Comparison of BD Surepath and ThinPrep Pap Systems in the Processing of Mucus-Rich Specimens

Transcription

1 Original Article Comparison of BD Surepath and ThinPrep Pap Systems in the Processing of Mucus-Rich Specimens Shelly Kenyon, MS, CT 1 ; Brenda J. Sweeney, MS, SCT 1 ; James Happel, BS, DLM, HTL 1 ; Gloria E. Marchilli, MS, SCT 2 ; Barbara Weinstein, MD 2 ; and Douglas Schneider, MD 1 BACKGROUND: Excessive mucus, as well as blood and inflammation, can be problematic in the processing and screening of liquid-based cervical Pap preparations by interfering in the process of cell retrieval onto specimen filters or slides. This study compares the capacity of the BD SurePath and ThinPrep liquid-based Papanicolaou (Pap) tests to handle mucus-laden specimens. METHODS: A 100 ll volume of pooled cervical epithelial cells was added to BD SurePath and ThinPrep liquid Pap test vials. Aliquots of cervical mucus that had been collected and pooled from previously processed mucus-rich Pap specimens were added to each series of test vials in amounts of 250 ll, 500 ll, 1000 ll, and 2000 ll. The vials were then routinely processed on their respective instruments and the test slides evaluated for cellularity of the squamous cell component. RESULTS: The BD SurePath test specimens showed no reduction of the squamous cell component resulting from the addition of any of the aliquots of mucus. The ThinPrep test specimens showed a marked loss of cellularity upon the addition of the first aliquot of mucus, with large areas of the filters showing nearly complete absence of squamous cells. CONCLUSIONS: Excessive mucus remains a potentially limiting factor in the ThinPrep Pap system. This appears to be the result of direct obstruction of the filtration membrane causing a markedly diminished retrieval of squamous epithelial cells and potentially impacting disease detection. The BD SurePath system, conversely, manages mucus by a cell enrichment process, which results in undiminished cell recovery as observed in this study s test conditions. Cancer (Cancer Cytopathol) 2010;118: VC 2010 American Cancer Society. KEY WORDS: Pap test, liquid-based cytology, BD SurePath, ThinPrep, Papanicolaou smear, mucus, squamous intraepithelial lesion (SIL). The Papanicolaou (Pap) test is subject to limitation by obscuring factors that include blood, inflammatory cells, and debris. 1,2 Theoretically, this situation applies to liquid-based Pap technologies as well as the conventional Pap smear. Excessive blood and inflammation are known to interfere with specimen adequacy in some liquid-based Pap testing. 3-7 The 2 commonly used liquid-based cytology technologies, ThinPrep (Hologic, Marlborough, Mass) and BD SurePath (BD Diagnostics TriPath, Burlington, NC), differ in their methods of producing thin-layer Pap slides. 8,9 Some authors, based on their experience with Corresponding author: Douglas Schneider, MD, Department of Pathology, St. Elizabeth s Medical Center, MMR 3, 736 Cambridge Street, Boston, MA 02135; Fax: (617) ; Douglas.Schneider@caritaschristi.org 1 Department of Pathology, St. Elizabeth s Medical Center, Boston, Massachusetts; Massachusetts Received: May 21, 2010; Revised: June 16, 2010; Accepted: June 17, 2010 Published online August 31, 2010 in Wiley Online Library (wileyonlinelibrary.com) DOI: /cncy.20096, wileyonlinelibrary.com 2 Department of Pathology, Tufts Medical Center, Boston, 244 Cancer Cytopathology October 25, 2010

2 Excessive Mucus in Liquid-Based Cytology/Kenyon et al these technologies, have raised questions regarding the relative capacities of ThinPrep and BD SurePath for handling potentially limiting factors, particularly with regard to specimen adequacy. 10,11,12 We have previously shown marked differences in the ability of these 2 technologies to effectively deal with the presence of excessive blood. 13 This study recreates test conditions of excessive mucus and examines the ability of the Thin- Prep and BD SurePath methods to process specimens containing various amounts of cervical mucus. MATERIALS AND METHODS Human cervical epithelial cells were collected by combining the cellular residues from 15 BD SurePath vials after routine Pap test processing was completed. The cells were washed twice in deionized water, and the resulting cell button resuspended in 6 ml of normal saline. This procedure yielded sufficient squamous cells to produce richly cellular SurePath and ThinPrep Pap test slides. The resuspended cells were added to 5 BD SurePath and 5 Thin- Prep liquid Pap test vials in 100 ll aliquots. This procedure was repeated for each of 10 test runs. A collection of pooled cervical mucus was prepared by harvesting mucus from the top of the Prep Stain Density Reagent material in mucus-rich Pap specimens found during the routine BD SurePath cell-enrichment process. The pooled mucus was stable at room temperature for several days, allowing time to collect sufficient amounts for testing. To simulate clinical conditions of excessive mucus, aliquots of the pooled mucus preparation were added to test vials containing the previously dispensed squamous cells in amounts of 250 ll, 500 ll, 1000 ll, and 2000 ll. These volumes were chosen to range from roughly physiological amounts encountered during the BD Surepath cell-enrichment process in mucus-rich specimens to an amount representing an attempt to overwhelm the BD SurePath system with mucus. All test vials were routinely processed on their respective BD SurePath and ThinPrep instruments. For the purposes of this study, coated slides prepared in the laboratory were used for the SurePath tests, rather than the commercially available SurePath Pre- Coat slides. The Pap slides were evaluated for adequacy of cellularity by the first author, who used recommendations provided in The Bethesda System for Reporting Cervical Cytology. 1 The number of squamous cells in high-power fields per slide were counted manually, reading across the horizontal diameter of the test area to include the center, and rendering an average cell count per high power field. The average cell counts resulting from the 2 Pap test systems in 10 test runs were compared. Because of the differing diameters of the BD SurePath and ThinPrep test areas,13 mm and 20 mm, respectively, the Bethesda System Guidelines for Estimating Cellularity of Liquid-based Preparations 1 were used to ensure that the cellularity values presented for BD SurePath and ThinPrep specimens were directly comparable. The Student t test was used in assessing differences in average cellularity between specimens. Endocervical cells were not evaluated because the focus of this study was to determine the effects of mucus on specimen adequacy as determined by the cellularity of the squamous component. RESULTS Representative cytology slide images resulting from the addition of various amounts of mucus are shown for BD SurePath and ThinPrep specimens in Figure 1. The numerical results of the average cell counts from the 10 individual test runs are given in Table 1, with the data presented in graphical form in Figure 2. The BD SurePath test specimens showed no diminishment of cellularity from the addition of any of the aliquots of mucus. Abundant cellularity was observed even in the specimens with the maximum amount of added mucus (2000 ll). In test runs number 3 and number 4, the initial SurePath slides (without added mucus) showed significantly lower cellularity than any of the other initial SurePath test slides. This appeared to be the result of a degree of deficiency in the in-house-prepared slide coating in these 2 instances. Interestingly, most of the BD SurePath specimens showed increases in cellularity with added mucus, presumably because of the release of squamous cells from the mucus (P ¼.01 for mucus aliquots of 1000 ll and 2000 ll). Unexpectedly, we were unable to overwhelm the BD SurePath system with the maximum amount of added mucus. The ThinPrep test specimens invariably showed a loss of cellularity upon the addition of mucus. The addition of the first aliquot of mucus (250 ll) resulted in a 61% average decrease in cellularity (P <.0001). The largest aliquot of mucus (2000 ll) produced Cancer Cytopathology October 25,

3 Original Article FIGURE 1. Gross and microscopic views of BD SurePath (SP) and ThinPrep (TP) slides (left and right columns, respectively) were prepared from pooled cervical epithelial cells with increasing amounts of added mucus. (A) SP slide with no added mucus (10); (B) TP slide with no added mucus (10) showing similar cellularity to slide A; (C) SP slide with 250 ll added mucus (40) showing abundant cellularity; (D) TP slide with 250 ll added mucus (40) showing obscured morphology from entrapment of cells in mucus; (E) SP slide (10) with 500 ll added mucus showing abundant cellularity; (F) TP slide (10) with 500 ll added mucus showing large areas with nearly complete absence of epithelial cells; (G) SP slide (40) with 1000 ll added mucus showing abundant cellularity; (H) TP slide (40) with 1000 ll added mucus showing reduced cellularity and entrapment of cells in mucus; (I) SP slide (10) with 2000 ll added mucus with an ample number of cells; (J) TP slide (10) with 2000 ll added mucus showing entrapment of cells in mucus and large areas with nearly complete absence of cells. 246 Cancer Cytopathology October 25, 2010

4 Excessive Mucus in Liquid-Based Cytology/Kenyon et al Table 1. Comparison of the Average Cell Count per High Power Field (HPF) Adjusted for Test Area Diameter Set Mucus Cells per HPF Added (ll) TP SP an average decrease in cellularity of 87%. ThinPrep slides with added mucus often showed large areas of the filters with nearly complete absence of epithelial cells. Other areas showed compromised morphology from entrapment of cells in mucus (Fig. 1). FIGURE 2. Comparison of SurePath and ThinPrep average cell counts per High Power Field (HPF) show increasing amounts of added mucus (adjusted for test area diameter). The mean and standard deviation of the average cell count per high power field for ThinPrep and SurePath cytology specimens with the addition of 0 ll, 250 ll, 500 ll, 1000 ll, and 2000 ll are plotted. The ThinPrep values were adjusted according to the Bethesda Guidelines for Estimating Cellularity. 4 The data are a compilation of 10 separate experimental runs. DISCUSSION Decreased unsatisfactory rates for the ThinPrep Pap test, as compared with conventional smears, have been reported Conversely, our laboratory experience has shown problems in the uniformity of cell deposition on ThinPrep slides in some conditions, suggesting regional clogging of ThinPrep filter pores with the potential for impacting the detection rate of cervical lesions. Published studies have also indicated the possibility of plugging of the ThinPrep filter membrane resulting in under-represented or unsatisfactory Pap tests. 3,4,18,19 Factors involved in the obstruction of the membrane include blood, inflammation, mucus, cellular debris, and tumor diathesis. Specific to this study, mucus can result in ThinPrep slides with holes or areas devoid of cells. 4 One study looking at reprocessing of unsatisfactory ThinPrep Pap tests found that of the unsatisfactory Pap tests remaining after reprocessing, 20.6% were caused by mucus. 3 Another study reported 37.9% of unsatisfactory ThinPrep Pap tests to be due to mucus before reprocessing procedures were applied. 20 The issue of decreased cell retrieval or uneven cell distribution in ThinPrep Pap tests with excessive mucus, as well as blood and inflammation, was essentially eliminated in our laboratory upon switching to the BD Cancer Cytopathology October 25,

5 Original Article SurePath liquid-based technology. As a result we became curious to see to what extent these 2 technologies differ in their ability to handle mucus-rich specimens. This small study examines this issue using recreated mucus-rich specimens. The results show that specimens processed in the BD SurePath system had effectively no diminution of cellularity with any amount of added mucus. Even when mucus appeared on the slides after amounts were added sufficient to exceed the capacity of the cell enrichment process, there was no compromise of cellular adequacy and no obscuring of cellular morphology. Indeed, cellularity increased in some specimens, presumably from the release of cells from mucus. This finding raises questions about the practice of removing gross mucus from the specimen container before processing, suggesting the possible loss of entrapped cells. In contrast, the ThinPrep specimens invariably showed a loss of cellularity upon the addition of the first aliquot of mucus. The ThinPrep slides exhibited large areas devoid of cells consistent with filter membrane plugging. When mucus was transferred to the slide, those cells trapped in mucus were morphologically compromised. The results are understandable given the differences in technology between the BD SurePath and ThinPrep Pap test systems as posted on company internet sites. 21,22 BD SurePath uses no filters and employs a proprietary cell enrichment process (that) separates and reduces obscuring debris (eg, blood, mucous) and inflammatory cells. The obscuring debris is trapped in a gradient density material that is removed, after which a second centrifugation produces an enriched cell pellet. The cells settle onto coated slides called PreCoat slides. Conversely, the ThinPrep 2000 uses a computerized process and patented membrane technology that controls dispersion, collection, and transfer of diagnostic cells from the sample to the slide. After rotation within the sample vial to separate debris and disperse mucus, a gentle vacuum collects cells on the exterior surface of the Filter membrane. The cells are then transferred by gently pressing the membrane against the microscope slide. It is during the vacuum collection process that the ThinPrep technology appears vulnerable to excess mucus, inflammation, or blood, which can compete with diagnostic cells in the collection process and slow the flow rate, thus terminating collection. Recovery of unsatisfactory ThinPrep specimens has been described using off-label techniques, 3,4,6,20,23 but this entails added expense and labor. Nevertheless, some have found reprocessing of unsatisfactory ThinPrep Pap tests to be a necessity. 4 The issue of unevenly or deficiently coated slides in the SurePath system has been eliminated by the use of company s SurePath PreCoat slides. In our experience at present, the only unsatisfactory BD Sure- Path specimens are those that are truly hypocellular. Further studies appear indicated to determine the best practice for the prevention of potential false-negative results when processing mucus-rich specimens with the ThinPrep methodology. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. REFERENCES 1. Birdsong G, Davey D, Darraugh T, Elgert P, Henry M. Specimen Adequacy. In: Solomon D, Nayur R, eds. The Bethesda System for Reporting Cervical Cytology. 2nd ed. New York: Springer; Cox JT. ASCCP practice guidelines: management issues related to quality of the smear. J Low Genit Tract Dis. 1997;1: Bentz JS, Rowe LR, Gopez EV, Marshall CJ. The unsatisfactory ThinPrep Pap test: Missed opportunity for disease detection? Am J Clin Pathol. 2002;117: Chacho MS, Mattie ME, Schwartz PE. Cytohistologic correlation rates between conventional Papanicolaou smears and ThinPrep cervical cytology: A comparison. Cancer Cytopathol. 2003;99: Islam S, West AM, Saboorian MJ, Ashfaq R. Cancer. 2004; 102: Belinson JL, Pan QJ, Biscotti C, et al. Primary screening with liquid-based cytology in an unscreened population in rural China, with an emphasis on reprocessing unsatisfactory samples. Acta Cytol. 2002;46: Song LH, Goh EST, Phang LC, Poh WT, Tay SK. Technical aspect of ThinPrep. Singapore Med J. 2000;41: ThinPrep. (Product Insert). Volume Rev J. Boxborough, MA: Cytyc Corporation; TriPath Imaging I. PrepStain System Product Insert. 2004; Doc No. 61CR000021, Rev Nance KV. Evolution of Pap testing at a community hospital: a ten year experience. Diagn Cytopathol. 2007;35: McGrath CM, Rowe LR, Gupta PK, Bentz JS. Comparison of cytologic preparations in cervicovaginal cytopathology: Conventional smear vs. ThinPrep vs. AutoCyte PREP. Acta Cytol. 2002;46: Cancer Cytopathology October 25, 2010

6 Excessive Mucus in Liquid-Based Cytology/Kenyon et al 12. Fite SK. Impact of converting ThinPrep to SurePath on overall sensitivity and adequacy. Acta Cytol. 2002;46: Sweeney BJ, Haq Z, Happel JF, Weinstein B, Schneider D. Comparison of the effectiveness of 2 liquid-based Papanicolaou systems in the handling of adverse limiting factors, such as excessive blood. Cancer. 2006;108: Linder J, Zahniser D. The ThinPrep Pap test. A review of clinical studies. Acta Cytol. 1997;41: Lee KR, Ashfaq R, Birdsong GG, et al. Comparison of conventional Papanicolaou smears and a fluid-based, thinlayer system for cervical cancer screening. Obstet Gynecol. 1997;90: Bolick DR, Hellman DJ. Laboratory implementation and efficacy assessment of the ThinPrep cervical cancer screening system. Acta Cytol. 1998;42: Wang TY, Chen HS, Yang YC, Tsou MC. Comparison of fluid-based, thin-layer processing and conventional Papanicolaou methods for uterine cervical cytology. J Formos Med Assoc. 1999;98: Clark SB, Dawson AE. Invasive squamous cell carcinoma in ThinPrep specimens: diagnostic clues in the cellular pattern. Diagn Cytopathol. 2002;26: Dalton P, MacDonald S, Boerner S. Acetic acid recovery of gynecologic liquid-based samples of apparent low squamous cellularity. Acta Cytol. 2006;50: Papillo JL, St John TL, Mahoney PD, et al. Reprocessing of unsatisfactory ThinPrep samples: Procedures to improve specimen adequacy and increase detection of cervical abnormalities. Acta Cytol. 1998;42: BD Surepath Cell Enrichment Process for Liquid-based Pap Test. BD website: sp_cell_enrich.asp. Accessed June 8, The ThinPrep Pap Test: The ThinPrep ThinPrep website: system/thinprep_2000.html. Accessed June 8, Hecht S, Swinehart W, Linder J. Comparison of methods for reprocessing bloody gynecological specimens in Preserv- Cyt solution. Acta Cytol. 2004;48:681. Cancer Cytopathology October 25,