Search

Portal vein thrombosis represents one of the most frequent causes of portal hypertension in childhood. The aim of the present study was to describe the clinical and laboratorial characteristics of portal vein thrombosis in pediatric patient. We studied 26 children with diagnosis of portal vein thrombosis through splenoportography (two patients) and ultrasound scan (24 patients) which ages varied from 2 months to 11 years and 4 months (median-5 years and 3 months). Data of the patient history, physical and laboratories examination were used to a retrospective study which was done through medical record analysis. The main complaint of the examination was hematemesis, which was found in 57.6%. In 26.9% a possible risk factor for portal vein thrombosis was found [catheterization of the umbilical vein (four), sepsis (two), omphalitis (one)]. Splenomegaly was present in all cases and the associated illness to portal vein thrombosis were: hepatoportal sclerosis (three), cytomegalovirus infection (two), blastomycosis (two), virus C (two), virus B (one) and virus A (one). The time between the first bleeding and the examination at University of Campinas Hospital, in Campinas, SP, Brazil, varied from 0.23 months to 54 months with a median of 12 months. Only 11.5% of patients underwent the endoscopy with sclerotherapy before going to University of Campinas Hospital. Aminotransferases' activities were considered normal in 20 patients. We could conclude that: 1. The most frequent initial symptom was hematemesis. 2. The known risk factors for portal vein thrombosis were present in about 1/3 of the cases. 3. Laboratorial exams usually indicated absence of hepatocitic lesions. 4. The efforts towards sending the patient to a reference center were late with a delayed diagnostic and with delayed effective therapeutic conduct. 5. In about 50% of the cases there was PVT associated with other hepatic diseases.

Gastroesophageal reflux disease has been very often in children and one of its most critical complications is the peptic esophagitis disease. The diagnosis of peptic esophagitis is obtained based on the endoscopic changes and the histologic features. The pressure reduction of the lower esophageal sphincter is one of the elements of the esophagitis pathogenesis. The pressure is related with the length of the esophageal abdominal segment, which can be measured by ultrasonography. The research presented in this paper aims to relate ultrasonographic measurements of esophageal abdominal segment length with endoscopic changes and with esophageal biopsy obtained from children with gastroesophagel reflux disease. We submitted 16 children with gastroesophagel reflux disease, between 10 and 156 months old (median 63.5 months old), to endoscopy and to esophageal biopsy. We verified the following results from endoscopy and biopsy: six of the 16 patients had endoscopic peptic esophagitis and, in five of these (six patients), the same diagnosis was confirmed by biopsy. Ten of the 16 patients had normal endoscopy, but the biopsy of four of these 10 patients showed histologic changes compatible with esophagitis. We observed no relationship between endoscopy and biopsy, when all degrees of histologic change were used to diagnose esophagitis. However, we found high relationship when the mild histologic changes were considered as only a consequence of the reflux. We verified the following results from ultrasonography: five of the six patients with endoscopic peptic esophagitis and all patients with esophagitis determined by biopsy (excluding those with mild histologic changes) had reduced esophageal abdominal segment length. In conclusion, we have found relationship between endoscopic changes and histologic features in the diagnosis of esophagitis and correlation between the reduced esophageal abdominal segment length and the presence of esophagitis.

Hepatoportal sclerosis is the term used to name a clinicopathological condition responsible for non-chirrotic portal hypertension. A three cases report of children with hepatoportal sclerosis is presented associated with portal vein thrombosis. The first two patients presented as main complaint upper digestive hemorrhage and the third one was admitted for investigation of hepatosplenomegaly. The ultrasonographic exam revealed alterations indicative of extrahepatic portal vein thrombosis in the three cases. The patients underwent liver biopsy for they had presented altered liver enzymes. The main histological findings were: subintimal sclerosis, portal fibrosis and telangiectases of the intrahepatic venous branches, consistent with the diagnosis of hepatoportal sclerosis. The three patients showed good evolution, being the hemorrhage controlled in the first two cases through esclerotherapy of esophageal varices.

Neonatal cholestasis still presents a diagnostic challenge, both from the anatomic and etiologic point of view. Distinguishing intrahepatic from extrahepatic causes of cholestasis is of paramount importance since the latter may be treated by surgery but prognosis depends on the age at which operation is performed. Many tests have been proposed to help in differentiating these two entities, among which liver biopsy is the most frequently employed, and ultrasonography one of the most recently included. Our purpose is to present our experience with these two methods in the differential diagnosis of intra and extrahepatic causes of cholestasis. From January 1989 to July 1993, 35 patients with neonatal cholestasis were evaluated through a protocol which included liver biopsy and ultrasonography. The latter was performed after a 4 hour fast and was considered indicative of extrahepatic cholestasis when the gallbladder was not visualized, was hypoplastic or non-functioning, or if a cystic structure was seen at the extrahepatic biliary tree. Sensitivity, specificity, and accuracy were determined for these two tests. Seventeen patients were found to have extrahepatic cholestasis (all with biliary atresia) and 18 intrahepatic cholestasis, on the basis of clinical evolution or operative findings. Sensitivity was 100% for ultrasonography and 76% for liver biopsy in diagnosing extrahepatic cholestasis. Accuracy was 83% for ultrasonography and 86% for biopsy, rising to 96% when both tests were considered together. Based on these findings we strongly recommend ultrasonography with definite criteria as the initial investigation tool in the management of neonatal cholestasis, associated with liver biopsy.