Accuracy Assessment of a Blood Glucose Monitoring System for Self-Testing with Three Test Strip Lots Following ISO 15197:2013/EN ISO 15197:2015

The International Organization for Standardization (ISO) standard 15197:2013 and the European harmonized version EN ISO 15197:2015 are used to establish guidelines around acceptable performance for blood glucose monitoring systems (BGMSs). The minimum accuracy performance criteria are 95% of the system’s results shall fall within either ±15 mg/dL of the average measured values of the reference method at blood glucose (BG) concentrations <100 mg/dL or within ±15% for BG concentrations ≥100 mg/dL. In addition, 99% of the individual BG results shall fall within zones A and B of the consensus error grid.¹

The Accu-Chek® Instant BGMS (Roche Diabetes Care, GmbH, Mannheim Germany) has a simple-to-use 1-button meter, has a Bluetooth® Low Energy (BLE) communication capability in order to transfer data to a mobile device, and uses a new test strip architecture. Especially the latter provides many improvements that are not immediately visible to the user (eg, a high level of system accuracy). The study was performed at NB Research, Inc in Indianapolis, Indiana, USA, between January and February 2016. The study incorporated good clinical practices (GCPs), and the study protocol was approved by an Institutional Review Board.

Meters and test strips were provided by the manufacturer (Roche Diabetes Care, Inc, Indianapolis, IN, USA) since the meter and test strips are not yet available for purchase. Systems were set and maintained according to the manufacturer’s instructions and daily control measurements were performed.

Following the ISO 15197:2013/EN ISO 15197:2015 guidelines, capillary fingertip blood samples from 100 different subjects were measured in duplicate across 3 different lots of test strips. Comparison measurements were performed at Roche Diabetes Care, Inc, with a hexokinase laboratory method (cobas® 6000 c501, Roche Diagnostics GmbH), which is traceable to the National Institute of Standards and Technology standard reference material.

The differences between the BG meter result and the comparison measurement were calculated, and the number of values within the limits of ISO 15197:2013/EN ISO 15197:2015 was determined. The relative bias was also calculated according to Bland and Altman.²

All 3 tested lots of the system comply with the accuracy criteria described in ISO 15197:2013/EN ISO 15197:2015 (Figure 1).

Figure 1.

System accuracy for each individual lot showing the absolute difference between the BGMS results and the comparison measurement results. For each lot, 100 samples measured in duplicate (200 data points) are shown. Dotted lines indicate the ±15 mg/dL/±15% for glucose concentrations <100 mg/dL/≥100 mg/dL per the ISO 15197:2013/EN ISO 15197:2015. Standard and solid lines indicate the ±10 mg/dL/±10% for glucose concentrations <100 mg/dL/≥100 mg/dL.

Each lot was analyzed independently. For BG concentrations <100 mg/dL, 100% of the results were within ±15 mg/dL, 100% within ±10 mg/dL, and depending on the lot, 90.7%, 94.4%, and 98.1% were within ±5 mg/dL of the comparison method measurement results. For BG concentrations ≥100 mg/dL, 100% of the results were within ±15%, and for each lot 99.3% were within ±10%, and depending on the lot, 70.5%, 80.8, and 82.2% were within ±5%. In addition, 100% of the results fell within zone A of the consensus error grid. The relative bias according to Bland and Altman² ranged from −2.5% to −0.5%. This study demonstrated that the BGMS met and exceeded the ISO 15197:2013/EN ISO 15197:2015 accuracy requirements.