A Shandon PapSpin liquid-based gynecological test: A split-sample and direct-to-vial test with histology follow-up study

BACKGROUND: Studies for liquid-based Papanicolaou (Pap) tests reveal that liquid-based cytology (LBC) is a safe and effective alternative to the conventional Pap smear. Although there is research on ThinPrep and SurePath systems, information is lacking to evaluate the efficiency and effectiveness of systems based on cytocentrifugation. This study is designed to determine the sensitivity and specificity of the Shandon PapSpin (ThermoShandon, Pittsburgh, Pennsylvania, USA) liquid-based gynecological system. We used split-sample and direct-to-vial study design. MATERIALS AND METHODS: 2,945 women referred to prophylactic check-up were enrolled in this study. Split sample design was used in 1,500 women and residual cervical cytology specimen from all these cases was placed in fluid for PapSpin preparation after performing conventional smear. The direct-to-vial study was carried out in another cohort of 1,445 women in whom the entire cervical material was investigated using only the PapSpin technique. Follow up histological diagnoses for 141 women were obtained from both study arms following 189 abnormal cytology cases. 80 LBC cases from the split sample group and 61 LBC cases in the direct-to-vial group were correlated with the histology results. The sensitivity and secificity of the conventional smear and PapSpin tests in both study arms were compared. RESULTS: In the split sample group, conventional smears showed a higher proportion of ASC-US (atypical cells undetermined significance): 31 (2.1%) vs 10 (0.7%) in PapSpin (P = 0.001). A higher proportion of unsatisfactory samples was found in the conventional smear group: 25 (1.7%) vs 6 (0.4%) cases (P = 0.001). In the split sample group, the sensitivity of the conventional and PapSpin tests was 68.7% vs 78.1%, and the specificity 93.8% vs 91.8%, respectively. In the direct to vial group PapSpin sensitivity was 75.9% and specificity 96.5%. The differences in sensitivity and specificity were not significant. The positive predictive values for the conventional and PapSpin methods were not different in the split sample group: 88.0% vs 86.2% and 95.7% in the direct-to-vial group. Also, no differences were found for negative predictive value (82.1, 86.8% and 80.0% respectively). CONCLUSIONS: PapSpin showed good qualitative results in both study arms, even after the material splitting in the first study arm, and is a good alternative to the conventional Pap smear. Additionally, the PapSpin method offers several advantages such as the opportunity to prepare duplicate slides, option for HPV DNA testing and cell block preparations from residual material. Microscopic evaluation of thinner cell preparations is less time consuming than the conventional Pap smears.