Jeong Eun Song, Yeonjung Ha, Young Chang, Jeong-Ju Yoo
Ascites is a common complication of advanced liver disease, and accurate etiologic differentiation remains essential for clinical management. The serum-ascites albumin gradient (SAAG) is widely used to distinguish portal hypertensive from non-portal hypertensive ascites; however, its diagnostic performance across different clinical contexts has not been consistently established. We conducted a systematic review and meta-analysis of 47 studies involving 5,070 patients to evaluate the diagnostic accuracy of SAAG. Using a bivariate random-effects model, the pooled sensitivity and specificity were 0.889 (95% confidence interval [CI], 0.847 to 0.920) and 0.818 (95% CI, 0.752-0.870), respectively, with an area under the hierarchical summary receiver operating characteristic (HSROC) curve of 0.881. In subgroup analyses, SAAG showed excellent performance in identifying portal hypertension (area under the HSROC curve=0.936) and high specificity for malignant ascites, whereas its utility for predicting variceal bleeding was limited. Despite substantial heterogeneity and evidence of publication bias, the overall findings were consistent. These results support the continued use of SAAG as a practical and reliable first-line diagnostic tool in patients with ascites, particularly for confirming portal hypertension and excluding malignancy, while highlighting its limitations in other clinical settings.