Accuracy and interobserver agreement of small-caliber vs. conventional esophagogastroduodenoscopy for evaluating...

Advances in endoscope design have allowed high-quality imaging using small-caliber endoscopes (< 6 mm), and these have been proposed as providing an accurate modality for evaluating esophageal varices in several small studies. We aimed to evaluate the accuracy and interobserver agreement of small-caliber esophagogastroduodenoscopy (EGD) compared with conventional EGD for evaluating esophageal varices in a large prospective cohort.

A total of 115 patients with end-stage liver disease and/or portal hypertension were prospectively enrolled into the study. EGD procedures were performed using conventional (8.6-mm) and small-caliber (4.9-mm) endoscopes, back to back and under standard sedation, by two different endoscopists. Esophageal varices were graded at the time of EGD (the "real-time" grade); and by retrospective review of photographs by three endoscopists, when a "consensus" grade (i. e. a grading agreed by two out of the three endoscopists) was used as the final result.

Of the 115 patients, 33 patients (29 %) were classified as Child's class A, 47 patients (41 %) as Child's class B, and 35 patients (30 %) as Child's class C. The mean model for end-stage liver disease (MELD) score was 13.6. Thirty-six patients (31 %) had undergone previous ligation of esophageal varices. Compared with conventional EGD, the accuracy of small-caliber EGD for esophageal varices grading was 94 % (consensus grade) and 95 % (real-time grade). Excellent concordance was demonstrated between real-time grade and consensus grade, with a kappa of 0.95 for both types of EGD. There was excellent interobserver agreement between endoscopists, regardless of the type of EGD. The severity of hepatic dysfunction and the presence or absence of a history of previous esophageal varices ligation did not have any impact on the accuracy or interobserver agreement.

Small-caliber EGD performed under sedation via oral intubation is a highly accurate and reliable modality for evaluating esophageal varices in patients with end-stage liver disease and/or portal hypertension, regardless of the degree of hepatic dysfunction or history of previous esophageal varices ligation.