Reconciling conflicting diabetes guidance

The 73-year-old patient has had type 2 diabetes for years and also has a history of two heart attacks. On a recent office visit, his HbA1c was 7.3% on metformin alone. Before ACP issued new glycemic guidance in March, Jack Ende, MD, MACP, said he would have considered adding additional medications such as insulin or a sulfonylurea, or even more expensive medications such as sitagliptin, to get the patient's HbA1c to strictly adhere to existing recommendations from the American Association of Clinical Endocrinologists (AACE) for an HbA1c of 6.5% or lower.

But because ACP now recommends for most patients with type 2 diabetes an HbA1c level between 7% and 8%, Dr. Ende instead told the patient that at least for now, no additional medication is needed. Adding other medications, he explained, is no longer recommended because it won't necessarily help and could lead to problems such as hypoglycemia.

Using ACP's new guidance will help internists improve care and reduce costs, said Dr. Ende, Immediate Past President of ACP. But while there is general support for the ACP guidance statements that recommend individualizing care and avoiding intensive treatment in patients with a life expectancy of less than 10 years, the American Diabetes Association (ADA), the AACE, the American Association of Diabetes Educators, and the Endocrine Society have come out strongly against the statement recommending the new glycemic levels.

The recommendation of a higher HbA1c level in most patients with diabetes could cause more harm than good and could curtail long-term benefits of glucose control, the organizations contended in a joint press release in March. They also noted that ACP's guidance statements did not include research on newer drug classes, glucagon-like peptide 1 (GLP-1) receptors and sodium glucose co-transporter-2 (SGLT2) inhibitors, which they said are not associated with hypoglycemia or weight gain and might offer potential long-term cardiovascular benefit. The groups prefer to continue basing treatment on ADA's 2018 Standards of Care and the 2018 algorithm from AACE and the American College of Endocrinology on comprehensive type 2 diabetes management. The latter states in part, “An A1c level of ≤6.5% is considered optimal if it can be achieved in a safe and affordable manner, but higher targets may be appropriate for certain individuals and may change for a given individual over time.” The ADA's position is that an HbA1c of 7% or lower is an appropriate target in most nonpregnant adults based on the available evidence to date.

The current controversy centers on ACP's summary of the trials included in the guidelines that showed that pharmacologic management to HbA1c targets below 7% did not reduce long-term microvascular or macrovascular clinical outcomes compared to targets of 7% to 8% but did result in harms. Endocrinologists may be responding to their concerns about markers for microvascular changes such as albuminuria or serum creatinine, whereas ACP was looking for what's important to patients, such as blindness or amputation, Dr. Ende noted.

Amir Qaseem, MD, PhD, FACP, Vice President of Clinical Policy for ACP, emphasized that the new statements are guidance statements, not a guideline, and as such are based on systematically reviewing existing guidelines along with the evidence contained in the reviewed guidelines. He acknowledged that studies on the newer drugs were not included in the reviewed guidelines but said that ACP didn't want to wait, given that diabetes care is an important part of an internists' practice.

Endocrinologists are concerned, however, that the new targets will stymie efforts to rein in the number of patients with diabetes and its harmful long-term effects. “Increasing the targets gives people the impression that it's OK to have higher glycemic levels and that we are doing OK as far as diabetes is concerned. I don't think we're in that situation,” said Jonathan Leffert, MD, FACP, president of AACE. “A more aggressive stance is beneficial if it can be accomplished safely.”

But the guidance on raising the glycemic target levels as appropriate should not be misconstrued, said Dr. Ende, who is the Schaeffer Professor of Medicine at the Perelman School of Medicine at the University of Pennsylvania.

“The guidance statement shouldn't be taken to think ACP is relaxing on diabetes,” he said. “It's still a major source of morbidity, heart attacks, amputation, blindness, kidney failure, and neuropathy and deserves all of our attention.”

Controversy over targets

To develop its guidance, ACP reviewed guidelines from the National Institute for Health and Care Excellence (NICE), the Institute for Clinical Systems Improvement (ICSI), AACE/ACE, the ADA, the Scottish Intercollegiate Guidelines Network (SIGN), and the U.S. Department of Veterans Affairs and Department of Defense (VA/DoD). It also reviewed these large randomized trials: ACCORD (Action to Control Cardiovascular Risk in Diabetes), ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation), both UKPDS (United Kingdom Prospective Diabetes Study) trials, and the VADT (Veterans Affairs Diabetes Trial).

ACP used the AGREE instrument (Appraisal of Guidelines for Research and Evaluation II) to evaluate the quality of the guidelines it reviewed, looking at the recommendations and evidence presented, Dr. Qaseem explained. He said ACP found that the harms of treating to lower glycemic levels, including hypoglycemia, weight gain, and death, outweighed the benefits for most people. For example, he noted that the ACCORD study was stopped early because targeting HbA1c rates of less than 6.0% led to a 22% increase in all-cause mortality, a 35% increase in cardiovascular-related death, and a threefold increase in risk for severe hypoglycemia deaths.

But Dr. Leffert said that even though there were more hypoglycemic values in the intensive group in ACCORD, among those with severe hypoglycemia, mortality was higher in the conventional group. He also noted that the studies ACP cited included older populations, while he often sees relatively newly diagnosed patients who are younger than age 60 years. “So we feel [ACP's] targets are not adequate because studies such as ACCORD involved older, sicker patients with longer duration of diabetes evaluated in an era without today's improved medications and monitoring capabilities. This is like comparing apples with oranges,” he said. Instead, he noted that UKPDS 33 and 34 showed a benefit of lower glycemic targets for newly diagnosed patients with an average age of 54 years. “Most important is that these were decades-old studies, and in the sickest possible patients, i.e. poorly controlled, longstanding type 2 diabetes with already established cardiovascular disease or multiple cardiovascular disease risk factors, not reflecting the majority of patients, and using older drugs with side effects of hypoglycemia and weight gain,” he said.

Not everyone agrees. Patients enrolled in these studies are typical of most patients with type 2 diabetes, said Timothy J. Wilt, MD, MPH, MACP. Intensively treated patients in UKPDS 33 and 34 had HbA1c levels of 7.0% and 7.4%, within the range recommended by ACP, said Dr. Wilt, core investigator and staff physician at the Minneapolis VA Center for Chronic Disease Outcomes Research and professor of medicine at the University of Minnesota School of Medicine, who was a coauthor on the ACP guidance statements. Any clinical benefit through more than 10 years of follow-up in these two studies was 1% or less, he noted.

Regarding the balance of benefits and harms, physicians could use an HbA1c target of 7% or less in some younger healthier individuals if it can be done without adverse drug affects and if the patient wants more aggressive control, Dr. Qaseem said.

Dr. Leffert said endocrinologists feel comfortable with treating to lower targets. “Our interpretation is that the benefits are worth the risks and that we can mitigate the risks through a number of different treatment options and monitoring options,” he said.

But ACP warned that information from large long-term randomized trials provided convincing evidence that treating to HbA1c levels of less than 7% compared to between 7% and 8% for most individuals did not reduce death, or macrovascular or clinical microvascular events over 5 to 10 years, but did lead to harms. Dr. Wilt noted that the impact of those harms should not be underestimated.

“It's not just a little wooziness. Several studies showed a threefold increase in serious hypoglycemia where the person needed to be rescued by a bystander,” he said. Other important harms included weight gain, fluid retention, hospitalization, and burden and cost of care, he said. Furthermore, he noted that one trial conducted in the U.S. found that treatment to an HbA1c level of 6.4% increased all-cause and cardiovascular mortality compared to HbA1c levels greater than 7%.

A hypoglycemic event can lead to car crashes, loss of consciousness, and personal harm, said Marie T. Brown, MD, FACP, associate professor at Rush Medical College at Rush University Medical Center in Chicago. “Even one episode of hypoglycemia … can affect the life of that patient forever because they lose trust in medicine … for fear it could happen again,” she said.

Dr. Wilt said that ACP would have recommended treating to a lower target if evidence showed benefit, but “an HbA1c between 7% and 8% [is] where the science shows the best balance of benefits and harms.”

ACP's guidance stated that the phrase “most patients” means that there are exceptions based on individualized care. But subspecialists believe using “most” applies the new glycemic targets too broadly.

“By lumping ‘most’ people with type 2 diabetes into a 7% to 8% target range, ACP's new guidance may cause potential harm to those who may safely benefit from lower evidence-based targets,” said William T. Cefalu, MD, chief scientific and medical officer for the American Diabetes Association. As a result, he said the ADA is sticking to its recommendation that “a reasonable A1c goal for ‘many’ nonpregnant adults is less than 7% based on the available evidence to date.”

There's also the danger that a recommended HbA1c target of 7% to 8% may lead some physicians to become more comfortable with patients hitting the 8's, said ACP Member M. Sue Kirkman, MD.

“I don't want to get where 9 is okay because it's close to 8,” said Dr. Kirkman, medical director of the University of North Carolina Diabetes Care Center clinical trials unit in Chapel Hill. “The concern … is that the pendulum might swing too far to think that glucose control doesn't matter at all.”

For related reasons, the guidance to deintensify treatment if a patient's HbA1c is less than 6.5% was also controversial. Based on evidence in all of the guidelines reviewed, ACP said health outcomes are not improved at that level but that the patient could have potential harms. Dr. Leffert, meanwhile, said a patient on an oral agent may have an HbA1c of 6.5% precisely because he's on the agent. Deintensifying might make him hyperglycemic, he said. He noted that the ADA, the AACE, the American Association of Diabetes Educators, and the Endocrine Society support deintensification or simplification of complex regimens to reduce the risk of hypoglycemia in older adults if this can be done while adhering to each patient's individualized HbA1c target.

Dr. Wilt again noted that studies clearly show that reaching an HbA1c level below 6.5% with medications results in no clinical benefits but substantial harms. Therefore, physicians should consider medication deintensification in these individuals and aim to achieve an HbA1c level of between 7% to 8%, he said.

Dr. Leffert said that the newer GLP-1 receptors and SGLT2 inhibitors may help patients get to lower glycemic levels, which can decrease their risk of microvascular complications. Because the ACP guidance was based on previous guidelines that didn't discuss these new drugs, ACP left decisions on their use to individual practitioners for now.

Dr. Wilt noted that the newer drugs have not been evaluated in “treat-to-target” studies and that none of the other guideline groups included studies of newer drugs when developing their treat-to-target recommendations. These drugs have typically been tested as “add-on therapies” in selected high-risk adults with HbA1c levels that are above the targets recommended by ACP, he said.

Dr. Wilt also discouraged adding a drug just to drive down HbA1c levels in the absence of clinical benefit and noted that the new medications can cost $1,000 per month. “The new drugs may have a role [in diabetes management], particularly in higher-risk individuals with HbA1c levels persistently above 8% despite use of other more established and lower-cost medications. They were not part of our review or the other guideline groups' HbA1c target recommendations,” he said.

Regarding cost, Dr. Leffert contended that few patients pay cash retail prices for the newer drugs and that retail costs are more like $400 to $500 per month, comparable to monthly costs of $300 for basal insulin and $300 for bolus insulin. And Irl B. Hirsch, MD, MACP, said many endocrinologists are “flabbergasted” by ACP's stance.

“To leave [the newer drugs] out of the analysis and discussion is a mistake in 2018 because the conclusion [ACP] reached mayhave been appropriate a decade ago,” said Dr. Hirsch, who is a professor of medicine at the University of Washington School of Medicine in Seattle. The new guidance should have been deferred until SGLT2 inhibitors and GLP-1 receptor agonists could have been considered, he said.

Again, Dr. Wilt noted that SLGT2 inhibitors and other newer drugs have not been evaluated in “treat-to-target” trials and have typically been studied in patients with HbA1c levels above 8%.

Common ground

Despite concern over glycemic targets, there's agreement that diet and lifestyle, including exercise, weight loss, blood pressure control, and smoking cessation, are the cornerstones of primary care treatment to control diabetes, Dr. Wilt said. There's also strong agreement that care should be individualized. ACP specifically noted the importance of that concept by including it in its first guidance statement, said Dr. Qaseem.

Individualized care includes taking into account the patient's risk for hypoglycemia, weight gain, and other drug-related side effects, along with age, life expectancy, other chronic conditions, functional and cognitive impairments, fall risk, ability to adhere to treatment, and medication burden and cost.

“Guidelines are not rules. Management decisions should be individualized and decisions made jointly with patients,” said Dr. Ende.

Dr. Cefalu said the ADA, like the ACP, has long recommended that treatment goals be individualized. And Dr. Leffert said endocrinologists believe that using guidelines with personalized care is best for patients. “ACP's [Guidance Statement 1] is in line with what we've felt for a long time,” he said.

There's also agreement that the harms of intensive glucose management in patients at high risk and with comorbidities may outweigh the benefits, particularly in those with limited life expectancy. Still, Dr. Cefalu urged caution given that someone with a chronic condition may still live for years.

“The ADA disagrees that this applies broadly to anyone over age 80, anyone living in a nursing home, or anyone with chronic conditions and who has a limited life expectancy,” he said. “Each specific case should be evaluated individually.”

Dr. Wilt noted that any benefit from more intensive glycemic control is small in absolute terms; that it requires 10 or more years to occur; and that harms, burden, and costs of care are particularly detrimental to older, sicker individuals.

Impact on practice

Some internists may already provide individualized, patient-centered care for their patients with diabetes that takes more than the HbA1c into account.

For example, Dr. Kirkman said that while the recent controversy prompted her to review the risks and benefits and the differing views on what to do next, she's staying with what's worked for her and her patients.

“Focus on the individualization and personalization and not so much on the exact [HbA1c] numbers from any organization,” she said.

Dr. Brown said that the discussion has reinforced her patient-centered care for patients with diabetes who also often have hypertension, obesity, and depression and who often smoke. “The important message that the guidance emphasizes is that treating the entire patient is of far greater importance than treating just the A1c,” she said. “Adding medicines that cause weight gain (insulin and sulfonylureas) to a patient already struggling with obesity to achieve a lower A1c goal does not serve our patients well.”

For those who will make changes based on the new guidance, Dr. Wilt suggested telling patients that medicine is continually changing. Dr. Ende suggested noting that medicine learns not just from basic science studies but also from synthesizing evidence.

Dr. Wilt acknowledged that changing patients' as well as physicians' long-held beliefs can be difficult. “But when science shows it's for better health, we have to find ways to change,” he said. “Many of my patients are pleased to learn that they can have similar health benefits with fewer medication harms, burden, and costs by aiming for an HbA1c target between 7% and 8%.”