Glycemic control protocols in the hospital vary depending on the acuity of the patient, the degree of dysglycemia, and the target blood glucose concentrations, but central to any protocol is the requirement to monitor circulating concentrations of glucose, sometimes frequently, as is the case when utilizing intravenous insulin. To enable this, many facilities use hand-held glucose meters, which provide rapid quantitation of glucose using small volumes (less than 10µL) of capillary, venous, arterial, or neonatal heel stick whole blood. The minimal sample volume required, rapid analysis time, ease of use, and portability make these glucose meters an ideal tool to help manage dysglycemia. Thus, glucose meters are used throughout hospitals encompassing patients presenting in the ED, stable patients on general medical floors, and patients in surgical intensive care who have just undergone cardiothoracic surgery.
However, on January 13, 2014, the New York State Department of Health issued a letter to all laboratory directors stating, “to date, FDA has approved (or cleared) glucose meters for the quantitative measurement of whole blood glucose for use by health care professionals or lay users as an aid in monitoring the effectiveness of a diabetes control program,” and that “laboratories that use glucose meters for purposes or in populations beyond the Intended Use in the package insert or device manual are considered to be engaging in off-label use.”1 Moreover, “off-label uses of glucose meters include the monitoring of glycemic control of non-diabetic patients in hospitals; use on critically ill patients; and use at health fairs and other community events to screen the public for diabetes.” The letter continues by pointing out, “…in the event of such off-label use, the glucose meter defaults to high complexity and the laboratory must meet CLIA requirements for high complexity testing.”
Approximately 10 months later on November 21, 2014, the Centers for Medicare and Medicaid Services (CMS) issued a similar letter reminding all state survey agency directors that failure to follow manufacturers’ instructions for waived devices constitutes off-label use, defaults the classification to high complexity, and requires labs to meet all high-complexity testing requirements.2 The Center for Improvement in Healthcare Quality (CIHQ) issued an accreditation alert on December 9, 2014, and the College of American Pathologists followed suit with an eAlert to all of its accredited laboratories on January 2, 2015 (see FIGURE 1 for a complete timeline).3 Consequently, meeting facility point-of-care glucose testing needs while remaining compliant can be a challenge.
Glucose Monitoring Options
There are numerous analytical methods available for measuring glucose at the point of care, including cartridge-based approaches, systems that use test strips or whole blood electrodes, and devices that employ spectrophotometric cuvettes. Systems that comprise test strips, along with handheld meters (commonly referred to as blood glucose monitoring systems), are the most widely used at the point of care. There are four brands available in the US and all are FDA-designated as waived devices. Sites performing waived glucose meter testing must, at a minimum, follow the manufacturer’s instructions (as provided in the package insert and instructions for use manual) and adhere to the intended use statement and any device limitations.
The intended use statement is important because it informs testing personnel how the results should be used (ie, screening versus diagnosis), the patient populations for which the device is suitable, and the settings or situations in which the device should not be used (ie, contraindications). A comparison of intended use statements is provided in TABLE 1.
The Practical Dilemma
When considering use of these systems, one must answer the following questions:
1. Are the glucose meters being used only as a means to monitor the effects of diabetes (ie, are the results being used to manage a diabetic patient?)
2. Are the glucose meters being used on critically ill patients (eg, are the glucose meters being used in the intensive care setting?)
An answer of yes to either of these questions, with more emphasis placed on the latter, is an indication that the glucose meter is being used off-label. The exception is for users of the Nova Biomedical StatStrip Glucose, which was cleared by FDA in May 2015 “for use in the quantitative determination of glucose in venous whole blood, arterial whole blood, neonatal heel stick, and neonatal arterial samples throughout all hospital and all professional health care settings.”4 This intended use label includes patients that are commonly thought to represent the critically ill.
With that exception aside, using glucose meters on critically ill patients constitutes off-label use of the device; thus, the device defaults to a high-complexity classification, and labs must meet all high-complexity testing requirements, meaning that only certain personnel are qualified to operate the meter according to CLIA ’88 regulations (see TABLE 2), and the laboratory must fully validate the off-label use.
Most often, those who perform glucose meter testing are medical assistants, nurses with 2-year associate degrees, and other relatively low-level technical personnel. These staff members no longer would be qualified to perform the testing if the device defaults to high complexity. In effect, the glucose meter becomes a laboratory-developed test, with the clinical laboratory making and having to substantiate performance claims (covering analytical measurement range, precision, sensitivity, specificity, etc) for the device in a given patient population and for specific specimen types.
Defining Critically Ill
At the center of this issue is how to define critically ill. Presently, this task is left to each institution, as regulatory and accrediting bodies (eg, FDA, CMS, CAP, TJC) have not defined the term for the purposes of glucose meter testing, leading to a great deal of confusion among labs when attempting to determine if they are performing off-label testing. Definitions and interpretations vary, with some institutions defining critically ill by location (eg, intensive care units) and some defining the term with a set of conditions. What is more important is to define critically ill within the context of how the glucose meter performs and with regard to conditions that can affect glucose meter results. Sources of errors with glucose meters are wide ranging:
- Extremes of hematocrit are a well-documented source of error for glucose meters. The observed effect is an inverse relationship; states of low hematocrit (anemia) may falsely elevate glucose meter results and vice versa. Manufacturers have addressed this by incorporating measurement of hematocrit and subsequent correction (Nova Biomedical) or electrochemical compensation of hematocrit (Abbott and Roche). Non-correcting hematocrit meters continue to demonstrate error in patients with severe anemia, such as those who are severely burned.
- Exogenous sources of error for glucose meters can include various drugs (eg, ascorbic acid, vasopressors, maltose). Of particular note, ascorbic acid administered intravenously as part of a resuscitation protocol for burn patients has been shown to interfere with glucose meters and to lead to erroneous results. Manufactures have evaluated many compounds and drugs to determine their effect on the measurement of glucose, but it is impractical to test every drug and metabolite, so caution is necessary in polypharmacy settings, such as the ICU. In these situations, test strip methods that can adjust for electrochemical noise are preferred.
- Any condition that results in impaired peripheral circulation can confound results when measuring whole blood glucose in fingerstick capillary samples. Thus, conditions that affect hemodynamics, particularly those with effects on peripheral vascular perfusion, should be considered in the definition of critically ill. In this setting, avoid capillary specimens; instead use arterial or venous whole blood to dose the meter or send samples to the central lab for measurement of glucose.
Proposed Definitions of Critically Ill
Essentially, there are three approaches for defining critically ill: by location, by conditions, or by some combination of the two. Defining the term using location has the advantage of easily identifying patients in a particular unit or location within the hospital. The disadvantage is that all patients in that unit or location will require their blood glucose to be measured by high-complexity testing personnel, and capillary blood testing will be restricted (unless validated).
Alternatively, a cohort of patients within a setting, such as the ICU, can be defined as critically ill for the purposes of glucose meter testing. Using this method, one possible definition of critically ill includes those patients requiring invasive vascular (hemodynamic) monitoring. This definition is supported by a 2013 publication in which a group of experts consisting of intensivists, an endocrinologist, and laboratory medicine professionals developed the recommendation in TABLE 3 regarding blood sampling:5
The strength of this approach is that it allows for relatively easy identification of those patients for whom a glucose meter or a capillary specimen should not be used. Other proposed definitions have included the use of vasopressors. However, patients requiring vasopressors are usually in critical condition and tend to have unstable hemodynamics.
Definitions should be developed in collaboration with clinical colleagues, preferably in a multidisciplinary group setting with all stakeholders present because the definition will affect multiple areas and can have significant impact on nursing workflows. The most important priority is to develop a definition that best serves patient care. Therefore, the goal is to find a definition that maximizes patient care and outcomes while meeting testing requirements and regulations.
Once the definition for critically ill is determined, a decision must be made to either validate the meter or restrict its use according to the definition. This decision is informed by determining the number of patients that would meet the definition of critically ill and by the amount of resources the facility has to validate the testing and meet high-complexity testing requirements. Restricting use of the device and requiring glucose measurements to be performed in the central lab for those patients meeting the definition of critically ill is advantageous from a management perspective. Testing in the central lab provides the most accurate measure of glucose, requires no additional validations, and assures that testing personnel are highly-qualified medical technologists. Potential disadvantages are increased blood loss due to frequent blood draws for central lab measurements, increased turnaround time, and a potentially increased risk of infection due to frequent line access.
If point-of-care glucose testing is permitted on those patients meeting your definition of critically ill, then the following steps must be completed:
1. Determine permissible specimen types
2. Develop and execute a laboratory validation protocol for those specimen types
3. Modify glucose meter standard operating procedure (SOP)
4. Identify operators meeting high-complexity testing
5. Document operator qualifications. Note that per CAP,
copies of licenses are insufficient; sufficient documentation includes copies of transcripts and diplomas
6. Retrain/certify operators
7. Implement monitoring systems to ensure the new SOP is adhered to
Central to this process is the development and execution of laboratory validations. Guidance is provided in the Clinical and Laboratory Standards Institute (CLSI) document titled: POCT12-A3 Point-of-Care Blood Glucose Testing in Acute and Chronic Care Facilities; Approved Guideline—Third Edition.7 Laboratory evaluations should include an assessment of accuracy (relative to central laboratory methods), precision, analytical measurement range/linearity, sensitivity, and specificity. Ideally, accuracy should be determined using a split-specimen approach, which compares the measurement of glucose on the meter to that achieved with central laboratory plasma methods. Although this is fairly straightforward with arterial and venous specimens, it is more difficult with capillary specimens, given the smaller sample volume routinely collected during a fingerstick. For this reason and for the analytical concerns listed above, some institutions opt not to validate capillary blood, but rather restrict its use for patients meeting the definition of critically ill.
Glucose meters have proven to be a useful tool in the management of a patient’s glycemic state. They do, however, have certain analytical and labeling limitations that must be addressed in order to provide optimal patient care and satisfy current regulations. Manufactures that have not already expanded their intended use to cover patients identified as critically ill are now in the process of collecting the necessary data for submission to the FDA. In the meantime, patient care must continue, requiring that decisions be made. Facilities must decide whether to keep and validate current meters, keep and restrict the use of current meters to certain patient populations, or switch to a technology without labeling limitations. No single solution will meet all facilities’ needs. Ultimately, the determination of how POC blood glucose meters will be used must take into account available institutional resources, including time, money, and personnel, and, most important, patient safety and outcomes.
T. Scott Isbell, PhD, DABCC, is medical director of clinical chemistry and point-of-care testing at SSM St. Louis University Hospital; medical director of point-of-care testing at SSM Cardinal Glennon Children’s Hospital; and assistant professor of pathology at St. Louis University School of Medicine.
Disclosure: T. Scott Isbell presently serves on the Advisory Council for
1. New York State Department of Health. Off-label Use of Glucose Meters. http://www.wadsworth.org/labcert/clep/files/Glucose_meters_off_label_use_1_13_14.pdf. January 13, 2014. Accessed December 1, 2015.
2. Department of Health and Human Services. Centers for Medicare and Medicaid Services. March 13, 2015. https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/SurveyCertificationGenInfo/Downloads/Survey-and-Cert-Letter-15-11.pdf. Accessed December 1, 2015.
3. College of American Pathologists. CAP eAlert: Notification on Glucose Monitoring Devices. http://www.cap.org/web/home/lab/eAlerts/eAlert?contentID=11070879&_afrLoop=597174465850213#%40%3F_afrLoop%3D597174465850213%26contentID%3D11070879%26_adf.ctrl-state%3D162h14jofa_17. Accessed December 1, 2015.
4. 510(K) Substantial Equivalence Determination Decision Summary Assay And Instrument Combination Template. http://www.accessdata.fda.gov/cdrh_docs/reviews/K132121.pdf. Accessed December 1, 2015.
5. Finfer S, Wernerman J, Preiser JC, et al. Clinical review: Consensus recommendations on measurement of blood glucose and reporting glycemic control in critically ill adults. Crit Care. 2013;17(3):229.
6. Karon B, Koch CD, Wockenfus AM, et al. Accuracy of whole blood glucose measurement when venous catheter blood samples are used on glucose meters. Diabetes Technol Ther. 2009;11(12):819-825.
7. CLSI. Point-of-Care Blood Glucose Testing in Acute and Chronic Care Facilities; Approved Guideline—Third Edition. CLSI document POCT12-A3. Wayne, PA: Clinical and Laboratory Standards Institute; 2013. http://shop.clsi.org/point-of-care-documents/POCT12.html. Accessed December 1, 2015.
Is Liquid Biopsy in Oncology Ready for Prime Time?
Causes and Prevention of Pseudohyperkalemia
Outreach as a Laboratory Service Line
Buyer's Guide: Temperature Monitoring
- In The Loop!
- Digital Edition
- Special Announcements