S1079 Diagnostic Yield of Esophagram Compared to Gold Standard Esophageal Manometry for Detection of Esophageal Motility Abnormalities

Le résumé fourni par la source

Introduction: High resolution manometry (HRM) is the gold standard to diagnose esophageal motility disorders. Barium esophagram (BE), often used to evaluate dysphagia or other esophageal symptoms, routinely includes esophageal motility assessment. There are limited data evaluating the accuracy of BE to assess esophageal motility. The very few available studies used outdated HRM analysis paradigms and showed discordance between BE and HRM findings, suggesting that the BE may not be adequate for diagnosing esophageal dysmotility. Our aim was to determine whether BE is reliable for esophageal motility assessment, compared to the current gold standard of HRM analyzed by Chicago Classification version 4.0 (CCv4.0). Methods: This is a retrospective analysis of adult patients that underwent BE and HRM within a 12-month interval. Patients with a history of fundoplication were excluded. Esophagram was performed with standard technique, and esophageal motility was reported as normal or abnormal. HRM was performed and analyzed following CCv4.0 protocol and reported as normal or abnormal. Concordance, sensitivity, specificity, and area under the curve (AUC) were calculated for esophagram, with HRM diagnosis as the gold standard. Results: One hundred twenty patients were included, 81 (67%) women, mean age = 64 years, mean body mass index = 27. Normal esophageal motility was found in 60 (50%) patients by HRM and 32 (27%) patients by BE. Concordance between BE and HRM was poor; the concordance correlation coefficient was 0.26, with a 95% confidence interval [0.1, 0.4]. Compared to HRM, BE had 86% sensitivity and 40% specificity for diagnosing esophageal dysmotility; the receiver operating characteristic curve showed an AUC score of 0.6. BE had a positive predictive value of 59%, and a negative predictive value of 75%. Conclusion: Our results showed poor concordance between BE and HRM. BE had reasonable sensitivity but very low specificity to detect esophageal dysmotility, indicating a high rate of false positive results for dysmotility by BE. Furthermore, the AUC value indicates poor discrimination, and thus limited ability to distinguish correctly between patients with and without esophageal motility disorders. Based on these results, BE should not be used to assess esophageal motility.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.