Glucose Tolerance Tests

See also:
Screening for diabetes,
Diabetes in pregnancy,
Insulin resistance - syndrome X,
Managing impaired glucose tolerance in primary care.

The information in this article is mainly concerned with the diagnosis of type 2 diabetes, where a lack of symptoms and insidious onset to the illness mean that diagnostic tests may be needed to confirm a clinical suspicion of the disease, or investigate significant risk factors for the illness. Type 1 diabetics are much more likely to present with symptoms and a fairly rapid onset of illness, glycosuria and significant random hyperglycaemia, making diagnosis easier.

Recent research into diabetes and impaired glucose metabolism has created a bewildering array of terminology and definitions relating to this area of study and clinical practice. There are a variety of differing definitions and diagnostic criteria for concepts such as impaired fasting glucose (IFG), impaired glucose tolerance (IGT), the metabolic syndrome and 'prediabetes'.

The most important concern for primary care practitioners is that they can identify those patients with frank type 1 or type 2 diabetes, and be able to advise and monitor patients with indicators of impaired glucose metabolism who are at risk of developing type 2 diabetes. This article will therefore use the current (2006) recommendations of the World Health Organisation/International Diabetes Federation, which is based upon being able 'to distinguish a group of patients with significantly increased premature mortality and increased risk of microvascular and cardiovascular complications',¹ rather than becoming tied up in differing definitions based on epidemiological data and variable clinical practice.

• The WHO does not currently recommend the use of HbA1c as a diagnostic test for diabetes.¹
• There is no clear-cut consensus on the role of the oral glucose tolerance test (OGTT) in both clinical practice and research, and for epidemiological purposes.^1,2
• It is thought by the American Diabetic Association that the OGTT is a valid method of diagnosing diabetes but that fasting plasma venous glucose measurements should be preferred as a routine clinical test, because the OGTT is deemed to be more inconvenient, cost more and be less reproducible.¹
• There is also the problem that various studies have shown that fasting plasma venous glucose measurements and OGTT do not identify the same patients as having diabetes.
• One study showed the following distribution of patients in terms of their fit for the criteria used to diagnose diabetes for both tests.¹
  • 40% meet criteria for diagnosis of diabetes based on fasting plasma venous glucose measurement
  • 31% meet criteria for diagnosis of diabetes based on 2-hour plasma venous glucose measurement, after OGTT
  • 28% meet both sets of criteria for diagnosis of diabetes.

Studies of outcome based upon the mode of diagnosis of diabetes show worse outcomes in terms of cardiovascular morbidity and mortality in those diagnosed on the basis of the 2-hour plasma glucose result, as part of the OGTT. For this reason, and the fact that OGTT is the only useful method for diagnosing patients with IGT, the WHO still recommends the retention of the OGTT as a diagnostic test for diabetes mellitus and/or IGT:

• An OGTT is not always necessary to diagnose diabetes.
• Diabetes may be diagnosed on the basis of one abnormal plasma glucose (random >11 mmol/l or fasting >7 mmol/l) in the presence of diabetic symptoms such as: thirst, increased urination, recurrent infections, weight loss, drowsiness and coma.
• In asymptomatic people with an abnormal random plasma glucose, at least one and preferably two fasting venous plasma glucose in the abnormal range (>7 mmol/l) are required for diagnosis.
• An OGTT is indicated where repeat testing is inconclusive, where random plasma venous glucose measurements are in the range 6.1–6.9 mmol/l, or if fasting blood glucose is normal but the patient is:
  • Older
  • Not significantly obese
  • Has had gestational diabetes, in post-partum testing situation
  • Of African-Caribbean/Asian origin where fasting plasma venous glucose measurement is thought to be less useful, and there are possible symptoms of diabetes/suspicion of IGT
  • The OGTT is not recommended as a screening test for diabetes mellitus, fasting plasma venous glucose being preferred (see separate article – screening for diabetes).

• There are no agreed criteria on the use of OGTT during pregnancy.
• Its reproducibility and diagnostic utility are uncertain.
• Women should be screened for glycosuria at each antenatal visit.
• If 2+ glycosuria is detected then a random plasma venous glucose should be checked.
• An OGTT should be carried out if result of plasma venous glucose is >5.5 mmol/l two hours or more after food, or >7 mmol/l within two hours of food.
• Current criteria for diagnosing gestational diabetes mellitus are fasting venous plasma glucose >5.5 mmol/l or 2-hour OGTT>9 mmol/l.
• Women presenting with glycosuria/hyperglycaemia during pregnancy may be suffering from:
  • Intercurrent illness disrupting euglycaemia
  • IGT associated with pregnancy that will resolve post-partum
  • Newly diagnosed type 1 or type 2 diabetes
  • Gestational diabetes mellitus.
• OGTT during pregnancy should be interpreted with care, and the advice of the local diabetes service sought on test results and the patient's subsequent management.
• It is imperative that patients who present with de novo abnormalities of glucose regulation during pregnancy are assessed subsequently for post-partum OGTT. There is a significant 'drop-out' rate in the post-partum period and significant numbers of women are exposed to the risks of undiagnosed type 2 diabetes, due to a failure of follow-up.^4,5

• Test preceded by ≥3 days of normal, unrestricted diet (>150 g carbohydrate daily) with normal physical activity.
• Carbohydrate rich meal (30–50 g) on night before test.
• Overnight fast of 8–14 hours; drink only water.
• Record any factors that may affect interpretation of test such as medication, inactivity, infection, gestation of pregnancy, acute psychological stress etc.
• Collect fasting (and all other) samples in tube that permits measurement of plasma glucose (e.g sodium fluoride tube).
• Timing of test (0 hours) starts at beginning of glucose drink.
• Adults ingest 75g glucose in 250–300 ml water over 5 minutes.
• Children ingest 1.75 g/kg body weight in a similar volume of water by ratio (max 75g as for adults).
• No smoking during test.
• Take blood sample at 2 hours; some schema suggest taking a 1-hour sample but this is not strictly necessary in terms of diagnosing diabetes.
• Ideally take sample from warmed vein on back of hand (antecubital fossa samples may be artificially lower).
• An indwelling 'butterfly' or conventional cannula can be left in situ throughout the test (affix in place and dress); flush with saline after taking fasting sample, then draw at least 10 ml and discard before drawing sample for assay tube.
• Glucose should be measured immediately after collection by near-patient testing, or if a blood sample for a laboratory is collected, plasma should be immediately separated, or the sample should be collected into a container with glycolytic inhibitors and placed in ice-water until separated prior to analysis.¹
• An extended glucose tolerance test may be conducted to detect cases of reactive hypoglycaemia or other abnormalities of glucose metabolism with samples taken at 0, 30, 60, 90, 120, 150 and 180 minutes.
• The extended test may also be used to diagnose acromegaly when samples are also taken for growth hormone levels.

• It is important not to overinterpret the results of the OGTT, and to realise that it has variable reproducibility in and between individual patients.
• Use the WHO criteria above to interpret whether or not the test indicates a diagnosis of IFG, IGT or diabetes.
• It is rarely necessary to take and analyse anything other than 0 and 2 hour samples, especially in a primary care setting.
• If there is any uncertainty about the diagnosis of diabetes in potentially symptomatic individuals with equivocal test results, seek advice from your local diabetes service.
• Unusually low blood glucose results during OGTT should prompt consideration of referral for endocrinological assessment.
• If the test is conducted correctly and blood sampling performed appropriately, there are no causes of false-positive results, as such, other than factors that can provoke hyperglycaemia that should be checked for before performing the test:
  • Undisclosed medication changes (e.g. steroids)
  • Inactivity
  • Infection
  • Other acute illness
  • Pregnancy
  • Acute psychological stress
  • Failure to comply with the pre-test feeding/fasting regimen.