Usefulness of Automated Hematology System Sysmex XN‐3100 for the Quantification of Schistocytes and Early Detection of the Underlying Thrombotic Microangiopathy

Abstract

ABSTRACT Introduction Light microscopy is still the gold standard method for schistocyte counting in the blood smear. Some hematology analyzers can quantitatively determine fragmented red blood cells (FRC) as part of the full blood count. We examined the usefulness of two quantitative schistocyte methods on Sysmex XN‐3100, comparing them with the standard reference method (SCHISTO‐LM). Methods FRC count was performed by fluorescent flow cytometry in the reticulocyte channel. A locally designed method was applied on DI‐60 digital morphology analyzer (SCHISTO‐DI), allowing schistocyte quantification according to ICSH recommendations, using the RBC menu. FRC ≥ 1.0% and schistocytes > 2/high‐power field were considered positive. Measurements were performed in 91 blood samples/peripheral blood smears of 54 patients with a suspicion of thrombotic microangiopathy (TMA). Results Significantly higher FRC ( p = 0.004), SCHISTO‐DI ( p < 0.001), and SCHISTO‐LM ( p < 0.001) were found in TMA‐positive patients compared to TMA‐negative and patients in remission. Stratification of TMA‐positive patients by TMA type revealed no significant differences for all tests. Both positive SCHISTO‐LM and SCHISTO‐DI were recorded in 15/91 samples (sensitivity = 100%, 95% CI = 78.2–100.0 and negative predictive value (NPV) = 100%, 95% CI = 94.4–100.0). False‐negative FRC results were detected in 5/15 schistocyte‐positive samples (sensitivity = 66.7%, 95% CI = 38.4–88.2; NPV = 92.9%, 95% CI = 86.3–96.4). The highest agreement was revealed between SCHISTO‐DI and SCHISTO‐LM ( κ = 0.637). SCHISTO‐DI demonstrated the best AUC for discrimination between TMA‐positive and TMA‐negative patients (0.942; 95% CI = 0.843–0.987), and similar AUC (0.818; 95% CI = 0.700–0.905) as SCHISTO‐LM (0.820; 95% CI = 0.702–0.906) for discrimination between TMA‐positive and patients with TTP/aHUS in remission. Conclusion Despite the slightly lower sensitivity of FRC, the newly designed SCHISTO‐DI method demonstrated excellent sensitivity and NPV for schistocyte detection, concordant with SCHISTO‐LM results.