Global Socioeconomic Context and Brain Ageing in Epilepsy: an ENIGMA-Epilepsy study

This ENIGMA-Epilepsy study of over 2,000 individuals reveals that while epilepsy is associated with accelerated brain structural ageing, this effect is significantly exacerbated in countries with lower GDP per capita and higher income inequality, highlighting the critical influence of global socioeconomic context on neuroanatomical outcomes.

The Gist

The Big Picture: The Brain's "Biological Clock"

Imagine your brain has a built-in "biological clock" that tells you how old it feels based on its structure, rather than just how many birthdays you've had. Scientists can look at brain scans (MRIs) and use a computer program to guess your brain's age.

• Chronological Age: How old you are on your ID card.
• Brain Age: How old your brain looks on the scan.
• The Difference (BrainPAD): If your brain looks 5 years older than your ID says, that's a "gap." Usually, a bigger gap means the brain is under more stress or aging faster than it should.

The Study: A Global Check-Up

The researchers gathered data from over 3,000 people (2,109 with epilepsy and 1,041 without) across 12 different countries on five continents. They wanted to answer two main questions:

1. Does having epilepsy make your brain look older than it should?
2. Does the money and wealth of the country you live in change how much older your brain looks?

What They Found

1. The Epilepsy Effect: A "Wear and Tear" Gap
People with epilepsy generally had brains that looked older than healthy people of the same age.

• The Analogy: Think of a car. A healthy car with 30,000 miles on it looks like a 30,000-mile car. A car with epilepsy might look like it has 35,000 miles on it, even if it's the same age.
• The Details: On average, the brains of people with epilepsy looked about 4.2 years older than they should.
  • The gap was biggest for people with specific types of epilepsy involving scarring in the memory center of the brain (Mesial Temporal Sclerosis), where the brain looked up to 6 years older.
  • The gap was smaller (but still present) for other types of epilepsy.
  • Men vs. Women: Men's brains looked about 1 year older than women's brains, regardless of whether they had epilepsy or not.

2. The Money Factor: The "Safety Net" Analogy
This is the most interesting part. The researchers looked at two economic numbers for each country:

• GDP per capita: How much money the average person makes (National Wealth).
• Gini Index: How unevenly that money is shared (Income Inequality).

The Surprise: The country's wealth by itself didn't change how old a healthy person's brain looked. However, it did change how much epilepsy aged the brain.

• The Analogy: Imagine epilepsy is a heavy backpack.
  • In wealthy countries with low inequality (a strong safety net), the backpack is heavy, but the ground is solid. The person can carry it, and their brain doesn't age too much faster.
  • In poorer countries or those with high inequality (a weak safety net), the ground is rocky and uneven. Carrying the same heavy backpack causes much more damage. The brain ages significantly faster.
• The Result: The "gap" between a person's real age and their brain age was largest in countries with less money and higher inequality. The epilepsy hit harder in these environments.

Why Does This Happen?

The study suggests that while epilepsy itself wears down the brain, the environment matters. Living in a place with less economic support might mean:

• Less access to consistent healthcare.
• More stress.
• Fewer resources to manage the condition.

This extra stress seems to pile on top of the epilepsy, making the brain age even faster.

Important Limitations (What the Study Didn't Say)

• It's a Snapshot, Not a Movie: This study looked at one moment in time. It can't prove that poverty caused the faster aging, only that they are linked.
• The "Training" Bias: The computer program used to guess brain age was mostly trained on data from wealthy Western countries (like the UK and USA). The researchers admit this might make the results slightly less accurate for people from different backgrounds, though they tried to account for this.
• No Individual Details: They used country-level money stats, not individual family income. They also didn't have data on specific medications or head injuries.

The Bottom Line

Having epilepsy makes your brain look a bit older than it should. But, where you live matters. If you live in a country with less money and more inequality, the "aging effect" of epilepsy is stronger. It's as if the brain has to work harder to cope with the disease when the surrounding world offers fewer resources.

This study tells us that to understand brain health, we can't just look at the disease; we have to look at the economic world the patient lives in, too.

Technical Summary

1. Problem Statement

Chronic epilepsy is associated with accelerated brain atrophy and structural differences that suggest an "older-appearing" brain compared to chronological age. While machine learning models can quantify this discrepancy (termed Brain Predicted Age Difference or BrainPAD), the extent to which these differences are driven by clinical epilepsy characteristics versus broader socioeconomic contexts remains unclear. Previous studies have established links between low socioeconomic status and adverse health outcomes, but it is unknown whether macroeconomic factors (e.g., national wealth, income inequality) modulate the structural brain ageing associated with epilepsy. This study aims to investigate whether country-level socioeconomic indicators modify the relationship between epilepsy and brain ageing.

2. Methodology

• Study Design & Cohort:

  • Data Source: A retrospective analysis of data from the ENIGMA-Epilepsy consortium, comprising 26 epilepsy centers across 12 countries (Americas, Australia, Europe, Asia, Africa).
  • Participants: 2,109 individuals with epilepsy and 1,041 healthy controls (HC).
  • Demographics: Median age of 35 years (range 17–83); 57% female.
  • Epilepsy Subtypes: Included Temporal Lobe Epilepsy (TLE) with and without Mesial Temporal Sclerosis (MTS), Genetic Generalised Epilepsy (GGE), and non-lesional extratemporal epilepsy.
  • Exclusion Criteria: History of neurosurgery, known structural lesions (for non-lesional groups), and neurological disorders in controls.

• Socioeconomic Variables:

  • GDP per capita: Obtained from World Bank data (2023 estimates).
  • Income Inequality: Measured via the Gini coefficient (most recent available data, 2014–2022).

• MRI Acquisition & Processing:

  • Imaging: 3T structural T1-weighted MRI scans acquired across various manufacturers (Siemens, GE, Philips).
  • Preprocessing: Images were processed using FreeSurfer to derive regional cortical thickness and subcortical/whole-brain volume estimates.
  • Quality Control: Image quality was assessed using the Euler number (a topological measure of surface defects); higher (less negative) values indicate better quality.

• Brain Age Prediction:

  • Model: A pre-trained machine learning model (PyBrainAge) trained on 29,175 healthy subjects (mostly from the UK Biobank) using regional cortical thickness and volume features.
  • Metric: BrainPAD = (MRI-predicted brain age) – (Chronological age). Positive values indicate accelerated ageing.

• Statistical Analysis:

  • Models: Linear mixed-effects models with site as a random intercept to account for clustering.
  • Covariates: Age and sex were included in all models.
  • Key Analyses:
    1. Model 1: Epilepsy diagnosis vs. BrainPAD.
    2. Models 2 & 3: Interaction terms between diagnosis and GDP per capita / Gini index.
    3. Model 4: Epilepsy subtype-specific differences.
    4. Model 5: LASSO regression to compare the relative contributions of age of onset vs. epilepsy duration.
    5. Model 6: Adjustment for MRI quality (Euler number).

3. Key Contributions

• Global Scope: This is the largest international study to date investigating the intersection of epilepsy, brain ageing, and macroeconomic factors, utilizing a diverse cohort from 12 countries.
• Socioeconomic Interaction: It is the first to demonstrate that the impact of epilepsy on brain structure is not uniform but is significantly moderated by national socioeconomic conditions.
• Methodological Rigor: Utilization of a harmonized processing pipeline (FreeSurfer) and a large, pre-trained brain age model across a multi-site, multi-scanner dataset.
• Clinical Insight: Differentiation of brain ageing effects across specific epilepsy subtypes (lesional vs. non-lesional) and identification of age of onset as a stronger predictor than disease duration.

4. Key Results

• Overall BrainPAD: Individuals with epilepsy had significantly higher BrainPAD than controls (+4.2 years, SE 0.4, t=10.6).
• Subtype Variations:
  • Highest Acceleration: TLE with Mesial Temporal Sclerosis (MTS) showed the largest effects (Left MTS: +6.0 years; Right MTS: +4.6 years).
  • Moderate Acceleration: Non-lesional TLE (~3.5–3.7 years) and GGE (+3.3 years).
  • Lowest Acceleration: Non-lesional extratemporal epilepsy (+2.5 years).
• Socioeconomic Interactions (Crucial Finding):
  • There were no main effects of GDP or Gini index on BrainPAD in the general population.
  • However, significant interactions were found between Diagnosis × GDP and Diagnosis × Gini.
  • GDP Effect: The gap in BrainPAD between epilepsy patients and controls was larger in countries with lower GDP per capita. Specifically, for every $10,000 increase in GDP, the epilepsy-related BrainPAD decreased by 0.2 years.
  • Inequality Effect: The epilepsy-related BrainPAD gap was larger in countries with higher income inequality (higher Gini coefficient).
• Sex Differences: Males exhibited a BrainPAD 1 year higher than females (t=3.12), suggesting comparatively better brain health in females.
• Clinical Predictors: LASSO regression indicated that age of epilepsy onset contributed more to BrainPAD than epilepsy duration. A 10-year increase in age of onset was associated with a 0.6-year decrease in BrainPAD.
• Image Quality: No significant difference in MRI quality (Euler number) was found between groups, ruling out scan quality as a confounder.

5. Significance and Implications

• Contextualizing Brain Health: The findings suggest that brain structural ageing in epilepsy is not solely a biological consequence of seizures but is exacerbated by socioeconomic disadvantage. Patients in lower-income or high-inequality countries experience a more severe "brain age" penalty.
• Health Disparities: The results highlight that macroeconomic factors (access to healthcare, nutrition, education, stress) likely act as cumulative stressors that interact with the pathology of epilepsy to accelerate neurodegeneration.
• Research Diversity: The study underscores the critical need for geographic and economic diversity in neuroimaging research. Models trained primarily on high-income populations (like the UK Biobank) may not fully capture the variance in brain health across the global socioeconomic spectrum.
• Clinical Relevance: Understanding that socioeconomic context modulates brain ageing could inform public health strategies, suggesting that improving socioeconomic conditions and healthcare access in epilepsy care could mitigate long-term neuroanatomical decline.

Limitations Noted:

• Cross-sectional design prevents causal inference.
• Retrospective data led to non-standardized MRI acquisition across sites.
• Lack of individual-level socioeconomic data (relying on country-level proxies).
• Underrepresentation of newly diagnosed or well-controlled cases (bias toward refractory epilepsy).
• The brain age model was trained predominantly on Western populations, potentially introducing bias in non-Western cohorts.