A study was carried out in 135 women of reproductive age to evaluate the role of the Papanicolaou smear performed in cervical and vaginal samples for the diagnosis of bacterial vaginosis (BV), to validate the method for this diagnosis and to evaluate the reproducibility of the criterion used. The cervical and vaginal smears were stained using the Papanicolaou method and evaluated by two separate observers. The finding of 20% or more clue cells was considered positive for the presence of BV. This diagnostic criterion presented a sensitivity of 87%, specificity of 94%, a positive predictive value of 81 %, and a negative predictive value of 96%, both in cervical and vaginal samples. There vvas excellent agreement in the diagnosis of BV between the two observers (kappa, 0.87) and between cervical and vaginal samples (kappa, 0.92). Therefore, the presence of 20% or more clue cells in the Papanicolaou smear is all accurate and reproducible criterion for the diagnosis of BV and may be used ill screening for this infection, eliminating the need for further vaginal sample collection.