Sandra M. Sulik, MD, MS Karin Kroeger, MD Jennifer K. Schultz, MSEd Jennie L. Brown, MD Lorne A. Becker, MD William D. Grant EdD Syracuse, Jordan-Elbridge, and Fayetteville, New York Submitted, revised, July 13, 2001. From the Department of Family Medicine (S.M.S., J.K.S., L.A.B., W.D.G.) and the Center for Evidence-Based Practice (J.K.S., L.A.B., W.D.G.), State University of New York (SUNY) Upstate Medical University, Syracuse; St. Joseph’s Family Practice Residency, Syracuse (S.M.S.); Family Care Medical Group, Jordan-Elbridge (K.K.); and St. Joseph’s Family Medicine, Fayetteville (J.L.B.). Reprint requests should be addressed to Sandra M. Sulik, MD, MS, SUNY Upstate Medical University, Department of Family Medicine and St. Joseph’s Family Practice Residency, 5820 Heritage Landing Drive, East Syracuse, NY 13057. Email: smsulik@aol.com.
References
OBJECTIVE: We compared fluid-based cytologies (FBC) with conventional Papanicolaou (Pap) tests to determine if either is superior.
STUDY DESIGN: This was a systematic review of original research reports evaluating both conventional Pap and FBC with respect to specimen adequacy, comparison with a reference standard, or both. Two reviewers independently reviewed the articles to determine inclusion status, with differences resolved by consensus with a third author. Risk differences (RD) between occurrence rates for FBC and Pap were used for the specimen adequacy data.
DATA SOURCES: Studies published between 1985 and November 1999 were identified from MEDLINE, Best Evidence, EMBASE, Biological Abstracts/RRM, and The Cochrane Library.
OUTCOMES MEASURED: Sensitivity, specificity, area under the receiver operating characteristic curve (AuROC), and the proportion of satisfactory, unsatisfactory, and “satisfactory but limited by” test results were measured.
RESULTS: There was no significant difference in AuROC (p=.37). FBC specimens were more likely to be satisfactory (RD=0.06; 95% confidence interval [CI], 0.03-0.09) or to have absent endocervical cells (RD=0.06; 95% CI, 0.02-0.10) but had 10% fewer “satisfactory but limited by—other” reports (RD = -0.10; 95% CI, -0.14 to -0.06). There was no difference in unsatisfactory Pap test results.
CONCLUSIONS: For most women there is no reason to replace Pap with FBC. For women at high risk of cervical cancer or who are screened infrequently, the possible increase in FBC sensitivity may outweigh the potential harms from additional false positives.
Despite mass Papanicolaou (Pap) test screening, approximately 12,800 women are given the diagnosis of cervical cancer in the United States each year, and approximately 4600 die of the disease.1 Fahey and colleagues2 found an average sensitivity of 58% (range = 11%-99%). Nanda and coworkers3 reported a sensitivity of 30% to 87%, with a specificity of 86% to 100%. Follen Mitchell and colleagues4 reported sensitivities of 67% and specificities of 77%. Multiple factors including sampling technique, patient preparation, test fixation and staining, and interpretation accuracy5 can increase the false-negative rates of conventional Pap.
Fluid-based cytology (FBC) procedures use a fluid medium to capture and preserve the collected cells from the cervical-sampling device. The collected sample is homogenized using an automated device, and a subsample of cellular material is placed on a glass slide in a circumferential area. Because the technique provides a uniform thin layer and excludes obscuring debris, it eliminates problems often encountered with Pap including poor fixation, uneven thickness of the cellular spread, air-drying artifact, and obscuring of cells by blood or inflammatory exudates.6
We performed a systematic review to evaluate the accuracy of FBC (by comparing its sensitivity and specificity with Pap) and the specimen adequacy of this new method (by comparing the proportion of FBC and Pap slides reported as unsatisfactory or “satisfactory but limited” by either absence of endocervical cells or other factors).
Methods
Search Strategy
Our literature search was designed to find studies comparing FBC and Pap. The search was assisted by a medical librarian and used medical subject headings (MeSH) and text words. The search terms included monolayer technology, ThinPrep, CytoRich, Cytoprep, Autoprep, AutoCyte, Papanicolau/pap smear, liquid-based cytology, fluid-based cytology, cervical cancer screening, and vaginal smears. MEDLINE, Best Evidence, EMBASE, Biological Abstracts/RRM and The Cochrane Library were searched to retrieve all potentially relevant English-language articles published between 1985 (first literature published on fluid-based cytology) and November 1999. An attempt was made to contact both FBC manufacturers to find any other available articles and abstracts in publication. The authors of articles with incomplete data were contacted in an attempt to obtain missing information. The reference lists of retrieved articles were manually searched for additional citations.
Study Inclusion
Articles were reviewed for inclusion if they contained reports of original research evaluating both conventional and fluid-based cytology samples (either ThinPrep or AutoCyte). Studies in which both tests were simultaneously applied to the same group of women (split-sample studies) and those in which one group of women who received Pap was compared with a group that received FBC (cohort studies) were included. Two reviewers independently reviewed the titles. When either reviewer felt the study might merit inclusion, the full article was retrieved. The same authors independently reviewed each article to determine whether it met the inclusion criteria. Differences were resolved by consensus with a third author.
We used the entire set of studies to address the question of specimen adequacy. To address FBC accuracy, we identified the subset of articles that compared the 2 tests with an external reference standard. We excluded articles from this analysis unless they reported colposcopy and biopsy results for at least 50% of the women with a finding of high-grade squamous intraepithelial lesions or higher on either Pap or FBC. Because few studies subjected all women with normal Pap tests to colposcopy, we also included studies that provided colposcopy to a random sample of women with normal Pap or FBC tests and those that subjected normal tests to an independent consensus review by a panel of experienced cytology professionals.7
