The applicability of capillary electrophoresis (CE) with UV and mass spectrometric (MS) detection for the determination of dopamine and methoxycatecholamines in urine was evaluated in comparison with the liquid chromatography-electrochemical detection (LC-EC) method widely used in catecholamine analysis. The catecholamines in urine were deconjugated with acid or enzyme hydrolysis, purified by cation exchange (CEX) or solid-phase extraction (SPE) with a copolymer of N-divinylpyrrolidone and divinylbenzene and analyzed by LC-EC, CE-UV, and CE-MS. Acid hydrolysis was more effective in the deconjugation than enzymatic hydrolysis with Helix pomatia. However, the recoveries of HMBA, DA and NMN from spiked samples were less than 30% after acid hydrolysis and SPE purification. The CEX purification was more efficient than SPE in removing matrix compounds from the urine samples. The limits of detection were lower in LC-EC analysis than in CE-UV or CE-MS. Many factors in the analytical procedure caused deviations in the concentrations measured for urinary dopamine and methoxycatecholamines. The recovery of HMBA, which was used as the internal standard, was poor after acid hydrolysis and SPE purification. The purification methods were validated in conjunction with the analytical methods and therefore cross analysis was unsuccessful. The LC-EC method was the most sensitive, but CE-UV and CE-MS were sensitive enough for the determination of dopamine and methoxycatecholamines even in healthy patient urine. The EC and MS detections were superior to the UV detection in specificity since, after acid hydrolysis, some matrix compounds were migrating close to I.S., DA and 3MT.