Endocervical Cytology: Inter- and Intra-Observer Variability in Conventional Pap Smears

Abstract

<b><i>Introduction:</i></b> Although the cytological diagnostic criteria for cervical squamous and glandular lesions are established by the Bethesda System for Reporting Cervical Cytology, the reproducibility of the diagnosis of these lesions has been shown to be variable in previous studies. At best, occasional good kappa (κ) values were reached both inter- and intra-observerly. Generally, consensus on high-grade lesions has been better compared to milder changes. <b><i>Methods:</i></b> Altogether, 167 conventional Pap smears from 50 patients with histologically confirmed endocervical adenocarcinomas (EAC) and adenocarcinomas in situ (AIS) and from 28 patients with histologically proven high-grade intraepithelial lesions were analyzed by four cytopathologists. Twenty of the smears were later re-evaluated by the same cytopathologists. κ-values between cytopathologists in the categories of squamous versus glandular, negative for intraepithelial lesion or malignancy (NILM), atypical, and preneoplastic/neoplastic were calculated. The diagnostic Pap smears of EAC and AIS with best and worst consensus between observers were then morphologically analyzed. <b><i>Results:</i></b> The reproducibility ranged from poor to substantial. The overall κ-values between the four cytopathologists were 0.412, 0.314, 0.272, and 0.082, respectively, in the categories of preneoplastic/neoplastic, squamous versus glandular, NILM, and atypical. Overall intra-observer κ-values were correspondingly 0.491, 0.616, 0.345, and 0.241. In the diagnostic smears of AIS and EAC, the nuclear size >2 times the normal and nuclear pleomorphism were the commonest features associated with good diagnostic consensus and the lack of nuclear enlargement and degenerative changes were associated with poor consensus. <b><i>Conclusions:</i></b> The reproducibility of preneoplasia/neoplasia diagnoses was better than that of atypia and NILM both in the inter- and intra-observer part in this study. In the smears from AIS and EAC patients, general neoplasia-associated features were more common in samples with good agreement by the four cytopathologists of the neoplastic nature and the endocervical origin of the lesion.

Validation of cytopathology specimens for digital pathology

“…Intrapersonal and interpersonal variability in cytology can be quite high, especially for atypical or borderline categories. [15][16][17] This should be accounted for when dealing with these types of specimens and is the reason why a one-degree variation was considered concordant in our final analysis. According to the latest CAP 2021 guidelines, 5 there are three SRs and nine GPSs.…”

“…As previously stated, because these recommendations do not address specific cytopathological issues, adjustments were made to allow for the practical considerations associated with the analysis of cytological samples, such as defining agreement and final concordance rates. Intrapersonal and interpersonal variability in cytology can be quite high, especially for atypical or borderline categories 15–17 . This should be accounted for when dealing with these types of specimens and is the reason why a one‐degree variation was considered concordant in our final analysis.…”

Inter‐ and intraobserver agreement in whole‐slide digital ThinPrep samples of low‐grade squamous lesions of the cervix uteri with known high‐risk HPV status: A multicentric international study

“…From that perspective, it will be valuable in the near feature to analyze our samples for particular features related to the highest and lowest agreement. Endocervical atypical lesions with less agreement are often impaired by degenerative changes and lack of nuclear enlargement 38 …”

“…Endocervical atypical lesions with less agreement are often impaired by degenerative changes and lack of nuclear enlargement. 38 …”

“…Endocervical atypical lesions with less agreement are often impaired by degenerative changes and lack of nuclear enlargement. 38 When the two rounds were compared, expert cytopathologists were not biased by HR HPV data generally. Nevertheless, personal scores showed better agreement in the second round with known HR HPV status with statistical significance (p = .004753 for Fleiss´ kappa and p = .0005778 for Kendall's coefficient).…”