Fabry disease(FD) is a rare X-linked recessive hereditary lysosomal storage disease. One of the most commonly used biomarkers for detecting FD is Lyso-GB3, a globotriaosylsphingosine, which is usually analyzed by liquid chromatography-tandem mass spectrometry(LC-MS/MS), because of its high specificity and sensitivity. Due to the complexity of LC-MS/MS technology, efficient and reliable diagnosis of FD by LC-MS/MS was unavailable in most hospitals in China. Therefore, patients’ samples need to be prepared at local hospitals and then sent to central laboratories for LC-MS/MS analysis. As a result, samples undergo a lengthy shipment and storage process, potentially compromising sample quality and thus precision of diagnosis. We first established and validated a novel LC-MS/MS-based method and to guarantee FD sample quality, we studied the effects of pre-analytical variables on the analysis of Lyso-GB3. These pre-analytical variables include storage temperature and time of plasma, freeze-thaw cycles of plasma, degree of hemolysis of plasma, storage temperature and time of whole blood, and centrifugation temperature. Finally, we analyzed 86 FD samples and 100 normal human samples and made correlation studies on the relationship between Lyso-GB3 level and FD classification. Results showed Lyso-GB3 level started decreasing post 4 days when plasma was stored at 20°C. In addition, hemolysis significantly influenced the FD sample quality and severe hemolysis caused the Lyso-GB3 level decreased by 57.8%. Other pre-analytical variables have little impact on FD sample quality. Under the conditions of the standard operational procedures before analysis, the sensitivity and specificity for screening FD is 100%, which ensures the accuracy of the method. Furthermore, we used plasma Lyso-GB3 level to distinguish classical FD patients from non-classical FD patients. When we set the cut-off value as 55.15 μg/L, the sensitivity and the specificity were 71% and 100%, respectively. The area under the receiver operating characteristic curve was 0.83. Our findings can help standardize the operational procedures in medical labs to guarantee high FD sample quality and provide a reference for identification of Chinese FD classification.