Rapid Responses to:

RESEARCH: Elizabeth Davey, Jefferson d'Assuncao, Les Irwig, Petra Macaskill, Siew F Chan, Adele Richards, and Annabelle Farnsworth Accuracy of reading liquid based cytology slides using the ThinPrep Imager compared with conventional cytology: prospective study BMJ 2007; 0: bmj.39219.645475.55v1 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Man or Machine?

Jennifer M Roberts, Julia K. Thurloe, Ron C. Bowditch, Suzanne G. Hyne, Merle Greenberg, Joanne M. Clarke, Clare Biro.   (20 July 2007)

Man or Machine? 20 July 2007
Jennifer M Roberts, Gynaecological Pathologist Symbion Laverty Pathology, 60 Waterloo Rd, Nth Ryde, 2113, NSW, Australia., Julia K. Thurloe, Ron C. Bowditch, Suzanne G. Hyne, Merle Greenberg, Joanne M. Clarke, Clare Biro. Send response to journal: Re: Man or Machine?	We read with interest the findings of Davey et al in their 2-armed trial, comparing ThinPrep (TP) Pap test slides read with the assistance of the ThinPrep Imaging System (TIS), with paired manually-read conventional Pap smears (PS) 1. In making this comparison, any effects due to Imager- assisted reading are confounded with those due to the TP slide-making process. One of their major findings was that the TIS arm had fewer unsatisfactory reports. The authors assert (based on their meta-analysis 2, which is highly controversial 3), that there is no good evidence that liquid-based cytology read manually results in a reduction in unsatisfactory specimens. Thus they attribute the reduction in unsatisfactory reports in the current study to reading by TIS. However, the method of reading the slide can have no bearing on whether the slide itself constitutes an unsatisfactory specimen. Any reduction in unsatisfactory specimens must be due to the method of preparation, that is TP, rather than TIS reading. The other major finding was the detection of 1.29 more cases of histologic high-grade squamous disease per 1000 women screened. Again the authors attribute this improvement to the imaging process rather than to the TP slide making process. In our 3-armed trial, comparing manually read TP, Imager-assisted TP and conventional PS 4, we found no significant difference between Imager- read TP and manually read TP for either the unsatisfactory rate or for detection of high-grade abnormality. Both TP arms showed significant improvement over the conventional PS for both outcomes. The reason for these improvements can therefore be attributed to the TP slide-making process rather than to reading the slide using the TP Imaging system. However, Imager-assisted reading of TP slides significantly enhanced productivity. It would be unfortunate if the study by Davies et al led people to believe that the benefits of TP technology are attributable only to Imager assisted reading. References 1.Davey E, Irwig L, Macaskill P, D'Assuncoa J, Richards A, Farnsworth A. Accuracy of reading liquid based cytology slides using the ThinPrep Imager compared with conventional cytology: prospective study. BMJ 2007:335:31-5. 2.Davey E, Barratt A, Irwig L et al. Effect of study design and quality on unsatisfactory rates, cytology classifications and accuracy in liquid-based versus conventional cervical cytology: A systematic review. Lancet 2006;367:122-132. 3.Leiman G. Liquid-Based Cytology: Under scrutiny down- under.Diagnostic Cytopathology 2007;35:379-80. 4.Roberts JM, Thurloe JK, Bowditch RC, Hyne SG, Greenberg M, Clarke JM, Biro C. A Three-armed trial of the ThinPrep Imaging System. Diagnostic Cytopathology 2007;35:96-102. Competing interests: None declared