Diagnostic utility of alarm features in predicting malignancy in patients with dyspeptic symptoms

Abstract

Abstract Background Clinical features are of modest benefit in determining the etiology of dyspepsia. Dyspeptic patients with alarm features are suspected to have malignancy; but the proportions of patients and true cutoff values of various quantitative parameters in predicting malignancy are explored to a lesser extent. Methods This is a prospective observational study of consecutive patients undergoing esophagogastroduodenoscopy (EGD) for dyspeptic symptoms. Patients’ alarm features and clinical details were recorded in a predesigned questionnaire. The diagnostic accuracy of alarm features in predicting malignancy was studied. Results Nine hundred patients, 678 (75.3%) males, with a mean (standard deviation [SD]) age of 44.6 (13.54) years were enrolled. Commonest indication for EGD was epigastric pain in 614 (68.2%) patients. Dyspepsia was functional in 311 (34.6%) patients. EGD revealed benign lesions in 340 (37.8%) and malignancy in 50 (5.5%) patients. Among the malignant lesions, gastric malignancy was present in 28 (56%) and esophageal malignancy in 20 (40%) patients. Alarm features were present in 206 (22.9%), out of which malignant lesions were seen in 46 (22.3%) patients. Altogether, the alarm features had a sensitivity of 92% and specificity of 81.2% for predicting malignancy. The sensitivity and specificity for weight loss were 76% and 90.8%, while that of abdominal mass were 10% and 99.9% respectively. Based on receiver operating characteristic curve, the optimal age for screening of malignancy was 46.5 years in this population. Conclusions Patients of age group 40 to 49 years with dyspeptic alarm symptoms (predominant weight loss) need prompt endoscopy to screen for malignancy. The alarm features are inexpensive screening tools, found to be useful in India, and should be utilized in countries with similar healthcare conditions and disease epidemiology.