Establishment and development of the personalized criteria for microscopic review following multiple automated routine urinalysis systems

Background: This study intends to develop the microscopic review criteria for automated urine chemistry analyzer and integrated urine chemistry and formed-element analyzer.

Methods: A total of 1058 samples were analyzed using chemistry analyzer (Siemens Atlas) for proteins (PRO), blood (BLD) and WBC. Cast, RBC and WBC were analyzed using 4 different instruments, IRIS IQ200, AVE-766, US 2026, and Sysmex UF-1000i. A phase-contrast microscopy was used as reference to evaluate false-negative rate (FNR) and review rate (RR).

Results: The optimized review criteria for Atlas were either PRO 2+, or BLD 2+, or WBC 2+, or specimen from nephrology department. FNR was 4.65% and RR was 40.41%. The optimized criteria for integrated Atlas and IQ200, or AVE-766 or US 2026 or UF-1000i were either BLD≥2+ (>2+ for females), or RBC count ≥ 2 times reference range, or different WBC results between chemistry and formed-element analysis, or PRO ≥ 2+ or CAST>the reference range. One additional rule for integrated Atlas and UF-1000i was different results between BLD and RBC counts. FNR was 1.94%, 2.03%, 1.74%, 1.65%, and RR was 41.09%, 40.12%, 44.86%, 50.39%, respectively.

Conclusions: These specific review criteria ensured a low missed-diagnosis rate and a reasonable workload in practice.