Cervical screening by visual inspection, HPV testing, liquid-based and conventional cytology in Amazonian Peru.
Almonte, Maribel;
Ferreccio, Catterina;
Winkler, Jennifer L;
Cuzick, Jack;
Tsu, Vivien;
Robles, Sylvia;
Takahashi, Rina;
Sasieni, Peter;
(2007)
Cervical screening by visual inspection, HPV testing, liquid-based and conventional cytology in Amazonian Peru.
International journal of cancer Journal international du cancer, 121 (4).
pp. 796-802.
ISSN 0020-7136
DOI: https://doi.org/10.1002/ijc.22757
Permanent Identifier
Use this Digital Object Identifier when citing or linking to this resource.
Cervical cancer is an important public health problem in many developing countries, where cytology screening has been ineffective. We compared four tests to identify the most appropriate for screening in countries with limited resources. Nineteen midwives screened 5,435 women with visual inspection (VIA) and collected cervical samples for HPV testing, liquid-based cytology (LBC) and conventional cytology (CC). If VIA was positive, a doctor performed magnified VIA. CC was read locally, LBC was read in Lima and HPV testing was done in London. Women with a positive screening test were offered colposcopy or cryotherapy (with biopsy). Inadequacy rates were 5% and 11% for LBC and CC respectively, and less than 0.1% for VIA and HPV. One thousand eight hundred eighty-one women (84% of 2,236) accepted colposcopy/cryotherapy: 79 had carcinoma in situ or cancer (CIS+), 27 had severe- and 42 moderate-dysplasia on histology. We estimated a further 6.5 cases of CIS+ in women without a biopsy. Sensitivity for CIS+ (specificity for less than moderate dysplasia) was 41.2% (76.7%) for VIA, 95.8% (89.3%) for HPV, 80.3% (83.7%) for LBC, and 42.5% (98.7%) for CC. Sensitivities for moderate dysplasia or worse were better for VIA (54.9%) and less favourable for HPV and cytology. In this setting, VIA and CC missed the majority of high-grade disease. Overall, HPV testing performed best. VIA gives immediate results, but will require investment in regular training and supervision. Further work is needed to determine whether screened-positive women should all be treated or triaged with a more specific test.