Beyond Hemoglobin A<sub>1c</sub>—Need for Additional Markers of Risk for Diabetic Microvascular Complications

IN 1993, THE DIABETES CONTROL AND COMPLICATIONS Trial demonstrated that intensive therapy lowered time-averaged blood glucose values (measured as hemoglobin A1C [HbA1C]) and significantly reduced development of microvascular complications in type 1 diabetes. For example, intensive therapy reduced the risk of sustained retinopathy progression by 73% compared with standard treatment. Such significant reductions led to the recommendation by professional societies that for microvascular disease prevention, the HbA1C goal for nonpregnant adults should be less than 7% or even less than 6.5% of total hemoglobin. However, few physicians recognized that only 6.6% of the variation in risk of retinopathy for the entire study cohort was explained by the difference in the treatment groups, although it was widely appreciated that nearly all of this treatment group effect was explained by differences in the mean level of HbA1C over time. The trial results also considered the instantaneous risk of retinopathy (ie, whether a patient would develop retinopathy at a particular point in time during the study) rather than eventual risk of retinopathy (whether a patient would develop retinopathy over his or her entire life). However, this latter outcome is not feasible to study because it would require lifetime follow-up of patients. Similarly, HbA1C and duration of diabetes (glycemic exposure) explained only about 11% of the variation in retinopathy risk for the entire study population, suggesting that the remaining 89% of the variation in risk is presumably explained by other factors independent of HbA1C. Given the magnitude of the effect of unmeasured elements in the Diabetes Control and Complications Trial, identification of these elements is critically important for designing more effective therapy for type 1 diabetes. What factors not captured by HbA1C measurements might explain the remaining 89% of microvascular complications risk? Possible factors unrelated to blood glucose levels include genetics, environmental toxins, and metabolic consequences of abnormal insulinization such as increased free fatty acid levels. Possible factors related to blood glucose levels most likely reflect the fact that since HbA1c represents the time-averaged mean level of glycemia, it provides no information about how closely the fluctuations of blood glucose levels around that mean mimic the normal narrow range of blood glucose excursion. In addition, patients with identical HbA1C values differ significantly in amplitude and duration of glycemic spikes. Thus, a potential determinant of microvascular complications not captured by HbA1C could be greater magnitude and frequency of glycemic excursions. This hypothesis is provisionally supported by experimental evidence that prior episodes of transient hyperglycemia trigger persistent increases in proinflammatory gene expression during subsequent periods of normal glycemia by inducing stable epigenetic changes in the promoter of NFB p65, which increase p65 expression. This in turn causes persistent increased expression of the proinflammatory proteins monocyte chemotactic protein-1, vascular cell adhesion molecule-1, intercellular adhesion molecule-1, interleukin-6, and inducible nitric oxide synthase. These persistent changes in gene expression are induced by spikes of hyperglycemia that have durations considered too short (6-16 hours followed by 6 days of normal glycemia) to influence HbA1C values. During the course of the Diabetes Control and Complications Trial, blood glucose was measured 1 day every 3 months at 7 time points throughout that day. However, calculation of within-day standard deviation and standard deviation of mean blood glucose values between these quarterly assessments revealed that neither of these measures of glucose variability was associated with increased risk of microvascular complications. This result is not surprising because recent continuous subcutaneous glucose monitoring data from patients with type 1 diabetes