A comparison of four screening methods for cervical neoplasia in a developing country

Objective: The purpose of this study was to compare the results of cytologic examination, cervicography, direct acetic acid test, and magnified acetic acid test (speculoscopy) with the results of histologic examination. Study design: The study was a prospective, descriptive, and analytic study that included 1286 women in a primary health facility who were screened by cytologic examination, cervicography, direct acetic acid test, and magnified acetic acid test (speculoscopy). Histologic evidence was obtained on all patients by means of cervical punch biopsies. The biopsy specimens were taken from aceto-white areas or areas suspicious of cancer. In normal individuals, the specimens were taken from the external cervical os at the 12 or 6 o’clock position, whichever was most convenient. Positive cases were treated by colposcopy and large loop excision of the transformation zone. Testing for human immunodeficiency virus infection was not done. Results: Of the 1286 women who were screened, 1093 women (85%) were eligible for analysis. The mean age of the women was 38.6 years. Because histologic evidence was obtained on all women, direct sensitivities and specificities could be calculated for each screening method. The sensitivity, specificity, and positive predictive values for cytologic examination were 53%, 95%, and 47%, respectively; for cervicography were 49%, 88%, and 26%, respectively; for the direct acetic acid test were 79%, 49%, and 12%, respectively; and for speculoscopy were 77%, 45%, and 11%, respectively. Of the 876 patients who were eligible for colposcopy, only 468 women (56%) eventually returned for colposcopy. Conclusion: Cytologic examination, cervicography, the direct acetic acid test, and speculoscopy are not suited for screening in developing countries because of the low sensitivity of the cytologic examination and the low specificity of the other tests. Two or more tests combined will increase the sensitivity, but the specificity remains low. Screening methods in developing countries remain an urgent problem. (Am J Obstet Gynecol 2003;188:395-400.)