Does using HbA1c inform diagnosis of diabetes in patients with coronary artery disease?

In recent years, using easy, quality-certified, and accurate glycated haemoglobin (HbA1c)measurements hasbecomeagloballyaccepted, if not recommended part of our everyday practical armamentarium in diagnosing diabetes mellitus (DM), although controversy continues over its precise position in the diagnostic algorithm. On the other hand, the very high and, in fact, increasing number of both diagnosed and undiagnosed cases of DM (besides other established cardiovascular risk factors) in patients withcoronary arterydisease (CAD) has also emerged as a worldwide phenomenon, and prompted the first joint guidelines in ‘cardio-diabetology’ released by the European Society of Cardiology and the European Association for the Study of Diabetes. Indeed, the view has been put forward in those guidelines that DM and CAD may represent the two sides of a coin, which was also maintained in the recent update. The publication in this issue based on the EUROASPIRE IV survey andaiming tocompare the performanceof the three different available biochemical tests, i.e. fasting plasma glucose (FG), 2 h glucose after an oral glucose tolerance test (2hPG), andHbA1c, indiagnosingpreviously unknown DM in patients with chronic CAD provides important and practicallyhighly relevantnew information in this context. First, it confirms the high prevalence of known diabetes, i.e. 27% or 2164 out of 7998 patients enrolled from 79 centres in 24 European countries during May 2012 to April 2013. Secondly, it finds another 29% of new DM cases in the finally investigated cohort of 4004 patients, after exclusion of patients not appropriately fasting or with insufficient or missing data. Allowing for extrapolation, this would add some 21% of new DM cases to the 27% cases of known DM in the EUROASPIRE IV cohort as a whole, i.e. close to 50% of a contemporary European CAD cohort has co-existing DM one way or the other, a figure similar to that reported recently from the German SWEETHEART registry evaluating 2767 consecutive patients presenting with acute myocardial infarction. Thirdly, it showed that using HbA1c measurements on top of FG determinations informed diagnosing diabetes in this CAD patient cohort at high risk for DM rather little, as the additional yield over and beyond the 75% of DM cases already identified by FG amounted to only 6%. Moreover—and this is of concern— the majority of these additional HbA1c-defined cases might actually represent a misclassification, not only since their corresponding FG was in the non-diabetic glucose range, but also because so was their 2hPG. In contrast, fourthly, it demonstrated that an oral glucose tolerance test (OGTT) was by far the better second diagnostic test in this cardiology population, identifying another 21% cases of previously undiagnosed diabetes, whereas relying exclusively on HbA1c measurements would have left 83% of patients with overt diabetes undetected in the current EUROASPIRE IV database. The rationale of using HbA1c measurements to monitor glycaemic control in persons with DM and—more recently—to diagnose DM is based on the fact that glycosylation of haemoglobin, i.e. attachment of glucose and irreversible transformation to HbA1c, is a result of the average glucose exposure of the haemoglobin molecules in the circulatingbloodduring theprevious6–8weeks.So, an increase inHbA1c is the end-result of prior increasing blood glucose concentrations, but short-lived glucose spikes as in the post-prandial state may not be sufficient to inducea lasting effect on HbA1c. It is awidelyheldconcept that DM often first develops post-prandially, followed by an increaseof FG, therise inHbA1cbeinga rather lateevent. In linewiththis,HbA1cdetectedDMrepresentedby far thesmallest groupofnewcasesofDM in EUROASPIRE IV. At the same time, there is very robust evidence that the surge in post-prandial glucosesignals increased cardiovascular (CV) complications in the future, so it should be captured when it comes to CV risk stratification. For example, at the 3-year followup of the SWEETHEART registry, the patients with OGTT-detected DM, and—emerging—the group with impaired glucose tolerance, showed a significantly adverse prognosis both for mortality and for MACCE (major adverse cerebrovascular and cardiovascular events), compared with non-diabetic patients at baseline.