Performance Evaluation of Blood Glucose Monitoring Systems (Polaris)

• Accuracy of the blood glucose meter in comparison to a standard reference [ Time Frame: within 10 min up to 300 min after start of experiment ] [ Designated as safety issue: No ]
  After start of the experiment, blood glucose will be measured by means of the test devices and blood will be drawn for assessment of venous blood glucose concentration with a standard laboratory reference method
  
  

• Intra-Assay precision [ Time Frame: within 10 min up to 300 min after start of experiment ] [ Designated as safety issue: No ]
  Blood samples from 5 individuals with different levels of blood glucose were each measured with 10 meters of both investigational devices, the BGStar® and iBGStar®, ten times resulting in a total number of 500 measurements.
  
  
• Hematocrit Interference (Helios) [ Time Frame: within 10 min up to 300 min after start of the experiment ] [ Designated as safety issue: No ]
  One patient provided blood and artificial samples are generated from venous heparinized blood. These blood samples were adjusted to five different BG levels (2.8 mmol/L, 5.6 mmol/L, 8.3 mmol/L, 13.9 mmol/L, 19.4 mmol/L) at five different Hct levels (35%, 40%, 45%, 50%, 55%) to fulfill the range required by the TNO 2001 and tested by all devices.
  
  

The main study (Polaris) investigated blood glucose measuring accuracy and intra-assay precision while the substudy (Helios) investigated potential haematocrit interference of sanofi-aventis BGMSs BGStar® and iBGStar® and partly of competitor BGMSs according to ISO 15197 [1] and TNO 2001 Quality Guideline [2], respectively.

The primary objective of the system accuracy evaluation is to identify the number of blood glucose readings of both sanofi-aventis devices within the ISO 15197 area of acceptable accuracy of ± 20% [DEV%] for blood samples with glucose concentration > 75 mg/dL (4.2 mmol/L) and within the ISO 15197 area of acceptable accuracy of ± 15 mg/dL (± 0.83 mmol/L) for blood samples with glucose concentration < 75 mg/dL (4.2 mmol/L) as compared to the StatStrip® Connectivity reference method. The primary objective of the haematocrit interference evaluation is to demonstrate that the BGStar® and iBGStar® devices meet the Hct interference requirements according to TNO 2001 Quality Guideline when used to measure BG of venous whole blood as compared to a venous plasma reference method (YSI 2300 STAT Plus™) at Hct levels between 35 - 55% in terms of:(1) a mean deviation not exceeding ± 10% [DEV%] for hyperglycaemic glucose concentrations (150 - 180, 250 - 280, 320 - 350 mg/dL) and (2)a mean deviation not exceeding ± 18 mg/dL (1.0 mmol/L) [DEVtotal] for hypo- and normoglycaemic glucose values (50 - 60, 100 - 120 mg/dL) compared to the reference method at Hct levels between 35 - 55%.