Comparison of Two Urinalysis Work Cells Particle Recognition, Flagging Efficiency, and Impact on Reflex Rates

Introduction/Overview

This prospective study evaluated the real-world performance of two fully automated urinalysis systems—the Sysmex UN-Series (utilizing fluorescent flow cytometry) and the Beckman Coulter DxU Iris Workcell—in identifying urine-formed elements and flagging abnormal findings. Conducted at Corewell Health William Beaumont University Hospital and Henry Ford Hospital, the study analyzed 272 adult emergency and inpatient specimens over four consecutive days to assess particle recognition accuracy, system agreement rates, and the impact on reflex testing workflows. The objective was to determine whether these automated systems perform equivalently in clinical practice, particularly regarding false-positive flagging rates that drive the need for manual review.

Key Study Findings

• Overall agreement between systems was high for normal specimens: RBC (86.4%), WBC (91.7%), and bacteria (77.3%)
• Normal specimens showed excellent concordance: Probability that a Sysmex-normal specimen would be DxU-normal was 95.8% (RBC), 98.1% (WBC), and 96.2% (bacteria)
• Similar calculations for abnormal screens yielded: RBC (51.9%), WBC (37.5%), and bacteria (64.6%)
• Clinical impact: Higher false-positive rates require additional operator time for image review or manual microscopy verification, affecting laboratory workflow efficiency