Aims: To test the ability of Ki-67 to detect cytological lesions in a screening setting and its use as a surrogate marker of human papillomavirus (HPV) infection. Methods: A study of liquid based cytology, HPV DNA testing by MY09/MY11 consensus polymerase chain reaction (PCR), type specific PCRs, and Ki-67 immunocytochemistry on a randomly selected series of 147 patients. Results: Comparison of the number of Ki-67 immunoreactive cells/1000 cells in the different cytological groups showed that the HSIL group yielded a significantly higher mean count than did the other groups. The number of Ki-67 immunoreactive cells/1000 cells was significantly higher in HPV-16 positive samples than in samples containing infections with other high risk types. Receiver operating characteristic curves indicated a test accuracy (area under curve) of 0.68, 0.72, and 0.86 for atypical squamous cells of undetermined significance (ASCUS), low grade squamous intraepithelial lesions (LSIL), and high grade squamous intraepithelial lesions (HSIL), respectively. Thresholds for 95% sensitivity were 0.07, 0.08, and 0.15 Ki-67 immunopositive cells/1000 cells for ASCUS, LSIL and HSIL, respectively. The threshold for 95% specificity was 1.9 Ki-67 immunopositive cells/1000 cells. Conclusions: Ki-67 immunocytochemistry can be applied to liquid based cytology. The accuracy and diagnostic indices of the test are good when compared with those of other techniques. As part of a panel of screening procedures, it could be used as an adjunct to liquid based cytology to identify HSIL, and as a surrogate marker of HPV-16 infection.T he causal relation between genital human papillomavirus (HPV) infection and cervical dysplasia/carcinoma is well established.1-3 HPV type 16 (HPV-16) is most frequently found in cervical squamous cell carcinoma, with more than 50% of these tumours harbouring this type. 4 However, even though the prevalence of HPV infection can be as high as 60%, only 10% of infected women will develop cervical dysplasia.