The Lancet Neurology SEPTEMBER 28, 2011

Effects of intensive glucose lowering on brain structure and function in people with type 2 diabetes (ACCORD MIND)

Launer LJ, Miller ME, Williamson JD, et al.

Bottom Line

The ACCORD-MIND study demonstrated that intensive glycemic control in patients with type 2 diabetes at high cardiovascular risk did not improve cognitive function, despite a significant, albeit clinically uncertain, preservation of total brain volume compared to standard glycemic control.

Key Findings

1. Intensive glycemic control (target HbA1c <6.0%) showed no significant difference in the primary cognitive endpoint—the 40-month Digit Symbol Substitution Test (DSST) score—compared to standard control (HbA1c 7.0–7.9%) (mean difference 0.32, 95% CI –0.28 to 0.91; p=0.2997).

2. Patients in the intensive therapy group exhibited a greater mean total brain volume (TBV) compared to the standard therapy group at 40 months (difference 4.62 cm3, 95% CI 2.0 to 7.3; p=0.0007), suggesting a potential structural benefit despite the lack of cognitive improvement.

3. The study findings do not support the use of intensive glycemic therapy to preserve cognitive function in patients with long-standing type 2 diabetes at high cardiovascular risk.

Study Design

Design

RCT

Open-Label

Sample

2,977

Patients

Duration

40 mo

Median

Setting

Multicenter, North America

Population Patients aged 55–80 years with type 2 diabetes, high HbA1c (>7.5%), and high risk of cardiovascular events.

Intervention Intensive glycemic control targeting HbA1c <6.0%.

Comparator Standard glycemic control targeting HbA1c 7.0–7.9%.

Outcome Performance on the Digit Symbol Substitution Test (DSST) at 40 months.

Study Limitations

• The study was an open-label substudy, which may have introduced assessment or reporting bias.

• The intensive glycemic control intervention was stopped prematurely (mean 3.5 years) due to increased mortality, which may have limited the ability to observe long-term cognitive effects.

• The DSST, while a validated tool for processing speed, may not capture all domains of cognitive decline affected by diabetes.

• The observed difference in total brain volume, while statistically significant, had uncertain clinical relevance given the parallel lack of benefit in cognitive test scores.

Clinical Significance

The results of ACCORD-MIND indicate that aggressive glycemic targets do not protect against cognitive decline in older adults with type 2 diabetes and high cardiovascular risk, reinforcing the need to prioritize safe glycemic targets that avoid both severe hypoglycemia and excessive cardiovascular mortality.

Historical Context

The ACCORD-MIND substudy was initiated to address the hypothesis that the chronic hyperglycemia associated with type 2 diabetes contributes to accelerated brain atrophy and cognitive impairment. Given that the main ACCORD trial famously reported increased mortality in the intensive glucose-lowering arm, ACCORD-MIND provided crucial data suggesting that the potential structural brain benefits of intensive therapy were insufficient to justify the clinical risks, leading to a shift in clinical practice toward more moderate glycemic goals in similar patient populations.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Chronic hyperglycemia is known to affect the central nervous system; what are the primary pathophysiological mechanisms by which type 2 diabetes leads to cognitive impairment and structural brain changes?

Key Response

Hyperglycemia contributes to cognitive decline through multiple pathways: the formation of advanced glycation end-products (AGEs) causing microvascular damage, increased oxidative stress leading to neuronal apoptosis, and impaired insulin signaling in the brain which can promote amyloid-beta accumulation. Understanding these basic mechanisms helps explain why researchers hypothesized that intensive glucose control might preserve brain health.

Resident

In a patient with long-standing type 2 diabetes and high cardiovascular risk, how does the evidence from ACCORD MIND influence your decision to target an HbA1c below 6.0% for the purpose of neuroprotection?

Key Response

The ACCORD MIND study demonstrated that intensive glycemic control (HbA1c < 6.0%) did not result in better cognitive function compared to standard therapy. Given the increased mortality risk associated with intensive control in the primary ACCORD trial and the lack of cognitive benefit found here, residents should avoid aggressive targets in high-risk patients solely for the purpose of preventing dementia or cognitive decline.

Fellow

How do you reconcile the discrepancy found in ACCORD MIND between the preservation of total brain volume (TBV) in the intensive group and the lack of a corresponding improvement in cognitive performance?

Key Response

While the intensive group showed a statistically significant preservation of TBV, this did not translate to functional cognitive gains. This suggests that 'brain volume' as a surrogate marker may not capture qualitative changes in neuronal health, or that the protective effect on volume was too small to manifest as a functional difference over the 40-month study period. It also raises questions about whether the intensive group's results were influenced by the osmotic effects of lower glucose or differences in cerebral blood flow.

Attending

The ACCORD MIND results highlight a 'surrogate-outcome gap' in metabolic trials. How should this study inform our clinical teaching regarding the use of imaging findings as evidence for therapeutic efficacy in neuro-diabetology?

Key Response

Attending physicians should emphasize that structural preservation (like TBV) does not always equate to clinical benefit (like improved memory or processing speed). The trial serves as a teaching point on why clinical outcomes must remain the gold standard for practice changes, as the intensive treatment actually posed higher safety risks (hypoglycemia, mortality) despite the 'prettier' MRI scans.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

Evaluate the sensitivity and specificity of the Digital Symbol Substitution Test (DSST) as the primary outcome for this trial; could the 'null' functional result be a consequence of the chosen psychometric instruments or the follow-up duration?

Key Response

Research methodology experts might argue that while the DSST is sensitive to general cognitive slowing, it may not have been specific enough to capture early executive dysfunction or memory changes specific to diabetic metabolic insult. Furthermore, since cognitive decline in T2DM is a slow process, a 40-month follow-up might be insufficient to demonstrate a divergence in trajectories between treatment arms, even if a structural difference was already emerging.

Journal Editor

Given that the primary cognitive endpoint was not met, should the findings regarding total brain volume be presented as a major discovery or relegated to a hypothesis-generating secondary analysis in the manuscript?

Key Response

A critical reviewer or editor would flag the risk of 'spin' if the authors over-emphasized the TBV findings. Since the primary functional endpoint failed, the structural findings must be reported with caution, acknowledging that they are secondary and lack a clear clinical correlate, which limits the study's impact on immediate clinical practice.

Guideline Committee

Based on the ACCORD MIND data, should current guidelines (such as those from the ADA or EASD) be updated to include intensive glycemic control as a strategy to reduce the risk of cognitive decline in elderly T2DM patients?

Key Response

Current guidelines, including those from the ADA (Standards of Care), recommend less stringent A1c goals (e.g., <8.0%) for older adults with complex comorbidities or high hypoglycemia risk. Because ACCORD MIND failed to show cognitive benefit and the main ACCORD trial showed increased mortality, guidelines should continue to recommend against intensive targets (<6.0%) for the prevention of cognitive impairment in this high-risk population.

Clinical Landscape

Noteworthy Related Trials

2008

ACCORD Trial

n = 10,251 · NEJM

Tested

Intensive glycemic control (HbA1c <6.0%)

Population

Patients with T2DM at high risk for cardiovascular events

Comparator

Standard glycemic control (HbA1c 7.0-7.9%)

Endpoint

Composite of nonfatal myocardial infarction, nonfatal stroke, or death from cardiovascular causes

Key result: Intensive therapy reduced the incidence of major cardiovascular events but increased mortality, leading to premature termination of the intensive arm.

2008

ADVANCE Trial

n = 11,140 · NEJM

Tested

Intensive glucose control (HbA1c target 6.5%)

Population

Patients with T2DM

Comparator

Standard glucose control

Endpoint

Composite of major macrovascular and microvascular events

Key result: Intensive glucose control significantly reduced the incidence of major microvascular events, primarily driven by a reduction in nephropathy, without a significant effect on macrovascular events or mortality.

2009

VADT Trial

n = 1,791 · NEJM

Tested

Intensive glycemic control

Population

Veterans with poorly controlled T2DM

Comparator

Standard glycemic control

Endpoint

Time to first occurrence of major cardiovascular event

Key result: Intensive glucose lowering did not significantly reduce the rate of major cardiovascular events or cardiovascular death in patients with long-standing T2DM.

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