OBJECTIVE: To assess the feasibility and performance of direct visual inspection (DVI) of the cervix as a primary tool for the detection of premalignant lesions of the cervix (HPV/CIN 1 and CIN 2,3). SETTING: The early cancer detection unit at the Ain Shams University Maternity Hospital in Cairo, Egypt. SUBJECTS AND METHODS: The study included 2049 women. Cervical smears were obtained from all women for cytologic evaluation followed by direct visual inspection (DVI) of the cervix after painting with 5% acetic acid. Women whose smear reports showed abnormal cells suggestive of squamous intraepithelial lesion (SIL) or human papillomavirus (HPV) infection or those who showed abnormalities or acetowhite areas on direct visual inspection subsequently were referred for colposcopy and biopsy when appropriate. Colposcopy also was performed for women with negative DVI and negative smears if they had contact bleeding or chronic per vaginal discharge. RESULTS: Mean (SD) age of women included in the study was 39.9 (10.2) years with their mean (SD) parity 2.9 (1.1). Results of DVI were normal in 1916 women (93.4%) and showed abnormal acetowhite appearance in 133 (6.6%) women. There were 458 (22.4%) colposcopic examinations and 130 biopsies (6.34%) were carried out, picking up 83 cases of premalignant lesions (4.0%). Premalignant lesions were 80 HPV/CIN 1 and CIN 2,3. Direct visual inspection detected 71 of the 83 premalignant lesions (sensitivity, 85.5%; specificity, 96.8%; positive predictive value, 52.6%). Direct visual inspection missed one of the three samples showing CIN 2,3. Cervical cytologic analysis showed abnormal cells in 60 (2.9%) and identified only 14 of the premalignant lesions (sensitivity, 16.9%; specificity, 97.8%; positive predictive value, 23.3%). Twelve of the premalignant lesions had positive smear results and a negative DVI, with none of them being of a high grade. CONCLUSIONS: Direct visual inspection is feasible and had superior sensitivity compare with cervical cytologic analysis in detecting premalignant lesions of the cervix. Direct visual inspection can be used as a primary screening tool with a satisfactory low biopsy rate in low-resource settings or where cytologic services are suboptimal.
OBJECTIVE: To assess the feasibility and performance of direct visual inspection (DVI) of the cervix as a primary tool for the detection of premalignant lesions of the cervix (HPV/CIN 1 and CIN 2,3). SETTING: The early cancer detection unit at the Ain Shams University Maternity Hospital in Cairo, Egypt. SUBJECTS AND METHODS: The study included 2049 women. Cervical smears were obtained from all women for cytologic evaluation followed by direct visual inspection (DVI) of the cervix after painting with 5% acetic acid. Women whose smear reports showed abnormal cells suggestive of squamous intraepithelial lesion (SIL) or human papillomavirus (HPV) infection or those who showed abnormalities or acetowhite areas on direct visual inspection subsequently were referred for colposcopy and biopsy when appropriate. Colposcopy also was performed for women with negative DVI and negative smears if they had contact bleeding or chronic per vaginal discharge. RESULTS: Mean (SD) age of women included in the study was 39.9 (10.2) years with their mean (SD) parity 2.9 (1.1). Results of DVI were normal in 1916 women (93.4%) and showed abnormal acetowhite appearance in 133 (6.6%) women. There were 458 (22.4%) colposcopic examinations and 130 biopsies (6.34%) were carried out, picking up 83 cases of premalignant lesions (4.0%). Premalignant lesions were 80 HPV/CIN 1 and CIN 2,3. Direct visual inspection detected 71 of the 83 premalignant lesions (sensitivity, 85.5%; specificity, 96.8%; positive predictive value, 52.6%). Direct visual inspection missed one of the three samples showing CIN 2,3. Cervical cytologic analysis showed abnormal cells in 60 (2.9%) and identified only 14 of the premalignant lesions (sensitivity, 16.9%; specificity, 97.8%; positive predictive value, 23.3%). Twelve of the premalignant lesions had positive smear results and a negative DVI, with none of them being of a high grade. CONCLUSIONS: Direct visual inspection is feasible and had superior sensitivity compare with cervical cytologic analysis in detecting premalignant lesions of the cervix. Direct visual inspection can be used as a primary screening tool with a satisfactory low biopsy rate in low-resource settings or where cytologic services are suboptimal.