|Encadrant||Lakhdar idrissi M|
Gastrointestinal bleeding is a frequent and serious complication of portal hypertension. The most common cause of bleeding is the rupture of eso-gastric varices. We present a retrospective descriptive and analytical study carried out at pediatric department of University Hospital Hassan II of Fez over a period of 3 years, from January 2011 until November 2014, on the study of bleeding profile of 43 children with portal hypertension in the purpose of: – Know the portal hypertension etiologies providers of a large bleeding risk. – Study rebleeding under treatment with beta blockers. – Study rebleeding compared to the number of ligatures. Our series of patients have: – Either presented gastrointestinal bleeding who discover a portal hypertension. – Either a known portal hypertension confirmed the basis of endoscopic data. The age of the patients in our series varies between 4 months and 15 years with an average age of 7 years and 8 months. Sex male / female ratio is 1.53. The gastrointestinal hemorrhage was present in 28 cases in our series (66%). The hematemesis is the most frequent mode of externalization; it is present in 15 cases. In half of cases, gastrointestinal bleeding is medium abundance (14 cases). Biologically: Severe anemia was noted in 8 cases (19%); liver function was impaired in 31 patients or 72 % of cases. The endoscopy has objectived oeso-gastric varices in 97.6 % of cases (23 cases of esophageal varices stage I, 27 cases for stage II, and 21 cases for stage III, the gastric varices are present in 5 sick and are always in association with esophageal varices ) , hypertensive gastritis was present in 19 cases ( 45.2 %). There is also only 1 case with duodenal varices. The causes are dominated by intrahepatic blocks with 33 cases (76.7 %), Sub – hepatic blocks (Portal cavernoma) come second with 9 cases ( 21%). Finally the above hepatic blocks (Budd-Chiari syndrome ) with only 1 case ( 2.3 %). The risk of occurrence of gastrointestinal bleeding was higher in the sub – hepatic block (Portal cavernoma) with 89% of cases of gastrointestinal bleeding, this difference was statistically significant (p = 0.042). For patients with portal hypertension of intrahepatic block, gastrointestinal bleeding occurred in 61 % of cases, especially indeterminate cirrhosis. However, the bleeding risk was zero in the above hepatic block (Budd-Chiari syndrome). In the sub -hepatic block, the patients showed especially hematemesis (p = 0.024). The gastrointestinal hemorrhage was significantly high abundance for the sub -hepatic block (p = 0.005), and low abundance for intrahepatic block ( p = 0.010 ) . In our series, recurrent bleeding is reported in 10 cases (23.3 %). The risk of rebleeding is high in the subhepatic block, this difference was statistically significant (p = 0.001). In the intrahepatic block, the evolution of patients is statistically favorable (p = 0.002) especially biliary cirrhosis and congenital hepatic fibrosis. Treatment with beta-blockers reduced the risk of recurrent bleeding, this difference was statistically significant in our study (p = 0.012). In our study, we found that the more ligation is realized; the more the evolution of our patients was favorable. In patients who received more than 3 ligation sessions; the evolution of patients is statistically favorable (p = 0.009).