Sasaki_2005_Surg.Today_35_385

PURPOSE: To clarify the correlation between portal hypertensive gastropathy (PHG) and clinical features after surgery for biliary atresia (BA).
METHODS: Routine upper gastrointestinal endoscopies were done over 3 years in 27 children who underwent surgery for BA. We reviewed the recorded endoscopic findings, and retrospectively diagnosed PHG according to McCormack's criteria. The differences in clinical features, such as endoscopically treated gastroesophageal varices and the results of routine laboratory tests, were compared between the children with PHG (PHG group) and those without PHG (non-PHG group).
RESULTS: Nine (33%) of the 27 children had PHG. Although there was no significant difference in age between the PHG and non-PHG groups, the frequency of past endoscopic variceal treatments was significantly higher in the PHG group (3.0 +/- 3.0 vs 0.6 +/- 1.5 times, P = 0.01). The PHG group also had lower white blood cell and platelet counts, at 3008 +/- 2411 vs 5527 +/- 2938/mm3 (P < 0.05) and 6.0 +/- 3.4 vs 13.9 +/- 4.7 x 10(4)/mm3 (P = 0.0001), respectively; higher serum aspartate aminotransferase, total bile acid, and total bilirubin levels at 80 +/- 31 vs 46 +/- 29 U/l (P < 0.05), 161 +/- 93 vs 64 +/- 88 U/l (P < 0.05), and 4.8 +/- 5.6 vs 1.0 +/- 0.8 mg/dl (P < 0.01), respectively; and lower prothrombin time, albumin, and cholinesterase levels, at 66 +/- 16 vs 79% +/- 14% (P < 0.05), 3.6 +/- 0.8 vs 4.1 +/- 0.5 g/dl (P < 0.05), and 2158 +/- 925 vs 3376 +/- 700 U/l (P < 0.001), respectively. CONCLUSION: Portal hypertensive gastropathy was found in 33% of children after surgery for BA. The factors contributing to the development of PHG were frequent endoscopic treatments of gastroesophageal varices, liver dysfunction, and hypersplenism.