Telecytology as a Tool To Screen Cervicovaginal Smears
Melody K Harrison, Leslie J Cloar, Nicole A Massoll, Murat Gokden, Ali G Saad, Shree G Sharma. University of Arkansas for Medical Sciences, Little Rock; Columbia University Medical Center, New York, NY
Background: Telepathology is rapidly becoming a part of routine practice. This technology is necessary to provide quality care, particularly for patients with limited access to medical care. The current study evaluates the application of this technology to screen cervicovaginal smears.
Design: 52 consecutive routine cervicovaginal smears were reviewed by a cytotechnologist, on a light microscope. These 52 cases were reviewed by the same cytotechnologist using a telepathology system comprised of a Nikon Eclipse 55i microscope and a Nikon DS-L2 digital camera. The slides were driven from a separate site by a medical student. The areas of interest and time spent on the slides were directed by the cytotechnologist on the computer screen. The results were noted for diagnoses (utilizing Betheseda system), concordance, and time taken for slide review. The study was retrospective, to avoid unnecessary delays in turn-around time.
Results: 47 cases (90.4%) screened via telepathology were concordant with the diagnoses made on the microscope. Of the 5 discrepant cases: 3 cases- appeared to have scant material on computer screen either due to excess blood or atrophic smear. Case 4: cytoplasmic fragments misinterpreted as Trichomonas vaginalis. Case 5: ASCUS diagnosed as AGUS. The average screening time: 5.27 ± 1.05 minutes on microscope and 6.02 ± 1.50 minutes remotely.
Conclusions: The difference in average screening time between the microscope and the computer screen was minimal (45 seconds). 2 of the 5 discordant cases pose potential problems for patient management: ASCUS and AGUS are managed differently; diagnosis of trichomoniasis, besides treatment, has significant social implications. The remaining 3 discordant cases would have required repeat smears but caused no harm in terms of management. Of note, the discordant cases were not reviewed by pathologist, which might have reduced discordance rate.
Our preliminary study (concordance rate 90%) is very encouraging. The reported rate of intra-observer agreement by light microscopy is approximately 78% and 58% for telecytology. In view of such wide variability, our study suggests real potential for implementing the technology for screening programs in remote areas. This study also indicates that the microscope can be operated effectively by a non-pathology trained person and the smears can be read on computer screen without prior training. With training the accuracy of diagnosis would improve. Additional study is needed to assess reproducibility and further assess intra-observer variability.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 232, Tuesday Afternoon