You are leaving Medscape Education
Cancel Continue
Log in to save activities Your saved activities will show here so that you can easily access them whenever you're ready. Log in here CME & Education Log in to keep track of your credits.



Accuracy of Point-of-Care Blood Glucose Monitoring in the Presence of Interfering Substances: Chronicle of a Medical Alert and Implications for Patient Safety in Different Clinical Settings

  • Authors: Irl B. Hirsch, MD; Bruce W. Bode, MD; Carol A. Verderese, BA
  • CME/CE Released: 1/22/2010
  • Valid for credit through: 1/22/2011
Start Activity

Target Audience and Goal Statement

This activity was developed for diabetologists, endocrinologists, nephrologists, rheumatologists, neurologists, hospitalists, primary care physicians, and allied health care professionals who treat patients with diabetes (registered nurses, pharmacists, nurse practitioners, physician's assistants).

The real-world implications of often overlooked substances that interfere with the accuracy of blood glucose readings in different clinical situations were recently brought to light by Food and Drug Administration (FDA) notifications of life-threatening/fatal hypoglycemia resulting from erroneous insulin dosing based on test results obtained from patients receiving products containing maltose. Maltose, a disaccharide composed of two glucose molecules, is used in a number of biological preparations, including intravenous immunoglobulin preparations and peritoneal dialysis solutions containing icodextrin (converted to maltose). Test strips based on glucose dehydrogenase pyrroloquinoline quinone (GDH-PQQ) methodology can falsely elevate blood glucose readings in patients receiving maltose or icodextrin. While other medications (eg, acetaminophen, L-dopa, tolazamide) and naturally occurring substances can affect performance of glucose monitoring systems, icodextrin can increase the glucose value reported by GDH-PQQ-based meters by more than 100 mg/dL. Fortunately, reasonable precautions may be implemented to protect diabetic patients who are practicing self-monitoring of blood glucose and receiving products containing maltose or other interfering substances. This on-line CME/CE educational activity will inform medical professionals about this and other clinical situations that warrant heightened vigilance.

Upon completion of this activity, participants should be able to:

  1. Cite the interfering substances that can cause inaccuracy of blood glucose readings
  2. Describe the clinical situations that warrant heightened awareness of the potential for inaccurate blood glucose measurements
  3. Recognize the effects, and risk level, of interfering substances as related to specific glucose measurement methodologies
  4. Develop systems-oriented patient safety protocols such as flagging charts of at-risk patients, affixing stickers to interfering products, and labeling affected meters
  5. Identify situations in which laboratory glucose monitoring is warranted to avoid the risk of false glucose determinations


The sponsors are committed to offering programs that promote improvements or quality in health care and are developed free of the control of commercial interests. Reasonable efforts have been taken to ensure that our programs are balanced, independent, objective, scientific, and in compliance with regulatory requirements. Faculty and course directors have disclosed all relevant financial relationships with commercial companies, and the sponsors have a process in place to resolve any conflict of interest. The sponsors also require that faculty disclose any discussion of off-label or investigational uses included in their presentations. Disclosure of a relationship is not intended to suggest or condone bias in a presentation, but is made to provide participants with information that might be of potential importance to their evaluation of a presentation.

The information presented at this CME program represents the views and opinions of the individual presenters, and does not constitute the opinion or endorsement of, or promotion by, Penn State College of Medicine, Postgraduate Institute for Medicine, or The Diabetes Education Group. Each participant must use his/her personal and professional judgment when considering further application of this information, particularly as it may relate to patient diagnostic or treatment decisions including, without limitation, FDA-approved uses and any off-label uses.


  • Bruce W. Bode, MD

    Atlanta Diabetes Associates; Associate Clinical Professor, Emory School of Medicine, Atlanta, Georgia


    Disclosure: Consultant/Advisor: AgaMatrix, Bayer, J&J Lifescan. Research Grants: Abbott, Bayer, J&J Lifescan.

  • Irl B. Hirsch, MD

    Professor of Medicine, University of Washington, Seattle, Washington


    Disclosure: Consultant/Advisor: Roche, Johnson & Johnson.

  • Carol A. Verderese, BA

    The Diabetes Education Group, Lakeville, Connecticut


    Disclosure: Stock ownership: Amylin Pharmaceuticals.

Planners and Reviewers

  • Paul F. Henrici

    Managing Director, The Diabetes Education Group, Lakeville, Connecticut


    Disclosure: Stock Ownership: Amylin Pharmaceuticals.

  • Linda Graham, RN, BSN, BA

    PIM Reviewer


    Disclosure: Nothing to disclose.

  • Robert Gabbay, MD, PhD

    Associate Professor of Medicine, Penn State College of Medicine, Hershey, Pennsylvania


    Disclosure: Speakers Bureau: Novo Nordisk & Merck. Consultant: Roche and Animas LifeScan.

  • Bonnie J Bixler, MEd

    Penn State College of Medicine, Continuing Education, Hershey, Pennsylvania


    Disclosure: Nothing to disclose.

  • Ruth Hirsch, RN (Nurse Planner)


    Disclosure: Consulting Fees: Johnson/Johnson (spouse), Roche (spouse); Contracted Research: Novo Nordisk (spouse).

Accreditation Statements

    For Physicians

  • This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of Penn State College of Medicine and The Diabetes Education Group (TDEG). Penn State College of Medicine is accredited by the ACCME to provide continuing medical education for physicians.

    Penn State College of Medicine designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit(s) TM. Physicians should only claim credit commensurate with the extent of their participation in the activity.

    Contact This Provider

    For Nurses

  • Postgraduate Institute for Medicine is accredited as a provider of continuing nursing education by the American Nurses Credentialing Center's Commission on Accreditation.

    This educational activity for 1.1 contact hours is provided by Postgraduate Institute for Medicine.

    Contact This Provider

    For Pharmacists

  • Postgraduate Institute for Medicine is accredited by the Accreditation Council for Pharmacy Education as a provider of continuing pharmacy education.

    Postgraduate Institute for Medicine designates this continuing education activity for 1.1 contact hour(s) (0.11 CEUs) of the Accreditation Council for Pharmacy Education.
    (Universal Activity Number - 809-9999-09-149-H01-P)

    The estimated time to complete this activity is 1 hour.

    Contact This Provider

For questions regarding the content of this activity, contact the accredited provider for this CME/CE activity noted above. For technical assistance, contact [email protected]

Instructions for Participation and Credit

There are no fees for participating in or receiving credit for this online educational activity. For information on applicability and acceptance of continuing education credit for this activity, please consult your professional licensing board.

This activity is designed to be completed within the time designated on the title page; physicians should claim only those credits that reflect the time actually spent in the activity. To successfully earn credit, participants must complete the activity online during the valid credit period that is noted on the title page.

Follow these steps to earn CME/CE credit*:

  1. Read the target audience, learning objectives, and author disclosures.
  2. Study the educational content online or printed out.
  3. Online, choose the best answer to each test question. To receive a certificate, you must receive a passing score as designated at the top of the test. In addition, you must complete the Activity Evaluation to provide feedback for future programming.

You may now view or print the certificate from your CME/CE Tracker. You may print the certificate but you cannot alter it. Credits will be tallied in your CME/CE Tracker and archived for 6 years; at any point within this time period you can print out the tally as well as the certificates by accessing "Edit Your Profile" at the top of your Medscape homepage.

*The credit that you receive is based on your user profile.


Accuracy of Point-of-Care Blood Glucose Monitoring in the Presence of Interfering Substances: Chronicle of a Medical Alert and Implications for Patient Safety in Different Clinical Settings

Authors: Irl B. Hirsch, MD; Bruce W. Bode, MD; Carol A. Verderese, BAFaculty and Disclosures

CME/CE Released: 1/22/2010

Valid for credit through: 1/22/2011


This CME/CE activity was developed to be distributed on Medscape.


Considering that over 4 million test strips for self-monitoring of blood glucose (SMBG) are used in the United States each day, relatively few adverse events (AEs) are voluntarily reported to the US Food and Drug Administration (FDA) by patients and clinicians using these products. A recent study of SMBG-related AE's received by the FDA's MedWatch reporting system determined that less than 1.0% came from users over a 3-year period.[1] This disproportionately low rate of spontaneous reporting constitutes a public health concern given that mandatory reporting requirements applying mainly to manufacturers do not usually illuminate the interplay of clinical factors that might lead to AEs under routine conditions.[2] Voluntary reporting systems, such as the FDA's MedWatch program, are thus considered crucial to ensuring the safety of patients using drugs and devices, such as blood glucose meters.[2-5]

Yet, the limitations of voluntary reporting were clearly illustrated when, in August, 2009, the FDA issued a repeat notification concerning interferences between products containing maltose (eg, icodextrin and certain intravenous immune globulin [IVIG] products) and blood glucose monitoring systems that use test strips containing the enzyme glucose dehydrogenase (GDH)-pyrroloquinoline quinone (PQQ).[6] The historical context for this notification is shown in Table 1. The first official warning appeared in 2003, when the Medicines and Healthcare products Regulatory Agency (MHRA) reported 3 AEs in the United Kingdom, including 1 death, from overestimation of blood glucose in patients treated with maltose-containing drugs.[7] Such overestimation can mask hypoglycemia and result in over-administration of insulin. Meters using test strips based on glucose dye oxidoreductase technology, as well as substances containing other non-glucose sugars (eg, galactose and xylose), are also subject to this interference; however, blood glucose monitoring systems based on glucose oxidase, GDH-flavin adenine dinucleotide (GDH-FAD), and GDH-nicotinamide adenine dinucleotide (GDH-NAD) enzymes are not affected and can be safely used with non-glucose-sugar-containing substances.[8] In 2007, the MHRA issued a second alert stressing this safety information.[9]

Table 1. History of Alerts Regarding Adverse Events Attributable to Interference Between Maltose-containing Medications and GDH-PQQ Monitoring Systems

Date Agency
April 16, 2003 MHRA Medical Device Alert
September 8, 2005 ISMP Safety Alert
September 22, 2005 ISMP Safety Alert Erratum
July 19, 2007 MHRA Medical Device Alert (update)
August 13, 2009 FDA Public Health Notification (Dear Healthcare Practitioner)
MHRA = Medicines and Healthcare Products Regulatory Agency
ISMP = Institute for Safe Medication Practices
FDA = Food and Drug Administration

In the US, the first official alert appeared in 2005, after an elderly immune-compromised patient with diabetes received the maltose-containing IVIG Octagam (Octapharma USA, Centreville, VA) while having his blood glucose monitored with a meter using GDH-PQQ methodology.[10] Although both the IVIG and meter package inserts noted the potential for drug-device interaction, the intensive care unit (ICU) staff administered high-dose IV insulin to correct elevated blood glucose levels later found to be erroneous. The patient experienced "profound" hypoglycemia, confirmed by a laboratory glucose measurement of 12 mg/dL, and developed irreversible neurological damage. The FDA subsequently required all manufacturers of products containing maltose to feature information about this potential interference in the Warnings sections of their package inserts. Despite this and other educational measures, however, the 2009 alert documented 13 deaths associated with GDH-PQQ-related falsely elevated blood glucose readings reported between 1997 and 2009.[6] Ten of the 13 patients were receiving icodextrin (Extraneal) peritoneal dialysis solution for renal failure, while 3 were receiving other maltose-containing substances. A list of the cited products and their indications appears in Table 2.[6]

Table 2. Interfering Products Containing Maltose

Product Approved Indication
Icodextrin Products
   EXTRANEAL Long-dwell during CAPD and APD in patients with ESRD
   Adept® Adhesion Reduction Solution (4% icodextrin) Reduce post-surgical laparoscopic adhesions in gynecological surgery
Immunologic Agents
   Orencia® (abatacept) Rheumatoid arthritis
   Octagam® Primary immunodeficiency
   WhinRho® SDF Liquid Immune thrombocytopenic purpura and suppression of Rh isoimmunization
   HepaGam B® Prevention of hepatitis B recurrence following liver transplant and post-exposure prophylaxis
   Bexxar® Non-Hodgkins lymphoma
Adapted from reference 6.

More recently, a systematic review of the medical literature and the FDA's Manufacturer and User Facility Device Experience (MAUDE) database confirmed that AEs associated with GDH-PQQ test strips have been underreported globally.[11] While thorough analysis of the possible reasons for underreporting of AEs is beyond the scope of this paper, among the most common are misconceptions that only major events warrant reporting and that causal relationships must be clearly understood before making a report.[2-5] Moreover, many health care professionals rarely look beyond the parameters of their own specialties when considering potential medical errors and the feasibility of preventing them. With this in mind, as well as the multiplicity of situations in which maltose-containing drugs such as icodextrin and IVIGs are administered to patients with diabetes, particularly in the hospital, this paper describes fundamentals of meter accuracy in relation to real-world medical practice, the clinical settings that can present heightened risk of substance interference when certain types of blood glucose meters are used, and practical measures for improving patient safety in this area.

Point-of-Care Blood Glucose Monitors: How Accurate Is Accurate?

The vast majority of the billions of point-of-care blood glucose tests performed around the world annually are electrochemical, and quality-control procedures to achieve infallible accuracy in every situation remain elusive. Most point-of-care meters use plastic or paper strips comprising electrochemical cells, which contain the enzymes glucose oxidase, GDH-PQQ, GDH-FAD, or GDH-NAD and a redox mediator.[12,13] Blood placed on the test strip undergoes a complex biochemical reaction. The plasma separates from the whole blood and diffuses through an internal layer of the test strip containing the enzymes and electrodes. The enzymes catalyze the conversion of glucose to gluconic acid and the electrons resulting from this reaction generate a current calibrated to measure the concentration of glucose in the sample.[14]

Electrochemical Methodologies, Potential Interferences, and Accuracy Standards

The classic methodology uses glucose oxidase, whereby the enzyme interacts with glucose, taking an electron and forming gluconic acid. It then interacts with water and oxygen, pushing electrons to the oxygen and forming hydrogen peroxide.[15,16] The oxygen is replaced by a mediator, which accepts the electron and passes it to an electrode to produce the current that generates the glucose signal. Because this mechanism of action requires oxygen and water, extremes of hydration or oxygenation can alter results.[16] Furthermore, the oxidized mediator is somewhat unstable and can be reduced, especially at high temperatures. Medications that can affect the readings from these point-of-care meters include acetaminophen, L-dopa, tolazamide, and ascorbic acid. However, these errors are usually not clinically significant due to the large doses of these drugs required to impact glucose oxidase.

The enzyme GDH is less specific than glucose oxidase and, therefore, is not as susceptible to variations in oxygen concentration. However, glucose oxidation can be catalyzed by one of three different co-factors in GDH-based systems: FAD, NAD, or PQQ. While systems using FAD or NAD do not exhibit interference by cross-reacting sugars, the disaccharide maltose, as well as the monosaccharides galactose and xylose, compete with glucose on GDH-PQQ test strips.[16] Thus, certain products containing these non-glucose sugars (Table 2), such as icodextrin and some IVIGs, can falsely elevate the reported blood glucose by >100% when test strips based on GDH-PQQ methodology are used. In one report, for example, a patient tested with a GDH-PQQ-based system had a blood glucose result of 200 mg/dL compared with a laboratory result of 19 mg/dL.[6] Another patient receiving icodextrin for peritoneal dialysis had a glucose result of 193 mg/dL with a GDH-PQQ meter versus a laboratory result of 8 mg/dL. In 8 of the 13 reports cited by the FDA, GDH-PQQ readings were 13 to 15 times higher than corresponding laboratory results, representing a clinically significant risk for patients receiving non-glucose-sugar formulations.[6]

Certain endogenous patient variables may also affect the accuracy of blood glucose readings. These include naturally occurring interfering substances, such as uric acid, oxygen, and triglycerides that might be present during disease states. While manufacturers maintain guidelines for evaluating potential interferences, international standards (International Organization for Standardization, or ISO) require that 95% of all differences in glucose values (eg, comparator glucose value minus new device glucose value) be within 15 mg/dL of glucose values <75 mg/dL and within 20% of glucose values ≥ 75 mg/dL.[17,18] In this country, the American Diabetes Association has suggested that meter systems should have an inaccuracy of <5%, although no meters currently meet this standard.[13,16,19] Some experts have proposed even more stringent criteria based on a hypothetical insulin dosing study showing that an inaccuracy of <2% is necessary to avoid excessive hypo- and hyperglycemia.[20]

Although many of today's meters average inaccuracies of only 5% to 6%, understanding common sources of error can lead to more accurate monitoring overall.[16] This is particularly true in the hospital environment where the potential for inaccuracy is greater due to numerous intertwining clinical variables, differences among measurement methodologies, and issues involved in the intensive management of hyperglycemia.[15]

An Object Lesson

The 2009 FDA public health notification referred to above provides an object lesson about the continuing knowledge gap concerning drug-device interferences when monitoring blood glucose in specific clinical situations.[1,2,5,6,21] Six of the 13 deaths cited by the FDA occurred since 2008 despite health alerts dating back to 2005 and a Letter to the Editor published in 1998.[6,10,22] Furthermore, at least two case reports and one case-series abstract warning of unrecognized hypoglycemia appeared in the medical literature during 2009, independent of the FDA notice.[23-25] In one of these, a 68-year-old man with diabetes hospitalized for sepsis, and receiving high-dose insulin infusion while on continuous ambulatory peritoneal dialysis (CAPD), developed tachyarrhythmia during sleep. Suspecting hypokalemia, the nurse in charge requested laboratory analysis for electrolytes. In addition to the finding of hypokalemia, the laboratory blood glucose measurement dramatically disagreed with the bedside capillary measurement (38 mg/dL versus 393 mg/dL).[24] Glucose was administered promptly, and the patient regained consciousness from a hypoglycemic coma. The next day, glucose monitoring of the same patient with several commonly used meters—two using the glucose oxidase method (Cobas® B221 [Roche Diagnostics], OneTouch® Ultra [LifeScan]), one using the GDH-FAD method (Contour® TS [Bayer]), and four using the GDH-PQQ method (Accu-Chek® Go [Roche], Accu-Chek® Inform [Roche], Glucocard X-Meter [Menarini Diagnostics], FreeStyle® Mini [Abbott])—confirmed that interference of the GDH-PQQ-based test strips with icodextrin had resulted in falsely elevated readings and a potentially life-threatening overdose of insulin. Results of two different blood samples tested twice ranged from 112 mg/dL to 149 mg/dL with the glucose oxidase and GDH-FAD meters versus 438 mg/dL to 521 mg/dL with the GDH-PQQ meters.[24]

A comprehensive search of the FDA's MAUDE database and the medical literature has provided further evidence of the potential scope of this drug-device interference.[11] This quantitative review uncovered 76 AEs (57 in the MAUDE database, 19 in the literature) related to falsely elevated glucose readings due to the non-specificity of GDH-PQQ test strips. Eleven (14%) were associated with death; 45 (59%) with severe hypoglycemia defined as requiring the assistance of a third party (4 resulting in permanent injury); 11 (14%) with non-severe hypoglycemia; and 9 (12%) were unclassified. Seventy-nine percent of the events involved use of icodextrin-based peritoneal dialysis solution, whereas 13% were associated with a maltose-containing IVIG. In 2 cases, the falsely elevated GDH-PQQ meter results occurred in the presence of galactosemia; in 3, a "maltose-containing substance"; and in 1, maltodextrin. Sixty percent of the AEs were confirmed inpatient; 25% confirmed outpatient; and approximately half (47%) occurred outside the US.[11] In 2008, the MAUDE database contained 2 fatal cases and 4 cases of severe hypoglycemia associated with this issue.

The FDA acknowledges that AEs recorded on the basis of spontaneous reports are a small proportion of those that actually take place.[5,26] Moreover, it has no data about the distribution of GDH-PQQ monitoring systems in clinical settings where patients would be likely to receive interfering medications. The agency also recognizes the difficulty of maintaining accurate lists of susceptible meters when brand names and technologies are constantly changing.[5] As such, clinicians should not rely on product lists to determine whether a meter is appropriate, but should instead become educated about known interferences, such as drugs containing non-glucose sugars, and consult professional package inserts when monitoring blood glucose in insulin-using patients exposed to these substances.[5]

Substance Interference in Clinical Context

Substance interferences refer to any concentration of a substance in the blood that affects the reliability of blood glucose results. These interferences may be introduced in the form of therapies administered by health care professionals or occur endogenously from substrates in the blood. A partial list of endogenous and exogenous interferences in blood that potentially give rise to false high or low glucose readings depending on the enzymatic reaction of the test strip being used appears in Table 3. Health care professionals who treat patients with diabetes should be aware of these substances and the potential for drug-device interactions in the clinical situations described below.

Table 3. Some Variables that Alter Blood Glucose Measurement*

Variable Effect on reading
   Anemia Increases (GO & GDH)
   Polycythemia Decreases (GO & GDH)
Oxygen concentration
   Hypoxia (high altitude) Increases (GO)
   Oxygen therapy Decreases (GO)
pH (6.8-7.55)
   Low pH May decrease (GO)
   High pH May increase (GO)
Hypothermia Increases (GO & GDH) or decreases (GDH)
Hypotension Increases (GO & GDH) or decreases (GDH)
Uric acid Decreases (GO)
   Ascorbic acid Increases (some GDH) and decreases (GO)
   Acetaminophen Variable (GO and GDH)
   Dopamine Variable (GO and GDH)
   Icodextrin Increases (GDH-PQQ only)
   Galactos Increases (GDH-PQQ only)
   Xylose Increases (GDH-PQQ only)
   Tolazamide Variable (GO)
   Mannitol Increases (GO)
*Consult product package inserts for updated information regarding meter accuracy and interfering substances.
GO = glucose oxidase
Adapted from references 15 and 16.

Maltose-containing Drugs

Maltose is used in certain biological products as a stabilizing agent and osmolality regulator (Table 2).[6,8] It is filtered through the renal glomeruli, hydrolyzed to glucose in the renal tubules, and then reabsorbed into the plasma circulation. Each saccharide chain generated by the metabolism of maltose has a free reducing group of glucose located at its end. Since maltose is a disaccharide derived from two units of glucose, its presence in the circulation may result in the overestimation of blood glucose levels with GDH-PQQ-based systems, which involve the oxidation of glucose and the measurement of reaction end products in direct proportion to glucose concentration. Pharmacokinetic studies of maltose suggest that rapid infusions yield high peaks followed by rapid elimination; whereas longer infusions peak at the end of the infusion period, with slow and prolonged elimination.[8] Yet, given the uncertainty about the specific serum maltose threshold required for interference with GDH-PQQ devices, the unpredictable length of time that levels exceed this threshold, and the associated patient safety issues, clinicians should avoid using GDH-PQQ systems when treating patients receiving maltose-containing drugs.

Icodextrin Used in Peritoneal Dialysis. Diabetes is now the most common cause of dialysis-dependent renal failure in the Western World.[27] The high prevalence of patients with end-stage renal disease (ESRD) has created an increasing demand for peritoneal dialysis.[28] Dialysis solution typically contains dextrose as an osmotic agent to allow water to pass across the peritoneum, but this osmotic effect may quickly diminish. Alternative dialysis solutions containing icodextrin, a glucose polymer, were developed to prolong ultrafiltration in patients on CAPD or automated peritoneal dialysis. Icodextrin is hydrolyzed by serum α-amylase to oligosaccharides, such as maltose, which are ultimately cleared by dialysis into the peritoneal cavity.[29] Because absorption of icodextrin from the peritoneal cavity is relatively slow compared with that of dextrose, it is ideal for ultrafiltration over long dwell times.[28]

In 1998, 6 CAPD patients undergoing one nocturnal exchange with Extraneal (icodextrin) for a minimum of 7 consecutive days were studied to determine the degree of discrepancy between glucose levels measured with a GDH-PQQ system versus a laboratory hexokinase method.[30] Overestimation of blood glucose was apparent in all 6 patients after a single nocturnal exchange lasting 10 hours. The overestimation was 60 mg/dL, on average, independent of the actual glucose value. Another study showed a similar degree of overestimation in results obtained with a GDH-PQQ meter used for patients with CAPD.[31] Given the increasing number of published case reports attesting to risks associated with such discrepancies, alternatives to GDH-PQQ point-of-care monitoring systems should be actively pursued when treating peritoneal dialysis patients in any clinical setting.[23-25,27,29,32-40] Furthermore, because renal failure and dialysis per se have a destabilizing effect on insulin and glucose metabolism, close collaboration between the renal and diabetes care teams is warranted for optimal treatment of these patients.[32]

Intravenous Immune Globulin. IVIG has been used for the treatment of immunodeficiency for more than 50 years. Preparations that interfere with GDH-PQQ-based blood glucose measurements are shown in Table 2. In primary and secondary immunodeficiency disorders, IVIG restores selected parameters of immune function by increasing antibody levels and possibly removing immunosuppressive complexes.[41] Although the mechanisms of action are not fully understood, they are thought to include anti-infective, immunoregulatory, and anti-inflammatory properties. Among the stabilizing agents used in these products are human albumin, glycine, and sugars such as sucrose, maltose, and glucose.[42]

IVIG has been approved by the FDA for treatment of 6 conditions: immune thrombocytopenic purpura, primary immunodeficiency, secondary immunodeficiency, pediatric HIV infection, Kawasaki disease, prevention of graft-versus-host disease, and infection in bone marrow transplant recipients. However, the FDA estimates that 50% to 70% of IVIG is used in off-label indications.[43] Over 150 unlabeled uses have been identified, including the treatment of multiple sclerosis and prevention of antiphospholipid syndrome in miscarriage. Table 4 lists frequently reported off-label uses, which involve many clinical disciplines, such as neurology, hematology, dermatology, infectious disease, and rheumatology.[42,43]

Table 4. Examples of Off-label Uses for Intravenous Immune Globulin*

Chronic inflammatory demyelinating polyneuropathy
Dermatomyositis, polymyositis
Guillain-Barré syndrome
Hematologic coagulation disorders
Hematologic immune-mediated cellular disorders
Infection prophylaxis in high-risk neonates
Inflammatory bowel disease
Miscarriage, recurrent
Myasthenia gravis
Multiple sclerosis
Multifocal motor neuropathy
Parvovirus-B19-associated anemia
Post-transfusion purpura
Rheumatoid diseases
Sepsis, toxic shock syndrome
Stiff-person syndrome
Systemic lupus erythematosus, systemic vasulitis
Toxic epidermal necrolysis (Lyell syndrome)
Transplantation: CMV-negative recipients of CMV-positive organs
Transplantation: renal graft rejection
Transplantation, solid organ: alloimmunization and hypogammaglobulinemia
*Many remain controversial and in need of large, controlled trials.
CMV = cytomegalovirus
Adapted from references 41 and 42.

According to FDA investigators, recommended doses of Gamimune N® 5% and Octagam® would likely result in falsely increased readings by GDH-PQQ methodology. Although Gamimune N® 5% has not been manufactured in the United States since 2005, both products have been distributed worldwide over a long enough period that patients receiving them would have included those hospitalized with diabetes subject to GDH-PQQ-based blood glucose monitoring.[5,6] A health care provider was key to reporting the index case that prompted the initial alert about this drug-device interference in 2005 and other cases have been published.[5,10,44,45] As such, the FDA encourages health care professionals across disciplines to submit adverse events reports, particularly for serious events involving FDA-approved, -licensed, or –regulated IVIG products that interfere with blood glucose readings generated by the GDH-PQQ monitoring method.

Acetaminophen and Other Medications

The pain reliever acetaminophen is one of the most common associated with both accidental and intentional poisoning and, at certain concentration levels, can generate analytical interferences on some blood glucose meters.[14] Although the interference of acetaminophen administered at therapeutic doses is negligible, all patients metabolize drugs differently so the concentration level at which an analytical interference may occur varies from patient to patient.

In one study of the effect of acetaminophen on the accuracy of four blood glucose meters, sample concentrations of 0, 5, and 10 mg/dL were added to donor sodium heparin blood with glucose concentrations that had been adjusted to 44, 145, 244, and 341 mg/dL, respectively.[46] No concentration of acetaminophen changed the mean baseline (no acetaminophen added) glucose level by more than 10 mg/dL (for experiments conducted at 44 mg/dL glucose) or 10% (for experiments conducted at 145, 244, and 341 mg/dL) on any of the meters. Thus, acetaminophen did not produce a clinically significant interference on any of the meter technologies studied, although glucose concentration must be verified by conventional laboratory techniques whenever an overdose is suspected.

Finally, given the potential for polypharmacy over the course of diabetes treatment and the wide variety of medications that interfere with the accuracy of blood glucose readings (Table 3), clinicians should review patients' drug regimens (including over-the-counter) and meter selections regularly and adjust data interpretation accordingly, recognizing that package inserts may not provide sufficient warning information for drugs used in critical care settings.[13] In one study, the effects of therapeutic and toxic levels of 30 different drugs on glucose readings from six different glucose meters was examined, with a comparatively low error threshold of ±6 mg/dL. Ascorbic acid interfered with the measurements on all six meters. Acetaminophen, dopamine, and mannitol interfered with glucose measurements on selected devices.[15]

Inpatient-specific Variables

The interferences listed in Table 3 are particularly relevant to inpatient care, where endogenous and exogenous variables as well as differences between assays are more likely to converge. Although the effect of each potential variable may be small, the net result could be clinically significant, especially at a narrow therapeutic target glucose range.[47,48] Moreover, accuracy standards for blood glucose meters do not discriminate between hospital and home use. Depending on the type of meter, the mean glucose value for a particular unknown test sample reported by an institution varies by >30% at glucose levels exceeding 150 mg/dL and by 60% at levels in the hypoglycemic range.[15] This variability may stem from analytical differences or user interfaces that are more prone to operator error. In one prospective study of 30 consecutive patients in a mixed medical/surgical ICU, agreement between fingerstick blood glucose measurements and laboratory reference values was only 56.8% overall.[48]

In another study set in a community hospital, retrospective chart review of 6885 patients conducted over 12 months revealed that 1.2% had an identified interfering substance with a meter using the GDH-PQQ method.[49] The reported interferences were serum uric acid >10 mg/dL, hematocrit <20% or >55%, total serum bilirubin concentrations >20 mg/dL, serum acetaminophen concentrations >8 mg/dL, and serum triglyceride concentrations >5000 mg/dL. The interference time interval was also calculated as the time period during which a substance was at a high enough concentration to interfere with the meter reading. Of the 84 patients identified with an interfering substance, 30 (36%) had an active order for an insulin product during the interference interval. These patients could have been at risk for inappropriate insulin administration if not for the hospital's automated interfering substance screening and notification program.[49]

Numerous inpatient variables influence point-of-care meter readings (Table 3). Common examples include hematocrit, hypotension, oxygenation (altitude), and temperature. For individuals in the hospital, factors that are unique to the patient must be considered in clinical context, particularly where differences between the bedside reading and the clinical circumstances are observed.

Hematocrit. Generally, increases in hematocrit lower glucose measurements and vice versa. Hematocrit may affect results of glucose testing because erythrocytes in the whole-blood sample can alter the ratio of blood glucose to plasma glucose, as well as the flow of plasma and delivery of oxygen into the test strip.[13] The presence of an increased number of red blood cells in the capillary blood could "mechanically" impede diffusion of the plasma through the layers of the strips, decreasing the volume of plasma available to the enzymatic reaction. So, for example, hypoglycemia could be masked in patients with anemia or glucose underestimated in patients with polycythemia.[15] Although manufacturers set adequate testing limits for hematocrit, the patient's hematocrit is not always known at the time of admission. Surgical patients may be most at risk for erroneous readings because of fluctuating hematocrit. Other situations where hematocrit and glucose changes occur rapidly include acute blood loss or transfusion.

Hematocrit may also be affected by microclot formulations in the samples or on the test strips, protein deposition, fibrin aggregation, and hemolysis.[13] It is particularly important to consider hematocrit when treating patients with diabetes who smoke, live at high altitude, or present with anemia, sickle-cell disease, dehydration, polycythemia vera, or end-stage renal failure. Because a given meter system will respond differently to glucose changes at fixed hematocrit levels, checking the hematocrit range specified by the manufacturer is an important safety measure, especially in the hospital setting.

Hypotension. Hypotension, or low blood pressure, may reduce perfusion and increase glucose utilization, potentially masking true blood glucose levels.[15] In one recently published study of 85 critically ill patients, low perfusion index reflecting peripheral hypoperfusion was associated with poor glucose strip performance.[50] Forty-one (15%) of 273 point-of-care blood glucose values differed from the laboratory reference value by more than 20% and conflicting values were associated only with low perfusion index values. Among the 41 discordant values, 12 (4.4%) were underestimates and 29 (10.6%) overestimates of laboratory results. In addition, an earlier study comparing fingerstick blood glucose measurements to laboratory values for 25 hypotensive patients versus 39 normotensive patients found that the fingerstick results showed an acceptable variation from the standard (±20%) in only 36% of the hypotensive patients compared with 89.7% of the normotensive patients.[51] A similar study found that 31.6% of the fingerstick glucose measurements were outside of the allowable 20% variance.[52] Finally, statistical agreement between an arterial blood gas point-of-care testing method and a fingerstick method still resulted in undetected hypoglycemia when a lower limit of 80 mg/dL was targeted in ICU patients with hypoperfusion.[53] Taken together, these data underscore the need for cautious interpretation of bedside blood glucose values in this setting.[53]

Oxygenation. High oxygen tension (ie, pO2 >100 mm Hg) can falsely decrease glucose readings on some glucose oxidase-based blood glucose meters, especially when patients are receiving oxygen therapy.[15] Moreover, glucose levels obtained at higher altitudes may be overestimated by as much as 15% with glucose-oxidase meters.[16] In a study of 6 glucose meter systems using whole blood and venous plasma, measurements at pO2 >100 mm Hg produced results outside of acceptable error limits. In general, lower oxygen tension (40 mm Hg) had a negligible effect.[54] Strips that use GDH as the enzyme are less affected by oxygen.[16]

Temperature. The influence of temperature on blood glucose readings is not predictable.[16] Low temperatures (common in operating rooms during certain procedures) decrease circulation to the skin and may produce either positive or negative errors regardless of the enzyme method. Active warming could reduce the potential for erroneous results; the significance of fever is not known.[15] Clinicians should remain aware that test strips exposed to extremes of temperature, humidity, or high altitude may produce falsely elevated readings.[13]

Safety Measures

Interfering substances and other factors affecting blood glucose meter accuracy are becoming increasingly relevant given the trend toward more individualized and, in some cases, more aggressive therapeutic targets in diabetes care.[19] As such, addressing frequent, often unrecognized, and largely unpublicized errors made under relatively common circumstances could significantly improve patient safety.[2] The health warnings featured in this article provide a good example of this type of AE, with achievable solutions that may be extrapolated to other patient-safety initiatives that depend on voluntary reporting. The FDA has done its part in issuing a notice summarizing reported AEs from 1997 to 2009, and manufacturers are redoubling efforts to educate health care professionals about this drug-device interference. Patient advocacy websites have also posted notices about this and other accuracy issues, reinforcing the role of the consumer in promoting safety standards.[55]

With this in mind, all clinicians who treat patients with diabetes should follow precautions to reduce the risk of false glucose meter readings in those receiving peritoneal dialysis solutions containing icodextrin, maltose-containing immune globulins, D-xylose for evaluation of reabsorption syndromes, and other maltose-containing drugs. While the FDA requires hospitals and other user facilities to report deaths and serious injuries related to glucose meters and test strips, individual clinicians can report adverse events either directly to the device manufacturer or to MedWatch, the FDA's voluntary reporting program. This can be done online at, by phone at 1-800-FDA-1088, by FAX at 1-800-FDA-0178, or by mailing FDA form 3500 (download from to MedWatch, 5600 Fishers Lane, Rockville, MD 20857-9787.

General Precautions

Recommendations issued by the FDA pertaining to the use of GDH-PQQ test strips in the presence of an interfering drug product may be accessed online at[6] Health care professionals should review the labeling for both the glucose meter and test strips used in their health care facilities, avoid using GDH-PQQ glucose test strips in these facilities if possible, and never use them in treating the following patients:

  • Those who are receiving interfering products (Table 2); in these patients, an alternative glucose testing method should be used
  • Those from whom or about whom there is little or no information regarding concomitant medication use (eg, patients who are unresponsive or cannot adequately communicate)
  • Those with an indwelling catheter in their abdomen or an indicative scar of recent dialysis[56]

Clinicians should be reminded to "treat the patient, not the glucose reading," and to correlate symptoms with glucose meter results, especially if the readings are extremely high or low.[56] If possible, clinicians should verify directly with patients that they are not exhibiting any symptoms of hypoglycemia before administering any dose of insulin, rather than relying on glucose meter readings alone. If uncertain, results should be verified with a serum laboratory test.[57] Additional systems-oriented safety measures are described below.

Label glucose meters. Meters and test strips should be labeled with warnings specifying the drug-device interaction and the need for alternative glucose monitoring methods in patients who are receiving interfering drugs.

Flag patients considered at risk. When taking the medication history, it is important to clearly identify at-risk patients, particularly those with diabetes who are immunocompromised or on peritoneal dialysis. (Note that the drug-device interaction can occur up to a week or more after maltose-containing medications are administered.) Members of the health care team should be made aware of the need to avoid using GDH-PQQ meters when treating these patients.

Increase awareness. Educate all staff about the risk of falsely elevated glucose readings in patients receiving medications that contain maltose or other non-glucose sugars. Initially, this can be accomplished by informing nurses working in ICUs, renal floors, and those areas of the hospital where IVIG is frequently used.

Develop protocols. Include the risk of false glucose determinations, the specific circumstances surrounding this risk, and the need for laboratory glucose monitoring in insulin administration protocols and order sets.

Establish warnings. Ask pharmacy to set up a special alert in the computer for maltose-containing medications so that a sticker can be attached to these products. A similar alert should be included in the medication administration record and diabetic flow sheets, as well as on the patient's chart.

Institute error-detection strategies. Establish signals for early recognition and treatment of a drug-device interaction or error (eg, discrepancies between laboratory and glucose meter readings, symptoms of hypoglycemia or hyperglycemia) and communicate these to the entire staff. Conduct periodic reconciliation of laboratory and glucose meter readings whenever an insulin infusion is running, and take steps to verify a suspicious reading.

Educate patients. Let patients know why certain glucose meters and test strips cannot be used for monitoring their glucose levels, and urge them to confirm the type of monitor being used whenever their blood glucose is tested during the course of treatment with maltose-containing products.

The use of visual aids, such as patient bracelets, chart stickers, pharmacy labels, and stickers for meter boxes containing GDH-PQQ-based meters may also be helpful in preventing errors.[56]

Summary and Conclusion

Test-strips based on GDH-PQQ methodology can falsely elevate blood glucose readings in patients receiving drugs containing, or metabolized to, non-glucose sugars, such as icodextrin and certain immune globulins. While other medications and naturally occurring substances can affect performance of glucose monitoring systems, these particular products can produce clinically significant increases in the glucose value reported by GDH-PQQ meters, leading to unrecognized hypoglycemia or insulin overdose. Fortunately, reasonable precautions may be taken to raise awareness and protect patients with diabetes who could be subject to this drug-device interference. Besides the practical measures enumerated above, health care professionals must take the initiative in submitting AE reports either to manufacturers or the FDA so the extent and nature of the issue, as well as appropriate responses, may be better understood. As one public health expert explained, "Health care practitioners are on the front line of the FDA's post-marketing program, and their reports are vital to the FDA's ability to protect consumers and patients."[3] Reports should ideally include a complete description of the adverse outcome, baseline status of the individual, temporal relationships involving the suspected drug or device, and information about confounding drugs or conditions.[3] Developing cross-specialty approaches to the recognition and prevention of AEs is especially important given the multiple practitioners involved in treating comorbidities of diabetes.

Post-Assessment: Measuring Educational Impact

Thank you for participating in the activity. Please take a few moments to read the following case and complete the questions that follow to help us assess the effectiveness of this medical education activity.

  • Print