Is self-irrigation smear effective for the detection of intraepithelial lesions of the uterine cervix?

To evaluate the accuracy of cervical smears obtained by the self-irrigation technique, 25 cases of intraepithelial lesions of the cervix were studied preoperatively using both self-irrigation and the ordinary sampling method. Four self-irrigation specimens were judged “unsatisfactory” because of very scant cellularity. All of the self-irrigation smears mainly consisted of superficial and intermediate squamous cells. Parabasal cells were inconspicuous even in postmenopausal women. The number of endocervical glandular cells and squamous metaplastic cells was also very low in self-irrigation smears. No dyskaryotic cells were observed in 18 self-irrigation specimens. A small number of dyskaryotic cells were noted only in 7 specimens, and most showed such degenerative changes as condensed or inconspicuous chromatin, thick cytoplasm, and cytoplasmic vacuoles. It is suspected that such glandular, metaplastic, and dyskaryotic cells observed in the self-irrigation smears were not scraped directly but had exfoliated spontaneously and that the majority of the squamous cells in the self-irrigation smears originated the vaginal epithelium. Since self irrigation smears are not “cervical” but “vaginal” smears, naturally they seem not to be an effective method for detecting intraepithelial lesions of the cervix.