Prediabetes, intermediate hyperglycemia, insulin resistance, or abnormal blood glucose.
No matter what you call it, the glucometabolic state that may precede diabetes is considered one that warrants preventive action. Internists know to screen patients for diabetes, but even if blood glucose levels fall short, they may still need to provide a diagnosis and treatment.
Patients with higher-than-normal blood glucose levels are at increased risk for developing type 2 diabetes, cardiovascular disease, and stroke, according to the CDC. By the agency's estimates, 86 million U.S. adults have prediabetes, and 90% of them don't know it.
This is not because nine out of 10 people aren't being screened, said Tannaz Moin, MD, MBA, MSHS, an assistant professor of medicine at the David Geffen School of Medicine at the University of California, Los Angeles. “People are being screened because doctors are thinking about diabetes, and we have pretty standardized guidelines on who should be screened. … But results in the prediabetes range is where I think there's confusion among patients and also clinicians,” she said.
The confusion is understandable, as there have been questions about the diagnostic accuracy of the available screening tests, as well as about the evidence behind using an off-label drug to prevent diabetes. Some clinicians have even considered the term “prediabetes” to be a misnomer, as not all patients with elevated blood glucose levels go on to develop diabetes.
Regardless of what term is used for the preceding condition, however, the goal of preventing diabetes remains worthwhile. Several experts offered their evidence-based tips for managing screening, diagnosis, and treatment.
As of 2015, the U.S. Preventive Services Task Force (USPSTF) recommends screening for glucose abnormalities in adults ages 40 to 70 with overweight or obesity as part of cardiovascular risk assessment, a B-grade recommendation. The Task Force's prior recommendation, published in 2008, had suggested screening for type 2 diabetes in asymptomatic adults with sustained blood pressure greater than 135/80 mm Hg; this was also a B-grade recommendation.
Of note, the USPSTF avoids the term “prediabetes” in favor of the more specific diagnoses of increased average blood glucose level, impaired fasting glucose (IFG), and impaired glucose tolerance (IGT).
In the recommendations, increased average blood glucose level is indicated by an HbA1c level of 5.7% to 6.4%, IFG is indicated by a fasting plasma glucose level of 100 to 125 mg/dL (5.6 to 6.9 mmol/L), and IGT is indicated by an oral glucose tolerance test result of 140 to 199 mg/dL (7.8 to 11.0 mmol/L). All positive results should be confirmed, preferably by repeating the same test on a different day, according to the Task Force.
The American Diabetes Association (ADA) recommends the same cutoffs to diagnose prediabetes in its 2017 guidelines. However, the ADA recommends screening a broader population: all asymptomatic adults ages 45 and older and adults of any age who are overweight or obese with one or more additional risk factors for diabetes. The guidelines specify that prediabetes should be viewed as a risk factor for diabetes and cardiovascular disease, not as a clinical entity in its own right.
The American Association of Clinical Endocrinologists (AACE) and American College of Endocrinology (ACE) are in slight disagreement with the 5.7% to 6.4% HbA1c threshold, said Yehuda Handelsman, MD, FACP, past president of both organizations. The 2015 AACE/ACE guidelines agree that an HbA1c of 5.7% to 6.4% signifies risk but stipulate that the risk actually starts with HbA1c values of 5.5%. The groups add that unlike its diagnostic role in diabetes, HbA1c should be used only as a screening tool for prediabetes, requiring further glucose testing to make a definitive diagnosis.
“We do agree that anybody with an A1c of 5.5% to 6.4% has glucose abnormalities, but we also know that many of them may already have diabetes based on glucose criteria,” said Dr. Handelsman, adding that the organizations view metabolic syndrome, the cluster of such cardiovascular risk factors as hypertension and central obesity, as equivalent to prediabetes in terms of its associated risk of developing diabetes and cardiovascular disease.
To quantify the accuracy of the HbA1c and fasting blood glucose tests, researchers assessed 49 studies of screening tests for prediabetes as part of a systematic review and meta-analysis published in January in The BMJ. Compared to the two-hour oral glucose tolerance test (which is considered the gold standard for diagnosis of diabetes), HbA1c had a sensitivity of 49% and a specificity of 79% for diagnosis of prediabetes, and fasting blood glucose had a sensitivity of 25% and a specificity of 94%.
Even though these figures may make the glucose tolerance test look like the screening superstar, experts said that, for most patients, its utility does not stack up in real-world practice. “We don't do tons of glucose tolerance tests because it's time-consuming, and you have to draw blood a couple times,” said endocrinologist Diana McNeill, MD, FACP, a professor of medicine at the Duke University School of Medicine in Durham, N.C.
Instead, HbA1c is widely considered the most common screening test in clinical practice. So rather than screening with a glucose tolerance test, which also requires patients to ingest a 75-g sugary drink, it's more practical to recognize the limitations in sensitivity of the HbA1c and fasting glucose tests, said Dr. McNeill.
“If you've got someone who's high-risk, don't become completely comfortable if the HbA1c does not meet the criteria of 5.7% to 6.4% for diagnosis of prediabetes. You need to take it in conglomerate with other risk factors for development of diabetes,” she said, adding that the glucose tolerance test could be considered if there is a high clinical index of suspicion for prediabetes.
Similarly, if a patient has a high HbA1c on screening, Dr. Handelsman recommends ensuring that he or she does not have frank diabetes by either checking fasting glucose as a next step or, better yet, doing a fasting and two-hour glucose tolerance test. “Two hours is more accurate than only fasting because there's a whole group of people that have IGT and they don't have IFG, particularly people with obesity,” said Dr. Handelsman, who is also medical director and principal investigator at the Metabolic Institute of America in Tarzana, Calif.
Making the diagnosis
After confirming a prediabetes diagnosis, an internist needs to get the patient up to speed with what exactly that means—and there are several aspects to cover.
First, even though patients with prediabetes do not have the same specified cardiovascular goals, such as blood pressure cutoffs, as those with diabetes, cardiovascular risk is still of paramount concern, said Dr. Moin, an endocrinologist at the VA Greater Los Angeles Healthcare System. “You start to see an increase in cardiovascular risk with a little bit of dysglycemia, and that risk increases as A1c goes up,” she said.
Because of these concerns, internists should perform a 10-year cardiovascular risk assessment after making a prediabetes diagnosis by using one of the main calculators (e.g., Framingham Coronary Heart Disease Risk Score), said Dr. McNeill. She added that in patients who smoke, quitting will always be the priority before weight loss.
Low-risk patients might not need any further intervention beyond lifestyle changes, she said, but if cardiovascular risk is in the mid-range (5% to 7%), clinicians could consider starting a baby aspirin and/or a statin. Both aspirin and a statin are indicated if risk is over 10%, but there is a “gray area” between 7.5% and 10% that requires clinical judgment, Dr. McNeill said.
Of note, studies have suggested that statins may cause hyperglycemia, but Dr. McNeill said the risk of developing diabetes from statins is very low and does not outweigh the risk of developing cardiovascular disease. “I'm not ignoring it, but it's not clinically significant to an endocrinologist,” she said.
There are several ways to communicate a prediabetes diagnosis, and the right approach depends on the internist's practice style and the patient. First, though, is consideration No. 1: using the word itself.
Although prediabetes as a term has caught on in the U.S., Dr. Handelsman noted that European organizations (e.g., the European Association for the Study of Diabetes) do not like it. “They did not think that somebody should get another [diagnosis] for a medical condition. Therefore, they never adopted the term, and they call these people at high risk for diabetes,” he said.
In an article published in 2014 in The BMJ, Victor Montori, MD, FACP, and John Yudkin, MD, argued against diagnosing prediabetes because of the harms of overdiagnosis: problems with self-image, insurance, employment, costs of care, and drug side effects. “Pre-diabetes could be defined as a risk factor for developing a risk factor. With this label comes much of the same baggage as for diabetes, without evidence of long term benefit,” they wrote.
Dr. Handelsman noted that there is a conversion of about one-third of people with prediabetes back to normal glucose without any management. He also pointed out that participants in the landmark Diabetes Prevention Program (DPP) trial needed to have both IFG and IGT to be eligible. “These are not great scientific numbers, but the rate of progression to diabetes within five years for people that have IFG [using ADA criteria] is about 30%, and people that have IGT had a 40% risk for developing diabetes within five years,” Dr. Handelsman said. “Both groups together presented an 80% risk.”
He said that he is not an alarmist and that he uses a combination of terms, such as glucose abnormalities, when discussing diabetes risk with patients. “I don't give them a diagnosis; I say, ‘You have some abnormalities in your glucose, and for your body now, because you gained all this weight, it's getting difficult to manage your glucose.’ That's a good way to describe it to them,” he said.
Dr. McNeill said that she tells her patients that they have metabolic abnormalities or symptoms of metabolic syndrome that may lead to diabetes. “If you use the term ‘may lead to diabetes,’ that makes people think that they can prevent it,” she said.
Dr. Moin, meanwhile, is a proponent of the term. “In many cases, I feel like using the term and labeling it, making it tangible, [and] taking the time to talk about it can make it an opportunity, and a motivating one, for patients to make some of the changes,” she said. “I've yet to meet a patient who doesn't care if they get diabetes or not.”
But the term that's used to describe this condition may be less important than knowledge of the condition itself. Research has shown that communicating increased diabetes risk may increase patients' chances of successfully modifying their lifestyles.
For example, patients who were told that they were at higher risk for diabetes appeared more likely to adopt healthy lifestyle behaviors in a 2012 study published in Primary Care Diabetes. In another study, prediabetes-aware adults had higher odds of engagement in physical activity and weight management than those who were unaware of their diagnosis, according to results published in 2015 in the American Journal of Preventive Medicine.
These behaviors are important because the 27-center DPP trial, published in 2002 in the New England Journal of Medicine, showed that intensive lifestyle intervention reduced the incidence of diabetes by 58% compared to placebo.
Based on the evidence, the CDC has launched the National DPP, a network of organizations that offer evidence-based lifestyle change programs online and in various community settings. As part of these CDC-recognized programs, trained lifestyle coaches work with patients over the course of one year to help them implement healthier eating, movement, and coping habits, with the ultimate goal of preventing or delaying diabetes.
With the DPP being widely implemented and the CDC encouraging more physicians to identify and treat prediabetes, researcher Dina Griauzde, MD, wanted to see what was happening in clinical practice.
Dr. Griauzde, a VA fellow in the University of Michigan Robert Wood Johnson Clinical Scholars Program, and her research team interviewed 20 primary care physicians about their clinical decisions regarding 134 purposively sampled patients without diabetes who met ADA criteria for screening.
She found that, for 24 patients who had prediabetes, the doctors usually recommended weight loss and increased physical activity (58%) but never specifically recommended participation in a DPP, according to results published in April in BMC Family Practice.
They also never recommended metformin, which was shown in the original DPP trial to reduce the risk of developing diabetes by 31% compared to placebo. (The review published earlier this year in The BMJ also assessed 50 studies of interventions, determining diabetes risk reductions of 36% for lifestyle intervention and 26% for metformin, compared to placebo.)
Clinicians seem to be conflicted about the diagnosis of prediabetes and what it means for patients, and this is apparent in their referring and prescribing practices, Dr. Griauzde said. “I think a lot of people are hesitant to give patients the diagnosis, and as a result, they don't offer evidence-based treatment,” she said.
Indeed, other research has found gaps in metformin prescribing for prediabetes.
Between 2010 and 2012, only 3.7% of more than 17,000 insured adults with prediabetes were prescribed metformin, according to a 2015 study in Annals of Internal Medicine. The metformin prescribing rate for prediabetes is “abysmally small,” said Dr. Moin, lead author of the study. “I think there are some patients that would benefit, and we should be talking about this as an option with our patients.”
Experts said patients typically prefer to try to change their lifestyles before taking a pill, but the medication may provide a useful tool if behavior change proves challenging (as it often does).
A qualitative study of 35 adult primary care patients with prediabetes found that the majority considered both lifestyle intervention and metformin to be acceptable treatment options, according to results published in December 2016 in Diabetes Educator.
“When I talk to internists all over the country, what I hear is ‘Oh, but patients would never take metformin,’” said lead author Matthew J. O’Brien, MD, an assistant professor of medicine and preventive medicine at Northwestern University Feinberg School of Medicine in Chicago. “At least in a pretty diverse sample that we looked at, we found that really wasn't the case.”
Prescribing metformin to prevent diabetes is technically off-label use, as the FDA hasn't approved the medication for this indication. However, Dr. Moin pointed out that pharmaceutical companies do not seem to be interested in filing an application for relabeling an off-patent drug, as they don't stand to make money from such action.
She noted that there are nuances in the evidence behind metformin, which seems to work better in select subgroups of patients. Based on randomized clinical trial evidence, Dr. Moin said, the following groups would benefit from metformin more than others: patients younger than age 60, women with a history of gestational diabetes, and individuals with BMIs greater than 35 kg/m2.
She added that clinicians may choose to use both metformin and behavior modification, but there is no evidence to support the efficacy of combining them. “I think patients can use both,” Dr. Moin said. “And if the A1c is improving and if they're able to lose weight, then we can discuss stopping the metformin.”
Despite evidence that metformin delays progression to diabetes, Drs. Moin and O’Brien noted that a common argument among physicians is that prescribing the drug amounts to prematurely treating prediabetes as diabetes. However, Dr. O’Brien said, metformin does more than lower blood glucose, such as produce weight loss and other biochemical changes. “There's much more going on,” he said.
Another concern is that there is no evidence that metformin will improve long-term mortality or lower the risk of cardiovascular events, Dr. Moin noted, adding that two large, placebo-controlled trials are underway to address this lack of evidence.
“Metformin is a real opportunity space for us,” said Dr. O’Brien. He suggested starting at the lowest dose (500 mg twice daily) and increasing as needed. Gastrointestinal side effects may affect up to 10% of patients, he noted, but a long-acting extended-release form of the medication, taken once daily, may help with side effects at some added expense.
The costs of treating prediabetes are worth considering. Metformin is generally inexpensive, and CMS has announced plans to expand Medicare to cover the DPP (which varies in price but can cost up to $400 to $500 out of pocket) starting in January 2018.
More broadly, cost comes into play in terms of who is at risk for developing diabetes. Those of low socioeconomic status are less likely to be screened and are at greater risk for diabetes than those of high socioeconomic status, said Gabriela Spencer Bonilla, a medical student in Dr. Montori's Knowledge and Evaluation Research Unit at the Mayo Clinic College of Medicine and Science in Rochester, Minn.
“In fact, the diabetes epidemic has basically stabilized within the white, higher-income portion of the population, but in young, brown, and poor people, the epidemic of diabetes continues to grow,” she said. Although behavioral programs and medications can work well for those who have the resources, preventing diabetes for all will require societal change, Ms. Spencer Bonilla said.
“We should be creating societies that push people toward being healthier … [with] safe neighborhoods to walk in and access to fresh and healthy food,” she said.