Abstract

Public health and therapeutic measures to reduce cardiac and stroke risk may already be reducing risks of dementia in populations. In routine clinical care, dementia risk could be further reduced by optimized management of delirium, depressive disorders, traumatic brain injury and stroke. These are opportunities to minimize risk of progression to dementia during acute care and subsequent rehabilitation. Although interventions to protect against progress to dementia may be of small effect in each clinical situation, awareness of dementia risk and appropriate steps to reduce that risk should contribute to an overall reduction in the incidence of dementia.

Introduction

The lifespan of the authors has encompassed unprecedented changes in patterns of disease in Britain. Increased life expectancy from reduction in many infective, malignant and cardiac diseases has, however, led to increased prevalence of disabilities associated with ageing, especially progressive cognitive problems. While improved treatment has played an important role in increasing longevity, preventive public health strategies, including the use of vaccines and other therapy, have played major roles. Since integrated public health and personal prevention have shown benefits in heart disease and cancer; should the same strategy not also apply to dementias? Such conditions assume great importance in relation to the costs of care of an increasingly elderly population in times of economic austerity.

There are two complementary strategies for dementia risk reduction. One is to aim at general behavioural change across a population and the other is to focus on specific, at-risk groups of individuals. Both depend on an understanding of factors associated with increased risk. Here, we discuss the application of such strategies in prevention of dementias and consider the role of physicians in promoting this. In an ageing population, many have a relative with dementia so advice on familial risk and personal prevention is increasingly likely to be sought. When this is soundly based, it is more likely to be followed.

Dementia risk reduction in a population

Recent reports of decreasing late onset dementia prevalence1 and improved physical and cognitive health among the very old2 encourage the view that the onset of dementia is modifiable, although the means to achieve this are poorly understood. The typical neuropathology of Alzheimer’s disease (AD) probably explains no more than 50% of inter-individual variation in late life cognitive decline.3 Furthermore, it is unclear whether a continuing focus on the neurobiology of AD will fulfil optimistic predictions of effective dementia prevention.4,5 Other major causal influences on dementia will require better understanding of the many pathophysiological pathways leading to dementia some of which will be shared with senescence and stroke.6 An association between dementia and stroke mediated through vascular disease mechanisms encourages current clinical trials of dementia prevention using general behavioural change strategies with vascular risk reduction at their core.7 Among these are physical exercise, smoking cessation, moderate intakes of calories and alcohol and better control of hypertension and diabetes. To date, evidence-based public health policies to prevent dementia are not available although relevant stroke and heart disease prevention policies already exist and contribute to established benefits.

Predictive indices and risk factors

The prediction of late onset dementia is not based on molecular genetic tests; any such putative test would raise significant legal and ethical issues in its application. Meta-analyses of powerful molecular genome-wide association studies (GWAS) so far show17 genetic loci contribute to dementia susceptibility, of which 16 can be replicated in independent samples.8 The possibility that multiple associations discovered in GWAS could be combined into a single genetic susceptibility score (that includes Apoliprotein E, APOEε4)9 is of interest but does not so far contribute to personalized estimates of dementia risk in clinical practice.

Currently, prediction relies on the presence of five risk factors relating to: (i) senescence; (ii) family history; (iii) mid-life vascular risk factors; (iv) mild cognitive impairment; and (v) the detection of preclinical biomarkers associated with later clinical dementia syndromes. Available dementia predictive indices do not typically include all five groups but more usually devise scores based on just one or two groups.10,11 Ageing and family history are the strongest risk factors for dementia and related neurodegenerative diseases. The life course approach to dementia aetiology complements genetic discoveries.12 It incorporates a lifetime accumulation of multiple risk factor exposures that is sufficient to overcome an ageing individual’s capacity to repair and compensate for the presence of dementia. Table 1 summarizes these risk and protective factors stratified by opportunities to modify their effects. Many are appropriate to a public health approach and would be expected to have benefits beyond dementia prevention.

Table 1

Summary of risk and protective factors shown in observational studies to be associated with risk of late onset dementia

 Risk factor Protective factor 
 Evidence 1-weak 4-strong 
Non-modifiable APOEε4 [4] APOEε2 [3] 
Family history [4] 
Epigenetic markers [2] 
Below average IQ [4] 
Early life modifiable Low education [4] Enriched education [1] 
Bilingualism [3] 
Midlife modifiable Hypertension [3] BP control 
Smoking [2] Smoking cessation [?] 
Alcohol [1] Moderate alcohol [2] 
Obesity [2] Weight loss [?] 
Hypercholesterolaemia [2] Statins [2] 
Diabetes [2] Hypoglycaemic agents [2] 
Job complexity [1] Physical activity [3] 
Physical ‘inactivity’ [2] Mental effort [1] 
Depression [4] Antidepressants [2] 
TBI [4] Rehabilitation [?] 
Late life modifiable Stroke [4] Dietary fruits, vegetables [3] 
Delirium [4]  
TBI [4] Rehabilitation [?] 
 HRT [1] 
Loneliness [2] Social engagement 
Hyperhomocysteinaemia [4]  
Low folate [3] Folate/B12 [3] 
Pro-inflammatory processes NSAIDs [1] 
  Fish oil [2] 
 Risk factor Protective factor 
 Evidence 1-weak 4-strong 
Non-modifiable APOEε4 [4] APOEε2 [3] 
Family history [4] 
Epigenetic markers [2] 
Below average IQ [4] 
Early life modifiable Low education [4] Enriched education [1] 
Bilingualism [3] 
Midlife modifiable Hypertension [3] BP control 
Smoking [2] Smoking cessation [?] 
Alcohol [1] Moderate alcohol [2] 
Obesity [2] Weight loss [?] 
Hypercholesterolaemia [2] Statins [2] 
Diabetes [2] Hypoglycaemic agents [2] 
Job complexity [1] Physical activity [3] 
Physical ‘inactivity’ [2] Mental effort [1] 
Depression [4] Antidepressants [2] 
TBI [4] Rehabilitation [?] 
Late life modifiable Stroke [4] Dietary fruits, vegetables [3] 
Delirium [4]  
TBI [4] Rehabilitation [?] 
 HRT [1] 
Loneliness [2] Social engagement 
Hyperhomocysteinaemia [4]  
Low folate [3] Folate/B12 [3] 
Pro-inflammatory processes NSAIDs [1] 
  Fish oil [2] 

Levels of evidence (? - uncertain; 1-weak 4-strong) are arbitrary and require detailed evaluation; estimates are provisional and use data from references.10,11

Abbreviations: APOE, apolipoprotein E; TBI, traumatic brain injury; NSAID, non-steroidal anti-inflammatory drug; HRT, hormone replacement therapy.

Although Table 1 suggests that risk prediction may be feasible, this should be regarded cautiously. Although some authors11 provide predictive estimates for each risk factor, this is inappropriate when risk factors are inter-dependent (e.g. smoking and alcohol use, depression and loneliness). Furthermore, prediction scales are often based on epidemiological data that could not be satisfactorily adjusted for important confounders (e.g. socioeconomic status, education or living group). Prediction is much improved when preclinical mild cognitive decline is present and improved further when a biomarker (e.g. MRI-derived hippocampal atrophy) is detected. Their inclusion contributes to secondary prevention where pre-symptomatic dementia predicts progress to dementia.

From the physician’s point of view it is important to identify factors that contribute to individual risk of dementia that are amenable to modification based on current knowledge. Four dementia risk factors: depression, delirium, stroke and traumatic brain injury (TBI), are so commonly encountered in routine clinical practice that their improved general medical management could reduce the overall burden of dementia in the population.

The physician’s role in dementia prevention

There are opportunities in late life to modify individual risk of progression to dementia. While each individual measure will only make a small contribution, in combination these possess sufficient potential to make a significant impact on the overall prevalence of dementia. Indeed, the authors of recent reports of an apparent decline in dementia thought improvements in diagnosis and care of stroke were already implicated in dementia reduction.13

  • 1.  Delirium There is a well-established link between delirium and enduring cognitive disabilities.14 In a major outcome study of patients who had received critical care,15 ∼75% met criteria for delirium, 25% were cognitively impaired 12 months after critical illness and 33% had impairment typically associated with moderate TBI. Patterns of neuropsychological deficit were more diffuse than seen in AD of equivalent severity. Measures to identify, prevent and/or optimize the clinical management of delirium could reduce dementia risk. These include ‘best practice’ nursing guidelines with repeated measures to detect delirium in critical care. Post-graduate education of physicians in critical care should include evidence-based approaches to the diagnosis and management of delirium which could significantly reduce risks of subsequent premature death and chronic cognitive impairmen.16

  • 2.  Depression Throughout middle to late adulthood, associations are reported between recurrent depressive episodes and increased dementia risk. To a lesser extent, links between first depressive illnesses in adulthood and late onset dementia are also found.17 This is a relevant distinction because the preclinical stage of dementia probably develops 10–20 years before dementia symptoms. Dementia of predominantly cerebrovascular origin frequently presents with depressive symptoms, weakening a causal model of depression predisposing to dementia18 but a meta-analysis19 showed that when timing of onset of depressive illness and dementia are separated, a reliable association between depression and dementia remains.

    The neurobiological basis of a depression/dementia association may reflect the chronic effects of stress on brain.20 Genes affecting individual differences in stress responses are poorly understood and future research may explain how age-associated differences in brain compensatory systems can moderate neuropathological burden and bolster pathways that cope with psychological stressors. This is relevant to the neurobiology of ‘cognitive reserve’ in the face of ageing and adversity.21 Cognitive capacity depends on widespread neural networks rather than isolated brain areas. Neurodegeneration affects cognition once any compensatory capacity is exceeded. Genetic predispositions for depression, loss of efficiency of neural networks and the pervasive neurotoxicity of chronic exposure to hypercortisolaemia may be implicated. Effective antidepressant care could, therefore, contribute to reduced incidence of recurrent depression and reduced risk of dementia.18

  • 3.  Traumatic brain injuryThe association of dementia after head injury in contact sports is well-known but old people are also vulnerable through falls and accidents. Persistent cognitive impairment after TBI is common and its likelihood is increased when other risk factors for dementia are present.22 Although some meta-analyses estimate an increased risk of dementia in later life following earlier TBI, no consensus exists that is sufficient to advise individuals with a known susceptibility to dementia (e.g. a family history of dementia or APOEε4 status) to avoid occupations where TBI is frequent. The exact nature of the relationship between TBI and late onset AD seems likely to concern the efficiency of innate compensatory or regenerative systems that influence the hypothesized threshold (‘cognitive reserve’) below which the signs or symptoms of dementia become detectable. Rehabilitation programmes after TBI aim through directed training to increase ‘cognitive reserve’ and so reduce cognitive deficits. Lifestyle interventions are advisable to reduce exposures to neurotoxic substance abuse (excess alcohol, some illicit drugs) that may be implicated in the original TBI.

  • 4.  Stroke Dementia is a well-recognized outcome of stroke. When cognitive impairment develops at least 3 months after stroke,23 excluding previous dementia: (i) after first-ever stroke, pooled mean dementia prevalence estimates are ∼10%; (ii) after either first or recurrent stroke, pooled estimates are ∼20% and (iii) after recurrent stroke pooled estimates are ∼41%. Predictors of dementia after stroke are older age, female sex, lower educational attainment, non-white race, diabetes and atrial fibrillation. Factors that significantly predict later dementia include: haemorrhagic stroke, dysphasia, left hemispheric involvement and recurrent stroke. Seizures at the time of stroke are associated with higher risk of dementia, as are incontinence or delirium. MRI findings of leukoaraiosis or cortical atrophy (particularly of the medial temporal lobe) could also be linked to post-stroke dementia. These predictors are also shared with onset of sporadic late onset dementia not associated with stroke.

    Numerous problems bedevil detection of dementia after stroke and the complex issues arising in the prevention of post-stroke cognitive impairment.23,24 Their investigation raises the possibility that within a spectrum of dementias following delirium, stroke or TBI, there are both immediate and delayed opportunities to intervene to reduce dementia risk. Potential immediate interventions will probably be informed by individual characteristics of the acute illness episode rather than linked specifically to one or other clinical syndrome. For example, when increased pro-inflammatory markers are present, the critical intervention to protect against dementia might be to increase anti-oxidant defences. Likewise, when cerebral thrombosis is detected, fibrinolytic therapy could be appropriate. Better assessments of cognitive function are needed in routine clinical practice in order to detect and treat impairment after stroke. Most delayed dementia risk reduction after acute stroke will be derived from reduction in risk factors for stroke. Future clinical trials will test interventions to prevent post-stroke cognitive impairment.25 Current research in neuroprotection/neuroregeneration aims to develop novel therapies that augment cerebral plasticity. Putative ‘plasticity-enhancing’ strategies encompass interventions as diverse as environmental enrichment, trans-cranial magnetic stimulation and oral antidepressant therapies.

Conclusion

Dementia research does not differ from other medical research in having the hope of both prevention and cure. Emphasis on AD has been on finding a drug but there is increasing evidence that both it and the other syndromes of dementia have important environmental risk factors, often in common with cardiovascular disease. There is evidence that public health and therapeutic measures to reduce cardiac risks may also be influencing the risk of dementia. However, there are specific medical conditions or events that increase risks in individuals and it seems likely that improved management of conditions such as hypertension, diabetes, stroke, depression, traumatic head injury and delirium would contribute to a lessening of the burden of dementia in the elderly. There are opportunities here for all physicians to play their part in this endeavour.

Conflict of interest: None declared.

References

1
Matthews
FE
Arthur
A
Barnes
LE
Bond
J
Jagger
C
Robinson
L
, et al.  . 
A two-decade comparison of prevalence of dementia in individuals aged 65 years and older from three geographical areas of England: results of the Cognitive Function and Ageing Study I and II
Lancet
 , 
2013
, vol. 
382
 (pg. 
1405
-
12
)
2
Christensen
K
Thinggaard
M
Oksuzyan
A
Steenstrup
T
Andersen-Ranberg
K
Jeune
B
, et al.  . 
Physical and cognitive functioning of people older than 90 years: a comparison of two Danish cohorts born 10 years apart
Lancet
 , 
2013
, vol. 
382
 (pg. 
1507
-
13
)
3
Boyle
PA
Wilson
RS
Yu
L
Barr
AM
Horner
WG
Schneider
JA
, et al.  . 
Much of late life cognitive decline is not due to common neurodegenerative pathologies
Ann Neurol
 , 
2013
, vol. 
74
 (pg. 
478
-
89
)
4
Mangialasche
F
Solomon
A
Winblad
B
Mecocci
P
Kivipelto
M
Alzheimer’s disease: clinical trials and drug development
Lancet Neurol
 , 
2010
, vol. 
9
 (pg. 
702
-
16
)
5
Corbett
A
Pickett
J
Burns
A
Corcoran
J
Dunnett
SB
Edison
P
, et al.  . 
Drug repositioning for Alzheimer’s disease
Nat Rev Drug Discov
 , 
2012
, vol. 
11
 (pg. 
833
-
46
)
6
Gorelick
PB
Scuteri
A
Black
SE
Decarli
C
Greenberg
SM
Iadecola
C
, et al.  . 
Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association
Stroke
 , 
2011
, vol. 
42
 (pg. 
2672
-
713
)
7
Mangialasche
F
Kivipelto
M
Solomon
A
Fratiglioni
L
Dementia prevention: current epidemiological evidence and future perspective
Alz Res Ther
 , 
2012
, vol. 
4
 pg. 
6
  
doi: 10.1186/alzrt104
8
Lee
SH
Harold
D
Nyholt
DR
Goddard
ME
Zondervan
KT
Williams
J
, et al.  . 
Estimation and partitioning of polygenic variation captured by common SNPs for Alzheimer's disease, multiple sclerosis and endometriosis
Hum Mol Genet
 , 
2013
, vol. 
22
 (pg. 
832
-
41
)
9
Lambert
J-C
Amouyel
P
Genetics of Alzheimer’s disease: new evidences for an old hypothesis?
Curr Opin Genet Dev
 , 
2011
, vol. 
21
 (pg. 
295
-
301
)
10
Jessen
F
Wiese
B
Bickel
H
Eifflande-Gorfer
S
Haller
F
Kolsch
H
, et al.  . 
Prediction of dementia in primary care patients
PLOS
 , 
2011
, vol. 
6
 pg. 
e16852
 
11
Barnes
D
Yaffe
K
The projected effect of risk factor reduction on Alzheimer’s disease incidence
Lancet Neurol
 , 
2011
, vol. 
10
 (pg. 
819
-
28
)
12
Whalley
LJ
Dick
FD
McNeill
GA
Life-course approach to the aetiology of late-onset dementia
Lancet Neurol
 , 
2006
, vol. 
5
 (pg. 
87
-
96
)
13
Qiu
C
von Strauss
E
Bäckman
L
Winblad
B
Fratiglioni
L
Twenty year changes in dementia occurrence suggest decreasing incidence in central Stockholm, Sweden
Neurology
 , 
2013
, vol. 
80
 (pg. 
1888
-
94
)
14
MacLullich
AM
Starr
JM
Passmore
AP
Delirium should be included in guidelines and curriculums
Br Med J
 , 
2007
, vol. 
334
 pg. 
968
 
15
Pandharipande
PP
Girard
TD
Jackson
JC
Morandi
A
Thompson
JL
Pun
BT
, et al.  . 
Long-term cognitive impairment after critical illness
N Engl J Med
 , 
2013
, vol. 
369
 (pg. 
1306
-
16
)
16
Maclullich
AM
Anand
A
Davis
DH
Jackson
T
Barugh
AJ
Hall
RJ
, et al.  . 
New horizons in the pathogenesis, assessment and management of delirium
Age Ageing
 , 
2013
, vol. 
42
 (pg. 
667
-
74
)
17
Brommelhoff
JA
Gatz
M
Johansson
B
McArdle
JJ
Fratiglioni
L
Pedersen
NL
Depression as a risk factor or prodromal feature for dementia? Findings in a population-based sample of Swedish twins
Psychol Aging
 , 
2009
, vol. 
24
 (pg. 
373
-
84
)
18
Barnes
DE
Yaffe
K
Byers
AL
McCormick
M
Schaefer
C
Whitmer
RA
Mid-life versus late-life depressive symptoms and risk of dementia: differential effects for Alzheimer’s disease and vascular dementia
Arch Gen Psychiat
 , 
2012
, vol. 
69
 (pg. 
493
-
8
)
19
Diniz
BS
Butters
M
Albert
S
Dew
MA
Reynolds
C
Late-life depression and risk of vascular dementia and Alzheimer’s disease: systematic review and meta-analysis of community-based cohort studies
Brit J Psychiat
 , 
2013
, vol. 
202
 (pg. 
329
-
35
)
20
Aznar
S
Knudsen
GM
Depression and Alzheimer’s disease. Is stress the initiating factor in a common neuropathological cascade?
J Alz Dis
 , 
2011
, vol. 
23
 (pg. 
177
-
93
)
21
Whalley
LJ
Deary
IJ
Appleton
CL
Starr
JM
Cognitive reserve and the neurobiology of cognitive aging
Aging Res Rev
 , 
2004
, vol. 
3
 (pg. 
369
-
82
)
22
Moretti
L
Cristofori
I
Weaver
SM
Chau
A
Portelli
JN
Grafman
J
Cognitive decline in older adults with a history of traumatic brain injury
Lancet Neurol
 , 
2012
, vol. 
11
 (pg. 
1103
-
12
)
23
Gottesman
RF
Hillis
AE
Predictors and assessment of cognitive dysfunction resulting from ischaemic stroke
Lancet Neurol
 , 
2010
, vol. 
9
 (pg. 
895
-
905
)
24
Pendlebury
ST
Rothwell
PM
Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis
Lancet Neurol
 , 
2009
, vol. 
8
 (pg. 
1006
-
18
)
25
Blackburn
DJ
Krishnan
K
Fox
L
Ballard
C
Burns
A
Ford
GA
, et al.  . 
Prevention of decline in cognition after stroke trials (PODCAST): a study protocol for a factorial randomised controlled trial of intensive versus guideline lowering of blood pressure and lipids
Trials
 , 
2013
, vol. 
14
  
doi: 10.1186/1745-6215-14-401