Abstract: To establish a liquid chromatography-tandem mass spectrometry(LC-MS/MS)for therapeutic drug monitoring(TDM)of serum digoxin. An isotope dilution LC-MS/MS was realized by tert-butyl methyl ether extraction, reversed-phase separation and positive electrospray ionization mass spectrometry detection. The results of external quality assessment(EQA)of digoxin TDM program of the National Center for Clinical Laboratory by LC-MS/MS were analyzed and compared with those of electrochemiluminescence immunoassay. The linear range was 0.05-5.00 ng/mL. The precision was 1.0%-4.3%, and the accuracy was 101.3%-108.9%. The recovery rate was 95.8%-97.5%, and the matrix effect was 95.7%-97.6%. The total run time was 3 min per injection. Good stability was observed on serum undergoing 3 cycles of frozen-thaw, at room temperature for 24 h and pretreating serum placed in autosampler for 36 h. The average bias was -4.09% between LC-MS/MS and the target value of EQA with a correlation formula of Y_LC-MS/MS=1.01X_{target value}-0.09, r=0.99. Electrochemiluminescence immunoassay results were 36.9% higher on average than those of LC-MS/MS. LC-MS/MS is simple and specific for the determination of serum digoxin. The results reported by LC-MS/MS are comparable with the national external quality assessment proficiency test. There is significant difference in digoxin results between LC-MS/MS and electrochemiluminescence immunoassay.

Key words: Digoxin, Liquid chromatography-tandem mass spectrometry, External quality assessment, Electrochemiluminescence immunoassay