Evaluation of the Analytical Specificity and Clinical Application of a New Generation Hospital-Based Glucose Meter in a Dialysis Setting

Abstract

Background: The use of hospital glucose meters is widely established; however, the reliability of glucose meters can vary according to the type of patient group tested. Significant error rates can occur with point-of-care glucose level measurements owing to hematocrit effect and/or chemical interferences associated with drug therapy and patient treatment protocols. In addition, chemical interference with some current glucose meters because of dialysate composition has been observed in patients with renal disease undergoing peritoneal dialysis. The new generation StatStrip glucose meter (Nova Biomedical, Waltham, Mass) has been designed to compensate for interference effects commonly associated with other currently available glucose meters. A previous laboratory evaluation of StatStrip in our hands demonstrated good precision and correlation to the central laboratory hexokinase reference method. The aims of this study were to assess the response of StatStrip to analytical interferences likely to be encountered in hospitalized patients and to evaluate the reliability of StatStrip for application to patients attending a specialized dialysis care center. Background: The interference response of StatStrip was compared to 3 conventional glucose meter technologies: Accu-Chek Aviva (Roche Diagnostics, Mannheim, Germany), Freestyle (Abbott Diabetes, Alameda, Calif), and Elite XL (Bayer, Leverkusen, Germany). Chemical interference factors that were assessed included β-hydroxybutyrate (βHB), bilirubin, lactate, and maltose monohydrate. Interference studies were performed by adding each of the interferants to whole blood at 3 different glucose concentrations for a range of hematocrit values of 26% to 65%. Immediately after analysis on the glucose meters, all samples were centrifuged to obtain plasma for analysis on the reference method Dimension RxL analyzer (Dade Behring, Deerfield, Ill). Methods: Within-run imprecision was studied using whole blood specimens spiked with glucose. A whole-blood specimen, spiked to yield samples with different glucose concentrations, was analyzed for glucose using the 4 strip-meter systems, and the results were compared to those from a reference hexokinase method. Common interferences, including hematocrit, βHB, bilirubin, lactate, and maltose, which have previously been shown to effect measurements from current glucose meter technologies, were tested on each of the 4 strip-meter systems at low, medium, and high blood glucose levels. Whole blood samples from 37 patients in the Nephrology Clinic's dialysis center were analyzed on each meter to determine the suitability of each in this patient care setting. Results: Regression analyses, comparing glucose values from each strip-meter system to the reference hexokinase method on a whole blood specimen, suggested that the StatStrip system's regression statistics, mean difference from the reference method, and percent bias were comparable to or better than similar statistics obtained from the other systems. Interferences studied included hematocrit, βHB, bilirubin, lactate, and maltose. Of the 4 strip-meter systems tested for interference, only the StatStrip system remained within 0.555 mmol/L of their initial value (at a glucose concentration 5.55 mmol/L) after the addition of bilirubin, βHB, lactate, or maltose. Maltose had a strong effect on the Freestyle and Accu-Chek Aviva systems. Hematocrit impacted all meter technologies except the StatStrip. Conclusions: The StatStrip glucose meter gave (within-run) precision comparable to that determined on the other 3 glucometer systems tested. It correlated well with a clinical laboratory reference hexokinase method, was not susceptible to hematocrit, βHB, bilirubin, lactate, or maltose interferences observed in 1 or more of the other blood glucose meters, and should minimize errors that are common to other glucometers. Our results indicate that StatStrip has good clinical reliability when used in a dialysis setting. An important consideration when selecting hospital glucose meters is to ensure that the specificity is optimal for the patient population with minimal interference effects. Maltose, a metabolite of Icodextrin or an additive in dialysis solutions, is a known interferant in certain glucose meter systems, making them unsuitable for use with patients on peritoneal dialysis. The new generation StatStrip glucose meter, which has been designed to compensate for hematocrit and chemical interferences, reduces the likelihood of erroneous results arising from these interference factors that influence current conventional glucose meters.