for extrahepatic portal vein obstruction (EHPVO), including congenital anomalies, hypercoagulable state due to deficiency of anticoagulants like protein Cand protein S deficiency or excessive production of procoagulants like factor V Leiden, trauma, sepsis, umbilical vein catheterization and malignancy [9]. Still, the most common cause was idiopathy, accounting for 70% of cases. In our study, out of 45 patients, only two had non-idiopathic causes. One patient was antithrombin III deficient, and the other one was JAK-2 mutation positive. In the rest of the patients (95%), the cause was unknown.
Hematemesis is the main presenting feature of non-cirrhotic portal hypertension (NCPH). In most studies, upper gastrointestinal bleeding is present in 80–90% of the patients with non-cirrhotic portal hypertension (NCPH) [10]. Forty patients (88.9%) in our study were presented with only hematemesis. Jaundice was the clinical presentation in 03 patients, and ascites was the leading complaint in 02 patients. In addition to the above features, other symptoms were abdominal pain, features of hypersplenism, and poor growth, especially in children.
Doppler ultrasound, computerized tomography (CT) scan, and magnetic resonance imaging (MRI) are the primary imaging modalities with sensitivity and specificity from 80%to100% for diagnosis purposes. In the suspected portal vein thrombosis, doppler ultrasound was the first-line modality used in our study group. Ultrasound has a diagnostic accuracy of 88–98%. A computerized tomography (CT) scan was then performed to determine the extent of the thrombus, the severity of the disease, and any other pathology correlating with portal vein thrombosis (PVT) and to determine the surgical options.
In our study, most patients had normal texture liver with smooth margins, but two patients had slightly altered texture. These two patients were thoroughly investigated to rule out primary liver disease. We additionally performed a fibro-scan and biopsy on these patients, which were negative for steatosis, fibrosis, and cirrhosis.
For managing variceal bleeding due to non-cirrhotic portal hypertension (NCPH), recommendations are the same as for patients with portal hypertension and liver cirrhosis. Exclusively for non-cirrhotic portal hypertension (NCPH) related variceal bleeding, no controlled studies are available. Endoscopic band ligation is recommended for varices > 5mm. B-blockers are recommended for both primary and secondary prevention. All patients in our study underwent upper gastrointestinal endoscopy and band ligation at least twice (average 2.5% ± 2.3) before they presented to the surgical team. The commonest were grade III esophageal varices (51.1%), followed by grade II (40%) and grade I (8.9%).
Portal hypertension can also cause abnormalities of the biliary tree called portal biliopathy. These changes can progress to cholangitis, choledocholithiasis, biliary stricture, and secondary biliary cirrhosis.80–100%of these patients have biliary changes of some extent, but only in 5–38% of patients, it becomes symptomatic [11]. In our study, 28 patients (62.2%) had biliary abnormalities on computerized tomography (CT) suggestive of portal biliopathy. Only 04 patients developed biliary stricture, causing recurrent cholangitis. ERCP and stenting were done in these patients to decompress the channels and to manage the sepsis. Surgically, hepaticojejunostomy was performed in addition to splenectomy and shunt formation.
Morbidity and mortality related to upper gastrointestinal bleeding (UGIB) have dramatically improved with advances in endoscopic and radiological procedures [10]. Nonetheless, endoscopic procedures do not address the actual cause and sequelae of the disease. Trans-jugular intrahepatic portosystemic shunt (TIPS) is a decompressive procedure and reduces the risk of re-bleeding. However, it is technically challenging with a higher risk of complications, and it may be required multiple times due to increased chances of stent blockage.
However, surgery can decompress the portal system by channeling the higher-pressure system to the lower one. Moreover, the surgeon can also directly manage the complications related to portal hypertension [12, 13].
Surgical intervention is indicated if patients have complications refractory to endoscopic and radiological intervention. In patients with failure to thrive or symptomatic hypersplenism, there is difficulty in accessing healthcare centers for definitive treatment. Surgery can be divided into decompressive surgery and ablative surgery. Decompressive surgery includes shunt formation, whereas ablative surgery is esophagogastric devascularization.
Surgically created shunts decrease portal hypertension, resolve upper gastrointestinal bleeding, and can gradually reduce the splenic size and pressure effects at porta hepatitis. A proximal splenorenal shunt is a good option for diverting blood flow directly into the systemic system in non-cirrhotic patients. The patency rate is 87%, and the chances of re-bleeding are low (10%) [14]. In our study, proximal SRS was performed in 36 patients. For proximal splenorenal shunt surgery, one limitation is a thrombosed splenic vein. If the portal venous system is completely thrombosed, only a devascularization procedure can be performed to reduce the bleeding chances from gastric varices, and splenectomy can be done for hypersplenism [15]. In nine patients, proximal splenorenal shunting was not possible. The splenic vein was also thrombosed in eight of nine patients, so shunting could not be done. While one patient had a good-sized patent spontaneous splenorenal shunt, only devascularization was performed.
The most performed devascularization surgeries are the Sugiura and Hassab procedures. In the Sugiura procedure splenectomy, devascularization of the superior two-thirds of the greater and lesser gastric curvature and the distal esophagus through the diaphragmatic hiatus (originally described with an abdominal and thoracic approach) and esophageal transection 4–6 cm above the gastroesophageal junction followed by end-to-end anastomosis is performed. In the modified Hassab procedure, splenectomy and devascularization of the esophagus and proximal part of the stomach, along with vagotomy and pyloroplasty, are performed [16]. Devascularization surgery was performed on all our patients, either with or without a shunt. Devascularization of the distal esophagus was performed if significant-sized esophageal varices were appreciated on a computerized tomography (CT) scan. Vagotomy and pyloroplasty were not performed in these patients.
Although surgical shunting effectively controls variceal bleedings and improves portal biliopathy and gastropathy, occlusion, re-bleeding, and encephalopathy are common complications. After a follow-up of 05 years, only two patients presented with re-bleeding after surgery, and both had a single episode of hematemesis without requiring a blood transfusion. Endoscopy was unremarkable in both. All patients were alive and asymptomatic throughout our study period.
The limitation of our study was that it was a single-center study and small sample size. However, from only one specialized hepatopancreatic biliary center in Pakistan, sharing our experience would pave the path for further research on surgical management of non-cirrhotic portal hypertension (NCPH) and its outcome.