Glucose meters have a rich history. Dating back to 1970, the first glucose meter is thought to have been the Ames Reflectance Meter (ARM), a portable meter weighing around three pounds and designed to detect the colorimetric signal formed on the Detrostix. Over the next 20 years the technology evolved into the small, lightweight, and portable meters we know today, sold in pharmacy and retail centers. Within this same timeframe that glucose meters migrated into the hospital, along the way the meters were redesigned to incorporate familiar features such as quality control (QC) lockout, operator login/management, and interface with the laboratory and hospital information systems. Fast forward to the present day, and glucose meters are ubiquitous in the hospital, found in almost every unit and used for a variety of reasons—from identifying hypoglycemia in neonatal wards to guiding intravenous insulin titration in tight glycemic control protocols aimed at controlling a patient’s blood glucose concentration.
Irrespective of their look and their location-dependent (home vs. hospital) features, glucose meters in the home and hospital share a common principle of measurement. When whole blood is applied to the test strip, enzymes embedded within sense the glucose and generate either an electrical or colorimetric signal which is directly proportional to the activity of glucose. This activity is subsequently converted into concentration and reported as plasma equivalents. Virtually all meters utilize either glucose oxidase or glucose dehydrogenase enzyme as the sensor.
Glucose meters are routinely used in the management of dysglycemia—disorders of blood sugar metabolism characterized by hypoglycemia and hyperglycemia. The most prevalent form of dysglycemia is diabetes mellitus (DM), today seen as a group of metabolic diseases characterized by hyperglycemia resulting from defects in insulin secretion, insulin action, or both. Diabetes is widespread in the population, with the Centers for Disease Control and Prevention (CDC) estimating that around 25.6 million adults (age >20 years) have diabetes.1 Moreover, diabetes is a significant cause of morbidity and mortality. The goal of diabetic treatment is glycemic control with lifestyle modification (diet and exercise) and glucose lowering medications such as sulfonylueras, biguanides, and insulin.
The American Diabetes Association (ADA) recommends regular testing or self monitoring of blood glucose (SMBG) for patients using multiple insulin injections or insulin pump therapy.2 For patients using less frequent insulin, noninsulin therapy, or medical nutrition therapy alone, SMBG may be useful as a guide for success. The ADA does not at this time recommend use of SMBG for patients with type 2 diabetes due to lack of conclusive evidence that it is linked to improved outcomes. Nevertheless, it is routine practice for clinicians treating patients with type 2 DM to prescribe SMBG as a tool to help patients manage their blood glucose in the home setting. For evidence of this practice, one need look only at the size and growth of the consumer blood glucose meter market.
In the hospital setting, both diabetic and non-diabetic patients can experience another form of dysglycemia called stress hyperglycemia—an acute hyperglycemia secondary to a stress on the body from illness, surgery, or trauma. Stress hyperglycemia has been demonstrated to lead to poorer outcomes.3 The management of stress hyperglycemia through the use of “tight glycemic control” protocols sometimes using insulin infusion is controversial and currently being debated. However, irrespective of the debate on how tightly we control a patient’s blood glucose concentration during his or her stay in the hospital, it can be argued that all patients are on some sort of glycemic control while in the hospital. This practice frequently incorporates glucose meters in the management of inpatient normoglycemia.
There is increasing concern about the reliability of glucose meters in the hospital, especially in critical care settings. This concern stems from observed poor performance in the presence of interferences common in hospitalized patients. This article will review the challenges associated with using glucose meters in the hospital and discuss the current performance expectations from a regulatory and clinical perspective.
Challenges associated with using glucose meters in the hospital
Glucose meters provide a tool to rapidly measure glucose. Low sample volume plus rapid results allow for frequent serial measurements which in turn allow for more aggressive management of dysglycemia. Despite the many advantages, there remain challenges associated with the measurement of glucose at the hospital bedside. Factors and or interferences reported in the literature to affect the quality of glucose meter results are summarized in Table 1.
Table 1. Factors or interferences reported to affect glucose meters
Operator dependent errors are thought to represent the most significant source of glucose meter errors.4,5 Commonly reported operational errors have included, but are not limited to, the use of expired test strips, mishandling of the test strips leading to environmental exposure, use of incorrectly coded strips, and strip misdosing (application of too much or too little blood). Manufacturers have addressed these issues through implementation of barcoded vials and/or test strips that generate appropriate error codes, provision of individually wrapped test strips to address strip handling problems, modification of specimen application technology such as capillary action filling, and on-meter application to minimize strip misdosing. These improvements have reduced, but not eliminated, the influence of operational errors in the provision of a quality glucose meter result.
Glucose test strips have been demonstrated to be liable to changes in temperature, humidity, and altitude, which can affect the quality of glucose meter results. However, within a temperature-controlled hospital with staff dedicated to monitoring test strip storage, these environmental factors are reduced. Nevertheless, glucose meter testing in harsh climates can continue to present challenges. For example, a recent study by Cembrowski et al evaluated the seasonal variations (e.g., temperature and humidity) on glucose results measured in five hospitals in Edmonton, Alberta. Differences between 5% and 10% were reported for glucose measured during the winter and summer months, with the glucose meter results being negatively correlated with the outside temperature.6 Another potential environmental interference may arise from the disinfection of glucose meters, as reported in a study demonstrating a significant overestimation of glucose concentration hours after disinfecting a GO-based photometric meter with a hydrogen peroxide containing solution.7 Meters and test strips should be stored and used according to manufacturers’ recommendations.
A variety of electrochemically active or structurally similar exogenous substances have been reported to interfere with point-of-care glucose meters. Examples include acetaminophen, ascorbic acid, and dopamine. Patients may be at considerable risk given that polypharmacy is common in the hospitalized patient. One study estimates that patients in the ICU on average may have 13 to 22 different drugs administered to them.8 The most notable exogenous interference of late is that caused by maltose, primarily affecting glucose meters utilizing a glucose dehydrogenase enzyme coupled to the co-factor pyrroloquinoline quinine (PQQ). Thirteen deaths between 1997 and 2006 were reported to the FDA based on insulin overdose due to overtreatment of false hyperglycemia caused by maltose.9
Hospitalized patients, particularly those in the intensive-care setting, have a wide range of medical problems such as hypotension, anemia, and acidemia. Each has been demonstrated to affect glucose meter results. Hypotension leads to changes in peripheral circulation that can compromise the integrity of a capillary specimen. Hematocrit abnormalities (anemia and polycythemia) are quite common in the hospital and have been demonstrated on some meters to have an inverse effect on glucose meter results; that is, patients with anemia can have falsely high readings and patients with polycythemia can have falsely low readings.10-13 Acidemia has been shown to lead to glucose underestimation on some glucose oxidase meters. Pathological elevations in endogenous substances such as uric acid and trigylercides have also been shown to interfere with glucose meters.14 High levels of uric acid have been shown to falsely lower glucose results when analyzed with primarily glucose oxidase based meters. High triglyceride levels have been reported to influence meter accuracy because they can cause significant volume displacement.14 Patients with high oxygen tensions (pO2) secondary to oxygen supplementation can have inaccurate glucose meter measurements. Dungan et al15 reported that pO2 levels >100 mm Hg, as seen with surgical patients or patients on oxygen therapy, can also falsely lower glucose concentrations with some glucose oxidase based meters.
Relative performance of glucose meters
When evaluating the analytical performance of glucose meters it is important to recognize that there are a variety of methods available to measure glucose, each with a different degree of accuracy and precision. The relative analytical performance of the different glucose assays can be conceptualized as a measurement vector (Figure 1) with the positions of the different assays placed on the vector some relative distance to the definitive or “gold standard” measurement of glucose.
Figure 1. Glucose measurement vector. Isotope dilution mass spectrometry (IDMS) is the definitive method for glucose determination. An example of a true reference method for glucose is a hexokinase (HK) method aligned to IDMS using a percholoric acid (PCA) deproteinated sample. Point of care meters are routinely compared to central laboratory methods such as plasma hexokinase and/or blood gas analyzers to assess analytical performance.
Glucose meters are generally thought to be less accurate and precise than central laboratory and blood gas methodologies and therefore would be placed to the right of these methods on the measurement vector shown in Figure 1. Manufacturers of hospital glucose meters have made significant design impr ovements which have arguably increased quality, mainly by decreasing operator-dependent errors. Nevertheless, there are continued calls for improvements in the accuracy and precision of the measurement. But to what extent? How well should glucose meters be expected to perform and relative to which method?
Performance expectations for glucose meters
In order to determine the accuracy of glucose meters, ideally one compares the results from the meter to a reference or comparative method from the same sample. The reference or comparative method should have documented accuracy, low susceptibility to interferences, and high precision. Routinely, glucose meters are compared to central laboratory plasma hexokinase and sometimes blood gas analyzers. When determining the accuracy of glucose meters, it is important to understand the limitations of the methodology to which the results are being compared.
To interpret the results (i.e., absolute or percent bias) from studies designed to evaluate the accuracy of glucose meters, one may turn to various organizations such as the International Standards Organization (ISO), the Clinical and Laboratory Standards Institute (CLSI), and the American Diabetes Association (ADA). Each has proposed various standards which manufacturers, regulatory bodies, and end users can use as a guide in the development, oversight, and implementation of glucose meters (Table 2).
Table 2. Table 2. Various published performance expectations (relative to a reference method) for glucose meters
Meters intended for use in the home or in the hospital are presently evaluated by the FDA using the ISO15197 criteria. However, given the increased use of glucose meters in the hospital and the observed interferences common in hospitalized patients, many are calling into question the safety and application of the current performance expectations. The FDA, ISO and CLSI are currently re-examining performance expectations for glucose meter use in both community and hospital environments. Representation for these committees includes specific government bodies, professional organizations, the diagnostic industry, healthcare professionals, and end users. Establishment of new performance expectations will be achieved through consensus accounting for current technological limitations, different clinical needs for community versus hospitalized patients, and the balance between provision of a service and patient safety.
Glucose meters do provide a valuable testing service at the bedside and in the home. There are challenges associated with the use of glucose meters in a hospital given the observation that interferences and pathophysiological changes common in the critically ill patient can affect glucose meter results. The effect of these interferences varies and will depend on the technology of the device. It is critical therefore to evaluate glucose meters with these interferences in mind. Ultimately, one must determine if the technology meets clinical need by critically evaluating a technology within the context of where it will be implemented.
- National Diabetes Fact Sheet. Centers for Disease Control and Prevention. 2011.
- Standards of medical care in diabetes—2011. Diabetes Care. 2011;34(Suppl 1):S11-61.
- Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet. 2009:373.
- Ginsberg BH. Factors affecting blood glucose monitoring: sources of errors in measurement. J Diabetes Sci Technol. 2009;3(4):903-913.
- Nichols JH. Blood glucose testing in the hospital: error sources and risk management. J Diabetes Sci Technol. 2011;5(1):173-177.
- Cembrowski GC, Smith B, O’Malley EM. Increases in whole blood glucose measurements using optically based self-monitoring of blood glucose analyzers due to extreme Canadian winters. J Diabetes Sci Technol. 2009:3(4):661-667.
- Desmeules P, Ethier J, Allard P. Disinfectant wipes containing hydrogen peroxide induce overestimation of glucose results obtained with Lifescan SureStep Flexx® glucose meter. Clinical Biochemistry. 2010;43(18):1472-1474.
- Biswal S, Mishra P, Malhotra S, Puri GD, Pandhi P. Drug utilization pattern in the intensive care unit of a tertiary care hospital. Journal of Clinical Pharmacology. 2006;46(8):945-951.
- Schultz DG. FDA public health notification: potentially fatal errors with GDH-PQQ glucose monitoring technology. U.S. Department of Health and Human Services. 2009.
- Balion C, Grey V, Ismaila A, Blatz S, Seidlitz W. Screening for hypoglycemia at the bedside in the neonatal intensive care unit (NICU) with the Abbott PCx glucose meter. BMC Pediatrics. 2006;6:28.
- Dimeski G, Jones BW, Tilley V, Greenslade MN, Russell AW. Glucose meters: evaluation of the new formulation measuring strips from Roche (Accu-Chek) and Abbott (MediSense). Annals of Clinical Biochemistry. 2010;47(Pt 4):358-365.
- Karon BS, Griesmann L, Scott R, et al. Evaluation of the impact of hematocrit and other interference on the accuracy of hospital-based glucose meters. Diabetes Technology & Therapeutics. 2008;10(2):111-120.
- Mann EA, Salinas J, Pidcoke HF, et al. Error rates resulting from anemia can be corrected in multiple commonly used point-of-care glucometers. The Journal of Trauma. 2008:64(1):15-20; discussion 20-11.
- Bode B. The accuracy and interferences in self-monitoring of blood glucose. U.S. Endocrine Disease. 2007:47-48.
- Dungan K, Chapman J, Braithwaite SS, Buse J. Glucose measurement: confounding issues in setting targets for inpatient management. Diabetes Care. 2007:30(2):403-409.
- (ISO) ISO: ISO15197: In vitro diagnostic test systems—requirements for blood-glucose monitoring systems for self-testing in managing diabetes mellitus. 2003.
- Point-of-care blood glucose testing in acute and chronic care facilities; Approved guideline—second edition. Vol. 22. No. 17. Clinical and Laboratory Standards Institute (CLSI). 2002.
- Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. The National Academy of Clinical Biochemistry – Laboratory Medicine Practice Guidelines 2011.
- ADA. Self-monitoring of blood glucose. Diabetes Care. 1996;19(Suppl.1):S62-66.
T. Scott Isbell, PhD, DABCC, FACB, is the Director of Medical and Scientific Affairs for North America at Waltham, Massachusetts-based Nova Biomedical. Martha E. Lyon, PhD, DABCC, FACB, is a member of the Department of Pathology and Laboratory Medicine, Royal University Hospital, Saskatoon Health Region, Saskatoon, Saskatchewan, Canada.