[Comparison of conventional PAP smears with thin layer specimen (liquid-based PAP test) and correlation with...

Cervical smears of 554 outpatients of a hospital were examined using a blinded, split sample match pair protocol for which a conventional PAP-smear (CS) was first prepared with Cervex brush and the reminder of the sample was used for the thin-layer-preparation (TLP) according to the manual CytoRich System. The preparations of the two methods were compared with respect to quality and to sensitivity for atypias. In addition the HPV status was determined on the same cell suspension in cases with borderline changes (BLC) and dysplasias including carcinoma using the Hybrid Capture System. The use of TLP reduced the proportion of suboptimal preparations by more than 50% (14.6% vs. 35%) and eliminated the only inadequate preparation registered in CS. The DSP detected more than twice as many dysplasias of all degrees as CS (3.4% vs. 1.4%) and reduced the proportion of BLC to one third (3.2% vs. 9.6%). The percentages of cases positive for high- and intermediate-risk HPV in preparations with BLC, LSIL and HSIL were 17, 62.5% and 100% respectively. The TL-method improves significantly the efficiency of PAP-smears and allows the typing of HPV which is of clinical importance for the management of low grade squamous intraepitelial lesions and borderline changes. The findings speak against the further use of CS for cervical screening.