Do we need more than one ThinPrep to obtain adequate cellularity in fine needle aspiration?

Dear Dr. Bedrossian: Here in, we report our study regarding the value of additional or minimal number of ThinPrep slides per case needed to establish a correct diagnosis on cytology specimens. The objective of this study is to explore whether preparing a second ThinPrep (TP) slide that presumably contains additional clusters would improve the diagnostic accuracy of fine needle aspiration (FNA)? To the best of our knowledge there have been no studies to establish the minimal number of TP slides on aspirated materials. Over the last 10 years, TP an automated technique has been used to prepare slides not only for cervical–vaginal samples but also for a wide range of nongynecologic specimens including fine needle aspirates. It has been shown by several studies that TP has improved limitations previously experienced on conventional smears such as obscuring material (blood, exudate), difference in smear thickness, and air drying artifacts. In July 1995, the TP method was implemented at the University of Michigan for nongynecologic specimens. Only one TP slide has been prepared for each aspirate in our laboratory. While it is known that TP provides a random representative preparation of the obtained sample, it has not been established whether additional preps in low cellular specimens would provide additional information to establish a conclusive diagnosis. The literature suggests that in the Pap test, an additional TP slide carries the chance to detect a potential abnormality. The utilization of liquid based preparation (LBP) for FNA provides several advantages for both the laboratory and the clinicians performing those FNAs. In addition to the known advantages such as simpler collection, better preservation and elimination of obscuring elements, the concept is particularly attractive to large laboratory corporations with multiple centers throughout the nation as well as small private laboratories. The ability to reliably transport a specimen and to prepare a well preserved slide without investing on onsite assistance or technical support is an attractive principle to these institutions. Moreover, LBP shifts the control over the specimen to the laboratory. In FNAs, such as thyroid and breast, cellularity and number of clusters are crucial to consider the specimen adequate. However, it has been our observation that even in low cellular sample, not all clusters are represented in one TP. Whether additional TP, in suboptimal samples would provide additional cells/clusters and assist in establishing a diagnosis is not well established. In a study by Frost et al., they concluded that 1.43 slides were adequate. In the majority of our FNA biopsies, the conventional smears are prepared on site by pathologists or cytotechnologists. In theses cases, the needle is rinsed in CytoLyt solution after making the smears, and at least one pass is dedicated to make a TP slide. In contrast, when the clinicians perform the FNA in their clinic, they are instructed to rinse the needle directly in CytoLyt and send it to the laboratory. In this setting, a TP and a cell block are prepared. The specimen is transferred to 50 ml tube and centrifuged at 1500 rpm. Two to three drops of the cell pellet are transferred to PreservCyt and one TP slide is prepared by the Cytyc 2000 according to the manufacture’s brochure. A cell block is prepared by the histogel method from the remaining cell pellet. In this study, a total of 100 consecutive FNAs were processed with 2 TP slides prepared for each case. FNAs were obtained from the following sites: thyroid (40 cases), breast (17 cases), salivary gland (10 cases), skin lesions (10 cases), head and neck lesions (eight cases), lymph node (eight cases), gastrointestinal tact (seven cases), and ovary (two cases). All cases were included after the final diagnosis was reported *Correspondence to: Farnaz Hasteh, M.D., 200 West Arbor Dr, Rm 2-120, San Diego, CA 92103-8720. E-mail: [email protected] Received 2 March 2007; Accepted 22 August 2007 DOI 10.1002/dc.20735 Published online in Wiley InterScience (www.interscience.wiley.com).