Background: Organized cervical cytology screening programs are not feasible in many developing countries where cervical carcinoma is an important cause of mortality among adult women. This study compared visual inspection of the cervix after application of 3-4% acetic acid (VIA, or cervicoscopy) with cytology as methods for the detection of cervical carcinoma and its precursors.
Methods: Three thousand women were examined by both VIA and cytology. Those positive on one or both of the screening tests (n = 423) or those who had clinically suspicious lesions even if the tests were negative (n = 215) were invited for colposcopy. Directed biopsies were obtained from 277 of 573 women at colposcopy. Those with moderate dysplasia or worse lesions diagnosed by histology were considered true-positives. Those with no lesions or with reactive or reparative changes at colposcopy and those for whom histology revealed no pathology, reactive or reparative changes, atypia, or mild dysplasia were considered false-positives. The detection rate of true-positive cases and the approximate specificity of the two tests were compared.
Results: VIA was positive in 298 women (9.8%), and cytology was positive (for atypia or worse lesions) in 307 women (10.2%). Of the 51 true-positive cases (20 cases of moderate dysplasia, 7 of severe dysplasia, 12 of carcinoma in situ, and 12 of invasive carcinoma), VIA detected 46 (90.1%) and cytology 44 (86.2%), yielding a sensitivity ratio of 1.05. VIA detected five lesions missed by cytology, and cytology detected three missed by VIA; both missed two lesions. The approximate specificities were 92.2% for VIA and 91.3% for cytology. The positive predictive value of VIA was 17.0%, and that of cytology was 17.2%.
Conclusions: These results indicate that VIA and cytology had very similar performance in detecting moderate dysplasia or more severe lesions in this study. VIA merits further evaluation as a primary screening test in low-resource settings.