False-Negative Interpretation of Adenocarcinoma In Situ in the College of American Pathologists Gynecologic PAP Education Program.
Histologically, AIS is characterized by endocervical glands that exhibit a crowded or cribriform pattern and are lined with atypical pseudostratified columnar epithelial cells. Mitoses are commonly present in the glands and the amount of cellular cytoplasm is decreased. By definition, stromal invasion is absent but can be difficult to assess histologically. Patients with AIS are nearly always asymptomatic and the lesions are generally not visible upon gross examination.
Cytologically, typical AIS cells are in groups and strips with rosette formation, feathering, crowding and/or stratification with indistinct cell borders. The nuclei are described as "cigar shaped" or elongated in most cases and nucleoli are inconspicuous. The nuclear to cytoplasmic ratio is increased. However, these characteristic features are not always present. Papanicolaou (Pap) cytology detects only 38% to 50% of AIS cases, indicating that cytology alone often is ineffective for detecting AIS. (5-7) This low detection rate is caused not only by sampling difficulty because of the location of AIS within the endocervical canal, (8,9) but also by an inability to recognize AIS on Pap slides. (10) Either glandular or squamous cytologic abnormalities may precede a histologic diagnosis of AIS and include glandular abnormalities (50%-69%), squamous abnormalities (26%-31%), mixed squamous and glandular abnormalities (15%), and negative cytologic results (4%). (11,12) About 1.1% to 4.7% of women with atypical glandular cells on cervical cytology have a histologic diagnosis of AIS at follow-up. (13-19) Liquid-based cytology has not improved the ability to accurately predict AIS. (20)
Since 1988, the College of American Pathologists (CAP) has been providing educational gynecologic cytology glass slide challenges through the Interlaboratory Comparison Program in Cervicovaginal Cytology (PAP program). (21) In 2006, the CAP, in response to an emerging need in the cytology community for additional Centers for Medicare & Medicaid Services-approved and well-validated proficiency programs, transformed the PAP program into 2 separate but interrelated programs: PAP Education Program (PAP-Edu) and PAP Proficiency Test Program (PAP-PT). PAP-Edu provides laboratories performing gynecologic cytology an option to meet the CAP Laboratory Accreditation Program Phase I requirement of enrollment in an educational interlaboratory comparison program. As noted above, it is difficult to specifically recognize AIS on Pap slides. In earlier studies of PAP-Edu, significantly more AIS Pap slides were interpreted as negative than as other significant lesions including high-grade squamous intraepithelial lesion (HSIL), squamous cell carcinoma, and adenocarcinoma. (10) The false-negative rate for AIS has been reported as 3.8% to 11.7%. (10,12,22) The high false-negative interpretation rate for AIS is an important problem in daily practice. The aim of this study was to investigate the false-negative interpretation of AIS and the factors related to false-negative interpretations through responses in an educational environment by analyzing the responses to AIS Pap slides in the PAP-Edu program.
MATERIALS AND METHODS
Participants in the PAP-Edu program and members of the CAP Cytopathology Committee contribute slides to the programs. Submitted slides with a diagnosis of HSIL, AIS, or higher must be confirmed by subsequent or concurrent biopsy. The slides and accompanying clinical information are reviewed and agreed upon by a minimum of 3 anatomic pathology board-certified members of the CAP Cytopathology Committee. PAP-Edu slides that become field validated can be used for PAP-PT slide sets; however, AIS cases currently remain in PAP-Edu sets. (23-25) In both PAP-Edu and PAP-PT programs, the interpretive menu includes selection categories A, B, C, and D. PAP-Edu directs participants to specifically choose category D subcategories that include HSIL, AIS, squamous cell carcinoma, adenocarcinoma-not otherwise specified (NOS), HSIL/carcinoma and/or carcinoma-NOS, and nonepithelial malignant neoplasm. However, the category D in PAP-PT has participants select only 1 option, HSIL and above.
We retrospectively analyzed 11 337 responses from 173 AIS slides that were reported in the CAP PAP-Edu program between 2011 and 2015. The 173 slides included 164 liquid-based preparation slides, including 135 ThinPrep slides (Hologic, Bedford, Massachusetts) and 29 SurePath slides (BD Diagnostics, Franklin Lakes, New Jersey), and 9 conventional preparation slides. A misinterpretation response was defined as an AIS slide interpreted under a category other than AIS, while a false-negative response was defined as an AIS case undercalled as negative for intraepithelial lesion or malignancy (NILM)/benign. Unsatisfactory responses were very low, with fewer than 5 responses during the study period. These results were excluded from the analysis.
Our analysis examined the factors associated with the falsenegative rate of the AIS slides. A nonlinear mixed model was fit by using 2 factors--participant type (pathologist versus cytotechnologist) and preparation type (ThinPrep versus SurePath versus conventional). The interaction term for these factors was also included in the model. There was no performance trend, so program year was not included in the model (Cochran-Armitage trend test; P = .10). The model included a repeated measures component to model the slide factor correlation structure, which controls for slide-specific performance. A significance level of .05 was used for this analysis. All analyses were run with SAS v9.3 (SAS Institute, Cary, North Carolina).
The overall responses to AIS slides among laboratories, pathologists, and cytotechnologists are summarized in Table 1. The correct interpretation rate of AIS slides was only 50.0% (5667 of 11 337). High-grade intraepithelial lesion, carcinomas, or malignant neoplasms accounted for 42.7% (4842 of 11 337) of responses and low-grade squamous intraepithelial lesion accounted for 0.4% (44 of 11 337). Among the responses of HSIL and invasive malignancies, adenocarcinoma was the most common interpretation, accounting for 73.8% (3571 of 4842). The overall false-negative rate was 6.9% (784 of 11 337). Overall, 92.7% (10 509 of 11 337) of responses were HSIL and above. We first analyzed the false-negative interpretation data by participant types and slide preparation types (Table 2). Overall, pathologists' responses showed a significantly higher false-negative rate than cytotechnologists' responses (8.3%, 403 of 4835 versus 5.7%, 275 of 4816; P < .001). The participant-type performance differences were significant for both ThinPrep (pathologist-cytotechnologist, P < .001) and SurePath preparations (pathologist-cytotechnologist, P < .001) (Table 2).
When the false-negative rates were compared among different preparations (ThinPrep, SurePath, and conventional), no significant difference was identified overall. The false-negative rate was highest for conventional preparations (Table 2), but not significantly different. This is likely due to the small number of responses for conventional slides, which may affect the analysis. In addition, cytotechnologists had more AIS interpretations than pathologists in both ThinPrep (53.0%, 2175 of 4105 versus 47.6%, 1960 of 4117) and SurePath slides (50.1%, 323 of 645 versus 44.6%, 291 of 653).
In the PAP-Edu program, category B (negative) allows participants to select several responses: NILM with no additional descriptors, fungal organisms consistent with Candida spp, Trichomonas vaginalis, cellular changes consistent with herpes simplex virus, reparative cytologic changes, atrophic vaginitis, and follicular cervicitis. The false-negative frequency distributions are summarized in Table 3. Among 784 false-negative responses, NILM was the most common, accounting for 61.5% (482 of 784), followed by reparative changes (24.1%, 189 of 784) and atrophic vaginitis (7.7%, 60 of 784).
The distributions of false-negative interpretation were further analyzed by the preparation types and participant types (Tables 3 and 4). Adenocarcinoma in situ was more frequently interpreted as atrophic vaginitis on SurePath than on ThinPrep slides, but less frequently interpreted as NILM or reparative changes (Table 3). When the distributions were further stratified by participant types (pathologist, cytotechnologist, and laboratory), no significant difference among the different false-negative interpretation was identified among these 3 participant types (Table 4).
Adenocarcinoma in situ is difficult to specifically recognize on Pap slides and the false-negative interpretation rate is significantly higher for AIS than that for other significant lesions including HSIL, squamous cell carcinoma, and adenocarcinoma. (10) To our knowledge, this is the largest study to investigate the false-negative interpretation rate for AIS. The false-negative interpretation rate for AIS from our study was 6.9%, which is within the reported range of 3.8% to 11.7%. (10, 12,22)
In the current study, AIS was specifically recognized by only 50.0% of participants, which was similar to the previously reported result from PAP programs. (10) Many AIS slides were misinterpreted as HSIL or malignancy, indicating the difficulty in distinguishing AIS from HSIL, squamous cell carcinoma, and especially adenocarcinoma, based on cytomorphologic features alone. Clinically, misinterpretation of an AIS Pap as HSIL or other malignancy would not cause a significant impact on the management of the patient, because patients with any of these diagnoses will undergo colposcopy and biopsy. However, misinterpretation of AIS Pap test as a negative result may result in serious clinical and medicolegal consequences unless the patient also has a positive human papillomavirus test result that leads to closer monitoring or evaluation. In the current study, we focused on investigating the false-negative responses and their related factors. The false-negative interpretation rate for AIS from our study was 6.9%, which was much higher than the false-negative interpretation rate for HSIL (0.9%) in the CAP PAP-Edu program reported in our recent study. (26) The false-negative rate for AIS Pap slides was reported to be as high as 11.7% in PAP programs previously. (10) The current results may indicate that the participants of the PAP-Edu program have improved on the interpretation of AIS over time. However, the slides used in the 2001 and 2002 CAP Interlaboratory Comparison Program in Cervicovaginal Cytology Program were conventional preparations and did not include liquid-based preparations. The false-negative rate for HSIL reported in that study was also much higher than what we found in our recent study (4.6%, 343 of 7535 versus 0.9%, 266 of 28 000). (10,26)
Glandular cell abnormalities in the Bethesda System include atypical endocervical cells; atypical endometrial cells; atypical glandular cells, not otherwise specified; atypical endocervical (or glandular) cells, favor neoplastic; AIS; and adenocarcinoma of various types. The atypical glandular and squamous categories are excluded in both the CAP PAP-Edu and PAP-PT program because of interobserver variability and lack of a gold standard biopsy correlation. The difference of reporting categories between daily practice and the PAP-Edu program may affect the false-negative rate of AIS in the PAP-Edu program.
In this study, we also analyzed the potential factors related to false-negative interpretations, including preparation types and participant types. Pathologists' responses showed a significantly higher false-negative rate than cytotechnologists' responses (8.3% versus 5.7%, P < .001) in both ThinPrep and SurePath preparations, indicating that cytotechnologists had higher sensitivity for AIS, possibly owing to their job emphasis on highly sensitive screening skills. Pathologists are responsible for the final interpretation and may be less likely to commit to a definitive AIS interpretation. The participants' attitude toward the PAP-Edu program may also contribute to the different false-negative rates between pathologists and cytotechnologists. Cytotechnologists might review the slides more carefully than pathologists and may be more likely than pathologists to have PAP-Edu results used as a part of their competency evaluation.
Our study showed that when the false-negative rates were compared among different preparations (ThinPrep, SurePath, and conventional), no significant difference was identified overall. ThinPrep slides overall had the lowest false-negative rates; however, numbers of participant responses for conventional slides (187 of 11 337; 1.6%) were too low to demonstrate statistical differences. Roberts and Thurloe (20) reported that liquid-based cytology implementation had not improved the ability to accurately predict AIS. Liquid-based preparations tend to cause cell groups to round up, and feathering can be difficult to assess in liquid-based preparations as opposed to conventional preparations. One recent study (27) evaluated the detection rate of glandular lesions (both AIS and adenocarcinoma) and found that the false-negative rate for Pap tests to detect glandular lesions was 8.8%. They further analyzed discordance scores among different methodologies and found that SurePath nonimaged cases had the lowest discordance score (0.2), followed by ThinPrep nonimaged cases (1.1), ThinPrep imaged cases (1.4), and conventional cases (1.6). However, the number of cases in this study was quite small with a total of 91 cases. (27) The discrepancy between our study and the previous study may result from many factors, such as quantitative and obscuring artifact differences among individual cases, staining variation secondary to preparatory techniques, and cytopathologists' comfort level with the various preparations. Therefore, further studies are warranted to compare performance of these different methods, not only for AIS, but also for other glandular abnormalities.
Among all false-negative responses, NILM with no other descriptors was the most common, accounting for 61.5% (482 of 784), followed by reparative changes (24.1%, 189 of 784), and atrophic vaginitis (7.7%, 60 of 784). All other negative categories including microorganisms accounted for only 6.8% (53 of 784) of the false-negative responses. These findings suggest that the main reasons for the false-negative interpretations were that AIS cells were either not identified or were misinterpreted as reactive or reparative changes. In daily practice situations, cytotechnologists and pathologists may choose to classify cases they perceive as not having classic AIS criteria as "atypical" to prompt further investigation.
In conclusion, the low correct interpretation rate and high false-negative rate for AIS reveals the difficulty in interpreting AIS on all types of cytology preparations and reflects room for improvement in the interpretation of AIS. The significantly higher false-negative rate from pathologists than from cytotechnologists suggests that cytotechnologists have better Pap cytology screening skills. The higher false-negative rate in SurePath than in ThinPrep preparations warrants further studies to compare performance of these 2 methods.
Please Note: Illustration(s) are not available due to copyright restrictions.
(1.) Wang SS, Sherman ME, Hildesheim A, et al. Cervical adenocarcinoma and squamous cell carcinoma incidence trends among white women and black women in the United States for 1976-2000. Cancer. 2004; 100(5):1035-1044.
(2.) Dunton CJ. Management of atypical glandular cells and adenocarcinoma in situ. Obstet Gynecol Clin North Am. 2008; 35(4):623-632.
(3.) Lee KR, Flynn CE. Early invasive adenocarcinoma of the cervix. Cancer. 2000; 89(5):1048-1055.
(4.) Plaxe SC, Saltzstein SL. Estimation of the duration of the preclinical phase of cervical adenocarcinoma suggests that there is ample opportunity for screening. Gynecol Oncol. 1999; 75(1):55-61.
(5.) Krane JF, Granter SR, Trask CE, Hogan CL, Lee KR. Papanicolaou smear sensitivity for the detection of adenocarcinoma of the cervix: a study of 49 cases. Cancer. 2001; 93(1):8-15.
(6.) Lee KR, Minter LJ, Granter SR. Papanicolaou smear sensitivity for adenocarcinoma in situ of the cervix: a study of 34 cases. Am J Clin Pathol. 1997; 107(1):30-35.
(7.) Muntz HG, Bell DA, Lage JM, Goff BA, Feldman S, Rice LW. Adenocarcinoma in situ of the uterine cervix. Obstet Gynecol. 1992; 80(6): 935-939.
(8.) Noda K, Kimura K, Ikeda M, Teshima K. Studies on the histogenesis of cervical adenocarcinoma. Int J Gynecol Pathol. 1983; 1(4):336-346.
(9.) Christopherson WM, Nealon N, Gray LA. Noninvasive precursor lesions of adenocarcinoma and mixed adenosquamous carcinoma of the cervix uteri. Cancer. 1979; 44(3):975-983.
(10.) Renshaw AA, Mody DR, Lozano RL, et al. Detection of adenocarcinoma in situ of the cervix in Papanicolaou tests: comparison of diagnostic accuracy with other high-grade lesions. Arch Pathol Lab Med. 2004; 128(2):153-157.
(11.) Shin CH, Schorge JO, Lee KR, Sheets EE. Cytologic and biopsy findings leading to conization in adenocarcinoma in situ of the cervix. Obstet Gynecol. 2002; 100(2):271-276.
(12.) Mitchell H, Hocking J, Saville M. Cervical cytology screening history of women diagnosed with adenocarcinoma in situ of the cervix: a case-control study. Acta Cytol. 2004; 48(5):595-600.
(13.) Geier CS, Wilson M, Creasman W. Clinical evaluation of atypical glandular cells of undetermined significance. Am J Obstet Gynecol. 2001; 184(2):64-69.
(14.) Schnatz PF, Guile M, O'Sullivan DM, Sorosky JI. Clinical significance of atypical glandular cells on cervical cytology. Obstet Gynecol. 2006; 107(3):701-708.
(15.) Zhao C, Florea A, Onisko A, Austin RM. Histologic follow-up results in 662 patients with Pap test findings of atypical glandular cells: results from a large academic womens hospital laboratory employing sensitive screening methods. Gynecol Oncol. 2009; 114(3):383-389.
(16.) Chin AB, Bristow RE, Korst LM, Walts A, Lagasse LD. The significance of atypical glandular cells on routine cervical cytologic testing in a community-based population. Am I Obstet Gynecol. 2000; 182(6):1278-1282.
(17.) Kim TJ, Kim HS, Park CT, et al. Clinical evaluation of follow-up methods and results of atypical glandular cells of undetermined significance (AGUS) detected on cervicovaginal pap smears. Gynecol Oncol. 1999; 73(2):292-298.
(18.) Zhao C, Florea A, Austin RM. Clinical utility of adjunctive high-risk human papillomavirus DNA testing in women with Papanicolaou test findings of atypical glandular cells. Arch Pathol Lab Med. 2010; 134(1):103-108.
(19.) Pradhan D, Li Z, Ocque R, Patadji S, Zhao C. Clinical significance of atypical glandular cells in Pap tests: an analysis of more than 3000 cases at a large academic women's center. Cancer Cytopathol. 2016; 124(8):589-595.
(20.) Roberts JM, Thurloe JK. Comparative sensitivities of ThinPrep and Papanicolaou smear for adenocarcinoma in situ (AIS) and combined AIS/high-grade squamous intraepithelial lesion (HSIL): comparison with HSIL. Cancer. 2007; 111 (6):482-486.
(21.) Nielsen ML. Cytopathology interlaboratory improvement programs of the College of American Pathologists Laboratory Accreditation Program (CAP LAP) and Performance Improvement Program in Cervicovaginal Cytology (CAP PAP). Arch Pathol Lab Med. 1997; 121(3):256-259.
(22.) Barroilhet L, Van Hanegem L, Bernstein M, Feldman S. Potential effects of updated pap test screening guidelines and adenocarcinoma in situ of the cervix. Obstet Gynecol. 2013; 121(4):759-764.
(23.) Renshaw AA, Wang E, Mody DR, Wilbur DC, Davey DD, Colgan TJ. Measuring the significance of field validation in the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology: how good are the experts? Arch Pathol Lab Med. 2005; 129(5):609-613.
(24.) Renshaw AA, Walsh MK, Blond B, Moriarty AT, Mody DR, Colgan TJ. Robustness of validation criteria in the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology. Arch Pathol Lab Med. 2006; 130(8):1119-1122.
(25.) Bentz JS, Hughes JH, Fatheree LA, Schwartz MR, Souers RJ, Wilbur DC; for the Cytopathology Resource Committee, College of American Pathologists. Summary of the 2006 College of American Pathologists Gynecologic Cytology Proficiency Testing Program. Arch Pathol Lab Med. 2008; 132(5):788-794.
(26.) Zhao C, Crothers BA, Ghofrani M, et al. Misinterpretation rates of high-grade squamous intraepithelial lesion in the College of American Pathologists Gynecologic PAP Education and PAP Proficiency Test Program [published online ahead of print August 23, 2016]. Arch Pathol Lab Med. doi:10.5858/arpa.20150446-CP.
(27.) Miller RA, Mody DR, Tams KC, Thrall MJ. Glandular lesions of the cervix in clinical practice: a cytology, histology, and human papillomavirus correlation study from 2 institutions. Arch Pathol Lab Med. 2015; 139(11):1431-1436.
Chengquan Zhao, MD; Barbara A. Crothers, DO; Z. Laura Tabatabai, MD; Zaibo Li, MD, PhD; Mohiedean Ghofrani, MD; Rhona J. Souers, MS; Mujtaba Husain, MD; Fang Fan, MD, PhD; Rulong Shen, MD; Idris Tolgay Ocal, MD; Christine N. Booth, MD; Kelly Goodrich, CT(ASCP); Donna Russell, CT(ASCP); Diane D. Davey, MD; College of American Pathologists Cytopathology Committee
Accepted for publication September 14, 2016.
Published as an Early Online Release March 16, 2017.
From the Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (DrZhao); the Department of Pathology, Walter Reed National Military Medical Center, Bethesda, Maryland (Dr Crothers); the Department of Pathology, University of California San Francisco and VA Medical Centers, San Francisco (Dr Tabatabai); the Department of Pathology, Ohio State University Medical Center, Columbus (Drs Li and Shen); the Department of Pathology PeaceHealth Laboratories, Vancouver, Washington (Dr Ghofrani); the Departments of Biostatistics (Ms Souers) and Survey (Ms Goodrich), College of American Pathologists, Northfield, Illinois; the Department of Clinical Sciences, University of Central Florida, Orlando (Drs Husain and Davey); the Department of Pathology, University of Kansas Medical Center, Kansas City (Dr Fan); the Department of Pathology and Laboratory Medicine, Mayo Clinic, Scottsdale, Arizona (Dr Ocal); the Department of Pathology, Cleveland Clinic Foundation, Cleveland, Ohio (Dr Booth); and the Department of Pathology, University of Rochester Medical Center, Rochester, New York (Ms Russell).
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Chengquan Zhao, MD, Department of Pathology, University of Pittsburgh Medical Center, 300 Halket St, Pittsburgh, PA 15213 (email: Zhaoc@upmc.edu).
Table 1. Overall Interpretations of Adenocarcinoma In Situ (AIS) Categories No. of Percentage Responses AIS 5667 50.0 Adenocarcinoma 3571 31.5 HSIL 762 6.7 HSIL/carcinoma/carcinoma, NOS 431 3.8 Squamous cell carcinoma 67 0.6 Nonepithelial malignancy 11 0.1 LSIL 44 0.4 Negative 784 6.9 Total 11 337 100 Abbreviations: HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NOS, not otherwise specified. Table 2. Comparison of False-Negative Response Rate of Adenocarcinoma In Situ for Participant Types and Preparation Types-Nonlinear Mixed Model Summary (a) Factors No. of False-Negative, P Responses No. (%) Value Participant type <.001 Pathologist 4835 403 (8.3) Cytotechnologist 4816 275 (5.7) Laboratory 1686 106 (6.3) Preparation type .36 ThinPrepb 9619 636 (6.6) SurePathc 1531 124 (8.1) Conventional 187 24 (12.8) Participant typea and .07 preparation type ThinPrepb Pathologist 4117 317 (7.7) Cytotechnologist 4105 232 (5.7) Laboratory 1397 87 (6.2) SurePathc Pathologist 653 76 (11.6) Cytotechnologist 645 37 (5.7) Laboratory 233 11 (4.7) Conventional Pathologist 65 10 (15.4) Cytotechnologist 66 6 (9.1) Laboratory 56 8 (14.3) (a) Significant test results: Pathologist-cytotechnologist (P = .002); no testing for laboratory participants; Thin Prep: pathologist-cytotechnologist (P < .001); SurePath: pathologist-cytotechnologist (P < .001). (b) Hologic, Bedford, Massachusetts. (c) BD Diagnostics, Franklin Lakes, New Jersey. Table 3. Specific Negative Participant Interpretations for Adenocarcinoma In Situ Slides by Preparation Types Participant Interpretation Total, Conventional, No. (%) No. (%) NILM 482 (61.5) 16 (66.7) Reparative changes 189 (24.1) 7 (29.2) Atrophic vaginitis 60 (7.7) 0 (0.0) Trichomonas vaginalis 21 (2.7) 1 (4.2) Fungal organisms consistent 15 (1.9) 0 (0.0) with Candida spp Cellular changes consistent 12 (1.5) 0 (0.0) with herpes simplex virus Follicular/lymphocytic 5 (0.6) 0 (0.0) cervicitis Total 784 24 Participant Interpretation ThinPrep, (a) SurePath, (b) No. (%) No. (%) NILM 400 (62.9) 66 (53.2) Reparative changes 158 (24.8) 24 (19.4) Atrophic vaginitis 41 (6.4) 19 (15.3) Trichomonas vaginalis 13 (2.0) 7 (5.6) Fungal organisms consistent 12 (1.9) 0 (0.0) with Candida spp Cellular changes consistent 8 (1.3) 7 (5.6) with herpes simplex virus Follicular/lymphocytic 4 (0.6) 1 (0.8) cervicitis Total 636 124 Abbreviation: NILM, negative for intraepithelial lesion or malignancy. (a) Hologic, Bedford, Massachusetts. (b) BD Diagnostics, Franklin Lakes, New Jersey. Table 4. Negative Participant Interpretations for Adenocarcinoma In Situ Slides by Participant Types Participant Interpretation Cytotechnologist, Pathologist, No. (%) No. (%) NILM 171 (62.2) 248 (61.5) Reparative changes 62 (22.5) 97 (24.1) Atrophic vaginitis 26 (9.5) 27 (6.7) Cellular changes consistent 7 (2.5) 6 (1.5) with herpes simplex virus Trichomonas vaginalis 6 (2.2) 12 (3.0) Fungal organisms consistent 2 (0.7) 9 (2.2) with Candida spp Follicular/lymphocytic 1 (0.4) 4 (1.0) cervicitis Total 275 403 Participant Interpretation Laboratory, No. (%) NILM 63 (59.4) Reparative changes 30 (28.3) Atrophic vaginitis 7 (6.6) Cellular changes consistent 2 (1.9) with herpes simplex virus Trichomonas vaginalis 3 (2.8) Fungal organisms consistent 1 (0.9) with Candida spp Follicular/lymphocytic 0 (0.0) cervicitis Total 106 Abbreviation: NILM, negative for intraepithelial lesion or malignancy.