Diabetes Care 26:S106-S108, 2003
© 2003 by the American Diabetes Association, Inc.

Tests of Glycemia in Diabetes

American Diabetes Association

	INTRODUCTION

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

This position statement presents the recommendations of theAmerican Diabetes Association on the tests used most widelyin monitoring the glycemic status of people with diabetes andaddresses both patient and physician/laboratory-based testing.It does not address tests for diabetes screening and diagnosis.The recommendations are based on the American Diabetes Association’stechnical review on the subject, which should be consulted forfurther information (1).

	BLOOD GLUCOSE TESTING BY PATIENTS

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

The subject of SMBG has been addressed extensively by two AmericanDiabetes Association Consensus Conferences, which provide acomprehensive review of the subject (2,3).

Recommendations

1. Based principally on the DCCT results, it is recommended thatmost individuals with diabetes should attempt to achieve andmaintain blood glucose levels as close to normal as is safelypossible. Because most patients with type 1 diabetes can achievethis goal only by using SMBG, all treatment programs shouldencourage SMBG for routine daily monitoring. Daily SMBG is especiallyimportant for patients treated with insulin or sulfonylureasto monitor for and prevent asymptomatic hypoglycemia. Frequencyand timing of glucose monitoring should be dictated by the needsand goals of the individual patient, but for most patients withtype 1 diabetes, SMBG is recommended three or more times daily.The optimal frequency of SMBG for patients with type 2 diabetesis not known, but should be sufficient to facilitate reachingglucose goals. When adding to or modifying therapy, type 1 andtype 2 diabetic patients should test more often than usual.The role of SMBG in stable diet-treated patients with type 2diabetes is not known.
   
2. SMBG is recommended for all insulin-treatedpatients with diabetes.SMBG may be desirable in patients treatedwith sulfonylureasor other insulin secretagogues and in allpatients not achievingglycemic goals. Data indicate that onlya minority of patientsperform SMBG. Efforts should be madeto substantially increaseappropriate use of SMBG. Barriersto increasing use of SMBGinclude cost of testing, inadequateunderstanding by both healthcare providers and patients aboutthe health benefits and properuse of SMBG results, patientpsychological and physical discomfortassociated with finger-prickblood sampling, and inconvenienceof testing in terms of timerequirements, physical setting,and complexity of the technique.
   Given the importance of SMBG to diabetes care, government,third-partypayers, and others should strive to make the procedurereadilyaccessible and affordable for all patients who requireit. Thus,SMBG should be an important component of any healthcare benefitspackage.
   
3. Because the accuracy of SMBG is instrumentand user dependent,it is important for health care providersto evaluate each patient’smonitoring technique, bothinitially and at regular intervalsthereafter. Use of calibrationand control solutions on a regularbasis by patients helps ensureaccuracy of results. In addition,because laboratory methodsmeasure plasma glucose, many bloodglucose monitors approvedfor home use and some test stripsnow calibrate blood glucosereadings to plasma values. Plasmaglucose values are 10–15%higher than whole blood glucosevalues, and it is crucial thatpeople with diabetes know whethertheir monitor and strips providewhole blood or plasma results.
   
4. Optimal use of SMBG requiresproper interpretation of the data.Patients should be taughthow to use the data to adjust MNT,exercise, or pharmacologicaltherapy to achieve specific glycemicgoals. Health professionalsshould evaluate at regular intervalsthe patient’s abilityto use SMBG data to guide treatment.Although a number of SMBGmethods store test results and witha computer interface canprovide sophisticated analyses of bloodglucose data, it isnot known whether use of these data managementsystems yieldsbetter glucose control than patient review ofresults recordedin a logbook.
   

	BLOOD GLUCOSE TESTING BY HEALTH CARE PROVIDERS FOR ROUTINE OUTPATIENT MANAGEMENT OF DIABETES

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

1. Blood glucose testing (e.g., laboratory glucose or finger-stickglucose) should be available to providers for use as needed.With the availability of SMBG and glycated protein testing,routine laboratory blood glucose testing by health care providersshould no longer be used to assess glycemic control except tosupplement information obtained from other testing methods andto test the accuracy of SMBG. When adjusting oral glucose-loweringmedication(s) in a patient not taking insulin, laboratory testingalso may be appropriate.
   
2. Comparisons between results frompatient self-testing of bloodglucose in the clinic and simultaneouslaboratory testing areuseful to assess the accuracy of patientresults. If such testingis performed by health care providersusing portable capillaryblood testing devices rather than standardhospital or cliniclaboratory methods, rigorous quality controlprocedures shouldbe used. Participation in the College of AmericanPathologistsvoluntary proficiency testing program for home-usetesting devicesis recommended.
   
3. Continuous ambulatory bloodglucose monitoring may be used todetermine 24-h blood glucosepatterns and to detect unrecognizedhypoglycemia; however, itsrole in improving diabetes outcomesremains to be established.
   

	URINE GLUCOSE TESTING

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

If patients choose to perform urine glucose testing, they shouldfully understand the test limitations. Specifically, patientsshould be taught that although urine glucose measurements correlatewith blood glucose measurements, urine glucose testing providesonly a rough estimate of prevailing blood glucose levels. Patientsshould be taught that urine glucose testing provides no informationabout blood glucose levels below the renal threshold, whichfor most patients is 180 mg/dl (10 mmol/l).

Urine/blood ketone testing
Ketone testing is an important part of monitoring in type 1diabetic patients, in pregnancy with pre-existing diabetes,and in gestational diabetes. The presence of ketones may indicateimpending or even established ketoacidosis, a condition thatrequires immediate medical attention. Patients with type 1 diabetesshould test for ketones during acute illness or stress or whenblood glucose levels are consistently elevated (e.g., >300mg/dl [>16.7 mmol/l]), during pregnancy, or when any symptomsof ketoacidosis, such as nausea, vomiting, or abdominal pain,are present.

Ketones are normally present in urine, but concentrations areusually below the limit of detectability with routine testingmethods. However, positive ketone readings are found in normalindividuals during fasting and in up to 30% of first morningurine specimens from pregnant women. Urine ketone tests usingnitroprusside-containing reagents can give false-positive resultsin the presence of several sulfhydryl drugs, including the antihypertensivedrug captopril. False-negative readings have been reported whentest strips have been exposed to air for an extended periodof time or when urine specimens have been highly acidic, suchas after large intakes of ascorbic acid.

Ketone testing materials should be available in the office/clinicsetting. Health care professionals should be aware, however,that currently available urine ketone tests are not reliablefor diagnosing or monitoring treatment of ketoacidosis. Bloodketone testing methods that quantify ß-hydroxybutyricacid, the predominant ketone body, are available and are preferredover urine ketone testing for diagnosing and monitoring ketoacidosis.Home tests for ß-hydroxybutyric acid are available.

	GLYCATED PROTEIN TESTING

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

Glycated hemoglobin (GHb) testing
GHb, also referred to as glycohemoglobin, glycosylated hemoglobin,HbA_1c, or HbA₁, is a term used to describe a series of stableminor hemoglobin components formed slowly and nonenzymaticallyfrom hemoglobin and glucose. The rate of formation of GHb isdirectly proportional to the ambient glucose concentration.Since erythrocytes are freely permeable to glucose, the levelof GHb in a blood sample provides a glycemic history of theprevious 120 days, the average erythrocyte life span. GHb mostaccurately reflects the previous 2-3 months of glycemic control.

Many different types of GHb assay methods are available to theroutine clinical laboratory. Methods differ considerably withrespect to the glycated components measured, interferences,and nondiabetic range. Glycated hemoglobin is often reportedas hemoglobin A_1c. HbA_1c has become the preferred standard forassessing glycemic control. In referring to this test, the term"A1C test" will be used.

The A1C test has been shown to predict the risk for the developmentof many of the chronic complications in diabetes, analogousto using cholesterol determinations to predict the risk fordevelopment of cardiovascular disease. However, optimal useof the A1C test for this purpose requires the standardizationof A1C test assays. Without standardization, reported resultsbetween laboratories may not be comparable, even if both laboratoriesuse the same assay method. The National Glycohemoglobin StandardizationProgram (http://web.missouri.edu/~diabetes/ngsp.html), sponsoredin part by the American Diabetes Association to standardizeA1C test determinations to DCCT values, began in mid-1996. Onan annual basis, manufacturers of A1C test assay methods areawarded a "certificate of traceability to the DCCT referencemethod" if their assay method passes rigorous testing criteriafor precision and accuracy. It is desirable that laboratoriesuse only A1C test assay methods that have passed certificationtesting. It is also desirable that all laboratories performingA1C testing participate in the College of American Pathologistsproficiency testing survey for A1C testing started in mid-1996,which uses whole-blood specimens. Regardless of the assay methodtype and specific analyte qualified, all results should be reportedas "% HbA_1c" or "% HbA_1c equivalents."

A1C testing should be performed routinely in all patients withdiabetes, first to document the degree of glycemic control atinitial assessment, then as part of continuing care. Since theA1C test reflects a mean glycemia over the preceding 2–3months, measurement approximately every 3 months is requiredto determine whether a patient’s metabolic control hasreached and been maintained within the target range. Thus, regularA1C testing permits detection of departures from the targetrange in a timely fashion. For any individual patient, the frequencyof A1C testing should be dependent on the treatment regimenused and on the judgment of the clinician. In the absence ofwell-controlled studies that suggest a definite testing protocol,expert opinion recommends A1C testing at least two times a yearin patients who are meeting treatment goals (and who have stableglycemic control) and more frequently (quarterly assessment)in patients whose therapy has changed or who are not meetingglycemic goals.

Proper interpretation of A1C test results requires that healthcare providers understand the relationship between test resultsand average blood glucose, kinetics of the A1C test, and specificassay limitations. Data from the DCCT relating A1C test resultsto mean plasma glucose levels appear in Table 1 (4), but thesedata should be used with caution if the A1C test assay methodis not certified as traceable to the DCCT reference method.

 
A1C test values in patients with diabetes are a continuum; theyrange from normal in a small percentage of patients whose averageblood glucose levels are in or close to the normal range tomarkedly elevated values, e.g., >9.5%, in some patients,reflecting an extreme degree of hyperglycemia. Specific treatmentgoals should be individualized, but one must take into accountthe results of studies, such as the DCCT, showing a direct relationshipbetween A1C test values and the risk of many of the chroniccomplications of diabetes. The American Diabetes Associationrecommends that the goal of therapy should be an A1C resultof <7% and that physicians should reevaluate and, in mostcases, significantly change the treatment regimen in patientswith A1C test results consistently >8%. Again, these specificA1C values apply only to assay methods that are certified astraceable to the DCCT reference method.

Glycated serum protein (GSP)
Because the turnover of human serum albumin is much shorter(half-life of 14–20 days) than that of hemoglobin (erythrocytelife span of 120 days), the degree of glycation of serum proteins(mostly albumin) provides an index of glycemia over a shorterperiod of time than does glycation of hemoglobin. Measurementsof total GSP and glycated serum albumin (GSA) correlate wellwith one another and with measurements of glycated hemoglobin(A1C test). In situations where the A1C test cannot be measuredor may not be useful (e.g., hemolytic anemias), the GSP assaymay be of value in the assessment of the treatment regimen.Several methods have been described that quantify either totalGSP or total GSA. One of the most widely used is called thefructosamine assay. Values for GSP vary with changes in thesynthesis or clearance of serum proteins that can occur withacute systemic illness or with liver disease. In addition, thereis continuing debate as to whether fructosamine assays shouldbe corrected for serum protein or serum albumin concentrations.

A single measurement of GSP provides an index of glycemic statusover the preceding 1-2 weeks, while a single A1C test providesan index of glycemic status over a considerably longer periodof time, 2-3 months.

Measurement of GSP (including fructosamine) has been used todocument relatively short-term changes (e.g., 1–2 weeks)in glycemic status, such as in diabetic pregnancy or after majorchanges in therapy. However, further studies are needed to determineif the test provides useful clinical information in these situations.

Simultaneous measurements of GSP and the A1C test might complementone another and provide more useful clinical information thanthe A1C test alone. However, additional studies are needed toconfirm the clinical utility of this approach.

Measurement of GSP, regardless of the specific assay method,should not be considered equivalent to the A1C test, since itonly indicates glycemic control over a short period of time.Therefore, GSP assays would have to be performed on a monthlybasis to gather the same information as measured by the A1Ctest three to four times a year. Unlike the A1C test, GSP hasnot yet been shown to be related to the risk of the developmentor progression of chronic complications of diabetes.

	Footnotes

The initial draft of this paper was prepared by David E. Goldstein,MD, Chair; Randie R. Little, PhD; Rodney A. Lorenz, MD; JohnI. Malone, MD; David M. Nathan, MD; and Charles M. Peterson,MD. The paper was peer-reviewed, modified, and approved by theProfessional Practice Committee and the Executive Committee,November 1996. Most recent review/revision, 2000.

Abbreviations: DCCT, Diabetes Control and Complications Trial;GSA, glycated serum albumin; GSP, glycated serum protein; MNT,medical nutrition therapy; SMBG, self-monitoring of blood glucose.

	References

TOP INTRODUCTION BLOOD GLUCOSE TESTING BY... BLOOD GLUCOSE TESTING BY... URINE GLUCOSE TESTING GLYCATED PROTEIN TESTING References

• Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan D, Peterson CM: Tests of glycemia in diabetes (Technical Review). Diabetes Care 18:896–909, 1995[Medline]
• American Diabetes Association: Self-monitoring of blood glucose (Consensus Statement). Diabetes Care 17:81–86, 1994[Medline]
• American Diabetes Association: Self-monitoring of blood glucose (Consensus Statement). Diabetes Care 10:93–99, 1987
• Rohlfing CL, Wiedmeyer HM, Little RR, England JD, Tennill A, Goldstein DE: Defining the relationship between plasma glucose and HbA_1c: analysis of glucose profiles and HbA_1c in the Diabetes Control and Complications Trial. Diabetes Care 25:275–278, 2002[Abstract/Free Full Text]