2.3.0.0 A1C and Mean Glu­cose

Table 6.1 shows the cor­re­la­tion be­tween A1C lev­els and mean glu­cose lev­els based on two stud­ies: the in­ter­na­tion­al A1C-‍De­rived Av­er­age Glu­cose (ADAG) study, which as­sessed the cor­re­la­tion be­tween A1C and fre­quent SMBG and CGM in 507 adults (83% non-‍His­pan­ic whites) with type 1, type 2, and no di­a­betes (6), and an em­pir­i­cal study of the av­er­age blood glu­cose lev­els at pre­meal, post­meal, and bed­time as­so­ci­at­ed with specified A1C lev­els using data from the ADAG trial (7). The Amer­i­can Di­a­betes As­so­ci­a­tion (ADA) and the Amer­i­can As­so­ci­a­tion for Clin­i­cal Chem­istry have de­ter­mined that the cor­re­la­tion (r = 0.92) in the ADAG trial is strong enough to jus­ti­fy re­port­ing both the A1C re­sult and the es­ti­mat­ed av­er­age glu­cose (eAG) re­sult when a clin­i­cian or­ders the A1C test. Clin­i­cians should note that the mean plas­ma glu­cose num­bers in the table are based on ˜2,700 read­ings per A1C in the ADAG trial. In a re­cent re­port, mean glu­cose mea­sured with CGM ver­sus cen­tral lab­o­ra­to­ry–mea­sured A1C in 387 par­tic­i­pants in three ran­dom­ized tri­als demon­strat­ed that A1C may un­der­es­ti­mate or over­es­ti­mate mean glu­cose (5). Thus, as sug­gested, a pa­tient’s CGM profile has con­sid­er­able po­ten­tial for op­ti­miz­ing his or her glycemic man­age­ment (5).

Table 6.1—Mean glucose levels for specified A1C levels (6,7)

Data in parentheses represent 95% CI, unless otherwise noted. A calculator for converting A1C results into eAG, in either mg/dL or mmol/L, is available at http://professional.diabetes.org/eAG.
*These estimates are based on ADAG data of ;2,700 glucose measurements over 3 months per A1C measurement in 507 adults with type 1, type 2, and no diabetes. The correlation between A1C and average glucose was 0.92 (6).