Questions and Answers
Action to Control Cardiovascular Risk in Diabetes (ACCORD) Trial
June 6, 2008
The ACCORD (Action to Control Cardiovascular Risk in Diabetes) Trial (www.accordtrial.org ) is a large clinical study of adults with established type 2 diabetes who are at especially high risk of cardiovascular disease (CVD). Type 2 diabetes is a complex metabolic disease characterized by high blood glucose (sugar) levels. People with this form of diabetes have insulin resistance and a progressive loss of the ability to produce insulin. Insulin is needed so that sugar in the blood (which comes from eating food) can enter the cells of the body in order for the body to function.
Type 2 diabetes increases the risk of a number of complications, especially CVD, which is the leading cause of early death in people with diabetes. Many people with diabetes are overweight and have high blood pressure and lipid or cholesterol problems-conditions that further add to their CVD risk. Adults with type 2 diabetes are two to four times more likely to die of heart disease and stroke than adults without diabetes, and about 65 percent of people with diabetes die from heart disease or stroke.
The ACCORD Trial is testing three treatment approaches to determine the best ways to decrease the high rate of major CVD events - heart attack, stroke or death from CVD - among people with type 2 diabetes who are at especially high risk of CVD. These treatment approaches are: intensive lowering of blood sugar levels compared to a more standard blood sugar treatment; intensive lowering of blood pressure compared to standard blood pressure treatment; and treatment of blood lipids by a fibrate plus a statin compared to a statin alone.
The study began enrolling participants in 2001 and is taking place in 77 clinical sites across the United States and Canada. A total of 10,251 adults with established type 2 diabetes are participating in the ACCORD trial. At enrollment, study participants were between the ages 40 and 79 (average age 62), had diabetes for an average of 10 years, and were at especially high risk for CVD events because they either already had diagnosed CVD or they had at least two CVD risk factors in addition to type 2 diabetes. The other CVD risk factors could be high cholesterol (high low-density lipoprotein cholesterol, or LDL), high blood pressure, smoking, or obesity.
ACCORD participant treatment is scheduled to end in 2009, and researchers plan to report the final results in 2010.
Return to "Contents"
The National Heart, Lung, and Blood Institute (NHLBI), which sponsors the ACCORD trial, stopped the intensive blood sugar lowering strategy in the study on February 6, 2008, and has transitioned participants in that treatment group to the same treatment strategy as participants in the standard group. NHLBI decided to make this change due to safety concerns after a thorough review of the available data and a recommendation by the study's Data and Safety Monitoring Board. The other two treatment questions being examined in the study - the blood pressure and lipid trials - will continue until June 2009 as planned.
Therefore, ACCORD participants are continuing to receive blood sugar treatment from their study clinicians until the planned trial conclusion in June 2009.
The results of intensive blood sugar lowering treatment strategy in the ACCORD
study are published by the New England Journal of Medicine
online June 6, 2008, and in the June 12, 2008, print edition.
The results were also presented June 10 at the American Diabetes
Association's 68th Annual Scientific Sessions in San Francisco.
Return to "Contents"
The ACCORD Data and Safety Monitoring Board (DSMB) is an independent group of 10 experts who regularly examine the study outcomes and safety data. The ACCORD DSMB includes expertise in diabetes, cardiovascular disease, statistics, ethics, epidemiology, and clinical trials. The DSMB is responsible for providing recommendations concerning starting, continuing, and/or stopping the clinical research study under review. The DSMB's recommendations are based on safeguarding the interests of study participants, assessing the safety and efficacy of study procedures, and monitoring the overall conduct of the study.
Since the trial's onset in 2001, the ACCORD DSMB has met regularly, generally every six months, to monitor study conduct and to review data from ACCORD.
Return to "Contents"
In its regular review of the available study data, the ACCORD DSMB noticed an unexpected increase in total deaths from any cause among participants who had been randomly (by chance) assigned to the intensive blood sugar strategy group compared to those assigned to the standard blood sugar strategy group. The data analyses showed that over an average of 3.5 years of treatment (ranging from about 2 years to about 7 years), 257 participants in the intensive group died, compared to 203 in the standard group - a difference of 54 deaths, or an excess of about 3 deaths per 1,000 participants treated for a year. This translates to a statistically significant 22% higher rate of death in the intensive than the standard group.
There was a trend toward lower (10% lower) rate of primary outcome events, primarily non-fatal heart attacks, in the intensive compared to the standard treatment group. However, the DSMB recommended discontinuing intensive blood sugar treatment because the harm of the intensive strategy outweighs the potential benefit. NHLBI accepted the DSMB's recommendation and decided to transition all participants to the standard blood sugar strategy. This transition enables ACCORD to fulfill its commitment to provide diabetes therapy to the participants until 2009, and, at the same time, to enable the other two questions (blood pressure and lipids) to be answered.
On the whole, the death rates in both blood sugar strategy groups were lower than those seen in similar populations. That is, although the death rate was higher in the intensive treatment group than the standard group, it was still lower than death rates reported in other studies of type 2 diabetes. This difference is likely because ACCORD participants have received and continue to receive excellent care for their diabetes, and because participants in clinical trials generally have excellent adherence to their treatment regimens.
Return to "Contents"
The ACCORD trial was designed to determine if a strategy of intensive lowering of blood sugar levels, intensive lowering of blood pressure, or treatment of blood lipids with a fibrate plus a statin, can reduce the risk of major CVD events in patients with type 2 diabetes who are at especially high risk of CVD. Thus, the ACCORD trial was designed to answer three complementary questions about treatment of people with diabetes.
The primary outcome measure for the trial is the first occurrence after randomization of a major CVD event, specifically nonfatal heart attack, nonfatal stroke, or CVD death. Secondary outcomes include total mortality (deaths), microvascular outcomes (eg, eye, kidney, and nerve complications), health-related quality of life, and cost-effectiveness.
Previous clinical trials in patients with type 2 diabetes have shown that
- blood sugar lowering reduces eye, kidney, and nerve complications of diabetes,
- blood pressure lowering reduces CVD, eye, and kidney complications of diabetes, and
- LDL-cholesterol lowering with statins reduces CVD events, like heart attacks and strokes.
However, no previous clinical trial has tested lowering blood sugar and blood pressure to near-normal levels in patients with diabetes, and no previous clinical trial has tested effects of treating multiple blood lipids compared to treating only LDL-cholesterol in persons with diabetes.
Return to "Contents"
After providing written informed consent, the participants were randomly (i.e., by chance) assigned to intensive blood sugar-lowering strategy or to standard blood sugar-lowering strategy. Blood sugar levels are measured by a blood test called A1C, which reflects a person's average level of blood sugar in the previous 2 to 3 months. Participants in ACCORD's standard group were treated to a target A1C level of 7 to 7.9 percent, which is about that achieved in the U.S. and lower than the participants' levels when they joined the study. Therapy for those in the intensive blood sugar group sought to achieve an A1C of less than 6 percent, a level of blood sugar that is similar to that of a person without diabetes.
Participants were also asked to perform home blood glucose monitoring and learned about preventing, recognizing, and treating hypoglycemia (very low blood sugar). Hypoglycemic events were monitored carefully and, if necessary, medications were adjusted to minimize severe hypoglycemic reactions.
In addition to being randomized to one of the blood sugar strategy groups, participants were also enrolled in one of two other randomized clinical trials. One trial is comparing intensive lowering of blood pressure - to near-normal levels - to standard blood pressure lowering. The other trial is examining the effects of treating multiple blood lipids by using a fibrate plus a statin - to raise HDL ("good") cholesterol, lower triglyceride levels, and lower LDL ("bad") cholesterol - compared to a statin to lower LDL cholesterol and a placebo that looks like the fibrate.
Return to "Contents"
A large body of research in which adults were followed for several years has shown that diabetes increases the risk of developing CVD. In addition, the higher the blood sugar, the greater the future risk of CVD, and the lower the blood sugar, the lower the risk of CVD. However, most studies were "observational", rather than more rigorous randomized clinical trials that test interventions or treatments, so they do not prove whether lowering blood sugar levels reduces CVD risk. Although observational studies are important to advancing our understanding of health and disease, well-conducted randomized clinical trials are considered the gold standard for determining safe and effective treatments.
Some clinical trials, however, have tested the effects of reducing blood sugar through medical treatments. The United Kingdom Prospective Diabetes Study (UKPDS), the NIH-supported Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study, and other clinical trials have demonstrated benefits of lowering blood sugar in persons with diabetes. Such studies have shown, for example, that
- For every percentage point drop in A1C blood test results (from 8.0 percent to 7.0 percent, for example), the risk of diabetic eye, nerve, and kidney disease is reduced by 40 percent. Lowering blood sugar reduces these microvascular complications in both type 1 and type 2 diabetes.
- Intensive blood sugar control in people with type 1 diabetes (average A1C of 7.4%) reduces the risk of any CVD event by 42 percent and the risk of heart attack, stroke, or death from CVD by 57 percent (source: DCCT/EDIC, reported in December 22, 2005, issue of the New England Journal of Medicine.) Note: Although type 2 diabetes is similar to type 1 diabetes (previously known as "juvenile" diabetes) in that both involve high blood sugar levels due to inadequate insulin, the two forms of diabetes are different in many ways, and results from studies in one type may not be relevant to the other type.
- In adults with more recent onset type 2 diabetes than the ACCORD participants, the UKPDS showed a trend toward reduced CVD events, particularly heart attacks. However, the results are not considered conclusive.
Moreover, no previous major clinical trial has studied whether lowering a raised blood sugar level to a level similar to that seen in people without diabetes reduces the risk of CVD. In addition, no previous clinical trial has studied the effects of intensive blood sugar lowering in people with longstanding type 2 diabetes who already have CVD or have multiple risk factors for CVD in addition to diabetes.
Return to "Contents"
All major classes of drugs approved by the FDA to treat diabetes were used to treat blood sugar in both groups of the ACCORD blood sugar trial - those in the intensive strategy group that targeted near-normal blood sugar level as well as those in the standard blood sugar strategy group.
Specific medications used included (in order of frequency of use): metformin; thiazolidinediones, or TZD's (rosiglitazone, pioglitazone); injectable insulins; sulfonylureas (gliclazide, glimepiride, glipizide, glyburide); and acarbose and exenatide. Combinations of medications could be used to achieve the A1C goals.
The type, number and dosages of these drugs varied, depending on participants' individual needs and their A1C goals. For example, metformin was used in about 95% of intensive-group participants, and in about 87% of standard group participants. Insulin was used in about 77% of intensive-group participants, and in about 55% of standard-group participants.
Participants in the intensive group were more likely to be on combinations of drugs than participants in the standard group. For example, 52% of participants in the intensive strategy group were on three oral medications as well as insulin, compared to 16% of those in the standard group.
For more information on diabetes medications, see http://diabetes.niddk.nih.gov/dm/pubs/medicines_ez/index.htm.
Return to "Contents"
ACCORD researchers have extensively analyzed the available data and have not been able to identify to date any specific cause for the higher death rate among the intensive blood sugar treatment group. Based on analyses done to date, there is no evidence that any medication or combination of medications is responsible for the higher risk. In addition, although the number of severe hypoglycemic events was greater in the intensive blood sugar strategy group, this does not appear to account for the difference.
Because of the recent concerns with rosiglitazone (Avandia), which is one of several medications used in ACCORD, researchers specifically reviewed data to determine whether there was any link between this particular medication and the increased deaths. To date, no link has been found. (For more on this issue, see http://public.nhlbi.nih.gov/newsroom/home/GetPressRelease.aspx?id=287 .)
ACCORD researchers will continue to monitor participants and conduct additional analyses to try and explain the findings. However, the main conclusion at this time is that the treatment strategy is responsible for the findings, not any specific component of the strategy.
Return to "Contents"
About half of the deaths were from cardiovascular diseases, such as heart attack, sudden cardiac death, stroke, heart failure, or another cardiovascular disease condition. Of the total 460 total deaths, there were 229 deaths from cardiovascular causes - 135 in the intensive treatment group and 94 in the standard treatment group. This difference resulted in a significant 35 percent higher cardiovascular death rate in the intensive strategy group. The remainder of the deaths was from other causes such as cancer. Differences between the intensive and standard treatment groups were present in overall death from any cause as well as deaths from combined cardiovascular causes.
In addition, death rates were consistently higher in the intensive strategy group regardless of participant characteristics at study enrollment, such as whether the participants had existing cardiovascular disease (versus multiple CVD risk factors), were male or female, were older or younger than 65 years, had an A1C level lower or higher than 8 percent, or were of white or non-white race/ethnicity.
Return to "Contents"
As noted above, participants' characteristics at enrollment did not impact their risk of death from the intensive strategy. However, for the primary outcome of any CVD event, including nonfatal events, there were some differences. Participants who did not already have a CVD diagnosis (that is, no history of heart attack or stroke) at enrollment were less likely to have a cardiovascular event (fatal and nonfatal heart attacks or strokes) if they were in the intensive group than if they were in the standard group. A similar result was seen in participants who had lower A1C levels at baseline (8.0% or lower). These findings, however, are only suggestive and are not definitive.
Return to "Contents"
In general, intensive-group participants achieved lower blood sugar levels than standard-group participants. An A1C level less than 6.4% was achieved in half of the intensive-group participants and a level less than 7.5% in half of the standard-group participants. These levels were maintained over the average 3.5 year study period. The level achieved in the intensive group required substantial effort on the part of ACCORD clinicians and patients, including multiple medications, frequent clinic visits (at least every 2 months), and frequent self-monitoring of blood sugar levels, sometimes multiple times a day.
Return to "Contents"
As noted above, analyses suggest that participants who were assigned to a strategy of blood sugar lowering aiming for an A1C less than 6.0 percent had a lower risk of a nonfatal heart attack.
However, it appears that, if a heart attack did occur, it may have been more likely to have been fatal.
ACCORD investigators are analyzing the effects of this therapeutic strategy on other serious health consequences of diabetes such as eye and kidney diseases.
Return to "Contents"
The level of care was excellent. The average LDL-cholesterol (the "bad" cholesterol) level was about 91 mg/dL in both groups, which is below the recommended level of 100 mg/dL for people with diabetes. Their systolic blood pressure was on average about 127 mm Hg, which is below the recommended level of 130 mm Hg for diabetes patients. About three-quarters were on aspirin, which is recommended to prevent heart attacks and strokes. These values compare very favorably with the level of care in the U.S. as a whole.
Return to "Contents"
There was a greater amount of seriously low blood sugar (called hypoglycemia) events in the intensive group than in the standard group: about 10% of intensive participants and about 3.5% of standard participants had a hypoglycemia event that needed care from a medical professional. There was more fluid retention in the intensive than the standard group: 71% compared with 67%. About 28% of participants in the intensive group gained more than 10 kg (22 pounds) of body weight, compared with about 14% of participants in the standard group.
Return to "Contents"
Before launching the full-scale ACCORD trial, NHLBI funded a Vanguard Phase of ACCORD to determine the safety and feasibility of implementing the study protocol. The ACCORD Vanguard Phase treated and followed 1,174 participants for at least one year, with an average treatment duration of about 1 and 1/2 years. There were no safety concerns identified in the Vanguard Phase.
From the time ACCORD was launched, the DSMB monitored trial data regularly (about every 6 months). Data on numerous safety parameters were collected and reviewed, including serious adverse events, measures of liver and kidney function, fluid retention, heart failure occurrence, and overall mortality.
Participants visited study sites at least every 2 months for the intensive group participants, and every 4 months for the standard group participants. They learned about managing diabetes, self-monitoring their blood sugar levels, and preventing and treating hypoglycemia (low blood sugar) episodes. They were provided diabetes care and monitored for hypoglycemia and heart failure. At these visits their health status was reviewed, any side effects of medications were discussed, adjustments in dosages were made, and medical tests were taken. They were provided results of their tests and notified of any abnormal findings, so they could take those results to their private physicians.
Return to "Contents"
The ACCORD study is continuing to treat all participants for their diabetes; however, all participants are now being treated according to the ACCORD protocol for standard blood sugar control. Participants formerly randomized to the intensive blood sugar strategy have been transitioned to the ACCORD standard blood sugar strategy, aiming for an A1C of 7 to 7.9 percent.
ACCORD will continue to study the effects of intensive compared with standard treatment of blood pressure and the effects of fibrate plus statin compared with placebo plus statin. All participants were randomized at enrollment to one of these trials, but not both, in addition to being randomized in the blood-sugar trial.
ACCORD will also continue to compare outcomes in the participants who were in the intensive blood sugar treatment group with those in the standard treatment group to determine whether the differences found to date persist with time.
As is standard in clinical trials, ACCORD participants have the option to stop volunteering for the study at any time.
Return to "Contents"
As soon as the NHLBI made the decision to stop the intensive blood sugar lowering strategy, the Institute and the ACCORD Coordinating Center implemented an action plan to alert all ACCORD investigators, staff, and participants, as well as the public. All study participants were sent a letter explaining that those in the more intensive blood sugar treatment group would be transitioned to the ACCORD standard treatment strategy due to safety concerns.
NHLBI also alerted the Food and Drug Administration, NIH institutes that co-fund the trial, and the Centers for Disease Control and Prevention (CDC). The investigators alerted their institution's Institutional Review Boards. In addition, companies that donate medications, supplies, or equipment were notified.
Return to "Contents"
Patients with type 2 diabetes should not be alarmed by these findings of the ACCORD trial. Most individuals with type 2 diabetes who are similar to the participants in ACCORD are not being treated to the very low goal of the ACCORD intensive treatment group. This goal of less than 6 percent A1C is very hard to reach for most patients who have had diabetes for several years, often requiring multiple medications. Even the achieved level of A1C of about 6.4% is often difficult to achieve in such patients. The ACCORD participants were at particularly high risk of a CVD event or already had CVD at enrollment, and the results may not apply to patients who are not like the ACCORD participants.
Current guidelines for diabetes treatment recommend maintaining an A1C level of less than 7 percent for most people with type 2 diabetes, based on the proven benefits of blood sugar control in preventing eye, kidney, and nerve diseases. However, treatment should be tailored to the individual patient.
As always, patients with diabetes should not make any treatment changes without first consulting with their physician.
All people with diabetes are at increased risk for cardiovascular diseases, such as heart attack and stroke. Adopting a heart-healthy lifestyle has been proven to help prevent diabetes and improve diabetes control. Weight loss through a healthy diet and increased physical activity helps prevent or delay type 2 diabetes, and a heart-healthy lifestyle can improve diabetes control, lower blood pressure and cholesterol, and reduce the risk of CVD. Controlling elevated blood pressure and blood cholesterol have been proven to reduce the risk of CVD in persons with diabetes.
Return to "Contents"
These results also do not apply to patients with type 1 diabetes.
The risk of heart disease is about 10 times higher in people with type 1 diabetes than in those without diabetes. The NIH-funded Diabetes Control and Complications Trial (DCCT) in type 1 diabetes patients showed that intensive blood sugar control aimed at a near-normal A1C level reduced complications of the eyes, nerves, and kidneys. In 2005, the long-term follow-up of DCCT participants showed that intensive blood sugar control also lowers by about half the risk of heart disease and stroke in people with type 1 diabetes. These results strongly suggest that patients with type 1 diabetes should begin intensive blood sugar control as early as possible. This recommendation has not changed. For young people without complications, the guidelines recommend an A1C goal as close to normal as possible without significant hypoglycemia.
It should be noted that participants in the DCCT received intensive treatment early in the course of disease and before the onset of CVD. Researchers have not studied the risks and benefits of intensive therapy in type 1 diabetes patients with established CVD.
Return to "Contents"
The National Diabetes Education Program (NDEP), sponsored by the NIH and the CDC, promotes the American Diabetes Association's guidelines for diabetes care. Under these guidelines, the A1C goal for most patients with type 2 diabetes is less than 7 percent, with treatment tailored to individual needs. For example, the guidelines state: "Less stringent A1C goals may be appropriate for patients with a history of severe hypoglycemia, patients with limited life expectancies, children, individuals with comorbid conditions, and those with longstanding diabetes and minimal or stable microvascular complications."
Individualization of treatment can now take into account the important findings from the ACCORD trial. Based on the ACCORD findings, the increased risk of death may outweigh the benefits from medical-treatment strategies aiming to achieve blood sugar control at levels that are near those who do not have diabetes (A1C to less than 6 percent) among patients with established CVD or who are otherwise at especially high risk of CVD (two or more risk factors in addition to diabetes).
These results may not apply to patients with type 2 diabetes who are younger, who have less-established diabetes, or who have a lower risk of CVD than participants studied in the ACCORD trial. It is not known what effect more intensive therapy might have on CVD in younger people with type 2 diabetes or in patients with a lower risk of CVD than were studied in the ACCORD trial.
Other clinical trials in type 2 diabetes, such as the ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron-MR Controlled Evaluation) international trial and the VADT (Veterans Administration Diabetes Trial), have recently reported their findings. The findings from ACCORD will be taken together with findings from these and other studies to help determine appropriate evidence-based treatment approaches for people with diabetes.
Return to "Contents"
The total cost of the trial is about $300 million. The NHLBI sponsors the study and funds the sites through contracts and inter-agency agreements. Other components of the NIH, including the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute on Aging, and the National Eye Institute contribute funding. The CDC funds sub-studies within ACCORD on cost-effectiveness and health-related quality of life. The following companies provided study medications, equipment, or supplies: Abbott Laboratories, Amylin Pharmaceuticals, AstraZeneca Pharmaceuticals LP, Bayer HealthCare LLC, Closer Healthcare Inc., GlaxoSmithKline Pharmaceuticals, King Pharmaceuticals, Inc., Merck & Company, Inc., Novartis Pharmaceuticals, Inc., Novo Nordisk Pharmaceuticals, Inc., Omron Healthcare, Inc., and Sanofi-Aventis U.S.
Return to "Contents"
This finding reinforces the importance of large randomized, controlled clinical trials for determining optimal treatment for diabetes and other diseases and for determining effective interventions for preventing CVD. There are still unanswered questions about optimal blood sugar control, as well as many other questions about how best to treat diabetes to prevent its medical complications. The NIH will consult with experts in diabetes and its complications to determine the most compelling opportunities for further research to improve the treatment of diabetes.
The ACCORD Blood Pressure and Lipid trials are continuing because of the very important questions these two components are testing.
Return to "Contents"
For more information about ACCORD, visit www.accordtrial.org/ or visit www.nhlbi.nih.gov and click on "Clinical Trials," then ACCORD.
Return to "Contents"
For information about the prevention, diagnosis, and treatment of diabetes, see the websites of the National Diabetes Information Clearinghouse http://diabetes.niddk.nih.gov/ and the National Diabetes Education Program http://ndep.nih.gov/ . For information about the prevention, diagnosis, and treatment of cardiovascular disease, see www.nhlbi.nih.gov.
Return to "Contents"
Last updated: June 6, 2008
|
HOME
Department of Health
and Human Services
National
Institutes of Health
National Heart, Lung, and Blood
Institute