Accuracy of Endoscopy in Predicting the Depth of Mucosal Injury Following Caustic Ingestion; a Cross-Sectional Study.

Introduction: Esophagogastroduodenoscopy (EGD) is currently considered as the primary method of determining the degree of mucosal injury following caustic ingestion. The present study aimed to evaluate the screening performance characteristics of EGD in predicting the depth of gastrointestinal mucosal injuries following caustic ingestion.

Methods: Adult patients who were referred to emergency department due to ingestion of corrosive materials, over a 7-year period, were enrolled to this diagnostic accuracy study. Sensitivity, specificity, positive and negative predictive values as well as negative and positive likelihood ratios of EGD in predicting the depth of mucosal injury was calculated using pathologic findings as the gold standard.

Results: 54 cases with the mean age of 35 ± 11.2 years were enrolled (59.25% male). Primary endoscopic results defined 28 (51.85%) cases as second grade and 26 (48.14%) as third grade of mucosal injury. On the other hand, pathologic findings reported 21 (38.88%) patients as first grade, 14 (25.92%) as second, and 19 patients (35.18%) as third grade. Sensitivity and specificity of endoscopy for determining grade II tissue injury were 50.00 (23.04-76.96) and 47.50 (31.51-63.87), respectively. These measures were 100.00 (82.35-100) and 80.00 (63.06-91.56), respectively for grade III. Accuracy of EGD was 87.03% for grade III and 48.14% for grade II.

Conclusion: Based on the findings of the present study, endoscopic grading of caustic related mucosal injury based on the Zargar's classification has good accuracy in predicting grade III (87%) and fail accuracy in grade II injuries (48%). It seems that we should be cautious in planning treatment for these patients solely based on endoscopic results.