Lower blood pressure targets and combination lipid therapy failed to
reduce the overall risk for cardiovascular events in patients with type 2
diabetes compared with standard treatment strategies, according to the results
of two new Action to Control Cardiovascular Risk in Diabetes trials, leading
many health care professionals to reason that intensive therapy may not be
better in diabetes.
The landmark ACCORD trial is one of the largest studies ever conducted
in adults with type 2 diabetes who are at especially high risk for CV events.
The multicenter, clinical trial tested three potential strategies to reduce the
risk for major CV events: intensive glycemic control, intensive BP control and
intensive lipid control. In 2008, results of the ACCORD glycemia trial revealed
an increased risk for death with intensive glycemic control compared with
standard control. Now, new data failed to show benefit of intensive control in
the BP and lipid trials.
In the ACCORD BP trial, targeting therapy to achieve a BP of less than
120 mm Hg had no significant effect on the combined rates of nonfatal
myocardial infarction, nonfatal stroke or CV death compared with conventional
BP control aimed at a systolic BP of less than 140 mm Hg. Similarly, intensive
lipid therapy with fenofibrate (TriCor, Abbott) plus simvastatin (Zocor, Merck)
failed to reduce the rate of fatal and nonfatal CV events compared with
simvastatin monotherapy in the ACCORD lipid trial. Results were presented in
March at the American College of Cardiology 59th Annual Scientific Session and
simultaneously published online in The New England Journal of
Henry N. Ginsberg, MD, presented the results of the ACCORD lipid trial.
Photo courtesy of:
Despite the negative findings of both trials, We need trials like
ACCORD because it illustrates that our previous beliefs about preventing CV
outcomes in patients with diabetes may not have been founded on the best
evidence, Bernard Zinman, MD, director of the Leadership Sinai
Centre for Diabetes and professor of medicine at the University of Toronto,
told Endocrine Today.
The data will provide evidence to help guide treatment recommendations
for adults with type 2 diabetes who are at high risk for CVD, according to
Susan B. Shurin, MD, acting director of the National Institute of
Healths National Heart, Lung, and Blood Institute, which funded the
This information provides guidance to avoid unnecessarily
increasing treatment that provides limited benefit and potentially increases
the risk of adverse events, Shurin said in a press release.
Endocrine Today interviewed ACCORD researchers and experts
in endocrinology and cardiology to provide insight on whether intensive therapy
is better when considering CV outcomes in diabetes.
The ACCORD trial enrolled 10,251 high-risk patients aged 40 to 79 years
with type 2 diabetes and CVD or related risk factors. All participants were
randomly assigned to intensive or standard glycemic control for the ACCORD
glycemia trial. In addition, 4,733 patients were also randomly assigned
in a two-by-two factorial design to a strategy to target intensive
(≤120 mm Hg) or standard (≤140 mm Hg) systolic BP control, using a
variety of BP-lowering medications, for the ACCORD BP trial.
Intensive BP treatment did not reduce the overall risk for CV events
compared with standard treatment. Systolic BP levels averaged 119.3 mm Hg in
the intensive-therapy group compared with 133.5 mm Hg in the standard-therapy
group during the course of the trial. After five years, the researchers
observed no significant differences in the combined rate of nonfatal MI,
nonfatal stroke or CV death in the intensive-therapy group (1.87%) vs. the
standard-therapy group (2.09%; HR=0.88; 95% CI, 0.73-1.06). The rate of death
from any cause was 1.28% with intensive therapy vs. 1.19% with standard therapy
(HR=1.07; 95% CI, 0.85-1.35).
Intensive therapy did not show a benefit in getting the systolic
BP down to 120 mm Hg, even though there was an effect on a secondary
outcome, William C. Cushman, MD, chief of the preventive medicine
section, Veterans Affairs Medical Center, Memphis, Tenn., said during a press
conference at the ACC meeting.
At five years, the risk for stroke was significantly lower in the
intensive-therapy group (36 strokes vs. 62 strokes). The annual rate of stroke
was 0.32% in the intensive-therapy group vs. 0.53% in the standard-therapy
group (HR=0.59; 95% CI, 0.39-0.89).
Cushman said the reduced stroke risk is consistent with that reported in
previously published studies. Since it was not a primary outcome, from a
statistical perspective, it could have happened by chance. But we do think it
is biologically plausible and shouldnt be ignored.
Franz H. Messerli, MD, director of the hypertension program at
St. LukesRoosevelt Hospital and professor of clinical medicine,
Columbia University College of Physicians and Surgeons, New York, said he was
surprised by the lack of emphasis given to the significant reduction in stroke
rate with intensive BP treatment.
Stroke is the most devastating complication of hypertension,
Messerli said. Poll any physician audience if they would rather have a
stroke or MI nearly everyone will choose MI over stroke.
Further, more serious adverse events were reported with intensive
therapy vs. standard therapy (3.3% vs. 1.3%; P≤.001), including
hypotension and hyperkalemia. In addition, some laboratory measures of renal
function were worse in the intensive-therapy group, but the researchers
reported no difference in rates of renal failure.
It seems that this is almost like déjà vu all over
again, Elijah Saunders, MD, head of the division of hypertension
at University of Maryland School of Medicine, said during a panel discussion at
the ACC meeting. Most of us have been teaching for many years that the
lower the BP, the better the outcomes therefore, seek to get the BP
The problem is the differences between what we see and observe
epidemiologically. The data we have suggest that we still cannot recommend
bringing BP down to less than 120 mm Hg as a way of reducing risk. We need to
look further and see if we can find some differences in subgroups that may
explain this non-positive result.
Peter M. Nilsson, MD, PhD, of the department of clinical science
medicine at University Hospital in Malmö, Sweden, wrote an editorial that
was published in The New England Journal of Medicine. The
main conclusion to draw from this study must be that a systolic BP target
≤120 mm Hg in patients with type 2 diabetes is not justified by the
evidence, he wrote.
The results do not resolve the issue of the optimal BP target for
patients with diabetes. The Seventh Report of the Joint National Committee on
Prevention, Detection, Evaluation and Treatment of High Blood Pressure
recommends a goal of less than 130 mm Hg. However, the ACCORD researchers said
the goal of the study was not to test the BP goal of 130 mm Hg recommended in
the JNC 7.
The ACCORD lipid trial was the first large clinical trial to compare the
CV effects of combination therapy with a fibrate plus a statin with statin
monotherapy in patients with diabetes, according to Henry N. Ginsberg,
MD, director of the Irving Institute for Clinical and Translational
Research at Columbia University College of Physicians and Surgeons.
Ginsberg and colleagues evaluated a subgroup of 5,518 patients enrolled
in ACCORD who underwent randomization, again in a two-by-two factorial design,
to receive simvastatin plus fenofibrate or simvastatin plus placebo.
Overall, the results of the ACCORD lipid trial do not support use
of combination therapy with fenofibrate and simvastatin to reduce CVD in
high-risk patients with type 2 diabetes, Ginsberg said in a press
During a mean follow-up of 4.7 years, the researchers found that
combination therapy was safe but did not significantly reduce the primary
outcome of first occurrence of CVD death, nonfatal MI or nonfatal stroke
compared with simvastatin alone (annual rate, 2.2% vs. 2.4%; HR=0.92; 95% CI,
0.79-1.08). Combination therapy also had no significant effect on secondary
outcomes, including individual rates of CVD, nonfatal MI, nonfatal stroke,
total mortality, fatal stroke, revascularization or hospitalization for
congestive heart failure. The annual rate of death was 1.5% in the fenofibrate
plus simvastatin group compared with 1.6% in the simvastatin alone group
(HR=0.91; 95% CI, 0.75-1.10).
The baseline fenofibrate dose was 160 mg per day, but it was later
adjusted according to the estimated glomerular filtration rate with use of the
abbreviated Modification of Diet in Renal Disease equation because of a rise in
serum creatinine levels in some patients receiving this dose.
Pre-specified subgroup analyses also suggested heterogeneity in
treatment effect according to sex, with a benefit for men and suggestion of
harm for women (P=.01). The researchers also noted that patients who had
both triglyceride levels in the top one-third and HDL levels in the bottom
one-third at baseline may have benefited from the combination therapy, compared
with other participants. Although a similar effect has been reported in other
studies, researchers said more research is needed on the effects in this
subgroup, which comprised 17% of the ACCORD participants.
At the end of the study, mean LDL levels decreased from 100 mg/dL to
81.1 mg/dL in the combination therapy group and from 101.1 mg/dL to 80 mg/dL in
the simvastatin monotherapy group. Mean HDL levels increased from 38 mg/dL to
41.2 mg/dL in the combination therapy group and from 38.2 mg/dL to 40.5 mg/dL
in the simvastatin monotherapy group. Triglycerides decreased from a median
value of 164 mg/dL to 122 mg/dL in the combination therapy group and from a
median value of 160 mg/dL to 144 mg/dL in the placebo group.
Although our analysis suggests that certain patients may benefit
from combination therapy, this study provides important information that should
spare many people with diabetes unneeded therapy with fibrates, Ginsberg
said in a press release.
However, the lack of benefit from fibrates should not obscure the
proven value of statins in preventing CVD, which is well established from
earlier studies, he said.
A potential criticism of the trial is that it did not examine a very
dyslipidemic population. The decision was made that we should have a
study group from which the results could be widely extrapolated in a general
diabetic population, Ginsberg said at a press conference.
Previously published studies indicated that fibrate therapy reduced the
rate of CHD events in patients with diabetes in the Veterans Affairs HDL
Intervention Trial but not in the Fenofibrate Intervention and Event Lowering
in Diabetes (FIELD) trial. A post-hoc analysis of the FIELD trial suggested
benefit for patients with elevated triglycerides and low HDL. However, previous
fibrate studies have not addressed the role of fibrates in patients receiving
Messerli said he is reluctant to use fenofibrate in a patient with
The patient with diabetes is usually on numerous medications to
start with, and then to add a drug such as a fenofibrate, which is fairly
expense and the benefits of which are not that well documented, is a bit of a
stretch for the practicing physician, he said in an interview.
Zinman said these data really put fibrates at a lower priority in
the context of anti-lipid therapies. Statins reign high they are tried,
true and proven, and almost every patient with type 2 diabetes should be taking
On March 16, the FDA informed health care professionals that it will
review the primary data from the ACCORD lipid trial.
Currently, the FDA has made no new conclusions or recommendations
regarding the combination use of simvastatin or other statins and fenofibrate.
The agency will conduct a thorough review of the primary ACCORD data as soon as
they become available. The FDA will determine how the ACCORD data relate to the
approved indication and labeling for fenofibric acid (Trilipix, Abbott), which
was approved by the FDA in 2008 as an adjunct to diet in combination with a
statin to reduce triglycerides and increase HDL in patients with mixed
dyslipidemia and CHD to achieve LDL goals. In the meantime, patients should not
stop taking their current medications, according to the FDA.
Although the new ACCORD data do not resolve any issues, they make
the picture of diabetes management more complete, Nilsson wrote in
Owing to the factorial design of the overall ACCORD study and the
inclusion and exclusion of criteria that were applied, the studys
statistical power was reduced and the event rate was lower than expected,
he wrote, adding that it is possible that a larger trial or a trial conducted
in a higher-risk population may show a significant benefit.
Cushman said the ACCORD BP and lipid trials have a different implication
than the glycemia trial, which found a 22% higher relative risk for death among
patients assigned to intensive glucose lowering (HbA1c ≤6%) compared with
standard glucose control (7% to 7.9%).
Unlike the glycemia trial, neither the BP nor lipid trials suggest
serious harm from the interventions tested, whereas in the glycemia trial, the
intensive group actually had a significantly higher mortality, and that part of
the study was stopped early, he said.
Other experts said the hope is that these data do not result in health
care professionals drawing the wrong conclusions.
My concern is that some may jump to the conclusion that it
doesnt matter if hyperglycemia or CVD risk factors are not assiduously
controlled in patients with diabetes and that is incorrect, Zinman
According to Denise G. Simons-Morton, MD, PhD, director of the
National Heart, Lung, and Blood Institute division for the application of
research discoveries, and former project officer for ACCORD, the results
indicate no need for intensive treatment over standard practice.
The treatment strategies used in the ACCORD standard control
groups have previously been shown to be effective, Simons-Morton said in
a press release. So the findings in no way detract from the important
point that controlling blood pressure and LDL levels reduce CV risk not
only in patients with diabetes, but in all patients with elevated levels.
Most experts said the standards of diabetes care and CV risk factor
control have improved considerably over the past decade.
Such progress is reassuring, but now we learn from the completed
ACCORD study that flexible goals should probably be applied to the control of
hyperglycemia, BP and dyslipidemia in patients with type 2 diabetes, taking
into account individual clinical factors of importance, Nilsson wrote in
A period of three landmark studies (UKPDS, ADVANCE and ACCORD) has
now come to an end. New trials should be designed on the basis of our new
understanding, he said.
The ACCORD findings will also be of importance in determining whether it
will be useful to proceed with the National Institutes of Health-sponsored
Systolic Blood Pressure Intervention Trial (SPRINT), which will evaluate the
same BP targets (≤120 mm Hg and ≤140 mm Hg) in patients without diabetes.
The nine-year, $114 million study will be conducted at more than 80 clinical
sites in the United States. The multicenter, randomized, clinical trial will
enroll approximately 7,500 participants aged 55 years and older. Enrollment is
expected to begin in the fall, and the study is projected to end in 2018,
according to the NIH.
Franz H. Messerli
I would like to know whether the same ACCORD results hold true in
the non-diabetic population, Messerli said.
If SPRINT shows that there is benefit of a BP target ≤120 mm
Hg, then we may be looking back to the ACCORD BP trial in a few years,
One hypothesis, he added, is that it may take more than five years to
see the benefit of intensive interventions.
We have gotten used to having five-year studies show dramatic
benefits; it is possible that it would take longer to see these effects, but we
have no way of proving that yet, Cushman said.
Researchers for both trials will follow participants after the study,
but they will no longer be randomly assigned to any treatment.
Whatever the answer is, it is what we need to know in order to
take good care of our patients with diabetes, Cushman said. by
What is the optimal blood
pressure target for patients with diabetes?
For more information:
- Cushman WC. LBCT I. Presented at: American College of Cardiology
59th Annual Scientific Session; March 13-16, 2010; Atlanta.
- Ginsberg HN. LBCT I. Presented at: American College of Cardiology
59th Annual Scientific Session; March 13-16, 2010; Atlanta.
- Nilsson PM. N Engl J Med.
- The ACCORD Study Group. N Engl J Med.