Objective To investigate the value of chloride clearance test in differential diagnosis of Gitelman syndrome (GS). Methods For patients with hypokalemic metabolic alkalosis and highly suspected GS,clinical data were documented and SLC12A3 gene screening was performed as gold standard to diagnose GS. Hydrochlorothiazide (HCT) test and furosemide (FUR) test were performed according to the standard process. Baseline and maximal increasement of chloride excretion fraction (FECl,the net and relative increase measured as εFECl) were compared between patients and controls to evaluated the reaction to the corresponding diuretics. Receiver operating characteristic (ROC) curve was used to evaluate the sensitivity and specificity of HCT test in GS diagnosis. Results Totally 27 patients and 20 health controls received HCT test. Among those patients,23 were diagnosed with GS genetically. When using the net and relative εFECl to diagnose GS,the areas under the ROC curve were 0.987 (95% CI:0.963～1.000,P<0.001) and 0.984 (95%CI:0.950～1.000,P<0.001),respectively. When a reasonable cutoff value for εFECl was selected,the sensitivity and specificity were both higher than 95%. Eight patients received both HCT test and FUR test. Five of them showed decreased reaction to HCT(net εFECl≤2.86% or relative εFECl≤223%),while normal reaction to FUR.SLC12A3 mutations confirmed their GS. Three patients with blunt reaction to FUR showed normal reaction to HCT,finally they were diagnosed as BS clinically because no SLC12A3 gene mutation was detected. Conclusion Comprehensive application of HCT test and FUR test to evaluate the diuretic reaction can effectively differentiate GS and BS.