Methylmalonic acid measured in plasma and urine by stable-isotope dilution and electrospray tandem mass spectrometry.

BACKGROUND Liquid chromatography-tandem mass spectrometry (LC-MS/MS) with electrospray ionization is robust and allows accurate measurement of both low- and high-molecular weight components of complex mixtures. We developed a LC-MS/MS method for the analysis of methylmalonic acid (MMA), a biochemical marker for inherited disorders of propionate metabolism and acquired vitamin B(12) deficiency. METHODS We added 1 nmol of the internal standard MMA-d(3) to 500 microL of plasma or 100 microL of urine before solid-phase extraction. After elution with 18 mol/L formic acid, the eluate was evaporated, and butyl ester derivatives were prepared with 3 mol/L HCl in n-butanol at 65 degrees C for 15 min. For separation, we used a Supelcosil LC-18, 33 x 4.6 mm column with 60:40 (by volume) acetonitrile:aqueous formic acid (1 g/L) as mobile phase. The transitions m/z 231 to m/z 119 and m/z 234 to m/z 122 were used in the selected reaction monitoring mode for MMA and MMA-d(3,) respectively. The retention time of MMA was 2.2 min in a 3.0-min analysis, without interference of a physiologically more abundant isomer, succinic acid. RESULTS Daily calibrations between 0.25 and 8.33 nmol in 0.5 mL exhibited consistent linearity and reproducibility. At a plasma concentration of 0.12 micromol/L, the signal-to-noise ratio for MMA was 40:1. The regression equation for our previous gas chromatography-mass spectrometry (GC-MS) method (y) and the LC-MS/MS method (x) was: y = 1.030 x -0.032 (S(y|x) = 1.03 micromol/L; n = 106; r = 0.994). Inter- and intraassay CVs were 3. 8-8.5% and 1.3-3.4%, respectively, at mean concentrations of 0.13, 0.25, 0.60, and 2.02 micromol/L. Mean recoveries of MMA added to plasma were 96.9% (0.25 micromol/L), 96.0% (0.60 micromol/L), and 94.8% (2.02 micromol/L). One MS/MS system used only overnight (7.5 h) replaced two GC-MS systems (30 instrument-hours/day) to run 100-150 samples per day, with reductions of total cost (supplies plus equipment), personnel, and instrument time of 59%, 14%, and 75%, respectively. CONCLUSIONS This method is well suited for large-scale MMA testing (> or =100 samples per day) where a shorter analytical time is highly desirable. Reagents are less expensive than the anion-exchange/cyclohexanol-HCl method, and sample preparation of batches up to 100 specimens is completed in less than 8 h and is automated.