Objective To establish a LC-MS/MS system for the determination of Bletilla striata glucomannan in Bletilla striata hemostatic sponge. Methods The Waters X Bridge^TM Amide(3.5 μm, 2.1 mm×100 mm)column was used at 35℃; acetonitrile and 5 mmol/L ammonium acetate (containing 0.1% formic acid) (60:40, V/V) were used as the mobile phase at a flow rate of 0.3 mL/min, with the injection volume being 10 μL. Results The linear range for D-mannose was 50.0-5 000.0 ng/mL (r=0.999 5). The results of intra-day and inter-day precisions, limit of detection and limit of quantitation were all within the normal ranges. The recovery rate (n=5) was 98.4%, RSD=1.03%. Conclusion The method in this study is simple, selective and sensitive; it can be used for the quantitative determination and quality control of Bletilla striata glucomannan.