As required by the Food and Drug Administration (FDA), manufacturers of POC glucose meters test their devices against known interfering substances. Unfortunately, evaluating all known substances is difficult. Similarly, it can be difficult to replicate complex in vivo physiologic variables that contribute to inaccurate meter performance in vitro. The development of autocorrecting POC glucose biosensors provides a means of overcoming these interferences. To produce more accurate glucose measurements, these novel sensors correct for interferences by automatically measuring hematocrit and correcting for reducing/oxidizing substances. The clinical impact of using autocorrecting glucose meters versus noncorrecting devices has been shown to reduce hypoglycemic events and glycemic variability, indicating the direction of POC glucose monitoring technology in the future.
Given the importance of interfering substances and the high risk of critically ill populations, institutions should include the following steps in their implementation process:
Selection of a Comparative Method: Institutions must first define a suitable comparative method. The "definitive" method for glucose testing is isotope dilution mass spectrometry (IDMS); however, few healthcare facilities have access to these higher order analytical methods (16). IDMS alternatives include "true reference methods" and "comparative methods," which are more widely available. Comparative methods, which include clinical laboratory analyzers that use plasma/serum or whole blood with blood gas analyzers, may be more convenient. When deciding on a comparison method, facilities must consider feasibility, accessibility, and performance.
Patient Population Selection: Once a suitable comparison method has been identified, facilities should define appropriate patient populations to validate glucose meter performance. The Clinical Laboratory Standards Institute recommends a sample size of 40 for method comparisons, with samples spanning the analytical measurement range of the device. If testing in critically ill populations is to be performed, laboratories should also include samples from intensive care units and emergency departments, spanning the range of expected specimen types (e,g., arterial, venous, capillary, fingerstick). Ideally, samples should include metadata to determine disease severity and classify patients as critically ill.
Patient Population Selection: Once a suitable comparison method has been identified, facilities should define appropriate patient populations to validate glucose meter performance. The Clinical Laboratory Standards Institute recommends a sample size of 40 for method comparisons, with samples spanning the analytical measurement range of the device. If testing in critically ill populations is to be performed, laboratories should also include samples from intensive care units and emergency departments, spanning the range of expected specimen types (e,g., arterial, venous, capillary, fingerstick). Ideally, samples should include metadata to determine disease severity and classify patients as critically ill.
Evaluation of Interfering Substances: Healthcare facilities should identify interfering substances that glucose meters are likely to encounter. For example, there has been a surge in interest in high-dose ascorbic acid for a variety of diseases, including sepsis, in recent years.
Facilities that anticipate high-dose ascorbic acid therapy for patients should conduct studies to determine if glucose meter performance is compromised and develop alternative workflows for these scenarios. Non-glucose sugars should also be evaluated if testing in neonatal (e.g., galactose) and peritoneal dialysis populations (e.g., icodextrin/maltose) populations is planned. When necessary and feasible, laboratories may consider fabricated samples, particularly if devices are used "off label" and method validation is required.
The implementation of POC glucose monitoring is not as straightforward as one might expect. Glucose monitoring is required to properly administer insulin, one of the most dangerous drugs in clinical use. When implementing new devices, this necessitates greater scrutiny and diligence. Accuracy is critical for safe insulin administration, especially in hospitalized patients. The clinical laboratory's role is to ensure not only regulatory compliance, but also the safe use of in vitro diagnostic testing. Laboratories should thoroughly evaluate point-of-care glucose monitoring systems and be aware of any limitations that may have an impact on patient care.