A standardized and scalable pooled-sample approach for external quality assessment in cervical cytology

Objectives External Quality Assessment (EQA) is essential for standardizing laboratory practices; however, strict requirements for slide homogeneity and stability have limited the implementation of cytology EQA in Vietnam. This study aimed to standardize a pooled-sample preparation process and optimize the Papanicolaou staining protocol for liquid-based cytology (LBC) slide production, ensuring slide homogeneity and stability in accordance with ISO/IEC 17043:2023 requirements for EQA. Methods Residual LBC specimens were pooled to generate cytology slides. An experimental design was applied to optimize fixation and staining durations for Hematoxylin, Orange G (OG), and Eosin Azure (EA). Homogeneity and stability were evaluated according to ISO/IEC 17043:2023 and ISO 13528:2022 guidelines. Results The pooled sampling methodology achieved a mean cellular density of 43,282 ± 1,063 cells per slide, demonstrating high reproducibility (CV=2.46 %). Multivariate linear regression analysis established the following predictive model for specimen quality: Y=27.667 − 0.028X 1 + 0.347X 2 + 0.375X 3 − 0.014X 4 . Based on this model, the optimized procedural protocol was defined as: Absolute Ethanol fixation for 15 min, Hematoxylin staining for 59 s and OG, EA staining for 3 min each. The optimized protocol yielded significantly high specimen quality scores (29.00 ± 0.707) with a further reduction in variability (CV=2.43 %). Statistical analysis confirmed high levels of homogeneity across all specimens and diagnostic pathological sub-groups (Kruskal-Wallis test, p>0.05). Additionally, longitudinal assessment indicated that staining quality maintained stability over a 3-month follow-up period (Friedman test, p>0.05). Conclusions A standardized pooled-sample preparation and optimized Papanicolaou staining protocol was established, ensuring high homogeneity, stability, and staining quality for cytological EQA slide production.