Cervical cytology findings in a large urban screening cohort in Tashkent, Uzbekistan: a 3-year population-based analysis using the Bethesda 2014 system.

INTRODUCTION

Cervical cancer remains a leading cause of cancer-related mortality in Central Asia, yet Bethesda 2014-classified population-level cytological data from this region are virtually absent from the indexed literature. We describe the distribution of cervical cytological findings and identify independent risk factors for high-grade squamous intraepithelial lesion (HSIL) in a large consecutive urban cohort from Tashkent, Uzbekistan.

MATERIALS AND

METHODS

A retrospective cross-sectional study of 28,310 consecutive cervical cytology specimens collected at a single tertiary cytopathology referral center between January 2022 and December 2024. All specimens were classified according to the Bethesda 2014 system. Statistical analysis used IBM SPSS v.26; chi-square tests, Cramer's V, Mann-Whitney U, Spearman correlation, and binary logistic regression were applied. As histological follow-up was not available in this cytology-only dataset, diagnostic accuracy metrics for individual Bethesda categories are presented as literature-derived reference benchmarks.

RESULTS

Specimen adequacy was 99.5%. Negative for intraepithelial lesion or malignancy was found in 88.3% (n = 25,008) of cases. Squamous epithelial atypia was identified in 9.7% (n = 2751) and glandular abnormalities in 2.9% (n = 823); overall, at least one epithelial abnormality was present in 11.7% (n = 3302) of specimens. HSIL rates rose from 0.3% (<30 years) to 2.7% (>60 years; P < 0.001). Atypical squamous cells of undetermined significance detection nearly tripled from 4.1% (2022) to 11.3% (2024; linear-by-linear χ2 = 285.70, P < 0.001). On multivariate logistic regression, older age (OR = 0.105-0.463 for younger groups versus >60 years), cervical inflammation (OR = 16.890; 95% CI 8.795-32.438), and district of residence were independent associated or relevant factors of cytological HSIL. The atypical squamous cells of undetermined significance-to-SIL ratio was 2.6, below the established upper threshold of 3.0 derived from College of American Pathologists interlaboratory comparison data.

CONCLUSIONS

This is the largest Bethesda 2014-classified cervical cytology cohort reported from Central Asia. Older age and cervical inflammation are the strongest independent risk factors for HSIL. The elevated atypical glandular cell rate (2.8%) and rising atypia detection trends have direct implications for screening policy and human papillomavirus co-testing implementation in Uzbekistan.