We thank Dr. Takahara (1) for the comment on our recent article exploring the impact of HbA1c, followed from diabetes onset, on the development of severe microvascular complications (2). As suggested, we have validated our results with Cox hazards analysis with severe microvascular events, i.e., laser-treated proliferative retinopathy and macroalbuminuria as a dependent variable and HbA1c (mmol/mol) as a time-dependent covariate.
For laser-treated proliferative retinopathy, we found a hazard ratio of 1.038 (95% CI 1.025–1.052, P < 0.001) and for macroalbuminuria, a hazard ratio of 1.075 (95% CI 1.050–1.100, P < 0.001).
Analyzing our data with Cox hazards analysis thus shows the strong influence of long-term HbA1c on severe microvascular complications, in agreement with our previous conclusions.
In our article, we chose to analyze and present the results in a way that was perhaps easier for a clinician to interpret and apply in clinical routine. With life-table analysis we found that the incidence of both laser-treated proliferative retinopathy and macroalbuminuria increased sharply and occurred earlier with increasing long-term weighted mean HbA1c. In the same manner, the prevalence of microvascular complications increased steeply with higher long-term weighted mean HbA1c, categorized in different groups.
In conclusion, our study irrespective of statistical methods shows a strong association between development of late complications and long-term mean HbA1c, and keeping the average HbA1c below 7.6% (60 mmol/mol) seemed sufficient to prevent microvascular complications for at least up to 20 years.