Prediabetes intervention, education key to diabetes prevention

Approximately 86 million Americans — about one in three — are estimated to have a blood glucose level that is higher than normal, but not high enough for a diagnosis of diabetes. That state, known as prediabetes, sets the many adults who have it on a course for potential complications that range from heart disease and stroke to conversion to overt type 2 diabetes.

Without lifestyle changes to improve their health, 15% to 30% of people with prediabetes will develop type 2 diabetes within 5 years, according to the CDC.

But considerable debate remains about what prediabetes means and how it should be treated — and even how the condition should be labeled. Some experts view prediabetes as a risk factor for type 2 diabetes, but others see it as early detection of a disease that has already left its mark.

“For every person with overt diabetes, there are about 2.5 to 3 people who have prediabetes, which is a bad term,” Ralph DeFronzo, MD, of the University of Texas Health Science Center in San Antonio, told Endocrine Today.

“People who have prediabetes have all the pathophysiologic disturbances that people with diabetes have. They’re just not as severe.”

Three criteria are necessary for a prediabetes diagnosis, as defined by WHO and the American Diabetes Association:

• HbA1c between 5.7% and 6.4%;
• fasting plasma glucose between 100 mg/dL and 125 mg/dL (ADA criteria) or 110 mg/dL and 125 mg/dL (WHO criteria); and
• an oral glucose tolerance test measurement between 140 mg/dL and 199 mg/dL.

For a patient, no clear symptoms accompany prediabetes, but recent studies show that decreased beta-cell function and reduced insulin sensitivity are evident even before a prediabetes diagnosis. In a 2012 analysis appearing in The Lancet,Adam G. Tabak, MD, PhD, of University College London, and colleagues described a multistage model of diabetes development, beginning with a long compensatory period when insulin resistance is present and accompanied by increased rates of insulin secretion and increased beta-cell mass. During a second stage, beta cells are no longer fully compensating for increased insulin resistance, meaning fasting or postload glucose values are not completely maintained.

“This period is likely to start when fasting and postload glucose levels are within the normal range, and is usually accompanied by a decrease in acute insulin secretion that is present at FPG levels around 5.6 mmol/L,” Tabak and colleagues wrote. “Much of the first and second stages, therefore, occur before the prediabetic phase is achieved.”

By the time a person reaches the so-called prediabetic state, the researchers wrote, beta cells are unable to compensate for a given insulin resistance, and glucose levels begin to increase rapidly.

In other words, DeFronzo said, damage has already been done.

“If you are in the upper tertiles of impaired glucose tolerance — say, a 2-hour glucose between 180 mg/dL and 199 mg/dL — you have already lost 80% of your beta-cell function, you’re already maximally insulin resistant, and a number of studies show you have already lost 10% to 20% of your beta-cell mass,” DeFronzo said. “These people, pathophysiologically, have type 2 diabetes, and they should be treated.”

But there are big differences between prediabetes and type 2 diabetes, some experts say, and the biggest is one of timing and opportunity.

“There is the argument that prediabetes is just a redefinition of diabetes itself,” David Marrero, PhD, immediate past president of health care and education for the ADA and director of the Diabetes Translational Research Center at Indiana University, told Endocrine Today.

“However, the data are very clear. [With prediabetes], you can normalize your glucose levels by weight loss. You can really make a substantial change in your status. If you have [overt] diabetes, you can improve the clinical outcomes, but you’re always going to have that diagnosis. So it’s not 100% clear that prediabetes is just diabetes seen earlier. It’s a risk factor. It’s telling us, ‘Boy, chances are good that if you keep going in this direction, you’re going to get caught.’

“The old joke is, you catch the horse before it gets out of the barn,” Marrero said. “And you really want to do that in this case.”

How early to intervene

In a 2013 study appearing in Diabetes Care,Kai McKeever Bullard, PhD, of the Division of Diabetes Translation at the CDC, and colleagues reviewed secular trends in U.S. prediabetes prevalence using National Health and Nutrition Examination Surveys data from 1999 to 2010. The researchers found an increase in overall age-adjusted prediabetes prevalence from 27.4% in 1999-2002 to 34.1% in 2007-2010. The increase, the researchers noted, was solely the result of more individuals fulfilling the HbA1c criterion, which was met by 9.5% of adults with prediabetes in 1999-2002 and by 17.8% of adults in 2007-2010, whereas the prevalence of adults meeting the impaired fasting glucose criterion remained essentially unchanged (23.8%-25.9%).

Prediabetes prevalence increased in all BMI subgroups, but rates were highest among adults with overweight and obesity, according to Bullard and colleagues. Intervention, experts say, could start with that population before patients develop prediabetes.

“Someone who is simply obese, but doesn’t have the other criteria for prediabetes, is already worth intervening with around lifestyle,” Robert A. Gabbay, MD, PhD, FACP, chief medical officer and senior vice president at Joslin Diabetes Center, told Endocrine Today. “That message gets stronger and more profound once a person actually has prediabetes, and therefore has a number of risk factors that put them not only at cardiovascular risk, but risk for developing diabetes.”

Jenifer Swartzentruber, RDN, CDE, former director of diabetes education and outreach at William Sansum Diabetes Center in Santa Barbara, California, said early education combined with successful community-based programs can have an effect on the numbers.

“In Santa Barbara, we were identifying patients with HbA1c as low as 5.5% to 5.6% — that gray area — because we knew that those patients were also at risk for developing diabetes down the road,” Swartzentruber, now a consultant developing curricula for Sansum, told Endocrine Today.

“Everyone benefits from good nutrition and exercise. You can never begin too early. You can take it all the way back from a prevention perspective to preconception care for moms.”

Lifestyle changes are key in preventing prediabetes progression to diabetes, Swartzentruber said. “We’ve seen that in studies. It’s the lifestyle that is going to be more successful than just giving metformin.”

Lifestyle plus pharmacy

Intensive lifestyle intervention was shown to be successful in the Diabetes Prevention Program (DPP) study, a 27-center randomized clinical trial in 2002 designed to determine whether lifestyle intervention (intensive training in diet, physical activity and behavior modification) or pharmacologic therapy (850 mg metformin twice daily) would prevent or delay the onset of diabetes in adults with IGT. During the study, lifestyle intervention decreased the incidence of type 2 diabetes by 58% vs. 31% in the metformin-treated group.

“These lifestyle-related diseases can be prevented or treated with lifestyle intervention,” Samuel Klein, MD, professor of medicine and nutritional science and director of the Center for Human Nutrition and of the Weight Management Program at Washington University School of Medicine in St. Louis, told Endocrine Today.

“If you’re overweight or obese, and if you have prediabetes, you [must] reduce your body weight, but you don’t have to reduce body weight by very much. A 5% weight loss goes a long way to reversing prediabetes and reducing your risk for developing diabetes in the future.”

Klein said lifestyle changes can be difficult to implement. “Five percent weight loss is not so easy to achieve and keep off long term, but it is much easier than achieving 10% or 30% weight loss.” Marrero agreed.

“If it was easy for people to lose weight and keep it off, we wouldn’t have any fat people,” Marrero said. “It is an issue of sustaining it. However, the data are also clear: If you’re 250 lb, and you lose 7% of your body weight, you just improved your [diabetes] risk by 60%. You may gain the weight back, and that risk will come back, but you can bring the weight back down and that risk will come down.”

Marrero advocates pharmacologic interventions for patients who struggle with lifestyle changes alone.

“In those individuals, pharmacologic intervention should be considered. It is a reasonable alternative — however you get it done, you get it done.”

For adults with prediabetes, pharmacologic intervention should be a key component of treatment, DeFronzo said.

“Of course, we all talk about lifestyle intervention, and I’m a big advocate of that,” he said. “The problem is when the [DPP] study ended at 3 years ... even with a re-education of the patient, people regained virtually all the weight and came back to where they were with metformin. Metformin does not afford you beta-cell protection, it does not improve insulin sensitivity in the muscle, it does not affect beta-cell function. Everybody loves metformin, but it is not the ideal drug.”

Studies with thiazolidinediones, also known as TZDs, consistently have shown a 55% to 72% decrease in the conversion rate from IGT to diabetes over 3 years, DeFronzo said, with 50% to 60% of people reverting to normal glucose tolerance.

“In the ACT NOW study, we showed that in the people treated with pioglitazone, there was a 72% reduction in the conversion rate of IGT to diabetes, and 48% of the people went back to normal glucose tolerance. That’s pretty darn good,” DeFronzo said.

The number needed to treat to prevent one case of diabetes was 18 — much fewer than the 40 to 45 needed to treat with a statin to prevent one myocardial infarction, according to DeFronzo.

“An excellent [therapy] choice would be low-dose pioglitazone [combined] with low-dose metformin,” he said.

New, long-term data have shed more light on agents, such as the glucagon-like peptide-1 receptor agonist Saxenda (liraglutide, Novo Nordisk), and their ability to reduce diabetes risk in adults with obesity and prediabetes. In study findings presented this year at the Endocrine Society Annual Meeting, researchers found more than half of adults with prediabetes and overweight or obesity randomly assigned liraglutide returned to normoglycemia at 3 years while also improving their cardiometabolic profile.

In the double blind, randomized controlled trial, participants assigned liraglutide took on average 2.7 times longer to progress from prediabetes to type 2 diabetes during 160 weeks than those assigned placebo (95% CI, 1.9-3.9), and participants assigned liraglutide reduced their risk for developing type 2 diabetes by 79.3% (P < .0001). Within the cohort, 66% of liraglutide participants regressed to normoglycemia vs. 36% in the placebo group (OR = 3.6; 95% CI, 3-4.4).

“In the last several years, there has been a real avalanche of medications,” Klein said. “There have been four — Qsymia (phentermine/topiramate, Vivus), Belviq (lorcaserin, Eisai Inc.), Contrave (naltrexone/bupropion, Takeda), Saxenda — all approved by the FDA to treat obesity, which is more than we have seen in the last 20 years. What is remarkable to me is that physicians and patients are so inclined to prescribe or take medicines for almost anything, except obesity. Why there is such a barrier for treating obesity [with medication], even though they improve health, is surprising to me. Is it the lack of reimbursement? Is it the lack of knowledge? Is it patients are afraid to take medications?”

One problem, DeFronzo said, is that pharmacologic treatments for prediabetes are off-label.

“We have treatment, but we don’t have any drugs that are officially approved,” he said. “Some way or another, the FDA needs to say, ‘We’re willing to entertain studies in which you treat prediabetic people to prevent diabetes.’ And preventing diabetes is a worthwhile endpoint because then you’re not going to develop the microvascular complications. The FDA simply has not come to that position.”

Raising awareness

In January, the ADA, AMA and CDC partnered with the Ad Council to launch the first national public service advertising about prediabetes. The ads, in English and Spanish, encourage people to take a short, online test to learn their risk and include an integrated text message component that provides ongoing support and lifestyle tips.

The effort is a good start, Gabbay said.

“Like any public health effort, it starts with raising awareness,” he said. However, “you can raise the awareness, and now I go to my doctor and say, ‘I want to be tested for prediabetes.’ And, yes, you have it. Now what?

“I’m not sure all providers have a great sense of what to do, and that’s where the endocrine community can certainly play an important role,” Gabbay said.

According to Swartzentruber, certified diabetes educators and nutritionists must be brought into the prediabetes conversation early on.

“The CDE dietitian plays such a critical role because we work with people on such a personal level,” Swartzentruber said. “Expanding programs where people have access to an RD, CDE is important. The broad statements of eat better, move more — people don’t know how to apply those to their daily life. Working toward getting reimbursement for prediabetes so that these programs would be paid for is key. It’s the RD, CDE that is going to take the time that the doctor does not have.”

Testing efforts for prediabetes also must be improved, Klein said.

“Just getting a simple blood glucose test isn’t enough,” he said. “[Prediabetes] does require getting an HbA1c in addition to blood glucose, but even, preferably, a 2-hour OGTT because you will miss things with the others that you will pick up with an OGTT.

“All of these lifestyle-related diseases are diagnoses made on a continuum,” Klein said. “We have these artificial cutoff points ... but these are all along the range of deteriorating metabolic function. The key to it is prevention. Prediabetes is a red flag to let you know that you are at high risk for problems.” – by Regina Schaffer

• References:
• Bullard KM, et al. Diabetes Care. 2013;doi:10.2337/dc12-2563.
• CDC. 2014 National Diabetes Statistics Report. Available at: http://www.cdc.gov/diabetes/data/statistics/2014statisticsreport.html. Accessed April 15, 2016.
• Fujioka K, et al. OR36-1. Presented at: The Endocrine Society Annual Meeting; April 1-4, 2016; Boston.
• Tabak AG, et al. Lancet. 2012;doi:10.1016/S0140-6736(12)60283-9.

• For more information:
• Ralph DeFronzo, MD, can be reached at University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229; email: defronzo@uthscsa.edu.
• Robert Gabbay, MD, PhD, FACP, can be reached at Joslin Diabetes Center, One Joslin Place, Boston, MA 02215; email: robert.gabbay@joslin.harvard.edu.
• Samuel Klein, MD, can be reached at Center for Human Nutrition, Washington University School of Medicine, 6606 South Euclid Ave., St. Louis, MO 63110; email: sklein@wustl.edu.
• David Marrero, PhD, can be reached at the Diabetes Translational Research Center at Indiana University, 410 W. 10th St., Suite 1140, Indianapolis, IN 46202; email: dgmarrer@iupui.edu.
• Jenifer Swartzentruber, RDN, CDE, can be reached at William Sansum Diabetes Center, 2215 Bath St., Santa Barbara, CA 93105; email: smartfood805@gmail.com.

Disclosures: DeFronzo reports serving on advisory boards for AstraZeneca, Boehringer Ingelheim, Intarcia, Janssen and Novo Nordisk and on the speakers’ bureau for AstraZeneca and Novo Nordisk and receiving support from AstraZeneca, Boehringer Ingelheim, Janssen and Takeda. Klein reports being a shareholder in Aspire Bariatrics and serving on scientific advisory boards for Novo Nordisk and Takeda. Gabbay, Marrero and Swartzentruber report no relevant financial disclosures.

Is the HbA1c assay the best diagnostic test to determine prediabetes?

The HbA1c assay offers a number of advantages.

HbA1c provides an integrated average blood glucose measurement over the previous 2 months or so. The test is widely available, standardized, easy to perform and can be done at any time of day or night. It does not require a person to fast or a health care provider to trust that a person has fasted. It does not matter if a person has experienced some stress, like a cold or flu, which does change measures of glucose.

HbA1c has relatively little intra-individual variability, whereas a fasting glucose or the 2-hour oral glucose tolerance test has tremendous variability from day to day. The HbA1c also provides the best index of risk for complications in the long term. The average glucose level a person is exposed to over time has a much greater role in complications than the day-to-day measurement of blood glucose that changes frequently.

With those advantages in mind, most international groups have adopted the HbA1c assay as a diagnostic test. It is also the best way of diagnosing people at risk for prediabetes. In many ways, it is more efficient, especially since all patients with newly diagnosed diabetes will require an HbA1c test.

Disadvantages, however, depend on the country. There may not be widespread availability, and HbA1c is, on the surface at least, a somewhat more expensive test. The lab costs can be three times as much as a single blood glucose test. However, other costs must be factored in, including the cost and inconvenience of a patient taking time off from work for a fasting or glucose tolerance test.

The HbA1c value can be artificially higher or lower, depending on the presence of various hemoglobinopathies; however, most assays now provide accurate measurements in the presence of the most common hemoglobinopathies. The more common disadvantage is alterations in red cell turnover that can affect the accuracy of the HbA1c test — for example, in individuals with certain anemias — but these are generally easily recognized factors.

There has been some discussion about whether the HbA1c translates into a different average glucose value by race. The data on this are weak at best. Some are absolutely convinced that this is the case, but there is an equal body of literature that suggests that the HbA1c is a good indicator of average glycemia for every race. We need to have a careful, acceptable study that demonstrates whether there is a clinically important difference among races in average glucose levels calculated from the HbA1c.

David M. Nathan, MD, is director of the Diabetes Clinical Research Center at Massachusetts General Hospital and professor of medicine at Harvard Medical School. Disclosure: Nathan reports no relevant financial disclosures.

HbA1c values are misleading.

The HbA1c assay has emerged in recent years as a gold standard measure of glycemic control and as a robust predictor of risk for diabetes-specific microvascular complications, like retinopathy. However, numerous limitations exist. Although HbA1c is an integrated measure of chronic glycemic load, the correlation between HbA1c values and ambient blood glucose levels among healthy individuals is quite modest. In addition, nonglycemic factors can contribute to a variance in HbA1c, including anemia, hemolysis, hemoglobinopathies and blood transfusion-dependent disorders, among others. Subclinical hemolysis, occurring frequently in patients with HIV, can also render an HbA1c measurement invalid. Studies have also found pregnancy to be associated with a reduction in HbA1c levels that is out of proportion with ambient glucose levels, making the assay an unreliable method for diagnosing gestational diabetes. Age, too, is another biological variable that seems to alter HbA1c levels independent of glycemia. In the Framingham study, researchers found participants aged at least 70 years had an HbA1c approximately 0.4% higher than participants aged 40 years or younger, despite similar blood glucose levels.

However, the strongest argument against use of the HbA1c assay as a sole diagnostic method for determining prediabetes is the unresolved issue of ethnic disparities in the correlation between HbA1c and ambient blood glucose levels. The exact mechanism behind these now well-documented differences remains unclear, but evidence indicates that black, Hispanic, Asian, Pacific Islander and Native American individuals generate different HbA1c values at similar blood glucose levels when compared with white patients. These differences have been noted across all age groups and among those who are healthy, have impaired glucose tolerance or have type 2 diabetes. In studies, HbA1c distribution curves show a shift to the right for U.S. ethnic minority groups vs. white Americans. That means the HbA1c assay as a sole diagnostic method to determine prediabetes status could push disproportionately large numbers of black patients across diagnostic thresholds, a fact that has important clinical and public policy implications. In addition, the HbA1c assay remains an expensive test, putting it out of reach for many patients in developing countries.

The key question is this: What percentage of individuals with an HbA1c of at least 6.5% has diabetes according to the WHO standard of a fasting plasma glucose greater than 7 mmol/L, or a 2-hour oral glucose tolerance test value of at least 11.1 mmol/L? The percentage should be as close to 100% as possible; however, one study showed that, when compared with the WHO standard, the HbA1c assay misdiagnosed between 47% and 73% of minority participants in Denmark, Greenland, Kenya and India. This is unacceptable for a diagnostic test. Until the mechanisms for the ethnic, demographic and other variations in HbA1c are better understood, using HbA1c as the sole diagnostic criterion for prediabetes is premature.

Samuel Dagogo-Jack, MD, MB, BS, MSc, FRCP, FACP, FACE, is an Endocrine Today Editorial Board member and the A. C. Mullins professor in translational research, professor of medicine, and chief of the division of endocrinology, diabetes and metabolism at the University of Tennessee Health Science Center in Memphis. Disclosure: Dagogo-Jack reports no relevant financial disclosures.