Comparison of the effectiveness of two liquid-based Papanicolaou systems in the handling of adverse limiting factors, such as excessive blood

Authors


Abstract

BACKGROUND

Excessive blood may compromise gynecologic Papanicolaou (Pap) smears. Liquid-based cytologic techniques have been developed in part to address this problem. In the current study, conditions of excessive blood were simulated to compare the ability of two liquid-based systems, ThinPrep® and SurePath™, to satisfactorily process specimens in the presence of this potentially limiting factor.

METHODS

Equal volumes of washed epithelial cells derived from pooled residues of liquid Pap vials were added to a series of ThinPrep and SurePath vials. Increasing volumes of freshly drawn, packed erythrocytes were added to the vials in progressive amounts from 50 μL or 100 μL up to 3000 μL. The vials were processed on their respective instruments according to U.S. Food and Drug Administration-approved procedures for a total of six test runs. The cellularity of the slides was measured by averaging epithelial cell counts in a total of five 40× fields.

RESULTS

SurePath preparations were uncompromised by blood until aliquots from 1000 μL to 3000 μL were reached. The ThinPrep system invariably was overwhelmed by the first 50-μL or 100-μL aliquot of blood, with epithelial cell counts dropping immediately to near zero.

CONCLUSIONS

The cell enrichment process of the SurePath system capably handled significantly greater amounts of potentially obscuring blood than the membrane filtration method of the ThinPrep system, which was compromised by as little as ≤ 1 drop of packed erythrocytes (1 drop = 65 μL). Cancer (Cancer Cytopathol) 2006. © 2005 American Cancer Society.

Liquid-based thin-layer Papanicolaou (Pap) tests, such as SurePath™ (TriPath Care Technologies, Burlington, NC) and ThinPrep® (Cytyc Corporation, Boxborough, MA), are in common use now as replacements for the conventional Pap smear. Increased sensitivity and improved specimen adequacy are attributed to these technologies,1–4 in part because of reductions in technical factors that limit conventional samples. One issue that has not shown improvement in some studies and laboratories, at least with the ThinPrep system, is the “unsatisfactory for evaluation” rate.5 When the cytology laboratory at St. Elizabeth's Medical Center (Boston, MA) converted it's thin-layer Pap smear technology from ThinPrep to SurePath, a drop in the “unsatisfactory for evaluation” rate from 1.8% to 0.2% was observed. The possibility that there may be differences in the ability of these two methodologies to handle potentially limiting technical factors was tested in the current study by simulating conditions of excessive blood and observing the points at which slides were compromised or the technologies became overwhelmed.

MATERIALS AND METHODS

Liquid-based Pap specimens were prepared by combining the cellular residue from 15 SurePath gynecologic specimens, washing the cells twice in deionized water, resuspending the cells in normal saline to a volume of 6 mL, and dispensing 100 μL of the suspension into each of a series of 9 or 10 SurePath and ThinPrep specimen vials. Peripheral blood from healthy donors (80 mL) was collected in lavender-top tubes and centrifuged. The plasma was removed, and the buffy coat and packed erythrocytes were mixed by vortexing. Conditions of excessive blood were simulated by adding increasing volumes of the erythrocyte/buffy-coat mixture to the SurePath and ThinPrep specimen vials beginning with 50 μL or 100 μL and continuing up to 3000 μL, as shown in Table 1. The vials were then processed according to the U.S. Food and Drug Administration (FDA)-approved methodologies for each liquid-based system. The cellularity of the resulting slides was measured by averaging the epithelial cell counts present in five 40× microscopic fields. Six test runs were performed.

Table 1. Comparison of the Cellularity of SurePath™ and ThinPrep® Slides Prepared from Epithelial Cell Suspensions with Increasing Volumes of Added Spun Red Blood Cells and Buffy Coat-Derived Leukocytes
Volume of added blood (μL)Epithelial cell countsa
SurePathThinPrep
  • a

    The average number of epithelial cells per 40× field (5 fields were counted).

Test run 1  
 058.262.4
 5060.00.0–1.0
 10062.60.0–1.0
 25059.80.0–1.0
 50059.20.0–1.0
 75059.00.0–1.0
 100059.60.0–1.0
 150057.80.0–1.0
 200016.80.0–1.0
 30000.0–1.00.0–1.0
Test run 2  
 026.444.6
 5028.00.0–1.0
 10024.20.0–1.0
 25018.60.0–1.0
 5005.60.0–1.0
 75022.60.0–1.0
 10000.0–1.00.0–1.0
 15000.0–1.00.0–1.0
 20000.0–1.00.0–1.0
 30000.0–1.00.0–1.0
Test run 3  
 093.464.8
 10090.00.0–1.0
 25092.60.0–1.0
 50055.60.0–1.0
 75039.20.0–1.0
 100046.00.0–1.0
 150039.40.0–1.0
 20000.0–1.00.0–1.0
 30000.0–1.00.0–1.0
Test run 4  
 091.462.6
 10089.20.0–1.0
 25082.80.0–1.0
 50084.60.0–1.0
 75070.80.0–1.0
 100059.00.0–1.0
 15004.40.0–1.0
 20000.0–1.00.0–1.0
 30000.0–1.00.0–1.0
Test run 5  
 0105.864.2
 100105.20.0–1.0
 25099.60.0–1.0
 50093.40.0–1.0
 75075.40.0–1.0
 100018.20.0–1.0
 15000.0–3.00.0–1.0
 20000.0–1.00.0–1.0
 30000.0–1.00.0–1.0
Test run 6  
 066.659.8
 5066.80.0–1.0
 10071.20.0–1.0
 25066.00.0–1.0
 50068.20.0–1.0
 75066.60.0–1.0
 100066.40.0–1.0
 150045.00.0–1.0
 20006.20.0–1.0
 30000.0–1.00.0–1.0

RESULTS

The results are shown in Table 1 and are illustrated in Figure 1. In all six test runs, slides prepared from pooled epithelial cell residues without added erythrocytes were “satisfactory for evaluation” using both SurePath and ThinPrep methodologies.

Figure 1.

SurePath™ (SP) (TriPath Care Technologies, Burlington, NC) and ThinPrep® (TP) (Cytyc Corporation, Boxborough, MA) slides compared with increasing amounts of added packed erythrocytes. (A) A SP slide with 0 μL blood demonstrates adequate cellularity. (B) A TP slide with 0 μL blood demonstrates adequate cellularity. (C) A SP slide with 50 μL blood demonstrates adequate cellularity. (D) A TP slide with 50 μL blood demonstrates inadequate cellularity. (E) A SP slide with 50 μL blood demonstrates adequate cellularity with a clean background. (F) A TP slide with 50 μL blood demonstrates clumped erythrocyte debris and rare cells obscured by debris. (G) A SP slide with 1000 μL blood demonstrates adequate cellularity with a clean background. (H) A TP slide with 1000 μL blood demonstrates granular debris with no cells. Original magnification × 100 (A–D); ×400 (E–H).

SurePath slide preparations were uncompromised by added erythrocytes until aliquots that varied from 1000 μL to 3000 μL were reached. The ThinPrep system was overwhelmed by the first 50-μL or 100-μL aliquot of added blood, with epithelial cell counts dropping immediately to near zero.

DISCUSSION

A drop in the unsatisfactory rate for gynecologic cytology specimens occurred in our laboratory after a change from ThinPrep to the SurePath liquid-based system. Over time, the observation of many perfectly interpretable SurePath slides that showed some amount of nonobscuring blood, inflammation, or mucus suggested the possibility that these two technologies may differ in their capacities to handle possible adverse limiting factors, such as excessive blood. The results of the current study indicate that this indeed is the case. The SurePath system handled from 750 μL to 2000 μL of the packed erythrocyte/buffy-coat mixture without compromise, whereas the ThinPrep system quickly was overwhelmed by 50–100 μL, an amount equal to 1–2 drops.6 With the addition of blood to the test vials, the ThinPrep filter membranes macroscopically exhibited a coating of granular material consistent with erythrocyte breakdown debris, which was seen both on the ThinPrep Pap slides (Fig. 1) and on microscopic examination of the ThinPrep filtration membranes (Fig. 2). This debris caused the ThinPrep processor to end its cycle with the collection of almost no squamous cells. Conversely, the SurePath methodology, with its cell-enrichment process,7 provided for the removal of enough blood to extend markedly the capacity of the system to produce satisfactory slides.

Figure 2.

Microscopic examination of ThinPrep® (Cytyc Corporation, Boxborough, MA) filtration membranes with 0 μL added blood in Panel A (note clean filtration pores) and 100 μL added blood in Panel B (note membrane and pores covered by erythrocyte debris). Original magnification × 400 (A, B).

This study was designed to correspond as closely as possible to real office and laboratory situations; therefore, the vials were processed 1 day after blood was added. If they were processed immediately after the addition of blood, the ThinPrep slides then were compromised slightly less. However, increasing amounts of flocculent material were noted in the ThinPrep vials over several hours after the addition of blood. This material remained consistently present thereafter. With the SurePath system, we have found no technical or morphologic compromises secondary to storage of vials for several weeks.

A few publications have raised concern regarding possible limitations in liquid-based thin-layer Pap smear technologies. One study, without explicitly making the case, implied the possibility of incomplete representation of pathologic material on ThinPrep slides by examining the tissue residue in specimen vials in a series of specimens that were diagnosed as “atypical glandular cells of undetermined significance.”8 In that series, it was found that 81% had interpretable residual material in the vial. Of these, 54% were identified as high-grade or malignant lesions. Minimal adequacy criteria presumably were met in the initial slide preparation in those specimens. Another study noted problems with ThinPrep slides described as “patchy cells,” “halo effect” (in which cells were collected only in the periphery of the sample area), and “obscuring blood and amorphous debris.”9 In still another study,5 examination of ThinPrep filter membranes from bloody specimens showed a thick layer of debris as well as membrane pores plugged with amorphous debris, a finding analogous to the results of the current study. In addition, problems with ThinPrep technology can be concluded from publications describing methods for the reduction of Pap unsatisfactory rates using off-label techniques, such as a CytoLyt® glacial acetic acid wash to reduce obscuring blood.5, 10, 11 Even Cytyc Corporation recently presented in abstract form four reprocessing methods for bloody gynecologic specimens using glacial acetic acid.12

Our own laboratory experience has shown problems in the uniformity of cell deposition on ThinPrep slides in the presence of blood, inflammation, or mucus, suggesting regional clogging of ThinPrep filter pores that could limit adequate representation of lesionable cellular material on the resulting slides. Bolick et al. have shown optimum detection rates for low-grade and high-grade squamous intraepithelial lesions when sample cellularity is > 20,000 squamous cells.13 Thus, technical factors that limit the cellular yield in thin-layer liquid-based Pap systems can impinge on the detection rate of cervical lesions.

Although the current study was small, the sharply contrasted results demonstrate systematically a technical limitation of the ThinPrep system in coping with excessive blood while using FDA-approved procedures. Further investigation of technologic differences in liquid-based Pap smear methodologies may yield additional data regarding specimen adequacy.

Ancillary