Diagnosis of Diabetes

Key Points
  • There are a number of tests that can be used to diagnose diabetes
    • Different tests are used depending on circumstances
    • The two key tests are Hba1c and OGTT
  • Complications have their own diagnostic investigations – see separate
Urine
    • The Urine Dipstick can check for three things
      • Glucose
        • Poorly sensitive and specific.
        • Glycosuria only occurs if BGL >10mmol/L
        • Only really used for screening
      • Albumin
        • As with glucose – Stick only detects very high albuminuria at >300mg/l
        • Ketones can be identified on urine if suspecting DKA in T1 diabeticcs
        • Also excludes DDx
      • Generally, the dipstick is not very sensitive for diagnosis
Blood tests
    • 2 main blood tests are used for diagnosis:
      • HbA1c
      • Oral glucose tolerance test (OGTT)
      • There are also two other options that are now rarely used
        1. Fasting plasma glucose (i.e. no food for ≥8 hours) (>7.0)
        2. Random plasma glucose (>11.1)
      • For Hba1c and plasma glucose, we usually require  2 samples but 1 may be enough in symptomatic patients
HbA1c (Glycated haemoglobin)
  • This test measures the amount of glycosylation of haemoglobin
    • Lasts over the lifetime of the erythrocyte
    • Gives a long term indication over ~3 months, and is therefore more reliable for monitoring and diagnosis.
  • This is the gold standard for diagnosis – preferred over OGTT/fasting glucose unless there is a contraindication  – see below
  • Threshold = > 6.5% or >48mmol/L depending on what assay is used (DCCT vs IFCC)
    • Measured twice for diagnosis
    • Currently, IFCC methods are preferred
  • Beware and avoid in certain scenarios where the constitution of haemoglobin is altered:
    • RBC abnormalities – Haemoglobinopathies, anaemia, blood transfusions, and CKD
    • Children
    • Pregnant women 
    • Acutely ill patients
Venous plasma glucose

This is typically measured using a fingerprick POC test (although lab tests can be used

  • Thresholds:
    • Random glucose > 11mmol/l
    • Fasting glucose > 7.0mmol/l
      • (To convert glucose from mmol/L to mg/dL multiply by 18)
  • Rarely used for diagnosis in asymptomatic patients
    • Typically used for patients for self-monitoring
    • If used for diagnosis
      • Would measure twice if asymptomatic
      • If symptomatic a single value >11 is enough
Oral glucose tolerance test (OGT)
  • It is very helpful in the diagnosis of Prediabetic States as well as diabetes
  • This is essentially a variant of plasma glucose

Process of OGT:

  • A fasting measurement is taken (as above)
    • This should be <6mmol/l in a normal patient
  • Patient is then given a 75ml solution of glucose (e.g. lucozade) and the BGL is measured again in 2 hours
    • This second reading should be <7.8mmol/L in a normal person
  • If either/both are high, the patient has IGT/IFG/Diabetes (See pre-diabetic conditions)

The OGT is mainly only used for diagnosis in two situations:

  1. Preferred over hba1c in certain situations when the Hba1c is contraindicated
  2. Used to confirm diagnosis if the fasting glucose or Hba1c result is uncertain
    • E.g. a patient is found to have a fasting plasma glucose of 6.1-7mmol/L or random glucose of 8-10mmol/L.
      • Anything over either or both of these values confirms diagnosis of diabetes
Other tests
  • These are mainly needed to distinguish Type 1 vs Type 2 diabetes
    • This is important in determining the use of insulin in management

Autoantibodies

  • Anti-GAD and Anti-Islet Autoantibodies are used to differentiate T1 from T2 
    • Anti-GAD is mainly used
    • Anti-Islet has poor sensitivity/specificity
  • Often used where the diagnosis is unclear
    • E.g. patients >30 who present with features of Type 1 DM

C-Peptide test

  • C-peptide is a byproduct of insulin synthesis and so is a useful marker of how much insulin is endogenously produced
  • Therefore it is absent in T1DM (opposite to the autoantibodies)
  • Indications:
    • Again distinguishes insulin resistance vs deficiency when it is clinically unclear (like antibodies above)
      • High levels = resistance
      • Absence = deficiency
    • Also useful in distinguishing endogenous vs exogenous insulin – e.g. in determining whether hypoglycaemia is due to sulphonylureas vs insulin
      • In exogenous insulin, C-peptide remains low