Impact of Cytologist‐Performed Rapid on‐Site Evaluation on Adequacy and Diagnostic Performance in Thyroid Fine Needle Aspiration Cytology

INTRODUCTION: Rapid on-site evaluation (ROSE) improves specimen adequacy in thyroid fine-needle aspiration biopsy (FNAB), but the conventional cytopathologist-attended model may be difficult to sustain in high-volume centers. Cytologist-performed ROSE has been proposed as a practical alternative. This study aimed to evaluate the impact of cytologist-performed ROSE on specimen adequacy, diagnostic category distribution, number of needle passes, number of prepared slides, and repeat cytology rates in thyroid FNAB. METHODS: We retrospectively compared thyroid FNAB cases performed during two sequential periods at a single institution: a pre-ROSE period, during which FNAB was performed without on-site cytological assessment, and a post-ROSE period, during which ROSE was performed by a cytologist who had completed a structured three-month training program under the supervision of a pathologist. Specimen adequacy, Bethesda System diagnostic categories, number of needle passes, number of prepared slides, and repeat cytology rates were compared between the two periods. RESULTS: A total of 916 patients were included, comprising 262 cases in the pre-ROSE period and 654 cases in the post-ROSE period. Specimen adequacy increased significantly from 49.2% in the pre-ROSE period to 85.8% in the post-ROSE period (p < 0.001). Bethesda category I diagnoses decreased from 50.4% to 14.8%, whereas Bethesda category II diagnoses increased from 39.7% to 73.5% after ROSE implementation (p < 0.001). The mean number of needle passes decreased from 3.27 ± 0.49 to 2.12 ± 0.92, and the mean number of prepared slides decreased from 11.86 ± 5.39 to 7.21 ± 3.13 (p < 0.001 for both). The repeat cytology rate also decreased from 8.8% to 5.2% (p = 0.043). CONCLUSION: Cytologist-performed ROSE was associated with significantly improved specimen adequacy, a marked reduction in nondiagnostic thyroid aspirates, fewer needle passes and prepared slides, and a lower repeat cytology rate. These findings suggest that a structured cytologist-performed ROSE model may be a feasible and effective approach for improving specimen adequacy in thyroid FNAB, particularly in high-volume centers with limited cytopathology resources.