ATRIAL FIBRILLATION IS ASSOCIATED WITH

REDUCED BRAIN VOLUME AND COGNITIVE

FUNCTION INDEPENDENT OF CEREBRAL
INFARCTS

1st JOURNAL READING

Saturday, 30 Augustus 2014
Presenter : dr. Steviyani
Moderator : dr. Iskandar Nasution Sp.S FINS
 INTRODUCTION

AF  risk factor for cognitive impairment &

dementia independent of stroke

3 previous studies  ≠ significant association

between AF and total brain atrophy

These studies did not report  difference between

patients with paroxysmal & permanent AF, or
whether the time from the first diagnosis of AF was
of any significancance
They also did not report differences in specific
brain tissue volumes

The aim  to examine the association between AF and brain

measures on MRI as well as cognitive function using data
from the large population based AGESReykjavik Study

The purpose  to assess the significance of the duration &

type of AF (paroxysmal VS persistent/permanent) on these
outcome measures
 MATERIAL AND METHODS
Reykjavik study
– Population : born between
1907 & 1935
– the greater reykjavik, 1967-
1994
– Longitudinal study
– Midlife data on
cardiovascular traits
AGES-Reykjavik study
– cohort
– 5764 subjects
(♂ 2438, ♀ 3326)
– 67 – 93 years
– September 2002 – february
2006
– Questionnaire, clinical
examination, images,
musculoskeletal system,
body composition,
vasculature, & heart
 Ascertainment of AF

Duration of
Hospital in AF :
Reviewing Reykjavik Excluded :
hospital & ICD-9 code from january Reviewing -Paroxysmal AF < 4 Calculated
private 427.9 / ICD- 1, 1987 – 12-lead - Persistent/ weeks from  date of
physician’s 10 code I48 day of the ECG open heart 1st dx/ -
permanent date of the
records study surgery
examination study
examination
 Potential Confounders
• Age
• Sex
Questionnare •
•
Education level
Smoking status
• Alcohol consumption

Geriatric
depression • High depressive symptomatology

scale

BMI • Measured high & weight

 • Total cholesterol level >6.0 mmol/L or
Hypercolesterolemia • Current use of lipid lowering drugs

• Self reported doctor’s diagnosis

Hypertension • Use of hypertensive medications
• Systolic BP ≥ 140 mmHg/ diastolic BP ≥ 90 mmHg

• Self-reported doctors history of Myocardial

infarction
Myocardial infarction • Evidence on ECG  possible/probable Myocardial
infarction
 • Self-reported doctor’s diagnosis
Diabetes • Use of diabetes mellitus related medications, or
mellitus • Fasting blood glucose > 7 mmol/L

Heart • Hospital discharge diagnosis codes from all

hospitals in reykjavik
Failure

Cerebral • Brain MRI

infarcts
MRI Image Acquisition and Image
Processing

All participants without contraindications were eligible for a

brain MRI performed on a study-dedicated 1.5-T Signa
Twinspeed system (General Electric Medical Systems)

Brain volumes were computed with an automatic image analysis

pipeline ( The AGES/MNI Pipeline)

Cerebral Infarcts were scored by trained radiographers

 Assessment of Cognitive
Function

Sex-spesific composite
Averaging the Z scores
The battery of cognitive measures were computed
across the tests in each
test (MEM, SP, EF) by converting raw scores
composite
 standarized z score

Ascertainment of dementia  Diagnostic and Statistical Manual of Mental

Disorders, Fourth Edition
 Statistical Analyses
ANCOVA :
- General characteristics and cardiovascular risk factors were compared
among participants with and without AF
- the association between AF and each brain MRI measure and the 3
cognitive composite scores

Adjustment for covariates was performed in 2 steps:

- basic model (age, sex, and education)
- multivariable model (basic model + previously described demographic +
cardiovascular risk factors + presence of cerebral infarcts)

To assess whether the association between AF and each cognitive

composite score was mediated by brain volume, we adjusted the
multivariable model for each brain MRI measure
The association of duration of AF (divided into tertiles) and type of AF
(paroxysmal versus persistent/permanent) was then assessed.

WMH volumes are reported as antilogs

In secondary analyses, we assessed for effect modification of cerebral infarcts

by including interaction terms in the multivariable model

Significance was set at P<0.05 for all models. All analyses were performed
using SAS version 9.2 (SAS Institute)
 Analytical Sample

3960
4251 ( 3 cognitive
(demented domain test
participants & composite
4569 scores)
missing
( Brain MRI) information
regarding
5706 cognitive
(inform status
consent)
5764
( AGES-
Reykjavik
study)
Results
 Discussion

In this large cross-sectional study of nondemented

elderly individuals in the general population, we
found:
• a significant association between AF and lower total
brain and gray and white matter volumes
• For both total brain and gray matter volume, the
association was stronger with persistent/permanent
compared with paroxysmal AF
• Participants with AF scored significantly lower on
tests of MEM
 AF and Brain volume

The Cardiovascular Health Study of 303 adults 65 to 95 years of

age  ≠ AF – markers of total brain athropy

the Framingham Offspring Study of 1841  ≠ AF – total brain

volume

Knecht et al  ≠ total brain volume – with/without AF

AF and Cognitive Function
After controlling for vascular factors  AF ≈ MEM, but not with
EP/SP

Having AF for longest time  ↓ test of MEM

Lower scores in those with persistent/permanent compared with

paroxysmal AF  did not statiscal significance

the Framingham Offspring Study of men free of symptoms of stroke

 AF ≈ lower scores on a number of cognitive tests ( mainly, on SP
& EF)

Knecht et al that included 87 patients with AF  AF ≈ MEM

 Potential Mechanims for the
Associations Observed

AF  multiple microembolisms to the brain  microinfarcts  atrophy

Additionally, altered cerebral blood perfusion, attributable to beat-to-

beat variation in stroke volume, also may play a part

Cerebral hypoperfusion  ↓ gray & white matter volumes of the brain

Strengths and Limitations of
the Study
Strengths Limitations
• any inference on direct cause and
• The main strength of this effect cannot be made
study is the large number • did not have sufficient information
on ejection fraction or stroke
of welldescribed volume to include it as a covariate
community-dwelling • controlling for previous clinical
subjects diagnosis of heart failure did not
affect the main conclusions of this
study
• did not have sufficient information
on frequency or length of previous
episodes of AF
• might have led to a selection bias
in the study
 Conclusions
• In the general elderly population, AF is associated with lower
total brain volume independent of cerebral infarcts.
• The association is stronger for persistent/permanent AF than
paroxysmal and with increased duration of the disease,
suggesting a cumulative effect.
• The difference is evident in the gray matter and in the white
matter of the brain.
• Future prospective studies are needed to determine whether
maintenance of sinus rhythm is of benefit to attenuate brain
atrophy and impaired MEM performance.
THANK YOU