Predictability of gastroesophageal caustic injury from clinical findings

Abstract

The aim of this study was to present our patients with corrosive ingestion retrospectively, to analyze the validity of clinical signs as predictors of outcome, and to emphasize the necessity of esophagogastroduodenoscopy. Data were evaluated from the medical records of patients admitted at the Mother and Child Health Care Institute, Serbia over a 10-year period. A total of 176 children, mean age 36.2 ± 18.1 months (range 9 months to 18 years), with corrosive ingestion were evaluated. The ingested substances were alkali in 96 cases (54.5%), acid in 41 (23.3%), and others in 39 cases (22.1%). In all, 116 patients (65.9%) were symptom free on admission. Positive clinical findings were observed in 60 (34.1%) patients. Upper endoscopy was performed in all children within the first 48 h. Ninety-five patients (54%) had normal endoscopic evaluation, 54 (30.6%) had mild lesions, and 27 (15.3%) had severe corrosive injuries. The validity of clinical findings in predicting the severity of esophageal and gastric injury was as follows: sensitivity - 74 and 75% and specificity - 73 and 68%, retrospectively. Eighteen patients (10.2%) developed esophageal stricture. Endoscopy is a mandatory technique in children with gastroesophageal caustic injuries, and should be performed to prevent unnecessary hospitalization and to plan future treatment. This study emphasizes that clinical signs and symptoms are not predictors of esophageal and gastric injury and that the absence of any clinical findings does not rule out a severe esophageal or gastric injury.