Smears diagnosed as ASCUS: Interobserver variation and follow‐up

Abstract

The purpose of this study was to apply atypical squamous cells of undetermined significance (ASCUS) criteria from the Bethesda System for Reporting Cervical/Vaginal Cytologic Diagnoses (TBS) to the rescreen of cases previously diagnosed as ASCUS, to compare initial and rescreen diagnoses, and to analyze agreement with follow-up (cytology or histology). Two cytotechnologists (S.B. and M.J.M.) and one cytopathology fellow (M.A.) rescreened 632 cervicovaginal specimens diagnosed as ASCUS between June 1, 1992-December 31, 1995. Age and LMP were provided. Rescreen diagnoses were categorized as within normal limits (WNL), ASCUS, low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL), or carcinoma (CA). Complete agreement was found in 200 specimens (32%): 31 (15%) WNL; 91 (45%) ASCUS; 77 (38.5%) SIL; and one (0.50%) CA. Follow-up revealed no abnormality in 67% of the cases reclassified as WNL, 49% of the cases reclassified as ASCUS, and 48% of the cases reclassified as squamous intraepithelial lesions (SIL). SIL was found in 29% of cases reclassified as WNL, 29% of specimens rediagnosed as ASCUS, and 34% of cases reclassified as SIL. Partial agreement was found in 391 specimens (62%). In 41 specimens (6%), rescreeners were in complete disagreement, and follow-up revealed 9/41 (22%) SIL or worse; 21/41 (51%) WNL; and 4/41 (10%) inconclusive. Applying established criteria, 14% (91/632) of cases diagnosed as ASCUS resulted in complete agreement, and 30% (190/632) resulted in partial agreement. Follow-up of cases initially diagnosed as ASCUS revealed SIL or CA in 30% of cases. ASCUS is a significant diagnosis warranting careful patient follow-up.

Implications of inter observer variability in cervical smear reporting

“…[5][6][7] Most studies that have been carried out, report maximum variations in the atypical squamous cell (ASC) category. [8][9][10][11][12][13][14] This has been attributed to many reasons amongst which, are the time allotted by screener for each smear, educational qualification of screeners, experience of screeners etc. 2,6,12 It has been observed by some authors that the experience of the observer and the grade of the lesions affects the reporting variability, i.e.…”

Visual estimates of nucleus-to-nucleus ratios

“…These results suggest that the participants were unable to differentiate between ASCUS and LSIL based on nuclear enlargement criteria, a finding which may implicate nuclear-enlargement determination as contributing to the well-documented poor reproducibility of the ASCUS diagnosis. [4][5][6][8][9][10]19 Furthermore, because ASCUS and LSIL are the major contributors to ASC and SIL, respectively, the difficulty in differentiating between ASCUS and LSIL is a likely explanation for the wide variation of ASC to SIL ratios between laboratories 1,3 and between cytotechnologists and pathologists, even when they practice at the same institution. [20][21][22][23][24][25][26][27] Previous studies have suggested that the variability among pathologists in the classification of squamous abnormalities may be related to differences in both diagnostic thresholds as well as in diagnostic accuracy.…”

“…poor interobserver reproducibility, [4][5][6][7][8][9][10] it is a very important practical diagnosis because it has been shown that most cases of significant squamous intraepithelial lesions (CIN 2/3) are identified during follow-up of Pap tests diagnosed as atypical squamous cells (ASCUS or atypical squamous cells cannot exclude high-grade squamous intraepithelial lesions [ASCH]). 11 After it was shown that the classification of atypical squamous cells has a large interobserver variability, 4 the Bethesda System has proposed and illustrated several subjective and objective morphologic quantitative and qualitative criteria that cells must fulfill to be classified as ASCUS.…”

Estudo das atipias indeterminadas em relação à prevalência e ao percentual de discordância nos casos do Programa de Prevenção do Câncer Uterino do Paraná

“…Gatsha et al (2001) realizaram estudo envolvendo 632 esfregaços com diagnóstico anterior de ASCUS, os quais foram revistos por outros três observadores. Entre os 200 casos (32%) que apresentavam concordância completa interobservadores, 91 (45%) diagnósticos primários de ASCUS foram confirmados, demonstrando a baixa reprodutividade dessa categoria diagnóstica (8) , as taxas de ASCUS e de lesão intra-epitelial escamosa são influenciadas pela rigidez na adoção de critérios morfológicos e pelo grau de experiência na interpretação dos espécimes citológicos, sendo muitas vezes desconhecidos os fatores que afetam a reprodutividade interobservadores (6,9,18) .…”