A-378 Performance Evaluation of the Roche Cobas® Pulse Point-of-Care Glucometer

Abstract

Background Point-of-care testing (POCT) for glucose monitoring plays a crucial role in enabling timely clinical decisions within healthcare settings. The Roche cobas® pulse is a POCT device for glucose measurement and is compatible with cobas® GLU test strips and cobas® GLU QC kit for internal quality control (Roche Diagnostics GmbH). This new generation glucometer utilises FAD-dependent glucose dehydrogenase (GDH) enzyme technology and is intended to replace the current Roche Inform II system. This study provided a real-world analytical and clinical accuracy evaluation of the Roche cobas® pulse glucometer compared to its predecessor Roche Inform II. Methods Between June to October 2025, we evaluated Roche cobas® pulse glucometer (FAD-GDH electrochemical method) against its predecessor Roche Inform II (mutant quinoprotein GDH electrochemical method) and referencing to central laboratory platform testing on Abbott CI hexokinase glucose method for method comparison, imprecision, linearity and accuracy. Within-day and between-day imprecision were assessed using two QC levels (Low level: 3.3 mmol/L; High level: 17.7 mmol/L). Linearity was verified using commercial linearity material ranging from 1.5 to 29.7 mmol/L. Method correlation studies were performed using 77 patient samples for each comparison (Alinity CI and Inform II) across the reportable range of 0.6 to 33.3 mmol/L. Accuracy was verified using CAP WBGQ-B 2025 EQA samples. Performance was evaluated using Royal College of Pathologists of Australasia (RCPA) allowable performance specifications (±0.4 mmol/L for glucose =5.0 mmol/L or ±8% for glucose >5.0 mmol/L) and Food and Drug Administration (FDA) recommendations of 98% accuracy within ±15% for glucose = 4.2 mmol/L or ±0.83 mmol/L for glucose < 4.2 mmol/L. Results The Roche cobas® pulse showed excellent precision with within-day CV of 1.0% and 0.5% and between-day CV of 1.3% and 0.4% for low and high levels respectively, fulfilling manufacturer claims. Linearity was verified across the tested range (R²=0.9998). Correlation with Abbott Alinity CI showed strong agreement (cobas® pulse = 0.9291[Alinity CI] + 0.2634, R²=0.9979, n=76), with 71 of 76 samples (93.4%) meeting RCPA allowable performance specifications (APS) after excluding 1 outlier. Comparison with Roche Inform II showed excellent correlation (cobas® pulse = 0.9693[Inform II] + 0.013, R²=0.9940, n=77), with 73 of 77 samples (94.8%) meeting RCPA APS. Both comparisons achieved 100% compliance with FDA hospital-use accuracy criteria, exceeding the 98% requirement. Accuracy verification using CAP EQA samples showed all results within ±2 SDI and CAP evaluation criteria (peer group mean ±12.5%). The Roche cobas® pulse glucometer demonstrated analytical equivalence to Inform II, with improved precision, equivalent linearity across the clinical range and comparable accuracy. Conclusion The Roche cobas® pulse glucometer demonstrated excellent analytical performance, meeting all manufacturer's claims, RCPA performance specifications, FDA hospital use recommendation and CAP evaluation criteria. It performed comparably to Roche Inform II and to central laboratory platform on Abbott Alinity, >93% of samples within RCPA APS criteria, confirms its suitability for replacing the Roche Inform II glucometer for point-of-care glucose monitoring in professional healthcare settings. Collectively, these findings reinforce cobas® pulse as a clinically dependable and technologically advanced successor to the Inform II in professional healthcare environments.