The New England Journal of Medicine APRIL 29, 2010

Effects of Combination Lipid Therapy in Type 2 Diabetes Mellitus

The ACCORD Study Group

Bottom Line

The addition of fenofibrate to simvastatin therapy in patients with type 2 diabetes at high cardiovascular risk did not significantly reduce the rate of major cardiovascular events compared to simvastatin alone.

Key Findings

1. The primary outcome (composite of nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death) occurred at an annual rate of 2.2% in the fenofibrate-simvastatin group compared to 2.4% in the placebo-simvastatin group (hazard ratio, 0.92; 95% CI, 0.79–1.08; P=0.32).
2. Fenofibrate therapy resulted in greater reduction of triglycerides (42 mg/dL vs. 16 mg/dL reduction; P<0.001) and a small, significant increase in HDL cholesterol (3.2 mg/dL vs. 2.3 mg/dL increase; P=0.01) compared to placebo.
3. A prespecified subgroup analysis showed a potential interaction by sex (P=0.01 for interaction), suggesting a benefit in men and possible harm in women, as well as a trend toward benefit in patients with high baseline triglycerides and low HDL cholesterol (P=0.057 for interaction).
4. No significant difference was observed in all-cause mortality (hazard ratio, 0.91; 95% CI, 0.75–1.10; P=0.33) between the two treatment groups.

Study Design

Design
RCT
Double-Blind
Sample
5,518
Patients
Duration
4.7 yr
Median
Setting
Multicenter, North America
Population Patients with type 2 diabetes and high cardiovascular risk currently receiving open-label simvastatin therapy.
Intervention Fenofibrate (145 mg/day) added to simvastatin therapy.
Comparator Placebo added to simvastatin therapy.
Outcome First occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke.

Study Limitations

The study did not achieve its original statistical power due to fewer enrolled participants and shorter follow-up than initially planned.
The trial was an add-on design where all patients were on open-label simvastatin, which may have limited the ability to isolate the specific benefit of fibrate monotherapy.
Prespecified subgroup analyses regarding gender and dyslipidemia profiles are hypothesis-generating and may be subject to chance given the lack of significance in the overall primary endpoint.
The trial utilized a specific fenofibrate dose (145 mg/day) which may not fully represent potential effects at different dosing regimens.

Clinical Significance

The results of this trial do not support the routine addition of fenofibrate to statin therapy for all patients with type 2 diabetes to reduce cardiovascular risk. It highlights the importance of statin monotherapy as the cornerstone of lipid management in this population.

Historical Context

Prior to ACCORD-Lipid, trials such as VA-HIT suggested potential cardiovascular benefits from fibrates in patients with type 2 diabetes, while the FIELD study showed neutral results. ACCORD-Lipid was specifically designed to test the additive value of fibrates in a high-risk population already established on modern statin therapy.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Fenofibrate is a PPAR-alpha agonist. Based on its molecular mechanism, how was it expected to modify the lipid profile of patients with type 2 diabetes, and why was this expected to improve cardiovascular outcomes?

Key Response

Peroxisome proliferator-activated receptor alpha (PPAR-alpha) activation increases the expression of lipoprotein lipase (LPL) and apolipoproteins A-I and A-II. This results in decreased triglycerides and increased HDL cholesterol. Historically, the 'lipid hypothesis' suggested that improving these surrogate markers would automatically reduce major adverse cardiovascular events (MACE), a premise that ACCORD-Lipid specifically tested and found lacking in a general diabetic population already treated with statins.

Resident
Resident

A 55-year-old male with type 2 diabetes and a BMI of 32 is currently on simvastatin 40mg. His LDL is 70 mg/dL, but his triglycerides are 250 mg/dL and his HDL is 30 mg/dL. Based on the ACCORD-Lipid subgroup analysis, would you add fenofibrate to his regimen for cardiovascular risk reduction?

Key Response

While the primary outcome of ACCORD-Lipid was negative, a pre-specified subgroup analysis suggested a potential benefit in patients with both high triglycerides (>204 mg/dL) and low HDL (<34 mg/dL). In this specific 'dyslipidemic' phenotype, the trial suggested a trend toward reduced MACE. However, for the general diabetic population, the trial demonstrated that adding fenofibrate to a statin does not provide incremental benefit, emphasizing that statins remain the first-line therapy.

Fellow
Fellow

Analyze the 'residual risk' paradox presented by ACCORD-Lipid: why did the significant improvement in non-HDL cholesterol and triglycerides in the combination group fail to translate into a reduction in the primary composite endpoint, unlike the results seen in trials like REDUCE-IT?

Key Response

ACCORD-Lipid showed that lowering triglycerides via the fibrate pathway (PPAR-alpha) does not necessarily yield the same clinical benefit as other TG-lowering pathways (e.g., purified icosapent ethyl in REDUCE-IT). This suggests that the 'residual risk' in diabetic patients is not purely a function of triglyceride levels, but perhaps the specific qualitative nature of the lipoproteins or the pleiotropic effects of the medications used. This nuanced interpretation helps fellows understand that surrogate marker improvement does not always equal clinical efficacy.

Attending
Attending

The ACCORD-Lipid trial reported a significant interaction between sex and treatment effect (p=0.01), suggesting potential harm in women receiving fenofibrate. How does this finding influence your approach to 'off-label' use of fibrates for residual risk in female patients?

Key Response

In the trial, men showed a trend toward benefit (HR 0.82) while women showed a trend toward harm (HR 1.38). This interaction highlights the danger of assuming uniform benefit across genders. An attending must use this to teach that clinical guidelines regarding 'subgroup benefits' (like the high TG/low HDL group) must be tempered by other demographic data; specifically, one should be extremely cautious or avoid adding fenofibrate to statins in female patients unless treating severe hypertriglyceridemia to prevent pancreatitis.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

ACCORD-Lipid utilized a 2x2 factorial design, randomizing patients to both glycemic and lipid interventions. Critically evaluate how the early termination of the intensive glycemia arm due to increased mortality might have introduced 'competing risk' or confounded the long-term observation of the lipid intervention's effect.

Key Response

The 2x2 design assumes no interaction between interventions. When the intensive glycemia arm was halted, many patients were transitioned to standard care, potentially altering their cardiovascular risk profile mid-trial. A PhD-level critique would focus on whether the excess deaths in the glycemia arm (a competing risk) or the changes in glucose management protocols masked a late-emerging benefit of fenofibrate, and how the statistical power for the lipid interaction was affected by these protocol changes.

Journal Editor
Journal Editor

If you were the primary reviewer for the ACCORD-Lipid manuscript, how would you address the discrepancy between the study's power calculations and the actual event rate observed in the control group?

Key Response

A common 'threat to validity' in large CV trials is an overestimation of the expected event rate. If the simvastatin-only group had fewer events than predicted (due to better-than-expected background care), the study may have been underpowered to detect a modest but clinically meaningful difference. An editor would flag this as a limitation, questioning whether the 'negative' result is definitive or merely a failure to reach statistical significance due to an overly optimistic effect size assumption.

Guideline Committee
Guideline Committee

Current ADA and AHA/ACC guidelines generally reserve fibrates for severe hypertriglyceridemia (TG >500 mg/dL) to prevent pancreatitis rather than for MACE reduction. How does ACCORD-Lipid support or refute the inclusion of a 'Class IIb' recommendation for combination therapy in the specific TG >204 and HDL <34 subgroup?

Key Response

The ADA Standards of Care currently allow for consideration of fenofibrate in the high TG/low HDL subgroup (Level C), largely based on the ACCORD-Lipid subgroup analysis and meta-analyses of FIELD and HHS. However, the committee must weigh this against the primary negative result and the potential for harm in women. ACCORD-Lipid serves as the primary evidence for 'de-escalating' the role of fibrates in general CV prevention, shifting the guideline focus toward icosapent ethyl or SGLT2 inhibitors for residual risk management in T2DM.

Clinical Landscape

Noteworthy Related Trials

2005

FIELD Study

n = 9,795 · Lancet

Tested

Fenofibrate 200 mg daily

Population

Patients with Type 2 Diabetes

Comparator

Placebo

Endpoint

CHD events (death or nonfatal MI)

Key result: Fenofibrate did not significantly reduce the primary outcome of CHD events, though it reduced the rate of nonfatal MI.
2011

AIM-HIGH Trial

n = 3,414 · NEJM

Tested

Extended-release niacin

Population

Patients with cardiovascular disease and low HDL-C levels

Comparator

Placebo on top of simvastatin

Endpoint

Composite of death from CHD, nonfatal MI, ischemic stroke, hospitalization for ACS, or symptom-driven revascularization

Key result: The addition of niacin to statin therapy did not provide significant incremental clinical benefit regarding cardiovascular outcomes.
2014

HPS2-THRIVE Trial

n = 25,673 · NEJM

Tested

Extended-release niacin/laropiprant

Population

Patients with established vascular disease

Comparator

Placebo on top of background statin therapy

Endpoint

Major vascular events

Key result: The addition of niacin/laropiprant to statin therapy did not significantly reduce the risk of major vascular events but increased serious side effects.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis