Long-term changes and variability in diabetes risk factors prior to the development of impaired glucose homeostasis.

AIMS: To determine whether rate of change and variability in risk factors provides insight into the development of impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or Type 2 diabetes (DM).
METHODS: In a nested case-control study, repeated risk factor measurements (mean number 4.3, duration 11.4 years) culminated in 59 men developing IFG (n = 37), IGT (n = 14) and/or DM (n = 11). For each case, two control subjects were matched for age and equivalence of follow-up. Rates of change and variability in diabetes risk factors prior to diagnosis were quantified by regression analysis. Changes between penultimate and diagnostic visits were also analysed.
RESULTS: The age-related rise in body mass index (BMI) was attenuated prior to IFG compared with control subjects (+0.102 vs. +0.772 kg/m2/decade, P = 0.02). There was also some evidence for this prior to IGT and DM (IGT: -1.530 vs. +0.158 kg/m2/decade, P = 0.09; DM: -1.146 vs. +0.332 kg/m2/decade, non-significant). Prior to onset, IGT cases were distinguished by higher inflammatory marker levels, a decline in insulinogenic index and greater variability in oral glucose tolerance test (OGTT) insulin, and DM cases by lower insulin sensitivity and higher liver enzyme activities. Fasting and OGTT glucose levels changed little during the mean 8.9 years prior to onset of IFG, IGT or DM. The transition to IGT or DM was accompanied by a fall in insulin sensitivity.
CONCLUSIONS: Except for BMI, change or variability in risk factor levels appears relatively unimportant in the development of clinically elevated glucose levels. Deterioration in glucose levels to IGT or DM occurs as a rapid, incremental increase accompanied by a decline in insulin sensitivity.