Early detection and treatment of preinvasive neoplasias decrease the incidence and mortality of the subsequent invasive cancers. This paper presents the results of a selective program to detect vulvar intraepithelial neoplasia (VIN). The program was selective because only "relative high risk" women were included, i.e. women with one or more of the following items: a) age more than 50 years; b) past history of epidermoid cervical or vaginal cancer (included intraepithelial stages); c) past history of genital radiation; d) past or actual history of genital condyloma; and e) past or actual history of hyperplasic or mixed vulvar dystrophy. Detection was made with the test described by Collins et al., staining the vulva with a toluidine blue aqueous solution and decoloring it with acetic acid. All positive sites (areas retaining the blue color) were biopsied under local anesthesia. Histopathology diagnosis served as gold standard for the program's evaluation. Patient with negative tests and those with NIV I were rescreened each 6 months. From March 1984 to September 1986, 212 patients were admitted in this program and 318 tests were performed. Individual tests varied from 1 (105 patients) to 5 (3 patients). The group was followed-up until March 1989, when the program was evaluated. There were 77 positive tests, among them 21 cases of NIV. Three women with NIV I progressed to NIV II during the observation period. NIV cases were classified as: NIV I, 7 cases (33.3%); NIV II, 10 cases (47.7%); and NIV III, 4 cases (19.0%).(ABSTRACT TRUNCATED AT 250 WORDS)