Primary human papilloma virus (HPV) screening is more effective than cytology in reducing the risk of cervical cancer, but screening intervals should be extended in HPV-negative women. However, some Markov models predicted that long intervals are associated with an excess risk of cervical cancer. The aim of this analysis was to estimate the real-life risks and benefits of annual Papanicolaou (Pap) screening in HPV-negative women with normal cytology.

Women with negative Hybrid Capture 2 (HC2) results and normal cytology at the time of inclusion in the Hannover HPV screening trial underwent annual Pap smears for 5 years. A subgroup was randomly selected for retesting with cytology, HC2, and colposcopy 60--68 months after recruitment.

Of 4236 women included, 3406 had at least one Pap smear, but only 1185 attended all five annual screening visits. The proportion of women with at least one abnormal smear was 14.4% in 60 months. The probability of abnormal smears increased continuously over time. No case of >= CIN2+ was observed during 5 years. Of 605 women selected for subgroup analysis, 292 agreed to be retested (48.3%). The rate of high-risk HPV at 60--68 months was 3.0% (9/296).

The long-term risk of high-grade neoplasia after an initial negative HC2 test and normal cytology result was low, while the rate of false-positive abnormal Pap smears was significant and increased constantly over time. Pap smear screening of HPV-negative women more frequently than every 5 years could be potentially harmful and seems to be of little clinical value.

Trial registration: Sub-analysis of European Union Biomed 5 contract HPV-based cervical cancer screening (QLG4-CT2000-01238)