A liquid concept--do classic preparations of body cavity fluid perform differently than ThinPrep cases? Observations from the College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology.
* Context.--Newer liquid-based preparations differ morphologically
from classic preparations (smears, filters, and cytocentrifuged
preparations). Is adenocarcinoma more readily detected in liquid-based
preparations? We reviewed responses from 16 750 fluid challenges of
adenocarcinoma distributed in 2005 in the College of American
Pathologists Interlaboratory Comparison Program in Nongynecologic
Cytology (CAP NGC).
Objective.--To compare the performance of body cavity fluid liquid-based preparations with adenocarcinoma to that in classic preparations in the CAP NGC.
Design.--Responses for ThinPrep challenges were compared with classic preparations for exact match diagnoses of adenocarcinoma from pelvic washes, pleural fluid, pericardial fluid, and peritoneal fluids in the 2005 CAP NGC.
Results.--A total of 13 690 pathologists, 8345 cytotechnologists, and 5958 laboratories submitted responses to fluid challenges in 2005. Adenocarcinoma comprised 16750 of the fluid challenges; 88% were classic preparations, and 12% were ThinPrep challenges. The exact match to the reference diagnosis of adenocarcinoma was seen in 77% of conventional preparations and 81% of ThinPrep challenges when a general category of "positive for malignancy" was assigned. When "suspicious for malignancy," an exact match diagnosis of adenocarcinoma was made in 5% and 4% of classic and ThinPrep challenges, respectively.
Conclusions.--ThinPrep challenges performed slightly better overall, but only pelvic washings and peritoneal fluids demonstrated statistically significant improved performance with ThinPrep challenges. Use of liquid-based preparation is widespread for nongynecologic preparations and performs as well, and sometimes better than, classic preparations in an interlaboratory comparison program.
(Arch Pathol Lab Med. 2008;132:1716-1718)
Body fluids (Medical examination)
Clinical chemistry (Methods)
Pathology, Cellular (Research)
Moriarty, Ann T.
Schwartz, Mary R.
Ducatman, Barbara S.
Booth, Christine N.
|Publication:||Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2008 College of American Pathologists ISSN: 1543-2165|
|Issue:||Date: Nov, 2008 Source Volume: 132 Source Issue: 11|
|Topic:||Event Code: 310 Science & research|
|Organization:||Organization: College of American Pathologists|
|Geographic:||Geographic Scope: United States Geographic Code: 1USA United States|
Body cavity fluid specimens are commonly encountered in daily
practice; differentiating adenocarcinoma from reactive mesothelial cells
is a daily dilemma. Since 1991, the ThinPrep (TP) processor has been
approved by the US Food and Drug Administration for processing
nongynecologic specimens and became commercially available in 2005. In
1999, the SurePath liquid-based technique was approved by the US Food
and Drug Administration and used to prepare nongynecologic specimens.
Many laboratories converted from their classic method of fluid
preparation (smears, filters, and/or cytocentrifuge, with or without a
cell block) to newer liquid-based processing in addition to or as a
replacement for their usual methods.
The College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology (CAP NGC) program began in 1997. During the following 8 years, the program grew to include 1750 laboratories and 7776 participants. This purely educational program consists of a quarterly glass slide mailing with 5 slides and a brief clinical history. Individuals and laboratories participate in the CAP NGC program for continuing education or to provide a measurement of their performance compared with other laboratories.
The members of the CAP Cytopathology Resource Committee submit cases used for the CAP NGC program. Cases are selected for the CAP NGC program by consensus at screening sessions. Two members, who agree on a general and specific interpretation, review all slides. The 2 members also agree that the slide is a good representation of the entity and that it is technically adequate for assessment. Body cavity fluids, respiratory specimens, cerebrospinal fluids, and fine-needle aspirations from a variety of sites are part of the CAP NGC. Each slide in the program has a specific reference diagnosis assigned to a body site.
In the field, the participants are able to choose first a general category in which to place the case (negative, suspicious, positive, unsatisfactory), and then a specific reference diagnosis in which participants hone their responses to a more specific diagnosis. After circulation in the field, all slides have a unique performance profile. The performance of each slide can be compared against the others in the same reference category.
We reviewed the responses from 16 750 fluid challenges of adenocarcinoma distributed in 2005 in the CAP NGC program. We reviewed the performance characteristics of liquid-based preparations of body cavity fluids with the reference diagnosis of adenocarcinoma compared with classic preparations of adenocarcinoma. When adenocarcinoma was the specific reference diagnosis, did participants more often recognize adenocarcinoma in modern liquid-based preparations? Do the preservation and presentation of the cells in modern liquid-based methods make a difference in identifying adenocarcinoma in an educational interlaboratory comparison program?
Cumulative histories of all responses for body cavity fluids with the reference diagnosis of adenocarcinoma were obtained through the CAP SCORES (CAP, Northfield, Ill) computer system for 2005. Body cavity fluids selected included pleural, pericardial, and peritoneal fluids and pelvic washes. Participant responses for the general category of "positive" for malignancy and "suspicious" for malignancy with the reference diagnosis of adenocarcinoma were included. The preparation type was noted. Liquid-based preparations were all TP in the CAP program during 2005. Classic preparations included direct smears and cytocentrifuged preparations.
The Fisher exact test was used to compare the population of TP body cavity fluids and classic preparations, including pelvic washes, peritoneal fluids, pleural fluids, and pericardial fluids. The responses of participants identifying the cases as malignant and suspicious were evaluated.
During the course of 2005, fluid challenges from pelvic washes, peritoneal fluid, pericardial fluid, and pleural fluid were answered by 22 035 individual participants (13 690 pathologists and 8345 cytotechnologists) and 5978 laboratories (Table 1). Positive body cavity fluid reference diagnoses included adenocarcinoma, squamous cell carcinoma, non-small cell carcinoma, small cell carcinoma, melanoma, lymphoreticular malignancy, mesothelioma, and sarcoma. Cases with the reference diagnosis of benign (negative) included nonspecific infection, specific infections (fungal organisms), inflammation, lymphocytosis, and normal/reactive mesothelial cells.
Adenocarcinoma comprised 16750 of the fluid challenges; 88% were classic preparations, and 12% were TP challenges. Table 2 demonstrates the findings both for the general category of suspicious and positive for the specific reference diagnosis of adenocarcinoma when adenocarcinoma was the reference diagnosis for the case. The TP challenges performed slightly better overall than the classic preparations. Performance varied from site to site. Pelvic washings and peritoneal fluids demonstrated statistically significant improved performance. Pleural and pericardial fluids did not achieve statistical significance. The suspicious for adenocarcinoma category in both pelvic washes and peritoneal fluids was also statistically significant.
Body cavity fluid cytologic preparations are a daily challenge. The practicing cytologist is constantly on guard when presented with a cellular specimen. Cytologic criteria for malignancy include a cellular specimen with 2 distinct cell populations. Several factors that complicate evaluation of the specimen include inflammation, blood, and the reactive mesothelium, which presents a continuum of morphologic changes, making distinction of malignancy difficult. (1) The introduction of liquid-based preparatory techniques allowed for enrichment of the cells in the preparations with a reduction in inflammatory cells and blood in the preparations. The presentation of cells in a uniform layer should theoretically aid in the identification of malignant cells. However, there are also "new artifacts" of the liquid-based preparation of which the cytologist must be aware. (2) The cells are smaller than those seen in cytocentrifuged specimens, the nuclear-cytoplasmic ratios seem higher, and the nuclear membranes and the nuclear features are very well preserved. The lack of air drying, flattening of cells, and increased visibility of nuclear features require appreciating new cellular criteria in the liquid-based preparations.
The introduction of liquid-based challenges into the CAP NGC allows us the opportunity to evaluate whether the liquid-based techniques make a difference in the diagnosis of adenocarcinoma on a single slide challenge. The large number of practicing cytotechnologists and pathologists who review cases of adenocarcinoma in body cavity fluids provides us with a large enough population for evaluation. Adenocarcinoma is the most common specific reference diagnosis in body cavity fluid challenges of the CAP NGC. This study demonstrates that there is a slightly better overall performance in liquid-based challenges compared with classic preparations, but it is not statistically significant.
There is a statistically significant advantage of liquid-based preparations over classic preparations in peritoneal fluid and pelvic washings. Previous studies from the CAP NGC program have demonstrated that those cases of adenocarcinoma in body cavity fluid that perform poorly fall into 3 categories: hypocellular adenocarcinoma, single-cell pattern of adenocarcinoma, and hypercellular adenocarcinoma without a distinct second population of cells. (3) We postulate that the adenocarcinoma that frequently involves the peritoneal cavity and pelvic washes is derived from ovarian or colon carcinoma, and the enhanced visibility of the nuclear features may aid in better performance, especially when there is a hypercellular population of malignant cells without distinct populations (Figure).
It is important to note that the CAP NGC program is a purely educational program in interlaboratory comparison. The slides originate from different laboratories, are not uniformly stained, and represent a variety of primary tumors. The participants come from a variety of backgrounds and experience. There are cytotechnologists and pathologists with different training experience and practices. The participants are given a single slide with limited clinical information and no access to ancillary studies. The performance of slides in this program is based on morphologic evaluation and therefore does not reflect usual clinical practice. However, it also isolates morphology as a single parameter and allows comparison of morphology of different preparations. The comparison of these populations is also based on a variety of cases of adenocarcinoma in fluids from different patients and primaries. It is not a split specimen comparison.
In summary, evaluation of the CAP NGC program in 2005 demonstrates no significant difference in performance of slides with adenocarcinoma in fluids that have been prepared with TP methods versus classic preparations. There may be an advantage to participants with the use of liquid-based methods in preparation from pelvic washes or peritoneal cavity sites.
This study and manuscript were prepared without financial support other than the resources provided by the College of American Pathologists.
Accepted for publication March 19, 2008.
(1.) Tao LC. Cytopathology of Malignant Effusions. Chicago, Ill: ASCP Press; 1996.
(2.) Smith RA. The cytology of effusions. In: ThinPrep[R] Non-Gyn Morphology Reference Atlas. Marlborough, Mass: Cytyc; 2005.
(3.) Moriarty AT, Stastny J, Volk EE, Hughes JH, Miller TR, Wilbur DC. Fluids--good and bad actors: observations from the College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology. Arch Pathol Lab Med. 2004;128:513-518.
Ann T. Moriarty, MD; Mary R. Schwartz, MD; Barbara S. Ducatman, MD; Christine N. Booth, MD; Jennifer Haja, CT(ASCP); Subhendu Chakraborty, MS; Beth Williamson, CT(ASCP)
From AmeriPath Indiana, Indianapolis (Dr Moriarty); the Cytopathology Resource Committee, College of American Pathologists, Northfield, Ill (Drs Moriarty, Schwartz, Ducatman, and Booth); the Department of Pathology, The Methodist Hospital, Houston, Tex (Dr Schwartz and Mr Chakraborty); the Department of Pathology, West Virginia University Hospital, Morgantown (Dr Ducatman); the Department of Anatomic Pathology, Cleveland Clinic Foundation, Cleveland, Ohio (Dr Booth); and the Surveys Department, College of American Pathologists, Northfield, Ill (Ms Haja and Ms Williamson).
The authors have no relevant financial interest in the products or companies described in this article
Presented in part at the 55th Scientific Session of The American Society of Cytopathology, Houston, Tex, November 2007.
Reprints: Ann T. Moriarty, MD, AmeriPath Indiana, 2560 N Shadeland Ave, Suite A, Indianapolis, IN 46219-1739 (e-mail: email@example.com).
Table 1. Body Cavity Fluid Challenges Circulated in 2005 Pathologist Cytotechnologist Laboratory Pelvic washings 1018 613 418 Pleural 8429 5072 3669 Pericardial 414 302 210 Peritoneal 3829 2358 1681 Total 13 690 8345 5978 Table 2. Comparison of Body Cavity Fluids With Adenocarcinoma as Reference * Responses, No. Adenocarcinoma, No. (%) Site C TP C TP Pelvic 1334 203 1150 (86) 195 (96) P < .001 Pleural 8035 890 5984 (75) 664 (75) P = .97 Pericardial 456 81 364 (80) 61 (75) P = .37 Peritoneal 4896 855 3867 (79) 722 (84) P < .001 Total 14721 2029 11 365 (77) 1642 (81) Suspicious, No. (%) Site C TP Pelvic 55 (4) 2 (1) P = .03 Pleural 390 (5) 51 (6) P = .25 Pericardial 6 (1) 1 (1) P = .60 Peritoneal 229 (4) 27 (3) P = .04 Total 680 (5) 81 (4) * The Fisher exact test. C indicates classic preparations; TP, ThinPrep specimens.
|Gale Copyright:||Copyright 2008 Gale, Cengage Learning. All rights reserved.|