XÁC ĐỊNH GIÁ TRỊ SỬ DỤNG VÀ ĐÁNH GIÁ ĐỘ TƯƠNG ĐỒNG CÁC XÉT NGHIỆM ĐỊNH LƯỢNG MEN GAN AST, ALT, GGT BẰNG PHƯƠNG PHÁP ĐỘNG HỌC ENZYME TRÊN HAI HỆ THỐNG MÁY ADVIA 1800 VÀ COBAS C702 TẠI BỆNH VIỆN CHỢ RẪY

Abstract

Objective: Liver enzyme assays (AST, ALT, GGT) are critical for diagnosing liver diseases. This study aimed to (1) Verify the performance characteristics (precision, bias, linearity, limit of quantitation) and assess the Sigma metric quality level of AST, ALT, GGT assays on the Advia 1800 and Cobas c702 systems at Cho Ray Hospital; and (2) Evaluate the comparability of these assays between the two systems. Subjects and Methods: A descriptive cross-sectional study was designed. Performance characteristics were verified according to CLSI guidelines: EP15A3 (precision, bias), EP06 (linearity), and EP17A (limit of quantitation). Acceptance criteria were based on Total Allowable Error (TEa) from desirable biological variation (Westgard). Comparability was assessed using 40 patient samples. Results: Both systems met the requirements for precision (CV < 0.33 TEa) and bias (Bias < 0.5 TEa). Linearity was acceptable ($R^2 > 0.99$). For the limit of quantitation (LoQ), the Advia 1800 system failed verification at 3 U/L for all three assays. The Cobas c702 system met LoQ for ALT and GGT, but failed for AST at 5 U/L. All assays had Total Error (TE) < TEa and achieved Sigma levels > 3. The Cobas c702 (Sigma 4.82 - 9.87) demonstrated higher performance than the Advia 1800 (Sigma 3.64 - 8.54). The comparability analysis revealed that the two systems were not comparable at clinically relevant concentrations (TE > TEa). Conclusion: Although the AST, ALT, and GGT assays on both the Advia 1800 and Cobas c702 systems are verified and acceptable for clinical diagnostics, the two systems are not comparable and cannot be used interchangeably for patient monitoring.