PREDICTION OF ESOPHAGEAL VARICES IN NON-ALCOHOLIC CIRRHOTIC PATIENTS WITH SERUM-ASCITES ALBUMIN GRADIENT: A CROSS-SECTIONAL STUDY

Abstract

Objective: Esophageal varices are a frequent and potentially life threatening consequence of portal hypertension in cirrhotic patients. Although several noninvasive indicators have been investigated, evidence regarding the usefulness of the serum–ascites albumin gradient (SAAG) in non alcoholic cirrhosis remains limited. The aim of our study was to determine the correlation between serum-ascites albumin gradient and esophageal varices in patients with ascites due to non-alcoholic cirrhosis and to identify an optimal cut-off for predicting varices. Methods: This analytic cross-sectional design enrolled 250 individuals having cirrhosis and ascites at Gujranwala Medical College Teaching hospital Gujranwala. Measurement of Serum-ascites albumin and upper GI endoscopy were performed among the patients. Statistical analysis included ROC curve analysis, Youden's index, multivariate logistic regression, and independent t-test. Results: Esophageal varices were present in 223 (89.2%) participants. Mean SAAG was raised in patients with varices than without varies (1.77±0.56 vs. 1.25±0.29 g/dl; p<0.001). An analysis with ROC curve, SAAG >1.6 g/dl predicted varices. Multivariate logistic regression confirmed serum-ascites albumin gradient >1.6 gm/dl as an independent predictor of varices (adjusted OR: 7.82; 95% CI: 2.89-21.14; p<0.001). Conclusion:  A SAAG value above 1.6 g/dl may serve as a useful noninvasive indicator for high risk of esophageal varices in cirrhotic patients who should undergo prompt endoscopic assessment. However, because of its poor negative predictive value, SAAG should not be used to exclude varices or replace endoscopic screening.