Biological monitoring of early genotoxic effects in urothelial cells using the urinary micronucleus (MNu) assay is promising for early detection of cancer, such as bladder carcinoma. But many problems are encountered, the major being the poorly differential staining of cells, particularly in women having an important amount of squamous cells. We have optimized the protocol and obtained a differential staining of the cell types present in urine on 10 subjects. Following Carnoy I fixation and Papanicolaou staining, urothelial cells were blue while most squamous cells were pink. This differential staining allowed for optimization of the MNu assay on a single urine void, for both females and males. Even if our MNu means were comparable to the literature, the great variation in reported MNu results could reside in the ability of scorers to distinguish correctly between urothelial and squamous cells. When monitoring exposed populations, this erroneous distinction could largely influence the results, even more in women's urine samples. Given a situation where exposure would not increase micronuclei frequency in vaginal squamous cells, their erroneous analysis in the MNu assay could mask an early genotoxic effect. Therefore, as transitional cell carcinoma of the bladder originates from transformed urothelial cells, restricting micronuclei analysis to urothelial cells could yield a more precise estimate of cancer risk in exposed populations. Moreover, it is hoped that the improvements proposed in this paper will allow for an easier implementation of the MNu assay in various set-ups and enhance its specificity, since MNu are considered a suitable biomarker.