Aim: The post-challenge glucose (PCG) level has been suggested to be superior to the fasting blood glucose (FG) level for predicting the risk of future cardiovascular disease (CVD); however, the extent of its superiority has not been consistently shown among previous cohort studies. Therefore, we conducted a meta-analysis to summarize the quantitative association of FG and PCG with CVD risk and compared the strengths of the two associations.
Method: Electronic literature searches using MEDLINE and EMBASE with an additional manual search were conducted for prospective observational studies of the association of FG and PCG with CVD risk. Studies were included if they were prospective studies in which the relative risk (RR) of CVD per 1 standard deviation increase in both FG and PCG could be estimated. Pooled relative risks for the incremental increase were calculated as RR(FG) and RR(PCG) using a bivariate random-effects model.
Result: Data were obtained from 14 eligible studies that included 70,889 participants and 2,927 cases. The pooled RR(FG) and RR(PCG) (95% confidence interval) were, respectively, 1.15 (1.06 to 1.26) and 1.24 (1.12 to 1.36); the difference was significant (P =0.001). The association of PCG with CVD risk was stronger in studies that targeted participants with a baseline mean FG < 100 mg/dl (P < 0.001) or mean age ≥ 55 years (P =0.004).
Conclusions: Overall, the association of PCG with CVD risk was stronger than that of FG by approximately 50% on a log scale. Measuring PCG is especially important in populations with relatively low FG levels or in the elderly, although it is often burdensome in routine clinical practice.