Abstract:
Objective:To evaluate the clinical application of four approaches to cervical cancer screening. Methods: A total of 1,137 women of15to 59years old selected from Huaqiaocheng community, Nanshan district of Shenzhen received a population-based cervical cancer screening between November 2004and December 2004. All of the subjects were investigated by four approaches, including visual inspection with acetic acid (VIA), col-poscopy, hybrid captureⅡ(HC-Ⅱ) human papillomavirus (HPV) test and liquid-based cytology test (LCT). Subjects with abnormal or suspicious findings during colposcopy received biopsies. HPV-positive subjects with atypical squamous cells of undetermined sign (ASCUS) based on LCT or HPV-negative patients with low grade squamous intraepithelial lesion (LSIL) based on LCT had biopsies during colposcopy. The pathology results were used as gold standards to evaluate the four approaches for screening for cervical cancer. Results:A total of 122 biopsy specimens were obtained. Pathological examination found no cases of cervical cancer, 3 cases of grade Ⅲcervical intraepithelial neoplasia (CIN), 11cases of grade ⅡCIN, 36cases of grade ⅠCIN, 69cases of chronic cervicitis and metaplasia of squamous epithelium, and3 cases of normal cervix. The HPV-positive rate was 14.0% by HC-II, the LCT-positive rate was 12.6%, the VIA-positive rate was12.5%, and the colposcopy positive rate was13.6%. The HPV-positive rate and LCT positive rate increased as the pathological grade of cervical lesions increased ( P<0.005 ). The sensitivity, specificity, accuracy, positive prevalue, negative prevalue, positive likelihood ratio and negative likelihood ratio were 100 %,87.1%,87.3%,8.8%,100 %,11.3 and 23.5% for HC-II HPV for detecting high grade cervical lesions; 92.9%,88.4%,88.5%,9.1%,99.9%,8.0 and 8.0% for LCT; 35.7%,96.0%,95.3%,10.0%, 99.2%,8.9, and 67.0% for VIA; and 50%, 86.8%, 86.4%, 4.5%, 99.3%,3.8, and 57.6% for colposcopy.Conclusion: HC- Ⅱhigh risk HPV detection and LCT are presently two promising approaches for cervical cancer screening. The sensitivity of VIA and colposcopy is lower for detecting high grade cervical lesions and is therefore not suitable for screening, but their specificity and negative prevalue are higher and readily applicable for clinical testing.