Abstract

• The incidence of cervical cancer and related deaths has fallen gradually since 1970 in several European countries. The following article examines the role of screening in this reduced mortality, and the potential harms of cervical screening, based on a review of the literature using the standard Prescrire methodology.
   
• Cervical cancer is mainly due to persistent infection with certain types of human papillomavirus. Progression to invasive cervical cancer is slow and infrequent.
   
• Cervical screening is based on taking a sample of superficial cervical cells for the detection of atypical cells associated with malignant transformation. Women with suspect findings initially undergo colposcopy, followed by more invasive procedures such as biopsy, conisation and surgery, if necessary.
   
• Detection of high-grade lesions is similarly efficient whether the sampled cervical epithelial cells are examined immediately on slides or first suspended in a preservative solution (liquid-based cytology).
   
• The sensitivity of cervical smear testing for detecting high-grade intraepithelial lesions depends on the sample quality and the experience and training of the person who collects the sample and the person who reads the slides. False-positive and false-negative results are frequent.
   
• There are no randomised trials comparing outcomes between women who have regular cervical screening and women who are never screened.
   
• Several case-control studies have shown a strong statistical correlation between screening and a reduced risk of cervical cancer. In particular, a British study showed that the risk of being diagnosed with invasive cancer between the ages of 40 and 65 years was 4 times lower among women who had been screened in the previous 5 years than in women who had not been screened.
   
• Historical studies, which can only provide a low level of evidence, suggest that the decline in cervical cancer mortality observed in several parts of the world is proportional to participation in screening. In England, the decline in cervical cancer accelerated when screening was better organised and participation increased markedly.
   
• Conisation consists of partial surgical excision of the cervix, sometimes for diagnostic purposes after a suspect smear result. It can cause bleeding and has been linked to a risk of preterm delivery in subsequent pregnancies.
   
• Most comparative studies have focused on organised screening programmes. In one case-control study, organised screening was more effective than opportunistic screening in reducing the incidence of cervical cancer.
   
• The results of case-control studies and epidemiological studies suggest that screening is unnecessary before the age of 21, or in the 3 years after first intercourse. Some data suggest that screening should continue until age 70, at least for some women.
   
• Pregnancy is associated with an increased frequency of atypical cells. 
   
• Immunodepression, due to HIV infection for example, increases the risk of cervical cancer, but the optimal screening frequency for these women is not known. 
  

©Prescrire September 2010

"Cervical cancer screening: organised screening to avoid unnecessary conisation" Prescrire Int 2010; 19 (108): 172-179. (Pdf, subscribers only).