Key points are not available for this paper at this time.
Abstract Background Precise and accurate measurements of whole blood glucose levels are important for the diagnosis and treatment of patients suffering from diabetes and hypoglycemia. This study demonstrates the precision of the i-STAT G cartridge with the i-STAT 1 System in arterial, venous, or capillary whole blood in point of care or clinical laboratory settings. Methods The precision of the i-STAT G cartridge was assessed at external clinical sites by intended end users with whole blood specimens across the measurement range of 20 - 700 mg/dL. The study included testing venous whole blood samples from 154 adult (≥ 18 years) patients and arterial whole blood samples from 173 adult patients at three clinical sites. Each whole blood sample was run in duplicate on the i-STAT G cartridge. In addition, 139 capillary whole blood specimens from both adult subjects and neonate subjects (birth - 28 days) were collected and tested for capillary whole blood precision. Duplicate fingerstick draws were collected in capillary tubes from 119 adult subjects at two clinical sites and duplicate heel-stick specimens were collected in capillary tubes from 20 neonate subjects at two clinical sites. Specimens used in this study were prospectively collected venous, arterial, or capillary specimens under an Institutional Review Board approved informed consent or were de-identified leftover venous or arterial specimens. For each specimen type, duplicates from each subject were grouped into subintervals based on their mean values. The subintervals included the glucose medical decision levels of 45 mg/dL, 120 mg/dL and 180 mg/dL. The estimates of the pooled standard deviation (SD) and pooled coefficient of variation (%CV), as well as their respective 95% confidence intervals (CI) were calculated for each subinterval (20-90 mg/dL, 90-150 mg/dL, 150-250 mg/dL, 250-700 mg/dL or 250-400 mg/dL and 400-700 mg/dL, as applicable). The precision estimates for each subinterval were compared against an acceptance criteria of the greater of 3 mg/dL or 5% of the mean (mg/dL) of the subinterval. Results The whole blood precision was found to be less than or equal to 5.00 %CV for all subintervals for each specimen type. For venous specimens, the %CV ranged from 0.35% to 0.43%; for arterial specimens the %CV ranged from 0.26% to 0.46% and for capillary specimens (adult and neonate combined), the %CV ranged from 1.38% to 2.71%. Conclusions The study demonstrated that the i-STAT Glucose test in the i-STAT G cartridge when tested using the i-STAT 1 System delivers precise results within ∼2 minutes for venous, arterial, and capillary whole blood patient specimens across a glucose range of 20 - 700 mg/dL when tested by end users in a point of care or clinical laboratory setting. The studies were funded by Abbott Laboratories.
Thibodeau et al. (Tue,) studied this question.