Aims/hypothesis  

The study was done to assess how well fasting and 2-h plasma glucose (FPG, 2-h PG) after a 75-g OGTT predict cardiovascular disease (CVD) and all-cause mortality in Asian subjects.

Methods  

People (n=6817) of Japanese and Asian Indian origin from five prospective studies in five countries were monitored for 5 to 10 years. Hazard ratios for death from all causes and CVD were estimated using Cox proportional hazard models, adjusting for FPG, 2-h PG and established risk factors.

Results  

Multivariate Cox regression analysis showed that an increase in FPG from 7.0 to 8.0 mmol/l (increase of 0.76 SD) increased relative risk (95% CI) by 1.14 (1.05–1.25) for all-cause and 1.24 (1.10–1.39) for CVD mortality. An increase in 2-h PG from 9.0 to 11.9 mmol/l (0.76 SD) increased relative risks by 1.29 (1.18–1.41) and 1.35 (1.19–1.54). Inclusion of 2-h PG in the FPG model improved the predictive value (p<0.001), whereas FPG did not influence the predictive value of 2-h PG (p>10). In a model containing FPG and 2-h PG, hazards ratios for 2-h PG in subjects with IGT or diabetes were 1.35 (1.03–1.77) or 3.03 (2.18–4.21) for all-cause and 1.27 (0.86–1.88) or 3.39 (2.14–5.37) for CVD mortality, compared with normal subjects. The respective hazards ratio for FPG in subjects with IFG or diabetes were 0.94 (0.68–1.31) or 0.88 (0.59–1.32) for all-cause and 1.05 (0.67–1.65) or 0.88 (0.51–1.51) for CVD mortality, compared with normal subjects.

Conclusions/interpretation  

For prediction of premature death, 2-h PG was superior to FPG in several Asian populations.