ROLE OF LABORATORY TESTS IN DIABETES MELLITUS
Diabetes mellitus (DM) is a metabolic group of disorders characterized by persistent hyperglycemia due to deficiency and/or diminished effectiveness of insulin. There are derangements of carbohydrate, protein, and fat metabolism due to failure of insulin action on target cells.
Typical features of DM are as follows:
- Fasting hyperglycemia
- Symptoms due to marked hyperglycemia: polyuria, polydipsia, weight loss, weakness, polyphagia, and blurred vision
- Long-term complications like atherosclerosis (leading to ischemic heart disease, cerebrovascular disease, and peripheral vascular disease) and microangiopathy (which can cause nephropathy with risk of renal failure; retinopathy with potential loss of vision; and peripheral neuropathy with risk of foot ulcers, amputations, or Charcot joints).
- Acute metabolic complications (hyperosmolar hyperglycemic state, diabetic ketoacidosis).
- Susceptibility to infections especially of skin, respiratory tract, and urinary tract.
In DM, applications of laboratory tests are as follows:
- Diagnosis of DM
- Screening of DM
- Assessment of glycemic control
- Assessment of associated long-term risks
- Management of acute metabolic complications.
LABORATORY TESTS FOR DIAGNOSIS OF DIABETES MELLITUS
Diagnosis of DM is based exclusively on demonstration of raised blood glucose level (hyperglycemia).
The current criteria (American Diabetes Association, 2004) for diagnosis of DM are as follows:
Typical symptoms of DM (polyuria, polydipsia, weight loss) and random plasma glucose ≥ 200 mg/dl (≥ 11.1 mmol/L)
Fasting plasma glucose ≥ 126 mg/dl (≥ 7.0 mmol/L)
2-hour post glucose load (75 g) value during oral glucose tolerance test ≥ 200 mg/dl (≥ 11.1 mmol/L).
If any one of the above three criteria is present, confirmation by repeat testing on a subsequent day is necessary for establishing the diagnosis of DM. However, such confirmation by repeat testing is not required if patient presents with (a) hyperglycemia and ketoacidosis, or (b) hyperosmolar hyperglycemia.
The tests used for laboratory diagnosis of DM are (1) estimation of blood glucose and (2) oral glucose tolerance test.
Estimation of Blood Glucose
Measurement of blood glucose level is a simple test to assess carbohydrate metabolism in DM (Figure 837.1). Since glucose is rapidly metabolized in the body, measurement of blood glucose is indicative of current state of carbohydrate metabolism.
Glucose concentration can be estimated in whole blood (capillary or venous blood), plasma or serum. However, concentration of blood glucose differs according to nature of the blood specimen. Plasma glucose is about 15% higher than whole blood glucose (the figure is variable with hematocrit). During fasting state, glucose levels in both capillary and venous blood are about the same. However, postprandial or post glucose load values are higher by 20-70 mg/dl in capillary blood than venous blood. This is because venous blood is on a return trip after delivering blood to the tissues.
When whole blood is left at room temperature after collection, glycolysis reduces glucose level at the rate of about 7 mg/dl/hour. Glycolysis is further increased in the presence of bacterial contamination or leucocytosis. Addition of sodium fluoride (2.5 mg/ml of blood) maintains stable glucose level by inhibiting glycolysis. Sodium fluoride is commonly used along with an anticoagulant such as potassium oxalate or EDTA. Addition of sodium fluoride is not necessary if plasma is separated from whole blood within 1 hour of blood collection.
Plasma is preferred for estimation of glucose since whole blood glucose is affected also by concentration of proteins (especially hemoglobin).
There are various methods for estimation of blood glucose:
- Chemical methods:
– Orthotoluidine method
– Blood glucose reduction methods using neocuproine, ferricyanide, or copper.
Chemical methods are less specific but are cheaper as compared to enzymatic methods.
- Enzymatic methods: These are specific for glucose.
– Glucose oxidase-peroxidase
– Glucose dehydrogenase
Chemical methods have now been replaced by enzymatic methods.
Terms used for blood glucose specimens: Depending on the time of collection, different terms are used for blood glucose specimens.
- Fasting blood glucose: Sample for blood glucose is withdrawn after an overnight fast (no caloric intake for at least 8 hours).
- Post meal or postprandial blood glucose: Blood sample for glucose estimation is collected 2 hours after the subject has taken a normal meal.
- Random blood glucose: Blood sample is collected at any time of the day, without attention to the time of last food intake.
Oral Glucose Tolerance Test (OGTT)
Glucose tolerance refers to the ability of the body to metabolize glucose. In DM, this ability is impaired or lost and glucose intolerance represents the fundamental pathophysiological defect in DM. OGTT is a provocative test to assess response to glucose challenge in an individual (Figure 837.2).
American Diabetes Association does not recommend OGTT for routine diagnosis of type 1 or type 2 DM. This is because fasting plasma glucose cutoff value of 126 mg/dl identifies the same prevalence of abnormal glucose metabolism in the population as OGTT. World Health Organization (WHO) recommends OGTT in those cases in which fasting plasma glucose is in the range of impaired fasting glucose (i.e. 100-125 mg/dl). Both ADA and WHO recommend OGTT for diagnosis of gestational diabetes mellitus.
Preparation of the Patient
- Patient should be put on a carbohydrate-rich, unrestricted diet for 3 days. This is because carbohydrate-restricted diet reduces glucose tolerance.
- Patient should be ambulatory with normal physical activity. Absolute bed rest for a few days impairs glucose tolerance.
- Medications should be discontinued on the day of testing.
- Exercise, smoking, and tea or coffee are not allowed during the test period. Patient should remain seated.
- OGTT is carried out in the morning after patient has fasted overnight for 8-14 hours.
- A fasting venous blood sample is collected in the morning.
- Patient ingests 75 g of anhydrous glucose in 250-300 ml of water over 5 minutes. (For children, the dose is 1.75 g of glucose per kg of body weight up to maximum 75 g of glucose). Time of starting glucose drink is taken as 0 hour.
- A single venous blood sample is collected 2 hours after the glucose load. (Previously, blood samples were collected at ½, 1, 1½, and 2 hours, which is no longer recommended).
- Plasma glucose is estimated in fasting and 2-hour venous blood samples.
Interpretation of blood glucose levels is given in Table 837.1.
|Parameter||Normal||Impaired fasting glucose||Impaired glucose tolerance||Diabetes mellitus|
|(1) Fasting (8 hr)||< 100||100-125||—||≥ 126|
|(2) 2 hr OGTT||< 140||< 140||140-199||≥ 200|
OGTT in gestational diabetes mellitus: Impairment of glucose tolerance develops normally during pregnancy, particularly in 2nd and 3rd trimesters. Following are the recent guidelines of ADA for laboratory diagnosis of GDM:
- Low-risk pregnant women need not be tested if all of the following criteria are met, i.e. age below 25 years, normal body weight (before pregnancy), absence of diabetes in first-degree relatives, member of an ethnic group with low prevalence of DM, no history of poor obstetric outcome, and no history of abnormal glucose tolerance.
- Average-risk pregnant women (i.e. who are in between low and high risk) should be tested at 24-28 weeks of gestation.
- High-risk pregnant women i.e. those who meet any one of the following criteria should be tested immediately: marked obesity, strong family history of DM, glycosuria, or personal history of GDM.
Initially, fasting plasma glucose or random plasma glucose should be obtained. If fasting plasma glucose is ≥ 126 mg/dl or random plasma glucose is ≥ 200 mg/dl, repeat testing should be carried out on a subsequent day for confirmation of DM. If both the tests are normal, then OGTT is indicated in average-risk and high-risk pregnant women.
There are two approaches for laboratory diagnosis of GDM
- One step approach
- Two step approach
In one step approach, 100 gm of glucose is administered to the patient and a 3-hour OGTT is performed. This test may be cost-effective in high-risk pregnant women.
In two-step approach, an initial screening test is done in which patient drinks a 50 g glucose drink irrespective of time of last meal and a venous blood sample is collected 1 hour later (O’Sullivan’s test). GDM is excluded if glucose level in venous plasma sample is below 140 mg/dl. If level exceeds 140 mg/dl, then the complete 100 g, 3-hour OGTT is carried out.
In the 3-hour OGTT, blood samples are collected in the morning (after 8-10 hours of overnight fasting), and after drinking 100 g glucose, at 1, 2, and 3 hours. For diagnosis of GDM, glucose concentration should be above the following cut-off values in 2 or more of the venous plasma samples:
- Fasting: 95 mg/dl
- 1 hour: 180 mg/dl
- 2 hour: 155 mg/dl
- 3 hour: 140 mg/dl
LABORATORY TESTS FOR SCREENING OF DIABETES MELLITUS
Aim of screening is to identify asymptomatic individuals who are likely to have DM. Since early detection and prompt institution of treatment can reduce subsequent complications of DM, screening may be an appropriate step in some situations.
Screening for type 2 DM: Type 2 DM is the most common type of DM and is usually asymptomatic in its initial stages. Its onset occurs about 5-7 years before clinical diagnosis. Evidence indicates that complications of type 2 DM begin many years before clinical diagnosis. American Diabetes Association recommends screening for type 2 DM in all asymptomatic individuals ≥ 45 years of age using fasting plasma glucose. If fasting plasma glucose is normal (i.e. < 100 mg/dl), screening test should be repeated every three years.
Another approach is selective screening i.e. screening individuals at high risk of developing type 2 DM i.e. if one or more of the following risk factors are presentobesity (body mass index ≥ 25.0 kg/m2), family history of DM (first degree relative with DM), high-risk ethnic group, hypertension, dyslipidemia, impaired fasting glucose, impaired glucose tolerance, or history of GDM. In such cases, screening is performed at an earlier age (30 years) and repeated more frequently.
Recommended screening test for type 2 DM is fasting plasma glucose. If ≥126 mg/dl, it should be repeated on a subsequent day for confirmation of diagnosis. If <126 mg/dl, OGTT is indicated if clinical suspicion is strong. A 2-hour post-glucose load value in OGTT ≥200 mg/dl is indicative of DM and should be repeated on a different day for confirmation.
Screening for type 1 DM: Type 1 DM is detected early after its onset since it has an acute presentation with characteristic clinical features. Therefore, it is not necessary to screen for type 1 DM by estimation of blood glucose. Detection of immunologic markers (mentioned earlier) has not been recommended to identify persons at risk.
Screening for GDM: Given earlier under OGTT in gestational diabetes mellitus.
LABORATORY TESTS TO ASSESS GLYCEMIC CONTROL
There is a direct correlation between the degree of blood glucose control in DM (both type 1 and type 2) and the development of microangiopathic complications i.e. nephropathy, retinopathy, and neuropathy. Maintenance of blood glucose level as close to normal as possible (referred to as tight glycemic control) reduces the risk of microvascular complications. There is also association between persistently high blood glucose values in DM with increased cardiovascular mortality.
Following methods can monitor degree of glycemic control:
- Periodic measurement of glycated hemoglobin (to assess long-term control).
- Daily self-assessment of blood glucose (to assess day-to- day or immediate control).
Glycated Hemoglobin (Glycosylated Hemoglobin, HbA1C)
Glycated hemoglobin refers to hemoglobin to which glucose is attached nonenzymatically and irreversibly; its amount depends upon blood glucose level and lifespan of red cells.
Hemoglobin + Glucose ↔ Aldimine → Glycated hemoglobin
Plasma glucose readily moves across the red cell membranes and is being continuously combined with hemoglobin during the lifespan of the red cells (120 days). Therefore, some hemoglobin in red cells is present normally in glycated form. Amount of glycated hemoglobin in blood depends on blood glucose concentration and lifespan of red cells. If blood glucose concentration is high, more hemoglobin is glycated. Once formed, glycated hemoglobin is irreversible. Level of glycated hemoglobin is proportional to the average glucose level over preceding 6-8 weeks (about 2 months). Glycated hemoglobin is expressed as a percentage of total hemoglobin. Normally, less than 5% of hemoglobin is glycated.
Numerous prospective studies have demonstrated that a good control of blood glucose reduces the development and progression of microvascular complications (retinopathy, nephropathy, and peripheral neuropathy) of diabetes mellitus. Mean glycated hemoglobin level correlates with the risk of these complications.
The terms glycated hemoglobin, glycosylated hemoglobin, glycohemoglobin, HbA1, and HbA1c are often used interchangeably in practice. Although these terms refer to hemoglobins that contain nonenzymatically added glucose residues, hemoglobins thus modified differ. Most of the studies have been carried out with HbA1c.
Glycated hemoglobin should be routinely measured in all diabetic patients (both type 1 and type 2) at regular intervals to assess degree of long-term glycemic control. Apart from mean glycemia (over preceding 120 days), glycated hemoglobin level also correlates with the risk of the development of chronic complications of DM. In DM, it is recommended to maintain glycated hemoglobin level to less than 7%.
|Box 837.1 Glycated hemoglobin
Spurious results of glycated hemoglobin are seen in reduced red cell survival (hemolysis), blood loss, and hemoglobinopathies.
In DM, if glycated hemoglobin is less than 7%, it should be measured every 6 months. If >8%, then more frequent measurements (every 3 months) along with change in treatment are advocated.
There are various methods for measurement of glycated hemoglobin such as chromatography, immunoassay, and agar gel electrophoresis.
Role of glycated hemoglobin in management of DM is highlighted in Box 837.1.
Self-Monitoring of Blood Glucose (SMBG)
Diabetic patients are taught how to regularly monitor their own blood glucose levels. Regular use of SMBG devices (portable glucose meters) by diabetic patients has improved the management of DM. With SMBG devices, blood glucose level can be monitored on day-to-day basis and kept as close to normal as possible by adjusting insulin dosage. SMBG devices measure capillary whole blood glucose obtained by fingerprick and use test strips that incorporate glucose oxidase or hexokinase. In some strips, a layer is incorporated to exclude blood cells so that glucose in plasma is measured. Aim of achieving tight glycemic control introduces the risk of severe hypoglycemia. Daily use of SMBG devices can avoid major hypoglycemic episodes.
SMBG devices yield unreliable results at very high and very low glucose levels. It is necessary to periodically check the performance of the glucometer by measuring parallel venous plasma glucose in the laboratory.
Portable glucose meters are used by patients for day-to-day self-monitoring, by physicians in their OPD clinics, and by health care workers for monitoring admitted patients at the bedside. These devices should not be used for diagnosis and population screening of DM as they lack precision and there is variability of results between different meters.
Goal of tight glycemic control in type 1 DM patients on insulin can be achieved through self-monitoring of blood glucose by portable blood glucose meters.
Semiquantitative urine glucose testing for monitoring of diabetes mellitus in home setting is not recommended. This is because (1) even if glucose is absent in urine, no information about blood glucose concentration below the renal threshold (which itself is variable) is obtained (Normally, renal threshold is around 180 mg/dl; it tends to be lower in pregnancy (140 mg/dl) and higher in old age and in long-standing diabetics; in some normal persons it is low), (2) urinary glucose testing cannot detect hypoglycemia, and (3) concentration of glucose in urine is affected by urinary concentration. Semiquantitative urine glucose testing for monitoring has now been replaced by self-testing by portable glucose meters.
LABORATORY TESTS TO ASSESS LONG-TERM RISKS
Urinary Albumin Excretion
Diabetes mellitus is one of the leading causes of renal failure. Diabetic nephropathy develops in around 20-30% of patients with type 1 or type 2 DM. Diabetic nephropathy progresses through different stages as shown in Figure 837.3. Hypertension also develops along the course of nephropathy with increasing albumin excretion. Evidence indicates that if diabetic nephropathy is detected early and specific treatment is instituted, further progression of nephropathy can be significantly ameliorated. Early detection of diabetic nephropathy is based on estimation of urinary albumin excretion. In all adult patients with DM, usual reagent strip test for proteinuria should be carried out periodically. Positive test means presence of overt proteinuria or clinical proteinuria and may be indicative of overt nephropathy. In all such patients quantitation of albuminuria should be carried out to plan appropriate therapy. If the routine dipstick test for proteinuria is negative, test for microalbuminuria should be carried out.
The term ‘microalbuminuria’ refers to the urinary excretion of albumin below the level of detection by routine dipstick testing but above normal (30-300 mg/ 24 hrs, 20-200 μg/min, or 30-300 μg/mg of creatinine). Albumin excretion rate is intermediate between normal (normal albumin excretion in urine is < 30 mg/24 hours) and overt albuminuria (> 300 mg/24 hours). Significance of microalbuminuria in DM is as follows:
- It is the earliest marker of diabetic nephropathy. Early diabetic nephropathy is reversible.
- It is a risk factor for cardiovascular disease in both type 1 and type 2 patients.
- It is associated with higher blood pressure and poor glycemic control.
Specific therapeutic interventions such as tight glycemic control, administration of ACE (angiotensinconverting enzyme) inhibitors, and aggressive treatment of hypertension significantly slow down the progression of diabetic nephropathy.
In type 2 DM, screening for microalbuminuria should begin at the time of diagnosis, whereas in type 1 DM, it should begin 5 years after diagnosis. At this time, a routine reagent strip test for proteinuria is carried out; if negative, testing for microalbuminuria is done. Thereafter, in all patients who test negative, screening for microalbuminuria should be repeated every year.
Screening tests for microalbuminuria include:
- Albumin to creatinine ratio in a random urine sample
- Urinary albumin excretion in a 24-hour urine sample.
Reagent strip tests to detect microalbuminuria are available. Positive results should be confirmed by more specific quantitative tests like radioimmunoassay and enzyme immunoassay. For diagnosis of microalbuminuria, tests should be positive in at least two out of three different samples over a 3 to 6 month period.
Abnormalities of lipids are associated with increased risk of coronary artery disease (CAD) in patients with DM. This risk can be reduced by intensive treatment of lipid abnormalities. Lipid parameters which should be measured include:
- Total cholesterol
- Low-density lipoprotein (LDL) cholesterol
- High-density lipoprotein (HDL) cholesterol
The usual pattern of lipid abnormalities in type 2 DM is elevated triglycerides, decreased HDL cholesterol and higher proportion of small, dense LDL particles. Patients with DM are categorized into high, intermediate and low-risk categories depending on lipid levels in blood (Table 837.2).
|Category||Low density lipoproteins||High density lipoproteins||Triglycerides|
|High-risk||≥130||< 35 (men)||≥ 400|
|< 45 (women)|
|Low-risk||< 100||> 45 (men)||< 200|
|> 55 (women)|
Annual lipid profile is indicated in all adult patients with DM.
LABORATORY TESTS IN THE MANAGEMENT OF ACUTE METABOLIC COMPLICATIONS OF DIABETES MELLITUS
The three most serious acute metabolic complications of DM are:
- Diabetic ketoacidosis (DKA)
- Hyperosmolar hyperglycemic state (HHS)
The typical features of DKA are hyperglycemia, ketosis, and acidosis. The common causes of DKA are infection, noncompliance with insulin therapy, alcohol abuse and myocardial infarction. Patients with DKA present with rapid onset of polyuria, polydipsia, polyphagia, weakness, vomiting, and sometimes abdominal pain. Signs include Kussmaul’s respiration, odour of acetone on breath (fruity), mental clouding, and dehydration. Classically, DKA occurs in type 1, while HHS is more typical of type 2 DM. However, both complications can occur in either types. If untreated, both events can lead to coma and death.
Hyperosmolar hyperglycemic state is characterized by very high blood glucose level (> 600 mg/dl), hyperosmolality (>320 mOsmol/kg of water), dehydration, lack of ketoacidosis, and altered mental status. It usually occurs in elderly type 2 diabetics. Insulin secretion is adequate to prevent ketosis but not hyperglycemia. Causes of HHS are illness, dehydration, surgery, and glucocorticoid therapy.
Differences between DKA and HHS are presented in Table 837.3.
|Parameter||Diabetic ketoacidosis||Hyperosmolar hyperglycemic state|
|1. Type of DM in which more common||Type 1||Type 2|
|2. Age||Younger age||Older age|
|3. Prodromal clinical features||< 24 hrs||Several days|
|4. Abdominal pain, Kussmaul’s respiration||Yes||No|
|6. Plasma glucose||> 250 mg/dl||Very high (>600 mg/dl)|
|7. Serum bicarbonate||<15 mEq/L||>15 mEq/L|
|8. Blood/urine ketones||++++||±|
|9. β-hydroxybutyrate||High||Normal or raised|
|10. Arterial blood pH||Low (<7.30)||Normal (>7.30)|
|11. Effective serum osmolality*||Variable||Increased (>320)|
|12. Anion gap**||>12||Variable|
|Osmolality: Number of dissolved (solute) particles in solution; normal: 275-295 mOsmol/kg
** Anion gap: Difference between sodium and sum of chloride and bicarbonate in plasma; normal average value is 12
Laboratory evaluation consists of following investigations:
- Blood and urine glucose
- Blood and urine ketone
- Arterial pH, Blood gases
- Serum electrolytes (sodium, potassium, chloride, bicarbonate)
- Blood osmolality
- Serum creatinine and blood urea.
Indications for testing for ketone bodies in DM include:
- At diagnosis of diabetes mellitus
- At regular intervals in all known cases of diabetes, during pregnancy with pre-existing diabetes, and in gestational diabetes
- In known diabetic patients: during acute illness, persistent hyperglycemia (> 300 mgs/dl), pregnancy, and clinical evidence of diabetic acidosis (nausea, vomiting, abdominal pain).
An increased amount of ketone bodies in patients with DM indicate impending or established diabetic ketoacidosis and is a medical emergency. Method based on colorimetric reaction between ketone bodies and nitroprusside (by dipstick or tablet) is used for detection of both blood and urine ketones.
Test for urine ketones alone should not be used for diagnosis and monitoring of diabetic ketoacidosis. It is recommended to measure β-hydroxybutyric acid (which accounts for 75% of all ketones in ketoacidosis) for diagnosis and monitoring DKA.
- Venous plasma glucose:
Fasting: 60-100 mg/dl
At 2 hours in OGTT (75 gm glucose): <140 mg/dl
- Glycated hemoglobin: 4-6% of total hemoglobin
- Lipid profile:
– Serum cholesterol: Desirable level: <200 mg/dl
– Serum triglycerides: Desirable level: <150 mg/dl
– HDL cholesterol: ≥60 mg/dl
– LDL cholesterol: <130 mg/dl
– LDL/HDL ratio: 0.5-3.0
- C-peptide: 0.78-1.89 ng/ml
- Arterial pH: 7.35-7.45
- Serum or plasma osmolality: 275-295 mOsm/kg of water.
Serum Osmolality can also be calculated by the following formula recommended by American Diabetes Association:
Effective serum osmolality (mOsm/kg) = (2 × sodium mEq/L) + Plasma glucose (mg/dl) / 18
- Anion gap:
– Na+ – (Cl– + HCO3–): 8-16 mmol/L (Average 12)
– (Na+ + K+) – (Cl– + HCO3–): 10-20 mmol/L (Average 16)
- Serum sodium: 135-145 mEq/L
- Serum potassium: 3.5-5.0 mEq/L
- Serum chloride: 100-108 mEq/L
- Serum bicarbonate: 24-30 mEq/L
- American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2004; 24:S5-S10.
- American Diabetes Association. Gestational diabetes mellitus. Diabetes Care 2004;27:S88-S90.
- American Diabetes Association. Hyperglycemic crises in diabetes. Diabetes Care 2004;27:S94-S102.
- American Diabetes Association. Screening for type 2 diabetes. Diabetes Care 2004;27:S11-S14.
- American Diabetes Association. Tests of glycemia in diabetes. Diabetes Care 2004;27:S91-S93.
- Lernmark A. Type 1 diabetes. Clin Chem 1999;45:1331-38.
- Lebovitz HE. Type 2 diabetes: An overview. Clin Chem 1999;45:1339-45.
- Reaven GM. The metabolic syndrome: Requiescat in pace. Clin Chem 2005;51:931-8.
- Reinauer H, Home PD, Kanagasabapathy AS, Heuck C. Laboratory diagnosis and monitoring of diabetes mellitus. Geneva. World Health Organization, 2002.
- Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002; 48:436-72.
- Trachtenbarg DE. Diabetic ketoacidosis. Am Fam Physician 2005;71:1705-14.