Laboratory Diagnosis of Diabetes Mellitus

OBJECTIVES

By the end of this session the reader should be able to:

• Describe the regulation of blood glucose
• Understand the analysis of blood glucose
• Identify appropriate laboratory tests for the diagnosis of diabetes
• Describe the major laboratory findings in diabetes mellitus type I
• Describe the major clinical findings in diabetes mellitus type I
• Describe the major laboratory findings in diabetes mellitus type II
• Describe the major clinical findings in diabetes mellitus type II
• Describe the differences between type 1 and type 2 diabetes mellitus
• Describe the laboratory and clinical findings in impaired glucose tolerance
• Discuss maturity onset diabetes of the young
• Discuss the hormonal regulation of blood glucose levels, noting what causes a decrease in concentrations and what causes an increase in concentrations
• Describe non-diabetic conditions in which hyperglycemia is noted
• Discuss appropriate sample collection for measurement of serum/plasma glucose
• Discuss the performance of the glucose tolerance test and the findings in diabetes mellitus and in impaired glucose tolerance; discuss its performance in pregnant women
• Discuss the determination of ketone bodies in serum
• Describe C-peptide and its diagnostic use
• Identify non-diabetic conditions in which glycosuria is not associated with hyperglycemia
• Identify non-diabetic conditions in which glycosuria is associated with hyperglycemia
• Discuss the use of glycated hemoglobin in the monitoring of treatment of diabetes mellitus
• Discuss the difference between glycated hemoglobin measured by ion-exchange high-performance liquid chromatography and that measured by affinity chromatography
• Describe the effect of various hemoglobin variants on the measurement of glycosylated hemoglobin
• Identify appropriate laboratory tests for monitoring diabetic therapy
• Discuss the use of glycated proteins in monitoring the treatment of diabetes mellitus
• Describe the measurement and use of albumin in urine in the monitoring of diabetes mellitus

KEY TERMS

Acromegaly - growth hormone excess in adults; characterized by enlargement of features such as the head, hands, and feet

Angiogenesis - abnormal proliferation of blood vessels in a tissue such as the eye lens; a complication of diabetes mellitus

Gestational diabetes - Glucose intolerance that occurs in some pregnancies.

Glucagon - a hormone produced by the α-cells of the pancreas; glucagon is primarily involved in energy release

Glucagonoma - excessive glucagon levels caused by a tumor

Gluconeogenesis - formation of glucose from molecules that are not themselves carbohydrates (e.g., amino acids, lactate and glycerol)

Glycated hemoglobin - hemoglobin that has been glycated at one or more amino acid residues

Glycation - the non enzymatic covalent addition of glucose to amino groups of proteins

Glycogenolysis - the breakdown of glycogen to form glucose

Hemoglobin A_1c - hemoglobin that has been glycated on the terminal valine of the beta chain

Hyperglycemia - high blood glucose concentrations

Hyperglycemic hyperosmolar non-ketotic coma (HHNC) - a complication of diabetes mellitus characterized by hyperglycemia, hyperosmolality, normal ketoacid levels, and lethargy or coma

Islet cell antibodies (ICA) - antibodies frequently found in type I diabetes that are suggestive of an autoimmune cause

Islet of Langerhans - group of cells in the pancreas composed of α-cells, which secrete glucagon; β-cells, which secrete insulin; and δ-cells, which secrete somatostatin

Ketonemia - excess of ketones and derived ketoacids in the blood

Ketonuria - excess of ketones and derived ketoacids in the urine

Lactic acidosis - acidosis (low blood pH) caused by excess lactic acid

Polydipsia - excessive thirst; a symptom of diabetes mellitus

Polyphagia - constant hunger; a symptom of diabetes mellitus

Polyuria - excessive urinary output; a symptom of diabetes mellitus.

Proinsulin - precursor to insulin

Retinopathy - a disorder of the retina, often a complication of diabetes mellitus, caused by cataract formation or proliferation of small blood vessels (angiogenesis)

BACKGROUND AND SIGNIFICANCE

Diabetes mellitus is a common complex group of metabolic diseases; it is estimated that 5% of the total U.S. population, and up to 20% of older adults, are affected by the disease and it is the 4th leading cause of death in America. Diabetes is characterized by under-utilization of glucose leading to hyperglycemia, fasting and postprandial, resulting from decreased action of insulin. This in turn may occur from insufficient production of the hormone (injury and destruction of pancreatic islet β cells, as in Type I diabetes) or from resistance of target tissues to the full activity of insulin (Type II diabetes).[1] These two pathogenic mechanisms are the basis of most forms of idiopathic diabetes. It should, however, be remembered that excessive production of the hormones that drive metabolic pathways of glucose production may also result in a diabetic picture.

The laboratory diagnosis of diabetes is critical as it is estimated that only half of the patients with diabetes have been diagnosed and that most patients have diabetes for 7 years prior to diagnosis. The correct diagnosis is essential for proper drug therapy. The hormonal regulation of blood glucose levels is summarized in Table 1.

Table 1: Hormonal Regulation of Blood Glucose Levels
Glucose decrease	Glucose increase
Insulin:	Growth hormone (insulin antagonist)
Glucose transport across cell membranes	ACTH (increase glucocorticoids)
Glucose metabolism	Glucocorticoids (increase gluconeogenesis)
Glycogen synthesis	Epinephrine (increase glycogenolysis + gluconeogenesis)
Lipogenesis	Glucagon (increase glycogenolysis + gluconeogenesis)
	Thyroxine (increase glycogenolysis + gluconeogenesis)
	Somatostatin (decrease insulin secretion)

CLASSIFICATION OF DIABETES MELLITUS (NATIONAL DIABETES DATA GROUP, AMERICAN DIABETES ASSOCIATION)

Type 1 Diabetes Mellitus

• Formerly known as “juvenile onset diabetes mellitus” and “insulin dependent”
• Caused by pancreatic islet β-cell destruction

Laboratory findings

• Elevated fasting glucose levels (≥ 126 mg/dL)
• 2-hour postprandial or casual glucose, >200 mg/dL
• Glycosuria
• Ketonuria and ketonemia frequently present
• Insulinopenia
• Elevated glycated hemoglobin
• Islet cell antibodies, mostly found within the first year after diagnosis
• Increased prevalence of the HLA antigens DR3 and DR4

Clinical findings

• Proneness to ketosis
• Lifelong dependence on injected insulin for control of ketosis and preservation of life
• Frequent onset at young age (peak at around puberty)
• Positive family history (variable)
• Possible association with failure of other endocrine organs on an autoimmune basis (hypothyroidism, hypoadrenalism)

Type 2 Diabetes Mellitus

• Formerly known as "adult onset diabetes mellitus" or "non-insulin dependent"
• Results from insulin resistance

Laboratory findings

• Normal or elevated fasting glucose levels (≥ 126 mg/dL)
• Elevated postprandial glucose levels
• Abnormal glucose tolerance test
• Variable glycosuria
• Elevated glycated hemoglobin
• Variable levels of insulin, often above normal
• Decreased number of insulin receptors on various tissues

Clinical findings

• Generally ketosis-resistant; however, ketosis may develop under special stress conditions
• Resistance to the action of insulin
• Onset more frequently after 40 years of age
• Often associated with obesity
• Strong family history

Impaired Fasting Glucose (IFG) and Impaired Glucose Tolerance (IGT)

Laboratory findings

• IFG = fasting glucose levels 100 - 125 mg/dL
• 2 hour plasma glucose levels from a 75-g OGTT: 140 - 199 mg/dL
• A1C: 5.7 - 6.4%
• Note that the risk for developing diabetes is continuous and extends below the lower part of the range while becoming greater at the higher limit of the range.

Clinical findings

• Increased prevalence of atherosclerosis, electrocardiographic abnormalities, hypertension, hyperlipidemia, obesity
• Higher age group
• At higher risk to develop diabetes mellitus

Gestation Diabetes/IGT

This class of patients is defined as women in whom during pregnancy diabetes or IGT become manifest. After pregnancy, the condition usually reverses to normal, but in some patients diabetes or IGT persists.

Previous Abnormality of Glucose Tolerance

This classification is applied to persons who, on previous occasions, have had diabetic hyperglycemia or IGT as demonstrated by a glucose tolerance test. These persons are not diabetic, but as a group have a great risk of becoming diabetic. Terms such as "latent diabetes" or "prediabetes" should not be used.

Maturity Onset Diabetes of the Young (MODY)

MODY is characterized by early onset, non-insulin dependent diabetes that is inherited in an autosomal dominant pattern. The major presentation of MODY is in patients under age 25 and with nonketotic hyperglycemia. Patients have a nonobese body habitus and may even be quite thin. It is estimated that 2 to 5% of all childhood cases of diabetes are caused by MODY. Treatment normally consists of diet modification and/or oral hypoglycemic agents. The use of insulin is rarely required.

The diagnostic criteria for MODY include:

• early onset diabetes
• autosomal dominant inheritance (multi-generational)
• not insulin dependent

Six different subclasses of MODY have been recognized (MODY-1 through MODY-6), and each subclass is associated with a mutation in a specific gene. The genes associated with MODY include glucokinase (phosphorylates glucose in beta-cells) and various transcription factors (e.g. HNF or IPF).

The most common forms of MODY are MODY-3 (hepatic transcription factor-1 gene; TCF1) and MODY-2 (glucokinase gene), and together account for 80% of all cases in Caucasians but are less common in other ethnic groups.

MODY-2 patients have a persistently raised fasting glucose and the increase in fasting glucose does not increase with age. The patients are often asymptomatic and the finding of hyperglycemia is incidental. MODY-2 patients generally respond to diet modification.

MODY-3 patients (mutation in HNF1-alpha) also have an elevated fasting glucose but the increase in fasting glucose levels increases with age, and glycosuria can develop. MODY3 is often misdiagnosed as type 1 diabetes due to the development of glycosuria and thin habitus of the patient. However, patients with MODY-3 still have detectable C-peptide whereas in type I diabetes patients the C-peptide is usually not detectable. Usually parents and/or grandparents of MODY-3 patients have been diagnosed as diabetic. MODY-3 patients often respond to diet modification, but with age require the additional use of oral hypoglycemic agents. A portion of MODY-3 patients will eventually require insulin for control of hyperglycemia.

Hyperglycemia Associated with Certain Conditions or Syndromes

• Postpancreatectomy
• Cancer of pancreas
• Pancreatic cysts
• Hemorrhagic pancreatitis
• Acromegaly
• Cushing’s syndrome
• Pheochromocytoma
• Glucagonoma
• Somatostatinoma
• Primary aldosteronism
• Hyperthyroidism
• Hemochromatosis
• Malnutrition
• Many (100+) drugs and chemical agents (e.g., thyroxine, phenytoin)
• Over 25 rare genetic syndromes (e.g., Down’s syndrome, Klinefelter’s syndrome etc.)

TESTS IN THE DIAGNOSIS, CLASSIFICATION AND MANAGEMENT OF DIABETES MELLITUS

Determination of Blood Glucose: Where, How and When

Glucose concentration is uniform in the water phase of plasma and erythrocytes. Since, however plasma contains per unit volume 27% more water than erythrocytes, glucose levels are higher in a given volume of plasma than in an identical volume occupied by erythrocytes. For this reason plasma glucose values are higher than whole blood glucose values.

Example: If glucose concentration is 120 mg/dL of water then:

	Water (%)	Glucose (mg/dL)
Erythrocytes	73	88
Whole blood (Hct 45%)	83	101
Plasma	93	112

In the fasting state, glucose levels in arterial and venous blood are similar. However postprandially, arterial and capillary blood have glucose levels about 20 mg/dL higher than venous blood. This is because the extraction of glucose by the tissues in the presence of insulin elevated in response to nutrient absorption in the gastrointestinal tract.

Quantitative determinations of glucose are based on a variety of chemical and enzymatic methods. The older chemical methods are less specific. Newer enzymatic methods (glucose oxidase, hexokinase) for glucose analysis are highly specific. Although they may occasionally tend to give values too low because of enzymatic inhibitors in the sample, hexokinase methods are the most accurate and widely used. In whole blood, clotted and kept at room temperature, glucose disappears at a rate of approximately 7% per hour owing to the ongoing glycolytic activity of leukocytes and red cells. It is thus preferable to collect blood in tubes containing fluoride, a strong inhibitor of glycolysis as well as citrate (acidity) to immediately inhibit glycolysis, or to separate the serum immediately[2]. The grey-stoppered Vacutainer tubes contain sodium fluoride in addition to potassium oxalate, the latter added to prevent clotting. Blood samples collected in these tubes cannot, however, be used for most other chemical examinations, particularly electrolytes, since the concentrations of sodium and potassium in this anticoagulant are: sodium = 85 mmol/L, and potassium = 30 mmol/L.

Fluoride has a weaker inhibitory effect on the glycolysis by white cells; consequently, a significant decrease of glucose may be seen in fluoridated blood upon standing when white counts are abnormally high (50,000/μL or greater).

Normal and abnormal values cited below reflect measurements of glucose performed in venous plasma. We shall become familiar with these standardized values, keeping in mind the rationale and factors for conversion between plasma and whole blood glucose, and between venous and capillary glucose (in the postprandial state).

Measurements of plasma glucose, in order to be easily interpretable, should be obtained under selected circumstances: (1) in the fasting state, or (2) during an oral glucose tolerance test, or (3) two hours after a 100 g carbohydrate meal (postprandial).

Fasting plasma glucose (FPG)

The measurement of fasting glucose yields information on the capability of basal levels of insulin to control glycogenolysis and gluconeogenesis. When basal production (or action) of insulin is insufficient, fasting hyperglycemia develops.

Determination of fasting plasma glucose is the very first procedure to be performed when entertaining a diagnosis of diabetes.

Patient preparation

The patient is kept on an average usual diet and, after an overnight fast (no food or sweetened drinks after midnight), blood is drawn in the morning.

Interpretation

Range	Interpretation
<100 mg/dL	Normal
100 - 125 mg/dL	May be indicative of IGT (perform glucose tolerance test)
≥ 126 mg/dL	On more than one occasion is indicative of diabetes mellitus; there is no need to perform a glucose tolerance test

Oral glucose tolerance test (OGTT) [3, 4]

This test is intended to measure the capability and timely response of the insulin-secreting cells to the integrated signals provided by GI hormones and rising blood glucose levels. Although 75 or 100 g of pure glucose obviously does not constitute a typical balanced meal, the test is designed to achieve maximal sensitivity for diagnostic and epidemiologic purposes.

Patient preparation

Put the patient for 3 days or more on a normal diet including at least 150 g of carbohydrates per day. In the morning after an overnight fast, 75 or 100 g of an aqueous solution of glucose is given. The various commercially prepared cola-flavored carbonated test drinks are convenient to use and are more palatable than a pure glucose solution. The patient should drink the solution within 5 minutes. If a patient vomits after the test drink, the test is invalid and must be repeated.

For the 75 g glucose tolerance test, blood is drawn at 0 (fasting), 1, and 2 hours after administration of glucose. For the 100 g glucose tolerance test, blood is drawn at 0 (fasting), 1, 2, and 3 hours.

In pregnant women, i.e. for the detection of gestational diabetes, the test dose is 100 g glucose. Blood is drawn at: 0, 1, 2, and 3 hours.

In patients with carbohydrate malabsorption, an intravenous glucose tolerance test replaces the oral glucose tolerance test. Reference ranges differ.

Interpretation

• Normal (in non-pregnant adult)
  • Fasting value: <95 mg/dL
  • Value at 1 hour: < 180 mg/dL
  • Value at 2 hours: < 155 mg/dL
  • Value at 3 hours: < 140 mg/dL
• Indicative of impaired glucose tolerance (IGT)
  • Fasting value: 110 - 126 mg/dL
  • At least one of the values at 30, 60, or 90 min > 200 mg/dL and value at 120 min between 140 and 200 mg/dL
• Indicative of diabetes mellitus
  • If the fasting glucose determination revealed diabetic values (>126 mg/dL), the OGTT should not be performed
  • If the fasting glucose fell into the IGT range (110-126 mg/dL) and an OGTT is performed, the results are indicative of diabetes mellitus if two or more of the venous plasma concentrations are reached or exceeded.

	NL	IGT	DM
F Gluc	<100	100 - 125	≥ 126
2 h OGGT	<140	140 - 199	≥ 200

The criteria for diagnosis of diabetes during pregnancy (gestational diabetes) are stricter than outlined above for non-pregnant adults. This is because even mild diabetes during pregnancy becomes a significant risk factor for fetal morbidity and mortality. Thus, the OGTT is performed with 100 g of glucose and it indicates gestational diabetes when two or more of the following values (in mg/dL) are reached or exceeded: fasting 110, 1 hour 190, 2 hours 165, 3 hours 145.

Two hour postprandial plasma glucose

This determination has no standardized role for diagnostic purposes. It is, however, often valuable when attempting to optimize patients' treatment. Normally, 120 min values are below 140 mg/dL.

Determination of Ketone Bodies

Ketone bodies (acetoacetate, β-hydroxybutyrate and acetone) are produced by the liver from oxidation of free fatty acids (FFA). Both release of FFA from adipose tissue stores (lipolysis) and their metabolism to ketones are highly accelerated by insulin deficiency. Thus, in the insulin-deficient state (characteristic of untreated type 1 diabetes) the liver produces ketone bodies and hydrogen ions resulting in ketoacidosis. This is often the presenting picture for juveniles who have developed type 1 diabetes. It will recur if insulin treatment is discontinued or if it is not increased when there is increased insulin demand (stress, infections, pregnancy). Ketoacidosis is rarely the presenting picture for type 2 diabetic patients; however, such patients may develop some degree of ketoacidosis under stressful circumstances.

The most convenient method for estimating the degree of ketonemia and ketonuria utilizes nitroprusside tablets (Acetest) or reagent strips (Ketostix). The test is performed by reacting 0.5 mL of serum or urine with a crushed Acetest tablet, or a Ketostix. If the reaction is negative, ketoacidosis is (with the caveat discussed below) excluded. If, however, the reaction is positive (purple color development), then serial dilution of the specimen is performed until the highest dilution is established that still gives a positive reaction.

Nitroprusside reacts with acetoacetate and, to a much lesser degree, with acetone (only 1/20 of the reactivity of acetoacetate), and fails to react with β-hydroxybutyrate. These differences in reactivity pose a potential source of problems: during mild ketosis (fasting, suboptimal insulin effect), the ratio of β-hydroxybutyrate to acetoacetate is ~2:1 and the latter ketone body is well represented, whereas in situations of prolonged, severe insulin deficiency and poor tissue oxygenation, β-hydroxybutyrate will be the prevalent or exclusive ketone body present. The nitroprusside reaction may thus be only weakly positive, or negative, in spite of the presence of dangerous quantities of β-hydroxybutyrate. The suspicion that this is so, however, is generally high because the concomitant hyperglycemia and the calculation of the anion gap will reflect excess anions requiring identification. The finding of a level of lactic acid increased enough to indicate compromised tissue oxygenation but not sufficiently increased to account for the anion gap suggests the presence of ketonemia due to β-hydroxybutyrate. It is interesting that, in such cases, the positivity of the nitroprusside test for serum ketones will become stronger as treatment with insulin and fluid is improving the patient’s status, due to formation of more acetoacetate.

From the foregoing it is obvious that "quantitation" of blood ketone bodies with nitroprusside is not straightforward. An estimate may be made by knowing that the reagent is sensitive to 0.5 - 1 mmol/L acetoacetate. Thus positivity at a 1:8 dilution will indicate approximately 4 - 8 mmol/L acetoacetate. This is multiplied by 3 to take into account the contribution of the unmeasurable β-hydroxybutyrate. Such a crude estimate can be compared with the calculated anion gap to verify that the latter is in fact accounted for by ketone bodies. Although serum acetone concentrations are markedly elevated in patients with diabetic ketoacidosis, this ketone, resulting from the spontaneous conversion of acetoacetate to acetone and carbon dioxide, does not contribute to the circulating anions and is readily excreted in breath and urine.

Determination of Insulin and C-peptide

Insulin is synthesized first as a precursor molecule, proinsulin. The A and B chains in proinsulin are held together by a connecting peptide, called C-peptide. Proinsulin is then converted in the beta cells to insulin which is secreted together with C-peptide in equimolar amounts. However, the t_1/2 of C-peptide is longer than that of insulin. C-peptide assays do not react with insulin antibodies. Human insulin differs from beef insulin by 3 amino acids and from porcine insulin by one amino acid, so that the latter two result in antibody formation (beef insulin more so than porcine insulin). Insulin has been synthesized commercially for clinical use by recombinant DNA technology and is now available in this form. This development will ultimately resolve the problem of antibody formation.

Measurements of serum insulin and C-peptide are not warranted for initial diagnosis of diabetes. They are mostly used to verify classification and for various investigational purposes. Measurements are performed by radioimmunoassay. The availability of a radioimmunoassay for C-peptide has been extremely useful in permitting studies of insulin secretory capacity of the pancreatic β-cells in individuals who had developed antibodies to the hormone following treatment with non-human insulin. Such antibodies interfere in the immunoassay procedures for insulin. Measurement of C-peptide is also useful in detecting surreptitious use of insulin since, with administration of exogenous insulin, serum insulin concentrations will be elevated without equivalent elevation of C-peptide levels. However, endogenous production of insulin is accompanied by elevation of C-peptide.

Glycosuria

Glycosuria appears when the blood glucose level exceeds the renal threshold for reabsorption of glucose. This normally lies at about 170 mg/dL, but it may be lower in renal tubular disease or elevated in diabetics to above 300 mg/dL. Thus, glycosuria may be absent in these patients despite markedly elevated blood glucose levels.

Non-diabetic causes of glycosuria are:

Glycosuria not associated with hyperglycemia

• Pregnancy
• Fanconi’s syndrome
• Renal glycosuria (“true renal diabetes”)
• Nephrotic syndrome

Glycosuria associated with hyperglycemia

• Thyrotoxicosis
• Cushing’s syndrome
• Damage to CNS
• Infections
• Anesthesia
• Excitement and stress

Semi-quantitative tests for glycosuria are described under routine urinalysis. They are based either on the use of glucose oxidase as in the dipstick or on copper reduction methods as in the tablet test (Clinitest).

Quantitative urinary glucose determinations are seldom necessary. The glucose concentrations in the urine of diabetics are commonly in the range of 0.5 to 2.0 g/dL, but values over 5 g/dL are occasionally encountered. The effect of high glucose concentration on urinary specific gravity should be kept in mind.

Glycylated-Hemoglobin

During the 120 day life span of the red cell, hemoglobin A and other forms become glycated due to the non-enzymatic, largely irreversible, post-translational attachment of a molecule of glucose to the terminal valine of the β-globin chain and to ε-amino groups of lysines of both α- and β-chains. The degree of glycation is directly proportional to the level of glucose in the blood, and it has been shown that the amount of glycated hemoglobin present in blood is a reflection of the average blood glucose level over the life span of a red cell. Thus, the quantitative determination of glycated hemoglobin has become a useful adjunct in the assessment of the efficacy of long-term therapeutic control of diabetic patients.

Glycated hemoglobin can be measured in several ways. The two most common methods are ion exchange and affinity chromatography. When measured by ion exchange the results are reported as HbA_1c and when measured by affinity chromatography, the results are GHb (total glycated hemoglobin).

Glycated hemoglobin results are reported as % HbA_1c.

HbA_1c by ion exchange

The analysis of Hemoglobin A_1c is performed by high performance liquid chromatography (HPLC) with a cation exchanger.

Reference Range: 3.8-6.3%; Target for therapy is < 7% [5]

• 6.3 - 8.3% = slightly elevated, treatment needs to be optimized
• 8.3 - 12.3% = elevated, treatment needs to be intensified
• over 12.3% = generally found in patients only at diagnosis

Total Glycylated Hemoglobin (GHb) by affinity chromatography

Measurement of total glycated hemoglobin using boronic-acid affinity chromatography has become a widely used method in the United States. Unlike the ion exchange method, this method detects all of the glycated hemoglobins (A, F, S, C, etc.) and does not distinguish among them, and thus it yields somewhat higher results than does the ion exchange procedure.

Both HbA_1c and GHb have strong advocates. Both methods provide results that permit patient management, but it does require knowledge of the method used.

Effect of Hemoglobin Variants on Measurement of Glycosylated Hemoglobin

Hemoglobin variants can affect the measurement of A_1c by ion exchange HPLC. The most common hemoglobin variants, AS and AC trait do not effect the measurement of A_1c by the HPLC methods, but the effect of other variants is unknown. When a patient has a hemoglobin variant other than AS or AC trait, hemoglobin A_1c reference ranges may not apply. In these cases A_1c should only be used to monitor therapy after obtaining a baseline value. If t_1/2 of red cells is shorter, as in some hemoglobinopathies (thalassemias and hemolytic anemias), the HbA_1c is underestimated.

Abnormal red cell survival will also affect interpretation of hemoglobin A_1c. Samples from patients with hemolytic anemia will have lower A_1c values due to the decreased lifetime of the red blood cells. Samples from patients post-splenectomy and patients with polycythemia may exhibit increased A_1c values due to the longer half-life of red blood cells in these patients. Hemoglobin variants that effect red blood cell survival will also effect interpretation of A_1c.

One formula for estimating average glucose (AG) based on HbA_1c values in stable diabetic patients is [6]:

(1)

\begin{align} AG \ = \ 28.7 \ (Hb {A}_{1c}) \ - \ 46.7 \end{align}

Glycated Proteins

Due to their shorter plasma half lives, glycated proteins such as glycated albumin may be useful in assessing the average blood glucose concentration over one to two weeks. The most popular method at this time is the fructosamine assay which is a measure of glycated plasma proteins (primarily Albumin). Since the t_1/2 of Hb is 60 days, HbA_1c and GHb values reflect glycemic control for the previous 6-8 weeks. For fructosamine, the period of glycemic control is based on albumin (t_1/2 = 18 days) and a variety of other proteins (t_1/2 range from 2.5 to 23 days). Therefore, fructosamine reflects control over the previous 2-3 weeks. While this method is still being critically evaluated it is being used at some sites. It should be used with caution as many investigators believe that the assay is not reliable.

Proteinuria

• Used to monitor progress of renal involvement of diabetes
• Angiotensin-converting enzyme inhibitors (ACE) appear to be beneficial in preventing further loss of function

Diabetic patients, even with recent onset, have increased urinary excretion and clearance of proteins with a MW higher than 40,000 when compared with non-diabetics. This increased glomerular permeability can become non-selective for the larger proteins and may progress to several grams of protein loss/24 hour. Since the appearance of proteinuria (albumin) is the earliest sign of renal involvement by the diabetic process, it is advisable to perform at least semi-quantitative tests for proteinuria eventually to be followed by precise quantitation (see Urinalysis) for albumin. The first signs of renal damage are signified by the appearance of microalbuminuria. (Microalbuminuria is a misnomer, which implies small amounts of albumin and not a small, abnormal molecule of albumin). Microalbuminuria may occur at a time when there is not yet any evidence of decreased glomerular filtration rate or evidence of glomerular lesions. It is a strong, early predictor of impending nephropathy in diabetic patients. Microalbuminuria is not detected with routine protein methods, and requires special techniques. The preferred way to measure proteinuria includes normalizing the value to the amount of creatinine excreted and reporting an albumin to creatinine ratio.

Anti-Insulin Antibodies

Before the availability of highly purified animal insulin preparations and recombinant human insulin, development of IgG antibodies to insulin could be demonstrated in nearly all patients receiving insulin therapy for more than a few weeks.The extent of antibody formation is a function of: (1) the patient's inherent immune response; (2) the species source of the insulin (beef is more immunogenic than pork); (3) the pharmaceutical form (long-acting preparations are more immunogenic than regular); and (4) purity of the insulin preparation. The clinical significance of these antibodies is not completely known, and in most patients they do not appear to be of consequence. In rare patients, however, insulin resistance may develop as a result of increased titers of IgG antibodies directed against beef insulin. In most laboratories, when this titer is 30 - 50 U/L, immunologic resistance is considered to be present. Occasionally an increased titer to pork insulin will be reported. Since this frequently represents cross-reaction between beef antibody and pork insulin in vitro, an increased titer to pork insulin does not necessarily preclude the clinical efficacy of pork insulin in a given patient.

With the currently increasing use of purified pork insulin and human insulin preparations, problems of immunologic resistance to insulin are quickly disappearing.

Table 2: Laboratory Tests in the Diagnosis and Management of Hyperglycemic Emergencies
Plasma glucose	Sodium
SUN	Potassium
Creatinine	Bicarbonate
Ketone bodies	Chloride
Osmolality	Blood pH

Table 3: Criteria for the Diagnosis of Diabetes [4]
Symptoms of diabetes and a casual plasma glucose > 200 mg/dl. Casual is defined as any time of day without regard to time since last meal. The classic symptoms of diabetes include polyuria, polydipsia, and unexplained weight loss
OR
FPG ≥ 126 mg/dl. Fasting is defined as no caloric intake for at least 8 h. Should be repeated on a separate occasion.
OR
2-h plasma glucose ≥ 200 mg/dl during an OGTT. The test should be performed as described by the World Health Organization, using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water.
OR
A1C ≥ 6.5%. The test should be performed in a laboratory using a method that is NGSP certified and standardized tot he DCCT assay.

CASE STUDIES

Case 1

(Diabetes Technol. Ther. 4(4):505-514, 2002)

The patient is a 54-year-old postmenopausal, morbidly obese Hispanic woman with a history of asthma, allergic rhinitis, hypertension, and type 2 diabetes for 10 years (baseline A_1c 7.1). Noncompliance with diet and exercise resulted in a 10 lb (4.5 kg) weight gain within an 8-month period. The patient also started to exhibit poor glycemic control, resulting in HbA_1c values of 6.7%, 7.9%, and 7.5% over 13 months. The patient adamantly refused insulin therapy; therefore, a trial of rosiglitazone 4 mg once a day was added to the existing combination therapy of extended-release glipizide 20 mg once daily and metformin 1 g BID. After 6 weeks of therapy with rosiglitazone, the dose was increased to 8 mg/day (given in two divided doses per day) to achieve better glycemic control as exhibited by an HbA_1c of 6.5% at 3 months. Within 3 months of the dose escalation, the patient began complaining of bilateral pedal edema below the knee all the way to the feet, accompanied by moderate pain upon walking or standing but no pain at rest. The dose of rosiglitazone was reduced to 4 mg, and diuresis therapy was not initiated. At the 2-month follow-up, bilateral pitting edema remained evident, and rosiglitazone was discontinued. An overall 11 lb (5 kg) weight gain was recorded for the 8-month period of rosiglitazone treatment. Rosiglitazone was discontinued for a 6-month period and only trace amounts of bilateral pedal edema remained. However, this patient’s glycemic control continued to worsen with an ensuing HbA_1c of 7.9%.

1. Based on this patient’s baseline HbA_1c of 7.1% calculate her mean blood glucose before receiving drug therapy.
2. What was her mean blood glucose when her HbA_1c was 6.5%?
3. What are possible drug-induced explanations for her edema? Is edema a common ADR for this class of drugs?
4. Since this patient is failing drug therapy what alternative drug therapies would you suggest?

Case 2

A 10-year-old boy, in excellent health until 1 month ago, is noted to have become irritable in school and at home and somewhat lethargic after meals. Because his 15 year old brother has recently developed diabetes, the mother is concerned and asks the pediatrician to evaluate the boy for diabetes.

Fasting plasma glucose on two occasions is 130 and 140 mg/dL. A random plasma glucose in the afternoon is 190 mg/dL, HbA_1c is 8.5%, and there is no ketonuria. A nearby laboratory is requested to analyze a serum specimen for islet-cell antibodies, which are found to be present at a titer of 1:4.

1. Should the boy be considered to have diabetes, and (if so) what type?
2. Should one perform further testing and (if so) what tests should be ordered?

Case 3

A 68-year-old obese woman is brought semi-comatose to the emergency room. She has a history of seizures and hypertension and has been treated with phenytoin and thiazide diuretics. She is not known to have diabetes.

On physical examination she exhibits a right hemiparesis. Initial laboratory results: plasma glucose 1,080 mg/dL, serum sodium 144 mmol/L, potassium 4.4 mmol/L, chloride 113 mmol/L, bicarbonate 20 mmol/L, SUN 60 mg/dL.

1. Calculate the osmolality.
2. Is the serum sodium appropriate for the degree of hyperglycemia?
3. What is the effect of the hyperosmolality on the hematocrit?
4. Do you expect ketoacidosis to be present?
5. Why are the SUN and chloride increased, and the bicarbonate decreased?

Case 4

A 40-year-old slightly obese woman has had Type II diabetes for seven years. Her serum glucose was relatively well controlled on a sulfonylurea drug. Over the last five months her fasting glucoses have progressively increased despite doubling the dose of the oral anti-diabetic agent.

At today's clinic visit she is noted to have lost 12 lbs. in two months, and she has moderate ketonuria. She complains of being nervous and of having difficulty concentrating on her work. On physical examination, the skin is warm and sweaty. The BP is 160/70 mm Hg and pulse is 100/min. A blood specimen is drawn and sent to the laboratory.

1. What laboratory tests should be ordered?
2. Should the patient be switched to insulin therapy?

Bibliography

1. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus, Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 26:S5-S20, 2003

2. Gambino R et al., Acidification of blood is superior to sodium fluoride alone as an inhibitor of glycolysis, Clin Chem 55:1019-1021, 2009

3. World Health Organization, Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation, Geneva: WHO Dept. of Noncommunicable Disease Surveillance, Publication No. WHO/NCD/NCS 99.2, 1999

4. American Diabetes Association, Diagnosis and Classification of Diabetes Mellitus, Diabetes Care 33:562-569, 2010

5. Skyler JS et al., Intensive glycemic control and the prevention of cardiovascular events: implications of the ACCORD, ADVANCE, and VA diabetes trials: a position statement of the American Diabetes Association and a scientific statement of the American College of Cardiology Foundation and the American Heart Association, Diabetes Care 32:187-192, 2009

6. Nathan DM, et al., Translating the A1C assay into estimated average glucose values, Diabetes Care 31:1473-1478, 2008