摘要
Background: Despite many reports on the diagnostic yield of cytology from endoscopic ultrasound-guided fine-needle aspiration(EUS-FNA), inter-slide differences are unknown. This prospective study aimed to compare diagnostic yield and cellular characteristics of bottom slides(BS) and top slides(TS) from EUSFNA cytology performed without an on-site cytopathologist. Methods: In patients with suspected pancreatic cancer on previous imaging explorations, a single endoscopist performed EUS-FNA and obtained 2 sets of cytology slide(8 BS and 8 TS), 1 cellblock slide, and 1 biopsy slide. Both slide sets were randomly assigned. A cytopathologist with more than 10 years of expertise in pancreatic cytopathology blindly inspected and compared two slide sets. Results: In total, 73 specimens [42 head(57.5%), 16 body(21.9%), and 15 tail(20.5%)] were acquired for final analysis. Seventy-one cases were finally diagnosed with pancreatic cancer. The sensitivity and specificity of BS were 80.3% and 100.0%; and of TS 78.9% and 100.0%, respectively. In analyzing inter-slide difference, 66 cases(90.4%) showed consistent results between BS and TS. However, seven(9.6%) were positive only in one slide sets(4 BS and 3 TS). The proportions of specimens more than moderate and high cellularity were 75.3% and 60.3% in both slide sets( P > 0.99), and the proportion of artifact-free sets were 50.7%, and 52.1% for the BS and TS, respectively( P = 0.869). Conclusions: Although BS and TS exhibited highly consistent diagnostic yields in cytologic smears from EUS-FNA, the proportion of inter-slide discordance is clinically considerable. Both slide sets need to be examined if there is no on-site cytopathologist.
Background: Despite many reports on the diagnostic yield of cytology from endoscopic ultrasound-guided fine-needle aspiration(EUS-FNA), inter-slide differences are unknown. This prospective study aimed to compare diagnostic yield and cellular characteristics of bottom slides(BS) and top slides(TS) from EUSFNA cytology performed without an on-site cytopathologist. Methods: In patients with suspected pancreatic cancer on previous imaging explorations, a single endoscopist performed EUS-FNA and obtained 2 sets of cytology slide(8 BS and 8 TS), 1 cellblock slide, and 1 biopsy slide. Both slide sets were randomly assigned. A cytopathologist with more than 10 years of expertise in pancreatic cytopathology blindly inspected and compared two slide sets. Results: In total, 73 specimens [42 head(57.5%), 16 body(21.9%), and 15 tail(20.5%)] were acquired for final analysis. Seventy-one cases were finally diagnosed with pancreatic cancer. The sensitivity and specificity of BS were 80.3% and 100.0%; and of TS 78.9% and 100.0%, respectively. In analyzing inter-slide difference, 66 cases(90.4%) showed consistent results between BS and TS. However, seven(9.6%) were positive only in one slide sets(4 BS and 3 TS). The proportions of specimens more than moderate and high cellularity were 75.3% and 60.3% in both slide sets( P > 0.99), and the proportion of artifact-free sets were 50.7%, and 52.1% for the BS and TS, respectively( P = 0.869). Conclusions: Although BS and TS exhibited highly consistent diagnostic yields in cytologic smears from EUS-FNA, the proportion of inter-slide discordance is clinically considerable. Both slide sets need to be examined if there is no on-site cytopathologist.
