Author Affiliations: University of Washington, Seattle (email@example.com).
In Reply: Dr Chalew and colleagues underscore the potential of HGI as a useful clinical tool in diabetes care. They suggest its value in predicting risk of diabetes-associated microvascular complications, as it proved a predictor of retinopathy and nephropathy independent of MBG level in analysis of data from the Diabetes Control and Complications Trial.1 We agree that this finding indicates HGI may reflect some component of microvascular risk that is not captured in MBG measurement and therefore warrants inclusion as a metric in future outcome trials. As Chalew et al also propose, HGI might be helpful in determining the relative contribution of MBG- and non–MBG-dependent components of HbA1c measurement in its overall predictive power. However, as the authors acknowledge, prospective studies are needed to determine whether HGI offers any additive predictive value to that of HbA1c level. One possibility is that HGI could be used to refine interpretation of HbA1c values; that is, a higher HGI might indicate lower predictive value for HbA1c level in an individual patient and demand greater vigilance for identifying clinical risk factors other than glycemic control. Although the application of HGI and other metrics clearly merits further investigation, no metric yet has offered predictive power beyond that of HbA1c testing.
Rubinow KB, Hirsch IB. Metrics for Monitoring Glycemic Control—Reply. JAMA. 2011;305(24):2522-2523. doi:10.1001/jama.2011.846