Aim of the study: Extrahepatic portal vein obstruction (EHPVO) is one of the main causes of childhood portal hypertension. We analyzed long-term outcomes of an institutional EHPVO cohort with or without shunt surgery.
Methods: After ethical approval 54 consecutive patients with EHPVO diagnosed between 1987 and 2017 were followed up through national health care registries and hospital records.
Main results: Median follow-up age was 17.6 (range 3.4-45) years. None of the patients were lost to follow-up or died. Esophageal varices (n=53) were diagnosed at the age of 5.5 (range 0.3-13) years. Patients underwent 9.5 (range 1-55) gastroscopies and 6 (range 0-29) sclerotherapies. Thirty-one (57%) patients had variceal bleeding (median 1, range 1-17) necessitating Sengstaken-Blakemore tube in three (6%). Overall, 39 (72%) patients underwent 46 shunt procedures with 82% (32/39) patency rate at the end of follow-up: Rex 72% (21/29), Warren 100% (4/4), side-to-side splenorenal 80% (4/5), proximal splenorenal 50% (1/2), mesocaval 25% (1/4), other 100% (2/2). Nine (17%) patients had undergone splenectomy (n=8) or splenic embolization (n=1). During the last five follow-up years 4 (7%) patients had experienced variceal bleeding episodes: 0/32 with patent shunt, 1/7 with occluded shunt and 3/15 without shunt (p=0.03). Resolution of hypersplenism was comparable after different shunt procedures (Rex 90% vs non-Rex 78%, p=0.6), although Rex shunt increased platelet levels more effectively [101 (6-233) vs 67 (-29-110) E9/L, p=0.04]. Five (9%) patients developed symptomatic portal biliopathy at the age of 22 (14-25) years. All of them had undergone splenectomy (5/8 vs 0/46, p<0.01), and had no shunt or occluded shunt (5/22 vs 0/32, p<0.01).
Conclusion: Different surgical shunts prevented variceal bleeding slightly better than endoscopic management in the long term, while Rex shunt alleviated hypersplenism most effectively. Splenectomy and the absence of a functioning shunt associated with development of portal biliopathy.