Postprandial Hyperglycemia: Actor or Understudy?

With the recent development of new methods to measure postprandial hyperglycemia and new treatments to modulate it, investigators have questioned whether postprandial hyperglycemia causes diabetic complications and whether it should be a target of therapy. To address these questions, the American Diabetes Association convened a consensus development conference in January 2000. The conference report is reprinted on the following pages.1 

In nondiabetic individuals, plasma glucose concentrations are extremely tightly regulated. Fasting and preprandial glucose concentrations range from 70 to 90 mg/dl. Glucose concentrations peak ∼60 min after the start of a meal, rarely exceed 120–140 mg/dl, and return to preprandial levels within 2–3 h. In diabetes, glucose concentrations vary widely both during a 24-h period and from day to day. For this reason, HbA_1c, which provides an integrated measure of the plasma glucose concentration over the preceding 2–3 months, has been widely adopted as an indicator of long-term glycemic control in diabetes.

In recent years, HbA_1c has been demonstrated to be an accurate predictor of risk of complications in long-term, randomized, prospective, controlled clinical trials. In a meticulous and exhaustive epidemiological assessment, the Diabetes Control and Complications Trial (DCCT) Research Group explored the association between levels of glycemic exposure and the risk of retinopathy, nephropathy, and neuropathy progression. They concluded that HbA_1c, duration of type 1 diabetes at screening, updated mean HbA_1c during the trial, and time in the trial were the dominant predictors of complications.2 In a similar analysis of the U.K. Prospective Diabetes Study (UKPDS), Stratton and colleagues demonstrated that baseline HbA_1c and mean updated HbA_1c were significantly associated with the incidence of any diabetes-related end point, diabetes-related deaths, and all-cause mortality.3 

Unfortunately, HbA_1c does not measure the magnitude or frequency of short-term fluctuations of blood glucose, and no clinical trials have prospectively assessed whether postprandial glucose, independent of other measures of glycemia, plays a unique role in the pathogenesis of diabetes-specific complications. More importantly, there are no clinical trials to prospectively assess whether therapies to specifically treat postprandial glucose levels prevent complications. In the debate about diagnostic criteria for diabetes, a number of studies have suggested that glucose levels after a glucose load may be more predictive of morbidity and mortality than fasting glucose concentrations.4,5 Although intriguing, these studies are observational and do not address the issue of day-to-day postprandial glucose levels.

The lack of clinical trials specifically addressing postprandial glucose levels independent of other measures of glycemia most probably relates to the fact that until recently, there were no effective ways of measuring or treating postprandial glucose levels. Studies to assess the contribution of postprandial glucose to the long-term complications of diabetes were not possible before the advent of feasible and acceptable methods to assess postprandial glucose levels, such as continuous glucose monitoring,6 and the development of alpha-glucosidase inhibitors, rapid-acting oral insulin secretagogues, and rapid-acting insulin analogs, which effectively and selectively modulate postprandial glucose levels. Now that these methods and agents are available, the necessary trials may be undertaken.

In the absence of data from such clinical trials and in light of the substantial gap between recommended and actual treatment as assessed by HbA_1c in the diabetic population, practitioners should remain focused on achieving HbA_1c levels as near to normal as possible. To the extent that newer agents reduce HbA_1c, they and other pharmacological therapies that lower HbA_1c should be part of our armamentarium. At the same time, we should not be distracted from our goal.