A procedure for the analysis of the oxygen-18/oxygen-16 isotope ratio of ethanol derived from the sugars of orange juice using the preparation steps of the SNIF-NMR method followed by pyrolysis-isotope ratio mass spectrometry is presented. The isotopic fractionation induced by the isotope effects of fermentation and distillation have been investigated, and it is shown that reproducible results can be obtained when appropriate analytical conditions are used. It is also shown that the oxygen isotope distribution in the water and organic matter pools of fruits remains quite stable during the harvest period and is not altered by the precipitation rate within the last few days before the fruits are picked. Due to the robustness of the method and the fact that most of the oxygen-18 enrichment from the initial sugars is still present in the end-product, ethanol appears as a convenient internal reference to circumvent the spatial and temporal variability observed for the oxygen-18/oxygen-16 isotope ratio of water. A very strong correlation is observed between the isotopic deviations of ethanol and water, which is altered in the event of a water addition, even at a low level. Combining the information brought by these two parameters leads to a more efficient authenticity testing tool, which avoids false positive cases and provides a lower detection limit for added water in juices not made from concentrate, whatever the origin of the sample tested.