Comparison of the conventional cervical smear and liquid-based cytology: results of a controlled, prospective study in the Abruzzo Region of Italy.

OBJECTIVE: To compare conventional cervical testing (CCT) and liquid-based cytology (LBC) within a randomized trial performed during 2001-2002 in the Abruzzo Region of Italy, including a cost-outcome comparative analysis. STUDY DESIGN: Study subjects were recruited in the framework of a controlled, randomized study organized in the Abruzzo Region. Women aged 2 6-64 years were randomized to an active arm (LBC) or control arm (CC1). The particip ating laboratories had no previous ex perience with LBC. RESULTS: The inadequacy rate was 4.3% in CCT and 1.3% in the LBC arm (D < 0.001). Atypical squamous cells of undetermined sign ifi cance and atypical glands of undetermined significance reports were more frequent at CCT vs. LBC. A small, insignificant excess of low grade squamous intraepithelial lesions or high grade squamous epithelial lesions+ reports was observed in the LBC arm. The cervical intraepithelial neoplasia 2+ (CIN2+) detection rate was not statistically different in the 2 arms (CCT=0.54%, LBC= 0.66%, p = 0.28). In the overall series positive predictive value was slightly but not significantly higher in the LBC arm. LBC increased costs by 4.2% per both screened women and CIN2+ detected. CONCLUSION: The study reflects the introductory phase of LBC in laboratories without prior LBC experience. In this setting LBC reduced the inadequacy rate and decreased reading and was at least as sensitive as and more specific than CCT. Utilization of LBC in organized screening programs will be based on local feasibility, considering that the high cost of LBC is only partially compensated for by other benefits, such as residual cellular material, available for molecular testing, including human papillomavirus testing.

Extension of organised cervical cancer screening programmes in Italy and their process indicators.

Since 1996, Italian national guidelines recommend that Regions implement organised screening programmes for cervical cancer. As in previous years since 1998, we have collected from Italian organised cervical screening programmes aggregated tables of data in order to centrally compute process indicators. In 2006, the target population of Italian organised screening programmes included 11,362,580 women, corresponding to 69% of Italian women aged 25-64 years. However, taking into account the proportion of women invited, the actual extension was 52.9%. Compliance to invitation was 38.5%, with a clear North-South decreasing trend. It should, however, be considered that many women are screened outside the organised programmes. The remaining process indicators deal with women invited during 2005 and screened up to April 2006 Of these, 6.1% were recommended to repeat cytology and 57% of them complied; 2.3% of screened women were referred to colposcopy. Compliance to colposcopy was 84.7% among women referred because of ASCUS or more severe cytology and 90% among those referred because of HSIL or more severe cytology. The positive predictive value (PPV) of referral because of ASCUS or more severe cytology for CIN2 or more severe histology was 16.8%. There was a relevant variability of both referral rate and PPV that were inversely related. The unadjusted detection rate of histologically confirmed CIN2 or more severe was 2.7 per 1,000 screened women (2.6 standardised on the Italian population, truncated 25-64).