Mild Cognitive Impairment Claudia Cooper UCL Clinical Reader Honorary consultant old age psychiatrist, Camden and Islington NHS FT
Talk plan Diagnosis Treating MCI: Evidence from RCTs Evidence from prospective cohort studies Preventing dementia in preclinical populations Future directions
Mild Cognitive Impairment a high risk group for dementia Around 20% of people over 70 have MCI (and 10% have dementia) 46% of people with MCI develop dementia within 3 years compared to 3% without MCI.
What prevents MCI converting to dementia?
RCT evidence
Cognitive training Specific strategies improve specific functions eg memory for specific information Effects on general cognition no better than active controls (eg discussion groups) for people with No impairment MCI dementia Part of some successful multicomponent interventions (eg Buschert, FINGER)
Exercise Younger and middle aged people who exercise have better cognition in older age Exercise neuroprotects: Releases BDNF Reduces cortisol Reduces vascular risk
Exercise Exercise did not improve cognition in Healthy older adults (Cochrane, 12 trials) Dementia (but did improve adls) (Cochrane,9 studies) Evidence for MCI inconsistent, in RCT group aerobic exercise no better than relaxation /flexibility control.
Pharmacological interventions for MCI ChEI and NSAIDs do not prevent dementia Cognition improved in single trials of: piribedil, a dopamine agonist over 3 months nicotine improved attention over 6 months. Inconsistent evidence: vitamins B & E, fish oils, Gingko biloba Nutritional (fish oils, vitamin E, tryptophan) improved cognition in one lower quality study Cooper et al, Br J Psychiatry. 2013 Sep;203(3):255-64. Treatment for mild cognitive impairment: systematic review.
Prospective cohort study evidence
Diabetes type 2 and prediabetes Increased risk of Alzheimer s disease and vascular dementia Increased risk of conversion from MCI to dementia Risk for amci, namci and any type MCI Higher rate for untreated diabetes Cooper et al, Modifiable predictors of dementia in mild cognitive impairment: a systematic review and meta-analysis AJP April 2015
Odds ratios for risk of dementia in people with and without diabetes Xu 2010 (Ep, MCI>D) Artero (Ep, MCI>D) Li 2011 (Ep, amci>ad) Prasad (Clin, amci>ad) Ravaglia (Clin, amci>ad) Solfrizzi 2004 (Ep, amci>d) Velayudhan (Clin, amci>d) combined [random] 1.65 (1.12, 2.43) 0.01 0.1 0.2 0.5 1 2 5 10 100 odds ratio (95% confidence interval)
Why does diabetes increase risk of dementia? Damage to blood vessels (atherosclerosis, glucosemediated toxicity) Brain infarcts High blood insulin levels (in type II diabetes and insulin replacement) cause vascular disease direct brain effects; insulin inhibits beta amyloid degradation
Other vascular risk factors: MCI to dementia conversion Pre MCI Hypertension X MCI Metabolic syndrome = 3+: obesity; plasma TG; HDL cholesterol; hypertension; fasting glucose) Cooper et al, AJP 2015 High cholesterol X Smoking X Metabolic syndrome
Figure 2b: Odds ratios for current hypertension Li 2011 (Ep, amci>ad) 1.61 (1.14, 2.26) Prasad (Clin, amci>ad) 0.92 (0.31, 2.71) Solfrizzi 2004 (Ep, amci>d) 1.86 (0.45, 10.97) Ravaglia (Clin, amci>ad) 1.23 (0.58, 2.57) Artero (Ep, MCI>D) 1.96 (1.25, 3.21) Oveisgharan (Ep, MCI>D) 0.81 (0.61, 1.08) Korf (Clin, MCI>D) 0.45 (0.13, 1.45) combined [random] 1.19 (0.81, 1.73) 0.1 0.2 0.5 1 2 5 10 100 odds ratio (95% confidence interval)
Neuropsychiatric symptoms depression increases dementia risk 2-5x tentative evidence antidepressants may have neuroprotective abilities
Figure 2f: Odds ratio, Neuropsychiatric symptoms Brodaty 2012 (Epi, MCI>D) 0.64 (0.07, 3.25) Teng (Clin, MCI>AD) 17.55 (1.60, infinity) Taragano (Clin, MCI>D) 7.51 (3.96, 14.29) Edwards 2009 (Clin, MCI>D) 1.61 (0.86, 3.10) Rosenberg 2013 (Clin, MCI>D) 1.86 (1.53, 2.26) combined [random] 2.52 (1.18, 5.37) 0.01 0.1 0.2 0.5 1 2 5 10 100 odds ratio (95% confidence interval)
Figure 2d: Odds ratios for depressive symptoms Richard 2013 (Ep, MCI>D) 1.70 (0.91, 3.08) Artero (Ep, MCI>D) 1.95 (1.36, 2.77) Chan (Clin, MCI>D) 2.18 (0.99, 4.57) Korf (Clin, MCI>D) 0.48 (0.17, 1.35) Chilovi (Clin, MCI>D) 0.28 (0.09, 0.78) Caracciolo (Ep, amci>d) 3.28 (0.60, 22.48) Panza 2008 (Ep, amci>d) 0.83 (0.23, 3.12) Velayudhan (Clin, amci>d) 1.59 (0.38, 6.24) Modrego (Clin, amci>ad) 11.91 (4.11, 38.57) Fellows (Clin, amci>ad) 0.76 (0.30, 1.94) Visser 2000 (Clin, amci>ad) 0.57 (0.14, 2.09) Palmer 2010 (Clin, amci>ad) 0.95 (0.23, 3.41) Richard 2012 (Clin, amci>ad) 1.86 (1.21, 2.87) combined [random] 1.35 (0.89, 2.06) 0.01 0.1 0.2 0.5 1 2 5 10 100 odds ratio (95% confidence interval)
Figure 2e Apathy Odds ratio Palmer 2010 (Clin, amci>ad) 8.67 (1.83, 39.44) Richard 2012 (Clin, amci>ad) 0.72 (0.47, 1.09) Robert (Clin, amci>ad) 2.39 (0.89, 6.66) Chilovi (Clin, MCI>D) 2.75 (1.03, 7.21) Chan (Clin, MCI>D) 0.34 (0.06, 1.14) combined [random] 1.62 (0.63, 4.17) 0.01 0.1 0.2 0.5 1 2 5 10 100 odds ratio (95% confidence interval)
Education Education predicts dementia in general population but not MCI.
Figure 2g: Effect size for years of education Mackin (Clin, amci>d) Velayudhan (Clin, amci>d) Visser 2000 (Clin, amci>ad) Ye 2012 (Clin, amci>ad) Fellows (Clin, amci>ad) Prasad (Clin, amci>ad) Rozzini 2007 (Clin, amci>ad) Hsiung 2008 (Clin, MCI>AD) Meyer (Ep, MCI>D) Chan (Clin, MCI>D) Devier (Clin, MCI>D) Korf (Clin, MCI>D) -1.5-1.0-0.5 0 0.5 1.0 DL pooled effect size = -0.117011 (95% CI = -0.26397 to 0.029947)
Diet Higher folate levels Adherence to Mediterranean diet
Mediterranean diet Decreases cognitive decline and reduces risk of Alzheimer s disease Lots of vegetables, legumes, fruits, cereals; fish, unsaturated fatty acids (olive oil) low intake of saturated fatty acids; dairy products, meat and poultry; Regular, moderate wine during meals Epidemiology. 2013 Jul;24(4):479-89.
Why might Mediterranean diet prevent dementia? It is associated with: Lower blood pressure Lower bad cholesterol Less diabetes Lower homocysteine levels (high homocysteine levels linked to heart disease) PREDIMED trial (Malaga): trained dieticians gave group and individual advice to people aged 55-80. Reduced heart attacks, stroke and death.
Alcohol Heavy alcohol use harmful In general population, moderate alcohol consumption might be helpful No evidence moderate alcohol use helps people with MCI
Preventing dementia preclinical stage
Factors increasing the risk of Alzheimer s disease: Physical inactivity (UK PAF 22%) Depression 8% Midlife hypertension 7% Midlife obesity 7% Smoking 11% Low educational attainment 12% Diabetes (5%) Norton et al, 2014, Lancet Neurology
What are the most important predictors of dementia (in the general older population) 1. crystallised intelligence (PAF 18%) 2. depression (10%) 3. Genetic risk (apolipoprotein E ε4 allele) 7% 4. fruit and vegetable consumption (6%) 5. diabetes (5%) BMJ. 2010; 341: c3885. Karen Ritchie
Social support Feelings of loneliness, not actual social isolation predicted dementia in older people (AMSTEL study) Larger social networks protected from dementia (US/Swedish studies)
FINGER trial (Lancet, 2015) 1260 people with high vascular risk scores Cognition improved with intervention: Diet (3 individual and 7-9 group sessions) Exercise (group and individual) Cognitive training (10 group sessions and individual computerised 3x per week) Regular nurse and physician management of vascular risk factors
What might a future intervention for MCI be like? Multimodal Cost-effective Decrease further damage: Maximise function: Treat vascular risk factors Diet, exercise Cognitive/ social stimulation Treat neuropsychiatric symptoms
What to tell people with MCI Looking after physical health and reducing excessive alcohol intake will help Eating well to reduce risk of diabetes and managing existing diabetes well very likely to help. Staying mentally well also important.