Ethylenethiourea (ETU) and propylenethiourea (PTU) are the main degradation products of dithiocarbamates fungicides, which are widely used in agriculture from several years ago. Their determination in water at low concentrations (e.g. sub-ppb levels) is highly problematic due to their polar character and low molecular size. In the present study, two analytical methodologies have been developed and compared for the selective and sensitive determination of ETU and PTU in various types of waters. Both approaches are based on liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) with electrospray ionization, using triple quadrupole analyzer. Whereas the first methodology used an on-line solid-phase extraction (SPE) step in order to reach the adequate sensitivity, the second one avoided sample treatment and was based on direct injection into an ultra high performance liquid chromatography (UHPLC-MS/MS) system, making use of a new-generation instrument in order to reach sub-ppb analyte levels in water. Strong matrix effects (typically leading to signal enhancement) were observed for most of the evaluated waters, especially when applying the on-line SPE method, surely due to the higher amount of sample injected into the system. The use of the own analyte (ETU-d₄)) as isotope-labelled internal standard (ILIS) allowed to compensate these effects and to achieve an accurate ETU quantification at low concentrations. Moreover, three simultaneous transitions, operating in selected reaction monitoring mode, were acquired for both ETU and ETU-d₄. This fact together with the evaluation of their relative intensity ratios assured the reliable identification of the analyte in the water samples. The two optimized methodologies were validated by analysis of six different samples (two drinking water, two groundwater and two surface water), spiked at two levels (0.1 and 1.0 μg/L), and analyzed each in quintuplicate. Satisfactory accuracy and precision, with recoveries ranging from 73 to 104% and RSDs lower than 20%, were obtained for ETU. Limits of detection for ETU were found to be 0.058 μg/L and 0.027 μg/L with direct injection and with the on-line methodology, respectively. No satisfactory recoveries were obtained, in general, for PTU despite using its own deuterium-labelled molecule for matrix effects correction. Notable differences in the chemical behaviour between PTU and PTU-d₆ were observed, which lead to significant variation in their chromatographic retention time and ionization efficiency. Thus, no satisfactory correction of matrix effects could be reached illustrating that the use of deuterated ILIS can be problematic in some particular cases. Despite the poor correction, a semi-quantitative analysis would be feasible for PTU at sub-ppb levels in water. To the best of our knowledge, this is the first article reporting the use of LC-MS/MS for the trace level determination of these problematic analytes in water.