WHO

• http://www.who.int/diabetes/publications/Definition and diagnosis of diabetes_new.pdf
• DM: FPG>=7.0mM/L[126mg/dL] or 2-h PG>=11.1mM/L[200mg/dL] (75g oral glucose load);="occurrence of diabetes-specific complications"
• IFG: FPG=6.1-6.9mM/L[110-125mg/dL] ->do OGTT;="icrd risk of developing DM"
• IGT: 2-h PG=7.8-11mM/L[140-199mg/dL] (75g oral glucose load) and FPG<6.1; ="icrd risk of developing DM; also icrd risk of cardiovascular disease"
• NB: PG approx. 11% higher [8%-15% if Hct 0.30-0.55] than BG; therefore conversion BG to PG is problematic
• HbA1c>6.1% - currently not sufficiently accurate diagnostic criterion [due to assay differences, effects of anaemia, Hb abnlties, pregnancy, uraemia, renal glucose threshold]
• see: Tests of Glycemia in Diabetes (2004) DiabetesCare [pubmed] [freepdf]):
• Correlation between HbA1c and mean PG in DCCT:
• 6%:7.5mM[135mg/dL]; 7%:9.5[170]; 8%:11.5[205]; 9%:13.5[240]; 10%:15.5[275]; 11%:17.5[310]; 12%:19.5[345];
  1% icr ~ 2mM/L icr; i.e. mean PG=7.5+2×(A1c-6%);
• NB: time to midpoint between starting level [0d] and new steady-state level [120d] ~ 30-35d, thus a large change in mean blood glucose is accompanied by a large change in HbA1c within a matter of 1-2wks not 3-4mths; in effect HbA1c a "weighted" measure, with last 30 days contributing ~50%;
• variations:
• [falsely] dcrd A1c with dcrd Ery survival, VitC/E (by ?inhibition of glycation); icrd A1c with IDA,
• interference with assay: HTG, Hyperbili, uraemia, chron alcoholism, chron salicylates, opiate addiction, Hb'opathies