Clinical and Laboratory Predictors of Esophageal Varices in Children and Adolescents With Portal Hypertension Syndrome

2008 
Objectives: To determine the clinical and laboratory parameters that may predict the presence of esophageal varices in children and adolescents with portal hypertension. Patients and Methods: Overall, 111 patients with portal hypertension and no previous history of digestive bleeding underwent esophagogastroduodenoscopy for detection of esophageal varices. A univariate analysis initially was carried out, followed by a logistic regression analysis to identify the independent variables associated with the presence of esophageal varices. Sensitivity and specificity rates, positive predictive value, negative predictive value, and the accuracy of the predictive variables identified among cirrhotic patients were calculated with the esophagogastroduodenoscopy as the reference test. Results: Sixty percent of patients had esophageal varices on the first esophagogastroduodenoscopy. Patients with portal vein thrombosis and congenital hepatic fibrosis were 6.15-fold more likely to have esophageal varices than cirrhotic patients. When we analyzed 85 cirrhotic patients alone, splenomegaly and hypoalbuminemia remained significant indicators of esophageal varices. Only spleen enlargement showed appropriate sensitivity and negative predictive value (97.7% and 91.7%, respectively) to be used as a screening test for esophageal varices among cirrhotic patients. Conclusions: In reference services and research protocols, endoscopic screening should be performed in all patients with portal vein thrombosis and congenital hepatic fibrosis. Among cirrhotic patients, the indication should be conditioned to clinical evidence of splenomegaly or hypoalbuminemia. For clinicians, the recommendation is to emphasize the orientations given to guardians of patients with portal vein thrombosis and congenital hepatic fibrosis as to the risk of digestive bleeding. Cirrhotic patients with hypoalbuminemia and splenomegaly should receive the same orientations.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    29
    References
    55
    Citations
    NaN
    KQI
    []