Accuracy evaluation of two systems for self-monitoring of blood glucose following ISO 15197.

Abstract

Today, an increasing number of systems for self-monitoring of blood glucose (SMBG) from different manufacturers are available. The ISO (International Organization for Standardization) 15197 standard stipulates requirements for SMBG systems. Manufacturers often use this standard to obtain the CE (Conformite Europeene) mark for their system, which is, in the European Union, required before the market approval. Regarding system accuracy, the following minimum criteria are defined in ISO 15197:20031 95% of a system’s measurement results shall fall within ±0.83 mmol/l (±15 mg/dl) of the comparison measurement results at blood glucose (BG) concentrations <4.2 mmol/l (<75 mg/dl) and within ±20% at BG concentrations ≥4.2 mmol/l (≥75 mg/dl). Recently, a revision was published (ISO 15197:2013)2 with more stringent limits. Following ISO 15197:2013, at least 95% of a system’s measurement results shall fall within ±0.83 mmol/l (±15 mg/dl) of the comparison measurement results at BG concentrations <5.55 mmol/l (<100 mg/dl) and within ±15% at BG concentrations ≥5.55 mmol/l (≥100 mg/dl), the evaluation shall be performed with 3 reagent system lots, and each tested lot shall fulfill minimum accuracy requirements. Mandatory compliance to ISO 15197:2013 is recommended after a transition period of 36 months. In this study, system accuracy of 2 systems with CE mark manufactured in China was assessed with 1 reagent system lot each following procedures and criteria of ISO 15197:2003. In addition, more stringent accuracy limits of ISO 15197:2013 were applied. The study was conducted in compliance with the German Medical Devices Act at the Institut fur Diabetes-Technologie Forschungs- und Entwicklungsgesellschaft mbH an der Universitat Ulm, Ulm, Germany on behalf of Roche Diagnostics GmbH, Mannheim, Germany. The study protocol was approved by the Ethics Committee and the Federal Institute for Drugs and Medical Devices. Both systems (“An Wen” [System 1] and “An Zhun” [System 2] blood glucose monitoring systems, Changsha Sinocare Inc, Changsha, China) were bought in a drugstore in China by the study’s customer and provided to the study site. For each system, 200 data points obtained on 100 capillary blood samples from different participants were evaluated. A glucose oxidase laboratory method (YSI 2300 STAT Plus™ glucose analyzer, YSI Inc, Yellow Springs, OH, USA) was used for comparison measurements. Regular internal and external quality control measures were performed to confirm trueness and precision of the comparison method. Accuracy criteria of ISO 15197:2003 were fulfilled by both systems with 96% (System 1) and 95% (System 2) of results within the required limits (Figure 1). Only 89.5% (System 1) and 92.5% (System 2) of results were within the limits of ISO 15197:2013. The relative bias according to Bland and Altman3 was −3.1% for System 1 and −0.5% for System 2. Figure 1. Difference plots of system 1 and system 2 with 1 reagent system lot each. For each system, absolute differences between the system’s measurement results and the mean comparison measurement results (200 data points, duplicate measurements on a ... Both systems fulfilled accuracy requirements of ISO 15197:2003 with the tested reagent system lot. The systems did not fulfill the more stringent criteria of the recently published ISO 15197:2013. Since only 1 reagent system lot was evaluated, an overall assessment following ISO 15197:2013 was not performed.