Current Psychiatry

Discuss this article

Pharmacologic treatments for Alzheimer’s disease (AD) may improve symptoms but have not been shown to prevent AD onset. Primary prevention therefore remains the goal. Although preventing AD by managing risk factors such as age or genetics is beyond our control (Box 1), we can do something about other factors.

This article summarizes the findings of many studies that address AD prevention and includes an online-only bibliography for readers seeking an in-depth review. The evidence does not support a firm recommendation for any specific form of primary prevention and has revealed hazards associated with estrogen therapy and nonsteroidal anti-inflammatory drugs (Box 2). Most important, it suggests that you could reduce your patients’ risk of developing AD by routinely supporting their mental, physical, and social health.

The potential benefits of modifying an individual’s AD risk factors likely will depend on his or her genetic makeup, environment, and lifestyle. Even so, counseling patients to exercise more and improve their diets—such as by eating more fish, fruits, and vegetables and less saturated fat—might help protect the brain. Your ongoing efforts to manage hypertension, hypercholesterolemia, and diabetes also may reduce their AD risk.

Box 1

Box 2

Cardiovascular risk factors

The risk of developing AD or vascular dementia appears to be increased by conditions that damage the heart or blood vessels. Recent evidence suggests that successfully managing cardiovascular risk factors may decrease the likelihood of dementia in later life.

Hypertension is associated with a higher risk of AD and all-cause dementia. Curiously, some studies have shown that low blood pressure also increases dementia risk, suggesting a U-shaped relationship between blood pressure and cognitive decline. Systolic hypertension in midlife may be associated with dementia 20 years later.

One might assume that antihypertensive therapy would help prevent dementia, but the data are conflicting. The Systolic Hypertension in Europe (SYST-EUR) study¹ showed a 53% reduction in vascular dementia or mixed dementia among patients receiving antihypertensive medication and a 60% reduction in AD. Similarly, the PROGRESS² clinical trial of prevention of recurrent stroke by antihypertensive treatment reported a 34% reduction in a composite measure of cognitive impairment and dementia. On the other hand, cognitive function neither improved nor worsened in the Hypertension in the Very Elderly Trial (HYVET-COG),³ whether patients received blood pressure treatment or placebo.

Hyperlipidemia. Lipid metabolism likely is an important pathway in amyloid beta-protein deposition, tau phosphorylation, and disruption of synaptic plasticity and neurodegenerative endpoints. Cognitive decline and incident dementia have been associated with higher dietary intake of saturated fats, partially hydrogenated unsaturated fatty acids (trans fats), and cholesterol. Not all studies have found this association, however. This could be because serum cholesterol levels may decrease in early dementia, limiting the ability to detect an effect of hypercholesterolemia on dementia risk when measurements are made later in life.

Using statins (3-hydroxy-3-methylglutaryl–coenzyme A reductase inhibitors) to treat hypercholesterolemia has been hypothesized to impede large vessel atherosclerosis and its consequences and to trigger metabolic effects in the brain related to AD pathogenesis. Mechanisms by which statins might help prevent dementia include:

• a direct association between amyloid processing and cholesterol in the brain
• an indirect effect by decreasing the risk of stroke, as even small cerebral infarcts worsen AD severity.

Nonrandomized epidemiologic studies such as the Cardiovascular Health Study⁴ and MRC/BHF Heart Protection Study⁵ suggested that statin treatment might reduce the incidence of dementia, the degree of age-related cognitive decline, and AD’s neuropathologic burden. Large, randomized, controlled trials have not supported these observations, however. Statins failed to reduce the incidence of dementia in:

• the Heart Protection Study, testing simvastatin for 5 years in 20,536 subjects age 40 to 80⁵
• the 3-year Preventive Study of Pravastatin in the Elderly at Risk (PROSPER) of 5,800 subjects.⁶

Similarly, patients receiving adjunctive atorvastatin or placebo showed no significant differences in cognition assessments after 72 weeks in the Lipitor’s Effect in Alzheimer’s Dementia (LEADe) study. This trial enrolled 640 subjects age 50 to 90 with mild-to-moderate dementia who were treated with donepezil.⁷ A recent Cochrane review concluded that high serum cholesterol may contribute to the development of AD and vascular dementia, but lowering cholesterol levels with statins does not prevent these problems.⁸

Diabetes mellitus. Diabetes and cognitive decline are closely associated. Diabetes is associated with a 50% to 100% increase in risk of AD and dementia overall and a 100% to 150% increased risk of vascular dementia. The mechanism by which diabetes increases dementia risk is uncertain but does not appear to be mediated entirely through vascular disease. High and low insulin levels may increase the risk of dementia, independent of diabetes and blood glucose. Increased peripheral insulin levels are associated with reduced brain atrophy and cognitive impairment in patients with early AD, suggesting a role for insulin signaling in AD pathophysiology. A possible relationship between insulin and beta amyloid metabolism is being studied.

Elevated postprandial plasma glucose has been associated with accelerated declines in cognitive performance.⁹ An inverse correlation has been noted between some cognitive measures and hemoglobin A1C levels.¹⁰ It is not clear that treating diabetes reduces the risk of dementia. In addition, in the prospective, population-based Rotterdam study, elderly patients with type 2 diabetes treated with insulin had the highest incidence of dementia.¹¹

Tobacco smoke directly affects neuronal function, integrity, and survival. Chronic smoking has been linked to decreased global cerebral blood flow, accelerated cerebral atrophy, and ventricular enlargement.

Some studies suggest an increased risk of dementia in middle-aged and elderly smokers, possibly through a cerebrovascular mechanism such as stroke. Other studies found no association between smoking and dementia risk, and 1 suggested that nicotine may protect against AD by reducing senile plaque formation. Any protective effect of smoking would be offset by increased risks of lung cancer, chronic obstructive pulmonary disease, and vascular dementia.

The apolipoprotein E epsilon 4 (APOE e4) gene may explain, at least in part, the conflicting results of these studies. In 2 population-based cohorts,^12,13 smoking was associated with memory decline in patients without, but not with, the APOE e4 genotype.

Dietary factors

Antioxidants. The brains of patients with AD contain elevated levels of endogenous antioxidants. In vitro studies show exogenous antioxidants can reduce the toxicity of beta-amyloid in brain tissue of persons with AD. These findings have led to interest in assessing the role of dietary antioxidants such as vitamins E and C for AD prevention.

High-dose alpha-tocopherol (vitamin E, 2,000 IU/d) may slow disease progression in patients with AD, but this association is not consistently found. Furthermore, a meta-analysis of 19 randomized controlled trials (RCTs) totaling >135,000 patients found an association between vitamin E doses >400 IU/d and increased all-cause mortality.¹⁴ High-dose vitamin E supplementation for primary or secondary prevention of AD may be dangerous and is not recommended.

The lack of consistent efficacy data for vitamin C in preventing or treating AD may discourage its routine use for this purpose.¹⁵

Homocysteine is a risk factor for stroke and heart disease. It also could play a role in vascular dementia through its association with large- and small-vessel disease.

Low folate and hyperhomocysteinemia have been associated with dementia or cognitive impairment, although a cause-effect relationship is not clear. In non-demented elderly populations, plasma homocysteine is inversely associated with poor performance in tests of global cognitive function, particularly in measures of psychomotor speed.

In a recent double-blind RCT, folic acid supplementation for 3 years significantly improved domains of cognitive function that tend to decline with age, especially information processing and sensorimotor speed.¹⁶ No other good evidence, however, has shown that homocysteine-lowering therapy using folic acid or other vitamin B supplements improves cognitive function or prevents cognitive decline.

Fish and omega-3 fatty acids. High total fat, saturated fat, and total cholesterol intake increases the risk for incident dementia. In epidemiologic studies, low omega-3 fatty acid serum levels have been linked to increased dementia risk.

Fish consumption may be beneficial in reducing the risk of dementia or cognitive decline. A prospective study of 815 elderly persons found 60% less risk of developing AD in those who ate ≥1 fish meal per week, compared with those who rarely or never ate fish.¹⁷ In the Framingham study, individuals who at baseline were in the top quartile of docosahexaenoic acid consumption had lower dementia rates over 9 years of follow-up.¹⁸ Results from cross-sectional and longitudinal studies have been inconsistent; some have shown that high intake of n-3 polyunsaturated fatty acids is associated with less cognitive decline,¹⁹ whereas others have not.²⁰

Although we cannot offer unequivocal advice regarding seafood or omega-3 fatty acid intake for primary prevention of dementia without evidence from RCTs, these uncontrolled studies show promise.

Mediterranean diet (MeDi) components include abundant fruits and vegetables, fish or shellfish at least twice weekly, very limited red meat, olive oil or canola oil instead of butter or margarine, tree nuts such as walnuts or pecans, red wine in moderation, and using herbs and spices instead of salt to season food. High adherence to the MeDi has been associated with a significantly lower risk for incident AD. The MeDi may affect the risk of developing AD²¹ as well as subsequent disease course, with a possible dose-response relationship in lower mortality.²²

Eating fruits and vegetables has been associated with improved cognitive performance²² and decreased incident dementia in elderly subjects.¹⁸

Alcohol. A U-shaped relationship exists between alcohol consumption and dementia risk. High alcohol intake is associated with clinical problem drinking and alcoholism and can lead to cognitive decline. Conversely, moderate wine consumption (250 to 500 mL/d) may be protective—compared with more or less than this amount—and is associated with approximately 50% less risk of dementia.

Alcohol use may increase the risk of dementia in persons carrying the APOE e4 allele, according to the population-based Cardiovascular Risk Factors, Aging and Dementia (CAIDE) study from Sweden.²³ After an average 21 years of follow-up of 1,449 individuals, researchers found that environmental factors—such as physical inactivity, dietary fat intake, alcohol consumption, and smoking at midlife—were associated with an increased risk of dementia at age 65 to 79 in APOE e4 carriers compared with noncarriers. The study also found that physical inactivity, dietary fat intake, and smoking at midlife increase AD risk, especially among APOE e4 carriers.

In the absence of evidence from RCTs, we cannot recommend alcohol to reduce the risk of AD.

Lifestyle and activity

Three components of lifestyle—social, mental, and physical activity—are inversely associated with the risk for dementia, AD, and cognitive impairment.

Physical exercise has been thought to enhance brain neurotrophic factor and modify apoptosis. Exercise can deter stroke and microvascular disease and improve regional cerebral blood flow. In the Cardiovascular Health Study, participants who expended the highest quartile of energy had a lower risk of all-cause dementia and AD compared with participants who expended the lowest quartile of energy.²⁴

Mental and social activity. Epidemiologic studies have shown associations between higher educational achievement and other socioeconomic factors and reduced AD risk. Advanced education is believed to represent a cognitive reserve that delays presentation of AD’s effects on memory and cognitive function, rather than providing a protective effect against accumulation of AD pathology. Higher-educated individuals appear to experience a somewhat more rapid rate of cognitive decline when AD does become apparent, perhaps because they have accumulated a greater degree of AD pathology at that point compared with less-educated persons.

Among 117 persons with dementia in the Bronx Aging Study, each additional year of formal education delayed the time of accelerated decline by 0.21 years. After accelerated decline began, each year of additional formal education was associated with a slightly faster rate of memory decline.²⁵

The longitudinal, population-based Kungsholmen Project in Stockholm, Sweden, found an association between daily mentally stimulating activities and decreased risk of all-cause dementia.²⁶ Similarly, higher levels of leisure activity were linked to reduced risk of all-cause dementia in a longitudinal study of 1,772 persons age ≥65 living in Manhattan, NY.²⁷ In a randomized, single-controlled study of the long-term effects of cognitive training, elderly individuals from 6 U.S. cities showed sustained improvement in specific cognitive performance up to 5 years after training sessions began, including memory, reasoning, and speed of processing.²⁸

It seems reasonable to encourage older patients to maintain or increase physical, cognitive, and leisure activities as well as social interaction. These interventions can improve the quality of life and lower the risk of depression, which may be a response to cognitive decline or an independent risk factor for dementia (Box 3). The Table lists “brain exercises” you can suggest to patients to increase their mental and social activity.

Head trauma. The Multi-Institutional Research in Alzheimer’s Genetic Epidemiology (MIRAGE) project found an association between AD risk and a history of head trauma, especially in persons with APOE e4 alleles.²⁹ Conversely, the Rotterdam Study showed no change in dementia risk for persons with a history of head trauma.³⁰

Even in the absence of conclusive evidence supporting AD prevention, protecting the head by buckling seat belts while driving, wearing helmets when participating in sports, and “fall-proofing” the home is recommended.

Box 3

Table

Brain exercises to suggest to patients

Related resource

• For an extensive bibliography of literature on Alzheimer’s disease risk factors and prevention, see this article at CurrentPsychiatry.com.

Drug brand names

• Atorvastatin • Lipitor
• Celecoxib • Celebrex
• Donepezil • Aricept
• Medroxyprogesterone • Provera
• Pravastatin • Pravachol
• Rofecoxib • Vioxx
• Simvastatin • Zocor

Disclosures

Dr. Bassil reports no financial relationship with any company whose products are mentioned in this article, or with manufacturers of competing products.

Dr. Grossberg receives research/grant support from and is a consultant to Bristol-Myers Squibb, Forest Pharmaceuticals, Novartis, Pfizer Inc., and Wyeth Pharmaceuticals. He also receives research/grant support from Baxter.

Discuss this article

Pharmacologic treatments for Alzheimer’s disease (AD) may improve symptoms but have not been shown to prevent AD onset. Primary prevention therefore remains the goal. Although preventing AD by managing risk factors such as age or genetics is beyond our control (Box 1), we can do something about other factors.

This article summarizes the findings of many studies that address AD prevention and includes an online-only bibliography for readers seeking an in-depth review. The evidence does not support a firm recommendation for any specific form of primary prevention and has revealed hazards associated with estrogen therapy and nonsteroidal anti-inflammatory drugs (Box 2). Most important, it suggests that you could reduce your patients’ risk of developing AD by routinely supporting their mental, physical, and social health.

The potential benefits of modifying an individual’s AD risk factors likely will depend on his or her genetic makeup, environment, and lifestyle. Even so, counseling patients to exercise more and improve their diets—such as by eating more fish, fruits, and vegetables and less saturated fat—might help protect the brain. Your ongoing efforts to manage hypertension, hypercholesterolemia, and diabetes also may reduce their AD risk.

Box 1

Box 2

Cardiovascular risk factors

The risk of developing AD or vascular dementia appears to be increased by conditions that damage the heart or blood vessels. Recent evidence suggests that successfully managing cardiovascular risk factors may decrease the likelihood of dementia in later life.

Hypertension is associated with a higher risk of AD and all-cause dementia. Curiously, some studies have shown that low blood pressure also increases dementia risk, suggesting a U-shaped relationship between blood pressure and cognitive decline. Systolic hypertension in midlife may be associated with dementia 20 years later.

One might assume that antihypertensive therapy would help prevent dementia, but the data are conflicting. The Systolic Hypertension in Europe (SYST-EUR) study¹ showed a 53% reduction in vascular dementia or mixed dementia among patients receiving antihypertensive medication and a 60% reduction in AD. Similarly, the PROGRESS² clinical trial of prevention of recurrent stroke by antihypertensive treatment reported a 34% reduction in a composite measure of cognitive impairment and dementia. On the other hand, cognitive function neither improved nor worsened in the Hypertension in the Very Elderly Trial (HYVET-COG),³ whether patients received blood pressure treatment or placebo.

Hyperlipidemia. Lipid metabolism likely is an important pathway in amyloid beta-protein deposition, tau phosphorylation, and disruption of synaptic plasticity and neurodegenerative endpoints. Cognitive decline and incident dementia have been associated with higher dietary intake of saturated fats, partially hydrogenated unsaturated fatty acids (trans fats), and cholesterol. Not all studies have found this association, however. This could be because serum cholesterol levels may decrease in early dementia, limiting the ability to detect an effect of hypercholesterolemia on dementia risk when measurements are made later in life.

Using statins (3-hydroxy-3-methylglutaryl–coenzyme A reductase inhibitors) to treat hypercholesterolemia has been hypothesized to impede large vessel atherosclerosis and its consequences and to trigger metabolic effects in the brain related to AD pathogenesis. Mechanisms by which statins might help prevent dementia include:

• a direct association between amyloid processing and cholesterol in the brain
• an indirect effect by decreasing the risk of stroke, as even small cerebral infarcts worsen AD severity.

Nonrandomized epidemiologic studies such as the Cardiovascular Health Study⁴ and MRC/BHF Heart Protection Study⁵ suggested that statin treatment might reduce the incidence of dementia, the degree of age-related cognitive decline, and AD’s neuropathologic burden. Large, randomized, controlled trials have not supported these observations, however. Statins failed to reduce the incidence of dementia in:

• the Heart Protection Study, testing simvastatin for 5 years in 20,536 subjects age 40 to 80⁵
• the 3-year Preventive Study of Pravastatin in the Elderly at Risk (PROSPER) of 5,800 subjects.⁶

Similarly, patients receiving adjunctive atorvastatin or placebo showed no significant differences in cognition assessments after 72 weeks in the Lipitor’s Effect in Alzheimer’s Dementia (LEADe) study. This trial enrolled 640 subjects age 50 to 90 with mild-to-moderate dementia who were treated with donepezil.⁷ A recent Cochrane review concluded that high serum cholesterol may contribute to the development of AD and vascular dementia, but lowering cholesterol levels with statins does not prevent these problems.⁸

Diabetes mellitus. Diabetes and cognitive decline are closely associated. Diabetes is associated with a 50% to 100% increase in risk of AD and dementia overall and a 100% to 150% increased risk of vascular dementia. The mechanism by which diabetes increases dementia risk is uncertain but does not appear to be mediated entirely through vascular disease. High and low insulin levels may increase the risk of dementia, independent of diabetes and blood glucose. Increased peripheral insulin levels are associated with reduced brain atrophy and cognitive impairment in patients with early AD, suggesting a role for insulin signaling in AD pathophysiology. A possible relationship between insulin and beta amyloid metabolism is being studied.

Elevated postprandial plasma glucose has been associated with accelerated declines in cognitive performance.⁹ An inverse correlation has been noted between some cognitive measures and hemoglobin A1C levels.¹⁰ It is not clear that treating diabetes reduces the risk of dementia. In addition, in the prospective, population-based Rotterdam study, elderly patients with type 2 diabetes treated with insulin had the highest incidence of dementia.¹¹

Tobacco smoke directly affects neuronal function, integrity, and survival. Chronic smoking has been linked to decreased global cerebral blood flow, accelerated cerebral atrophy, and ventricular enlargement.

Some studies suggest an increased risk of dementia in middle-aged and elderly smokers, possibly through a cerebrovascular mechanism such as stroke. Other studies found no association between smoking and dementia risk, and 1 suggested that nicotine may protect against AD by reducing senile plaque formation. Any protective effect of smoking would be offset by increased risks of lung cancer, chronic obstructive pulmonary disease, and vascular dementia.

The apolipoprotein E epsilon 4 (APOE e4) gene may explain, at least in part, the conflicting results of these studies. In 2 population-based cohorts,^12,13 smoking was associated with memory decline in patients without, but not with, the APOE e4 genotype.

Dietary factors

Antioxidants. The brains of patients with AD contain elevated levels of endogenous antioxidants. In vitro studies show exogenous antioxidants can reduce the toxicity of beta-amyloid in brain tissue of persons with AD. These findings have led to interest in assessing the role of dietary antioxidants such as vitamins E and C for AD prevention.

High-dose alpha-tocopherol (vitamin E, 2,000 IU/d) may slow disease progression in patients with AD, but this association is not consistently found. Furthermore, a meta-analysis of 19 randomized controlled trials (RCTs) totaling >135,000 patients found an association between vitamin E doses >400 IU/d and increased all-cause mortality.¹⁴ High-dose vitamin E supplementation for primary or secondary prevention of AD may be dangerous and is not recommended.

The lack of consistent efficacy data for vitamin C in preventing or treating AD may discourage its routine use for this purpose.¹⁵

Homocysteine is a risk factor for stroke and heart disease. It also could play a role in vascular dementia through its association with large- and small-vessel disease.

Low folate and hyperhomocysteinemia have been associated with dementia or cognitive impairment, although a cause-effect relationship is not clear. In non-demented elderly populations, plasma homocysteine is inversely associated with poor performance in tests of global cognitive function, particularly in measures of psychomotor speed.

In a recent double-blind RCT, folic acid supplementation for 3 years significantly improved domains of cognitive function that tend to decline with age, especially information processing and sensorimotor speed.¹⁶ No other good evidence, however, has shown that homocysteine-lowering therapy using folic acid or other vitamin B supplements improves cognitive function or prevents cognitive decline.

Fish and omega-3 fatty acids. High total fat, saturated fat, and total cholesterol intake increases the risk for incident dementia. In epidemiologic studies, low omega-3 fatty acid serum levels have been linked to increased dementia risk.

Fish consumption may be beneficial in reducing the risk of dementia or cognitive decline. A prospective study of 815 elderly persons found 60% less risk of developing AD in those who ate ≥1 fish meal per week, compared with those who rarely or never ate fish.¹⁷ In the Framingham study, individuals who at baseline were in the top quartile of docosahexaenoic acid consumption had lower dementia rates over 9 years of follow-up.¹⁸ Results from cross-sectional and longitudinal studies have been inconsistent; some have shown that high intake of n-3 polyunsaturated fatty acids is associated with less cognitive decline,¹⁹ whereas others have not.²⁰

Although we cannot offer unequivocal advice regarding seafood or omega-3 fatty acid intake for primary prevention of dementia without evidence from RCTs, these uncontrolled studies show promise.

Mediterranean diet (MeDi) components include abundant fruits and vegetables, fish or shellfish at least twice weekly, very limited red meat, olive oil or canola oil instead of butter or margarine, tree nuts such as walnuts or pecans, red wine in moderation, and using herbs and spices instead of salt to season food. High adherence to the MeDi has been associated with a significantly lower risk for incident AD. The MeDi may affect the risk of developing AD²¹ as well as subsequent disease course, with a possible dose-response relationship in lower mortality.²²

Eating fruits and vegetables has been associated with improved cognitive performance²² and decreased incident dementia in elderly subjects.¹⁸

Alcohol. A U-shaped relationship exists between alcohol consumption and dementia risk. High alcohol intake is associated with clinical problem drinking and alcoholism and can lead to cognitive decline. Conversely, moderate wine consumption (250 to 500 mL/d) may be protective—compared with more or less than this amount—and is associated with approximately 50% less risk of dementia.

Alcohol use may increase the risk of dementia in persons carrying the APOE e4 allele, according to the population-based Cardiovascular Risk Factors, Aging and Dementia (CAIDE) study from Sweden.²³ After an average 21 years of follow-up of 1,449 individuals, researchers found that environmental factors—such as physical inactivity, dietary fat intake, alcohol consumption, and smoking at midlife—were associated with an increased risk of dementia at age 65 to 79 in APOE e4 carriers compared with noncarriers. The study also found that physical inactivity, dietary fat intake, and smoking at midlife increase AD risk, especially among APOE e4 carriers.

In the absence of evidence from RCTs, we cannot recommend alcohol to reduce the risk of AD.

Lifestyle and activity

Three components of lifestyle—social, mental, and physical activity—are inversely associated with the risk for dementia, AD, and cognitive impairment.

Physical exercise has been thought to enhance brain neurotrophic factor and modify apoptosis. Exercise can deter stroke and microvascular disease and improve regional cerebral blood flow. In the Cardiovascular Health Study, participants who expended the highest quartile of energy had a lower risk of all-cause dementia and AD compared with participants who expended the lowest quartile of energy.²⁴

Mental and social activity. Epidemiologic studies have shown associations between higher educational achievement and other socioeconomic factors and reduced AD risk. Advanced education is believed to represent a cognitive reserve that delays presentation of AD’s effects on memory and cognitive function, rather than providing a protective effect against accumulation of AD pathology. Higher-educated individuals appear to experience a somewhat more rapid rate of cognitive decline when AD does become apparent, perhaps because they have accumulated a greater degree of AD pathology at that point compared with less-educated persons.

Among 117 persons with dementia in the Bronx Aging Study, each additional year of formal education delayed the time of accelerated decline by 0.21 years. After accelerated decline began, each year of additional formal education was associated with a slightly faster rate of memory decline.²⁵

The longitudinal, population-based Kungsholmen Project in Stockholm, Sweden, found an association between daily mentally stimulating activities and decreased risk of all-cause dementia.²⁶ Similarly, higher levels of leisure activity were linked to reduced risk of all-cause dementia in a longitudinal study of 1,772 persons age ≥65 living in Manhattan, NY.²⁷ In a randomized, single-controlled study of the long-term effects of cognitive training, elderly individuals from 6 U.S. cities showed sustained improvement in specific cognitive performance up to 5 years after training sessions began, including memory, reasoning, and speed of processing.²⁸

It seems reasonable to encourage older patients to maintain or increase physical, cognitive, and leisure activities as well as social interaction. These interventions can improve the quality of life and lower the risk of depression, which may be a response to cognitive decline or an independent risk factor for dementia (Box 3). The Table lists “brain exercises” you can suggest to patients to increase their mental and social activity.

Head trauma. The Multi-Institutional Research in Alzheimer’s Genetic Epidemiology (MIRAGE) project found an association between AD risk and a history of head trauma, especially in persons with APOE e4 alleles.²⁹ Conversely, the Rotterdam Study showed no change in dementia risk for persons with a history of head trauma.³⁰

Even in the absence of conclusive evidence supporting AD prevention, protecting the head by buckling seat belts while driving, wearing helmets when participating in sports, and “fall-proofing” the home is recommended.

Box 3

Table

Brain exercises to suggest to patients

Related resource

• For an extensive bibliography of literature on Alzheimer’s disease risk factors and prevention, see this article at CurrentPsychiatry.com.

Drug brand names

• Atorvastatin • Lipitor
• Celecoxib • Celebrex
• Donepezil • Aricept
• Medroxyprogesterone • Provera
• Pravastatin • Pravachol
• Rofecoxib • Vioxx
• Simvastatin • Zocor

Disclosures

Dr. Bassil reports no financial relationship with any company whose products are mentioned in this article, or with manufacturers of competing products.

Dr. Grossberg receives research/grant support from and is a consultant to Bristol-Myers Squibb, Forest Pharmaceuticals, Novartis, Pfizer Inc., and Wyeth Pharmaceuticals. He also receives research/grant support from Baxter.

Discuss this article

Pharmacologic treatments for Alzheimer’s disease (AD) may improve symptoms but have not been shown to prevent AD onset. Primary prevention therefore remains the goal. Although preventing AD by managing risk factors such as age or genetics is beyond our control (Box 1), we can do something about other factors.

This article summarizes the findings of many studies that address AD prevention and includes an online-only bibliography for readers seeking an in-depth review. The evidence does not support a firm recommendation for any specific form of primary prevention and has revealed hazards associated with estrogen therapy and nonsteroidal anti-inflammatory drugs (Box 2). Most important, it suggests that you could reduce your patients’ risk of developing AD by routinely supporting their mental, physical, and social health.

The potential benefits of modifying an individual’s AD risk factors likely will depend on his or her genetic makeup, environment, and lifestyle. Even so, counseling patients to exercise more and improve their diets—such as by eating more fish, fruits, and vegetables and less saturated fat—might help protect the brain. Your ongoing efforts to manage hypertension, hypercholesterolemia, and diabetes also may reduce their AD risk.

Box 1

Box 2

Cardiovascular risk factors

The risk of developing AD or vascular dementia appears to be increased by conditions that damage the heart or blood vessels. Recent evidence suggests that successfully managing cardiovascular risk factors may decrease the likelihood of dementia in later life.

Hypertension is associated with a higher risk of AD and all-cause dementia. Curiously, some studies have shown that low blood pressure also increases dementia risk, suggesting a U-shaped relationship between blood pressure and cognitive decline. Systolic hypertension in midlife may be associated with dementia 20 years later.

One might assume that antihypertensive therapy would help prevent dementia, but the data are conflicting. The Systolic Hypertension in Europe (SYST-EUR) study¹ showed a 53% reduction in vascular dementia or mixed dementia among patients receiving antihypertensive medication and a 60% reduction in AD. Similarly, the PROGRESS² clinical trial of prevention of recurrent stroke by antihypertensive treatment reported a 34% reduction in a composite measure of cognitive impairment and dementia. On the other hand, cognitive function neither improved nor worsened in the Hypertension in the Very Elderly Trial (HYVET-COG),³ whether patients received blood pressure treatment or placebo.

Hyperlipidemia. Lipid metabolism likely is an important pathway in amyloid beta-protein deposition, tau phosphorylation, and disruption of synaptic plasticity and neurodegenerative endpoints. Cognitive decline and incident dementia have been associated with higher dietary intake of saturated fats, partially hydrogenated unsaturated fatty acids (trans fats), and cholesterol. Not all studies have found this association, however. This could be because serum cholesterol levels may decrease in early dementia, limiting the ability to detect an effect of hypercholesterolemia on dementia risk when measurements are made later in life.

Using statins (3-hydroxy-3-methylglutaryl–coenzyme A reductase inhibitors) to treat hypercholesterolemia has been hypothesized to impede large vessel atherosclerosis and its consequences and to trigger metabolic effects in the brain related to AD pathogenesis. Mechanisms by which statins might help prevent dementia include:

• a direct association between amyloid processing and cholesterol in the brain
• an indirect effect by decreasing the risk of stroke, as even small cerebral infarcts worsen AD severity.

Nonrandomized epidemiologic studies such as the Cardiovascular Health Study⁴ and MRC/BHF Heart Protection Study⁵ suggested that statin treatment might reduce the incidence of dementia, the degree of age-related cognitive decline, and AD’s neuropathologic burden. Large, randomized, controlled trials have not supported these observations, however. Statins failed to reduce the incidence of dementia in:

• the Heart Protection Study, testing simvastatin for 5 years in 20,536 subjects age 40 to 80⁵
• the 3-year Preventive Study of Pravastatin in the Elderly at Risk (PROSPER) of 5,800 subjects.⁶

Similarly, patients receiving adjunctive atorvastatin or placebo showed no significant differences in cognition assessments after 72 weeks in the Lipitor’s Effect in Alzheimer’s Dementia (LEADe) study. This trial enrolled 640 subjects age 50 to 90 with mild-to-moderate dementia who were treated with donepezil.⁷ A recent Cochrane review concluded that high serum cholesterol may contribute to the development of AD and vascular dementia, but lowering cholesterol levels with statins does not prevent these problems.⁸

Diabetes mellitus. Diabetes and cognitive decline are closely associated. Diabetes is associated with a 50% to 100% increase in risk of AD and dementia overall and a 100% to 150% increased risk of vascular dementia. The mechanism by which diabetes increases dementia risk is uncertain but does not appear to be mediated entirely through vascular disease. High and low insulin levels may increase the risk of dementia, independent of diabetes and blood glucose. Increased peripheral insulin levels are associated with reduced brain atrophy and cognitive impairment in patients with early AD, suggesting a role for insulin signaling in AD pathophysiology. A possible relationship between insulin and beta amyloid metabolism is being studied.

Elevated postprandial plasma glucose has been associated with accelerated declines in cognitive performance.⁹ An inverse correlation has been noted between some cognitive measures and hemoglobin A1C levels.¹⁰ It is not clear that treating diabetes reduces the risk of dementia. In addition, in the prospective, population-based Rotterdam study, elderly patients with type 2 diabetes treated with insulin had the highest incidence of dementia.¹¹

Tobacco smoke directly affects neuronal function, integrity, and survival. Chronic smoking has been linked to decreased global cerebral blood flow, accelerated cerebral atrophy, and ventricular enlargement.

Some studies suggest an increased risk of dementia in middle-aged and elderly smokers, possibly through a cerebrovascular mechanism such as stroke. Other studies found no association between smoking and dementia risk, and 1 suggested that nicotine may protect against AD by reducing senile plaque formation. Any protective effect of smoking would be offset by increased risks of lung cancer, chronic obstructive pulmonary disease, and vascular dementia.

The apolipoprotein E epsilon 4 (APOE e4) gene may explain, at least in part, the conflicting results of these studies. In 2 population-based cohorts,^12,13 smoking was associated with memory decline in patients without, but not with, the APOE e4 genotype.

Dietary factors

Antioxidants. The brains of patients with AD contain elevated levels of endogenous antioxidants. In vitro studies show exogenous antioxidants can reduce the toxicity of beta-amyloid in brain tissue of persons with AD. These findings have led to interest in assessing the role of dietary antioxidants such as vitamins E and C for AD prevention.

High-dose alpha-tocopherol (vitamin E, 2,000 IU/d) may slow disease progression in patients with AD, but this association is not consistently found. Furthermore, a meta-analysis of 19 randomized controlled trials (RCTs) totaling >135,000 patients found an association between vitamin E doses >400 IU/d and increased all-cause mortality.¹⁴ High-dose vitamin E supplementation for primary or secondary prevention of AD may be dangerous and is not recommended.

The lack of consistent efficacy data for vitamin C in preventing or treating AD may discourage its routine use for this purpose.¹⁵

Homocysteine is a risk factor for stroke and heart disease. It also could play a role in vascular dementia through its association with large- and small-vessel disease.

Low folate and hyperhomocysteinemia have been associated with dementia or cognitive impairment, although a cause-effect relationship is not clear. In non-demented elderly populations, plasma homocysteine is inversely associated with poor performance in tests of global cognitive function, particularly in measures of psychomotor speed.

In a recent double-blind RCT, folic acid supplementation for 3 years significantly improved domains of cognitive function that tend to decline with age, especially information processing and sensorimotor speed.¹⁶ No other good evidence, however, has shown that homocysteine-lowering therapy using folic acid or other vitamin B supplements improves cognitive function or prevents cognitive decline.

Fish and omega-3 fatty acids. High total fat, saturated fat, and total cholesterol intake increases the risk for incident dementia. In epidemiologic studies, low omega-3 fatty acid serum levels have been linked to increased dementia risk.

Fish consumption may be beneficial in reducing the risk of dementia or cognitive decline. A prospective study of 815 elderly persons found 60% less risk of developing AD in those who ate ≥1 fish meal per week, compared with those who rarely or never ate fish.¹⁷ In the Framingham study, individuals who at baseline were in the top quartile of docosahexaenoic acid consumption had lower dementia rates over 9 years of follow-up.¹⁸ Results from cross-sectional and longitudinal studies have been inconsistent; some have shown that high intake of n-3 polyunsaturated fatty acids is associated with less cognitive decline,¹⁹ whereas others have not.²⁰

Although we cannot offer unequivocal advice regarding seafood or omega-3 fatty acid intake for primary prevention of dementia without evidence from RCTs, these uncontrolled studies show promise.

Mediterranean diet (MeDi) components include abundant fruits and vegetables, fish or shellfish at least twice weekly, very limited red meat, olive oil or canola oil instead of butter or margarine, tree nuts such as walnuts or pecans, red wine in moderation, and using herbs and spices instead of salt to season food. High adherence to the MeDi has been associated with a significantly lower risk for incident AD. The MeDi may affect the risk of developing AD²¹ as well as subsequent disease course, with a possible dose-response relationship in lower mortality.²²

Eating fruits and vegetables has been associated with improved cognitive performance²² and decreased incident dementia in elderly subjects.¹⁸

Alcohol. A U-shaped relationship exists between alcohol consumption and dementia risk. High alcohol intake is associated with clinical problem drinking and alcoholism and can lead to cognitive decline. Conversely, moderate wine consumption (250 to 500 mL/d) may be protective—compared with more or less than this amount—and is associated with approximately 50% less risk of dementia.

Alcohol use may increase the risk of dementia in persons carrying the APOE e4 allele, according to the population-based Cardiovascular Risk Factors, Aging and Dementia (CAIDE) study from Sweden.²³ After an average 21 years of follow-up of 1,449 individuals, researchers found that environmental factors—such as physical inactivity, dietary fat intake, alcohol consumption, and smoking at midlife—were associated with an increased risk of dementia at age 65 to 79 in APOE e4 carriers compared with noncarriers. The study also found that physical inactivity, dietary fat intake, and smoking at midlife increase AD risk, especially among APOE e4 carriers.

In the absence of evidence from RCTs, we cannot recommend alcohol to reduce the risk of AD.

Lifestyle and activity

Three components of lifestyle—social, mental, and physical activity—are inversely associated with the risk for dementia, AD, and cognitive impairment.

Physical exercise has been thought to enhance brain neurotrophic factor and modify apoptosis. Exercise can deter stroke and microvascular disease and improve regional cerebral blood flow. In the Cardiovascular Health Study, participants who expended the highest quartile of energy had a lower risk of all-cause dementia and AD compared with participants who expended the lowest quartile of energy.²⁴

Mental and social activity. Epidemiologic studies have shown associations between higher educational achievement and other socioeconomic factors and reduced AD risk. Advanced education is believed to represent a cognitive reserve that delays presentation of AD’s effects on memory and cognitive function, rather than providing a protective effect against accumulation of AD pathology. Higher-educated individuals appear to experience a somewhat more rapid rate of cognitive decline when AD does become apparent, perhaps because they have accumulated a greater degree of AD pathology at that point compared with less-educated persons.

Among 117 persons with dementia in the Bronx Aging Study, each additional year of formal education delayed the time of accelerated decline by 0.21 years. After accelerated decline began, each year of additional formal education was associated with a slightly faster rate of memory decline.²⁵

The longitudinal, population-based Kungsholmen Project in Stockholm, Sweden, found an association between daily mentally stimulating activities and decreased risk of all-cause dementia.²⁶ Similarly, higher levels of leisure activity were linked to reduced risk of all-cause dementia in a longitudinal study of 1,772 persons age ≥65 living in Manhattan, NY.²⁷ In a randomized, single-controlled study of the long-term effects of cognitive training, elderly individuals from 6 U.S. cities showed sustained improvement in specific cognitive performance up to 5 years after training sessions began, including memory, reasoning, and speed of processing.²⁸

It seems reasonable to encourage older patients to maintain or increase physical, cognitive, and leisure activities as well as social interaction. These interventions can improve the quality of life and lower the risk of depression, which may be a response to cognitive decline or an independent risk factor for dementia (Box 3). The Table lists “brain exercises” you can suggest to patients to increase their mental and social activity.

Head trauma. The Multi-Institutional Research in Alzheimer’s Genetic Epidemiology (MIRAGE) project found an association between AD risk and a history of head trauma, especially in persons with APOE e4 alleles.²⁹ Conversely, the Rotterdam Study showed no change in dementia risk for persons with a history of head trauma.³⁰

Even in the absence of conclusive evidence supporting AD prevention, protecting the head by buckling seat belts while driving, wearing helmets when participating in sports, and “fall-proofing” the home is recommended.

Box 3

Table

Brain exercises to suggest to patients

Related resource

• For an extensive bibliography of literature on Alzheimer’s disease risk factors and prevention, see this article at CurrentPsychiatry.com.

Drug brand names

• Atorvastatin • Lipitor
• Celecoxib • Celebrex
• Donepezil • Aricept
• Medroxyprogesterone • Provera
• Pravastatin • Pravachol
• Rofecoxib • Vioxx
• Simvastatin • Zocor

Disclosures

Dr. Bassil reports no financial relationship with any company whose products are mentioned in this article, or with manufacturers of competing products.

Dr. Grossberg receives research/grant support from and is a consultant to Bristol-Myers Squibb, Forest Pharmaceuticals, Novartis, Pfizer Inc., and Wyeth Pharmaceuticals. He also receives research/grant support from Baxter.

REM sleep behavior disorder (RBD) patients act out their dreams during sleep and could injure themselves or bed partner. In RBD, loss of skeletal muscle atonia during rapid eye movement (REM) sleep allows the patient’s motor activity to reflect dream content. During sleep, patients appear to punch, kick, or choke a bed partner or jump out of bed.

RBD is more common in elderly males and individuals with neurodegenerative disorders of alpha-synuclein accumulation, such as Parkinson’s disease, Lewy body dementia, and multiple system atrophy.¹ RBD may be a precursor to these diseases.

Most antidepressants can cause or increase RBD movements.² RBD also is associated with sedative-hypnotic withdrawal.

Differential diagnosis

When patients report striking out while asleep, differential diagnoses include RBD, periodic limb movement disorder (PLMD), sleepwalking disorder, and restless legs syndrome (RLS). Polysomnography with electromyography may distinguish among these disorders.

PLMD movements are repetitive, stereotyped motions of the foot and leg, and manifest as a repetitive partial flexion of the joints of the great toe, ankle, knee, and occasionally hip. Upper extremity movements are less common. Movements appear similar to myoclonus. Periodic limb movements occur in rhythmic patterns, every 20 to 60 seconds, continuing for 10 minutes to several hours.

Sleepwalking disorder movements occur without an associated dream during non-REM sleep. Individuals with RBD may jump out of bed, but usually don’t walk in their sleep.

RLS movement occurs prior to and in early stages of sleep, whereas in RBD, PLMD, and sleepwalking disorder, motor activity is limited to sleep. Patients perceive unpleasant sensations and an urge to move the feet and legs. Movement temporarily soothes these uncomfortable sensations. Patients are aware of these sensations before sleep; however, RBD patients are not conscious of movements until they wake and find themselves acting out a dream. RLS and PLMD often are comorbid.

Treatment

Clonazepam is most effective for RBD; however, also consider lorazepam, pramipexole, or melatonin. If clinically feasible, consider discontinuing antidepressants because this may decrease movements.³

To reduce risk of injury, recommend sleeping in separate beds, moving objects away from the bed, or padding the headboard and floor. Encourage patients with severe RBD to sleep in a zipped sleeping bag.

REM sleep behavior disorder (RBD) patients act out their dreams during sleep and could injure themselves or bed partner. In RBD, loss of skeletal muscle atonia during rapid eye movement (REM) sleep allows the patient’s motor activity to reflect dream content. During sleep, patients appear to punch, kick, or choke a bed partner or jump out of bed.

RBD is more common in elderly males and individuals with neurodegenerative disorders of alpha-synuclein accumulation, such as Parkinson’s disease, Lewy body dementia, and multiple system atrophy.¹ RBD may be a precursor to these diseases.

Most antidepressants can cause or increase RBD movements.² RBD also is associated with sedative-hypnotic withdrawal.

Differential diagnosis

When patients report striking out while asleep, differential diagnoses include RBD, periodic limb movement disorder (PLMD), sleepwalking disorder, and restless legs syndrome (RLS). Polysomnography with electromyography may distinguish among these disorders.

PLMD movements are repetitive, stereotyped motions of the foot and leg, and manifest as a repetitive partial flexion of the joints of the great toe, ankle, knee, and occasionally hip. Upper extremity movements are less common. Movements appear similar to myoclonus. Periodic limb movements occur in rhythmic patterns, every 20 to 60 seconds, continuing for 10 minutes to several hours.

Sleepwalking disorder movements occur without an associated dream during non-REM sleep. Individuals with RBD may jump out of bed, but usually don’t walk in their sleep.

RLS movement occurs prior to and in early stages of sleep, whereas in RBD, PLMD, and sleepwalking disorder, motor activity is limited to sleep. Patients perceive unpleasant sensations and an urge to move the feet and legs. Movement temporarily soothes these uncomfortable sensations. Patients are aware of these sensations before sleep; however, RBD patients are not conscious of movements until they wake and find themselves acting out a dream. RLS and PLMD often are comorbid.

Treatment

Clonazepam is most effective for RBD; however, also consider lorazepam, pramipexole, or melatonin. If clinically feasible, consider discontinuing antidepressants because this may decrease movements.³

To reduce risk of injury, recommend sleeping in separate beds, moving objects away from the bed, or padding the headboard and floor. Encourage patients with severe RBD to sleep in a zipped sleeping bag.

REM sleep behavior disorder (RBD) patients act out their dreams during sleep and could injure themselves or bed partner. In RBD, loss of skeletal muscle atonia during rapid eye movement (REM) sleep allows the patient’s motor activity to reflect dream content. During sleep, patients appear to punch, kick, or choke a bed partner or jump out of bed.

RBD is more common in elderly males and individuals with neurodegenerative disorders of alpha-synuclein accumulation, such as Parkinson’s disease, Lewy body dementia, and multiple system atrophy.¹ RBD may be a precursor to these diseases.

Most antidepressants can cause or increase RBD movements.² RBD also is associated with sedative-hypnotic withdrawal.

Differential diagnosis

When patients report striking out while asleep, differential diagnoses include RBD, periodic limb movement disorder (PLMD), sleepwalking disorder, and restless legs syndrome (RLS). Polysomnography with electromyography may distinguish among these disorders.

PLMD movements are repetitive, stereotyped motions of the foot and leg, and manifest as a repetitive partial flexion of the joints of the great toe, ankle, knee, and occasionally hip. Upper extremity movements are less common. Movements appear similar to myoclonus. Periodic limb movements occur in rhythmic patterns, every 20 to 60 seconds, continuing for 10 minutes to several hours.

Sleepwalking disorder movements occur without an associated dream during non-REM sleep. Individuals with RBD may jump out of bed, but usually don’t walk in their sleep.

RLS movement occurs prior to and in early stages of sleep, whereas in RBD, PLMD, and sleepwalking disorder, motor activity is limited to sleep. Patients perceive unpleasant sensations and an urge to move the feet and legs. Movement temporarily soothes these uncomfortable sensations. Patients are aware of these sensations before sleep; however, RBD patients are not conscious of movements until they wake and find themselves acting out a dream. RLS and PLMD often are comorbid.

Treatment

Clonazepam is most effective for RBD; however, also consider lorazepam, pramipexole, or melatonin. If clinically feasible, consider discontinuing antidepressants because this may decrease movements.³

To reduce risk of injury, recommend sleeping in separate beds, moving objects away from the bed, or padding the headboard and floor. Encourage patients with severe RBD to sleep in a zipped sleeping bag.

CASE: Stress and chest pain

A primary care physician refers Mr. J, age 40, to our mental health clinic for evaluation of anxiety symptoms. Almost a decade ago Mr. J presented to his primary care physician with anxiety and panic attacks that included chest pain and shortness of breath. Various pharmacologic treatments, including paroxetine, were only moderately successful until 4 years ago, when Mr. J began nighttime continuous positive airway pressure (CPAP) therapy and pramipexole, 0.25 to 0.5 mg/d, for obstructive sleep apnea (OSA), at which point his anxiety completely resolved.

Mr. J reported no anxiety for many years, but when shortness of breath, palpitations, and chest pain re-emerge, he consults his primary care physician. After a negative workup for myocardial infarction, Mr. J is started on short-term beta-blocker therapy and restarted on paroxetine, 20 mg/d. A sleep medicine specialist repeats polysomnography and makes slight adjustments to Mr. J’s CPAP therapy. Mr. J relocates to our city and his new primary care physician refers Mr. J to our mental health clinic.

In addition to OSA, Mr. J has mild anemia, hyperlipidemia, and vitamin D deficiency. Mr. J was adopted and has no knowledge of his family psychiatric or medical history. His mental status is normal. Mr. J is not obese, exercises regularly, and has slight micrognathia. His current medications include paroxetine, 20 mg/d, modafinil, 200 mg/d, and ergocalciferol, 50,000 units/week for vitamin D deficiency.

Mr. J says he experienced a single panic attack 7 months ago, but none since then. However, he complains of chronic chest pressure and mild intermittent anxiety associated with the stress of his new job and recent relocation.

The authors’ observations

Mr. J’s anxiety resolved fully only after receiving treatment for OSA, which is characterized by episodes of blocked breathing during sleep (Table 1).¹ Multiple studies show a significant association between OSA and panic attacks.^2-5 In a survey of 301 sleep apnea patients, Edlund et al⁶ demonstrated that OSA may cause nocturnal panic attacks. Untreated OSA can worsen anxiety symptoms. In a study of 242 OSA patients, those who were not compliant with CPAP therapy had significantly higher anxiety scores as measured on the Hospital Anxiety and Depression Scale_.⁷

OSA treatment options include CPAP, oral appliance, and surgery; weight loss and positional therapy may help. Thyroid function, B12, folate, ferritin, and iron studies, and complete blood count can rule out secondary causes of OSA.

Table 1

Obstructive sleep apnea risk factors, symptoms, and features

HISTORY: A succession of diagnoses

Approximately 9 years ago, Mr. J experienced several episodes of waking in the middle of the night from a bad dream with severe shortness of breath and chest pain. He also reported increasing fatigue, anxiety, and stress regarding work, graduate school, and his wife’s recent miscarriage. After negative cardiac workups, his primary care physician diagnosed panic attacks. He referred Mr. J to stress management classes and prescribed clonazepam, 1.5 mg/d, which was discontinued after 2 months.

One week after discontinuing clonazepam, Mr. J experienced chest pain, shortness of breath, and anxiety while awake. A cardiologist ruled out cardiac pathology. Mr. J’s primary care physician prescribed sertraline, 25 mg/d, and propranolol, 60 mg/d and 10 mg as needed, for anxiety.

Shortly after, Mr. J moved to a different city. His new primary care physician discontinued sertraline and propranolol and started paroxetine, titrated to 20 mg/d. A psychiatrist diagnosed Mr. J with panic disorder without agoraphobia, continued paroxetine, and added alprazolam, 1 mg/d as needed. Mr. J’s anxiety symptoms were moderately controlled for several years.

After his son was diagnosed with attention-deficit/hyperactivity disorder (ADHD), Mr. J also was evaluated and found to have ADHD and major depressive disorder, single episode. Mr. J received methylphenidate, 54 mg/d, and paroxetine was titrated to 40 mg/d, with moderate results.

Approximately 6 years before presenting to our clinic, Mr. J reported worsening daytime fatigue, which was treated with modafinil, 200 mg/d. He experienced significant improvement. The next year methylphenidate was switched to amphetamine/dextroamphetamine, then discontinued because of side effects. His physician started Mr. J on atomoxetine, which also was discontinued because of side effects.

Two years later, Mr. J complained of gradual worsening daytime sleepiness. Polysomnography revealed that Mr. J had severe OSA and periodic limb movement disorder. After he began nighttime CPAP and pramipexole, 0.25 to 0.5 mg/d, and continued modafinil, 200 mg/d, his anxiety symptoms completely resolved. Several months later Mr. J’s physician discontinued paroxetine because Mr. J reported it caused mildly decreased concentration.

The authors’ observations

The etiology of Mr. J’s anxiety is unclear; however, he does not meet criteria for:

• panic disorder, because he denies persistent concern about having more attacks or the implications or consequences of panic attacks, or significant change in behavior related to panic attacks (Table 2)⁸
• generalized anxiety disorder, because between panic attacks Mr. J’s baseline anxiety related to “real-world” stressors is mild, intermittent, and easily controllable⁸
• substance-induced anxiety disorder, because Mr. J denies using caffeine, tobacco, alcohol, or illicit drugs. Also, for many years he worked for a company that performed random drug screening.

Table 2

Diagnostic criteria for panic disorder without agoraphobia

Although it is difficult to draw a conclusion from a single case, Mr. J’s dramatic improvement with CPAP warrants speculation about possible etiologic relationships among daytime panic attacks, nighttime panic attacks, and OSA.

According to DSM-IV-TR, a panic attack has a distinct period of intense fear or discomfort (Table 3).⁸ Recurrent panic attacks can lead to anticipatory anxiety, which is an intense fear and/or dread of having another panic attack.⁹ According to Steven Reiss’ expectancy theory, anxiety sensitivity—ie, the fear of anxiety or fear of fear—may be a risk factor for panic disorder.¹⁰ Therefore, past panic attacks may increase the likelihood of future panic attacks.

Table 3

Diagnostic criteria for panic attack*

Mr. J’s panic symptoms may be caused by multiple OSA-induced nocturnal panic attacks. These nighttime panic attacks may predispose him to daytime attacks. It is possible that Mr. J had subclinical panic disorder before developing OSA. In this scenario, his OSA-induced nocturnal panic attacks may have worsened his panic disorder. Unfortunately, we do not know precisely how long Mr. J has had OSA—only that he was diagnosed with the condition 4 years before presenting to our clinic.

Mr. J responded moderately to paroxetine monotherapy but experienced rapid resolution of his panic attacks with a combination of paroxetine and CPAP. CPAP monotherapy sufficiently prevented panic attacks for 4 years. Finally, when Mr. J experienced a single panic attack several months before presenting to our clinic—at the end of a very stressful year—reintroducing paroxetine prevented subsequent attacks. This supports our hypothesis that OSA may predispose or indirectly cause patients to develop daytime panic attacks. Alternately, this case suggests that OSA may cause subclinical panic disorder to present as an acute condition.

We rule out anxiety disorder secondary to a general medical condition (OSA) and diagnose Mr. J with anxiety disorder not otherwise specified.

The authors’ observations

We continue paroxetine at 20 mg/d because it was working fairly well with minimal side effects. The sleep medicine specialist maintained modafinil, 200 mg/d. Laboratory studies—including a comprehensive metabolic panel, complete blood count with differential, and thyroid stimulating hormone—were within normal limits except a fasting blood glucose of 123 mg/dL, for which we referred Mr. J to his primary care physician.

OUTCOME: Discontinue paroxetine?

One month later, Mr. J denies panic attacks, other anxiety symptoms, or other psychiatric symptoms and is sleeping well. However, he reports that his mildly decreased concentration persists and he wants to stop paroxetine.

After discussing the risks and benefits, Mr. J and the treatment team decide to continue paroxetine at 20 mg/d. We cite peer-reviewed literature that recommends continuing antidepressants for at least 1 year and possibly indefinitely after symptom resolution to control panic disorder symptoms.⁹ In addition, we discuss the lack of studies comparing different lengths of treatment with SSRIs for apparent OSA-induced panic attacks that respond to SSRI/CPAP therapy. Because Mr. J was doing well and experiencing minimal side effects, he feels he would be better served with a longer period of psychopharmacologic treatment.

Six months later, Mr. J says his anxiety symptoms are well controlled and generally unchanged except for an occasional “little flutter” of anxiety every 3 or 4 days that lasts several seconds. For 1 year, he reports no recurrence of panic attacks, compliance with CPAP, and stable OSA.

Related resource

• Saunamäki T, Jehkonen M. Depression and anxiety in obstructive sleep apnea syndrome: a review. Acta Neurol Scand. 2007;116(5):277-288.

Drug brand names

• Alprazolam • Xanax
• Amphetamine/dextroamphetamine • Adderall
• Atomoxetine • Strattera
• Clonazepam • Klonopin
• Ergocalciferol • Calciferol
• Modafinil • Provigil
• Methylphenidate extended release • Concerta
• Paroxetine • Paxil
• Pramipexole • Mirapex
• Propranolol • Inderal
• Sertraline • Zoloft

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: Stress and chest pain

A primary care physician refers Mr. J, age 40, to our mental health clinic for evaluation of anxiety symptoms. Almost a decade ago Mr. J presented to his primary care physician with anxiety and panic attacks that included chest pain and shortness of breath. Various pharmacologic treatments, including paroxetine, were only moderately successful until 4 years ago, when Mr. J began nighttime continuous positive airway pressure (CPAP) therapy and pramipexole, 0.25 to 0.5 mg/d, for obstructive sleep apnea (OSA), at which point his anxiety completely resolved.

Mr. J reported no anxiety for many years, but when shortness of breath, palpitations, and chest pain re-emerge, he consults his primary care physician. After a negative workup for myocardial infarction, Mr. J is started on short-term beta-blocker therapy and restarted on paroxetine, 20 mg/d. A sleep medicine specialist repeats polysomnography and makes slight adjustments to Mr. J’s CPAP therapy. Mr. J relocates to our city and his new primary care physician refers Mr. J to our mental health clinic.

In addition to OSA, Mr. J has mild anemia, hyperlipidemia, and vitamin D deficiency. Mr. J was adopted and has no knowledge of his family psychiatric or medical history. His mental status is normal. Mr. J is not obese, exercises regularly, and has slight micrognathia. His current medications include paroxetine, 20 mg/d, modafinil, 200 mg/d, and ergocalciferol, 50,000 units/week for vitamin D deficiency.

Mr. J says he experienced a single panic attack 7 months ago, but none since then. However, he complains of chronic chest pressure and mild intermittent anxiety associated with the stress of his new job and recent relocation.

The authors’ observations

Mr. J’s anxiety resolved fully only after receiving treatment for OSA, which is characterized by episodes of blocked breathing during sleep (Table 1).¹ Multiple studies show a significant association between OSA and panic attacks.^2-5 In a survey of 301 sleep apnea patients, Edlund et al⁶ demonstrated that OSA may cause nocturnal panic attacks. Untreated OSA can worsen anxiety symptoms. In a study of 242 OSA patients, those who were not compliant with CPAP therapy had significantly higher anxiety scores as measured on the Hospital Anxiety and Depression Scale_.⁷

OSA treatment options include CPAP, oral appliance, and surgery; weight loss and positional therapy may help. Thyroid function, B12, folate, ferritin, and iron studies, and complete blood count can rule out secondary causes of OSA.

Table 1

Obstructive sleep apnea risk factors, symptoms, and features

HISTORY: A succession of diagnoses

Approximately 9 years ago, Mr. J experienced several episodes of waking in the middle of the night from a bad dream with severe shortness of breath and chest pain. He also reported increasing fatigue, anxiety, and stress regarding work, graduate school, and his wife’s recent miscarriage. After negative cardiac workups, his primary care physician diagnosed panic attacks. He referred Mr. J to stress management classes and prescribed clonazepam, 1.5 mg/d, which was discontinued after 2 months.

One week after discontinuing clonazepam, Mr. J experienced chest pain, shortness of breath, and anxiety while awake. A cardiologist ruled out cardiac pathology. Mr. J’s primary care physician prescribed sertraline, 25 mg/d, and propranolol, 60 mg/d and 10 mg as needed, for anxiety.

Shortly after, Mr. J moved to a different city. His new primary care physician discontinued sertraline and propranolol and started paroxetine, titrated to 20 mg/d. A psychiatrist diagnosed Mr. J with panic disorder without agoraphobia, continued paroxetine, and added alprazolam, 1 mg/d as needed. Mr. J’s anxiety symptoms were moderately controlled for several years.

After his son was diagnosed with attention-deficit/hyperactivity disorder (ADHD), Mr. J also was evaluated and found to have ADHD and major depressive disorder, single episode. Mr. J received methylphenidate, 54 mg/d, and paroxetine was titrated to 40 mg/d, with moderate results.

Approximately 6 years before presenting to our clinic, Mr. J reported worsening daytime fatigue, which was treated with modafinil, 200 mg/d. He experienced significant improvement. The next year methylphenidate was switched to amphetamine/dextroamphetamine, then discontinued because of side effects. His physician started Mr. J on atomoxetine, which also was discontinued because of side effects.

Two years later, Mr. J complained of gradual worsening daytime sleepiness. Polysomnography revealed that Mr. J had severe OSA and periodic limb movement disorder. After he began nighttime CPAP and pramipexole, 0.25 to 0.5 mg/d, and continued modafinil, 200 mg/d, his anxiety symptoms completely resolved. Several months later Mr. J’s physician discontinued paroxetine because Mr. J reported it caused mildly decreased concentration.

The authors’ observations

The etiology of Mr. J’s anxiety is unclear; however, he does not meet criteria for:

• panic disorder, because he denies persistent concern about having more attacks or the implications or consequences of panic attacks, or significant change in behavior related to panic attacks (Table 2)⁸
• generalized anxiety disorder, because between panic attacks Mr. J’s baseline anxiety related to “real-world” stressors is mild, intermittent, and easily controllable⁸
• substance-induced anxiety disorder, because Mr. J denies using caffeine, tobacco, alcohol, or illicit drugs. Also, for many years he worked for a company that performed random drug screening.

Table 2

Diagnostic criteria for panic disorder without agoraphobia

Although it is difficult to draw a conclusion from a single case, Mr. J’s dramatic improvement with CPAP warrants speculation about possible etiologic relationships among daytime panic attacks, nighttime panic attacks, and OSA.

According to DSM-IV-TR, a panic attack has a distinct period of intense fear or discomfort (Table 3).⁸ Recurrent panic attacks can lead to anticipatory anxiety, which is an intense fear and/or dread of having another panic attack.⁹ According to Steven Reiss’ expectancy theory, anxiety sensitivity—ie, the fear of anxiety or fear of fear—may be a risk factor for panic disorder.¹⁰ Therefore, past panic attacks may increase the likelihood of future panic attacks.

Table 3

Diagnostic criteria for panic attack*

Mr. J’s panic symptoms may be caused by multiple OSA-induced nocturnal panic attacks. These nighttime panic attacks may predispose him to daytime attacks. It is possible that Mr. J had subclinical panic disorder before developing OSA. In this scenario, his OSA-induced nocturnal panic attacks may have worsened his panic disorder. Unfortunately, we do not know precisely how long Mr. J has had OSA—only that he was diagnosed with the condition 4 years before presenting to our clinic.

Mr. J responded moderately to paroxetine monotherapy but experienced rapid resolution of his panic attacks with a combination of paroxetine and CPAP. CPAP monotherapy sufficiently prevented panic attacks for 4 years. Finally, when Mr. J experienced a single panic attack several months before presenting to our clinic—at the end of a very stressful year—reintroducing paroxetine prevented subsequent attacks. This supports our hypothesis that OSA may predispose or indirectly cause patients to develop daytime panic attacks. Alternately, this case suggests that OSA may cause subclinical panic disorder to present as an acute condition.

We rule out anxiety disorder secondary to a general medical condition (OSA) and diagnose Mr. J with anxiety disorder not otherwise specified.

The authors’ observations

We continue paroxetine at 20 mg/d because it was working fairly well with minimal side effects. The sleep medicine specialist maintained modafinil, 200 mg/d. Laboratory studies—including a comprehensive metabolic panel, complete blood count with differential, and thyroid stimulating hormone—were within normal limits except a fasting blood glucose of 123 mg/dL, for which we referred Mr. J to his primary care physician.

OUTCOME: Discontinue paroxetine?

One month later, Mr. J denies panic attacks, other anxiety symptoms, or other psychiatric symptoms and is sleeping well. However, he reports that his mildly decreased concentration persists and he wants to stop paroxetine.

After discussing the risks and benefits, Mr. J and the treatment team decide to continue paroxetine at 20 mg/d. We cite peer-reviewed literature that recommends continuing antidepressants for at least 1 year and possibly indefinitely after symptom resolution to control panic disorder symptoms.⁹ In addition, we discuss the lack of studies comparing different lengths of treatment with SSRIs for apparent OSA-induced panic attacks that respond to SSRI/CPAP therapy. Because Mr. J was doing well and experiencing minimal side effects, he feels he would be better served with a longer period of psychopharmacologic treatment.

Six months later, Mr. J says his anxiety symptoms are well controlled and generally unchanged except for an occasional “little flutter” of anxiety every 3 or 4 days that lasts several seconds. For 1 year, he reports no recurrence of panic attacks, compliance with CPAP, and stable OSA.

Related resource

• Saunamäki T, Jehkonen M. Depression and anxiety in obstructive sleep apnea syndrome: a review. Acta Neurol Scand. 2007;116(5):277-288.

Drug brand names

• Alprazolam • Xanax
• Amphetamine/dextroamphetamine • Adderall
• Atomoxetine • Strattera
• Clonazepam • Klonopin
• Ergocalciferol • Calciferol
• Modafinil • Provigil
• Methylphenidate extended release • Concerta
• Paroxetine • Paxil
• Pramipexole • Mirapex
• Propranolol • Inderal
• Sertraline • Zoloft

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: Stress and chest pain

A primary care physician refers Mr. J, age 40, to our mental health clinic for evaluation of anxiety symptoms. Almost a decade ago Mr. J presented to his primary care physician with anxiety and panic attacks that included chest pain and shortness of breath. Various pharmacologic treatments, including paroxetine, were only moderately successful until 4 years ago, when Mr. J began nighttime continuous positive airway pressure (CPAP) therapy and pramipexole, 0.25 to 0.5 mg/d, for obstructive sleep apnea (OSA), at which point his anxiety completely resolved.

Mr. J reported no anxiety for many years, but when shortness of breath, palpitations, and chest pain re-emerge, he consults his primary care physician. After a negative workup for myocardial infarction, Mr. J is started on short-term beta-blocker therapy and restarted on paroxetine, 20 mg/d. A sleep medicine specialist repeats polysomnography and makes slight adjustments to Mr. J’s CPAP therapy. Mr. J relocates to our city and his new primary care physician refers Mr. J to our mental health clinic.

In addition to OSA, Mr. J has mild anemia, hyperlipidemia, and vitamin D deficiency. Mr. J was adopted and has no knowledge of his family psychiatric or medical history. His mental status is normal. Mr. J is not obese, exercises regularly, and has slight micrognathia. His current medications include paroxetine, 20 mg/d, modafinil, 200 mg/d, and ergocalciferol, 50,000 units/week for vitamin D deficiency.

Mr. J says he experienced a single panic attack 7 months ago, but none since then. However, he complains of chronic chest pressure and mild intermittent anxiety associated with the stress of his new job and recent relocation.

The authors’ observations

Mr. J’s anxiety resolved fully only after receiving treatment for OSA, which is characterized by episodes of blocked breathing during sleep (Table 1).¹ Multiple studies show a significant association between OSA and panic attacks.^2-5 In a survey of 301 sleep apnea patients, Edlund et al⁶ demonstrated that OSA may cause nocturnal panic attacks. Untreated OSA can worsen anxiety symptoms. In a study of 242 OSA patients, those who were not compliant with CPAP therapy had significantly higher anxiety scores as measured on the Hospital Anxiety and Depression Scale_.⁷

OSA treatment options include CPAP, oral appliance, and surgery; weight loss and positional therapy may help. Thyroid function, B12, folate, ferritin, and iron studies, and complete blood count can rule out secondary causes of OSA.

Table 1

Obstructive sleep apnea risk factors, symptoms, and features

HISTORY: A succession of diagnoses

Approximately 9 years ago, Mr. J experienced several episodes of waking in the middle of the night from a bad dream with severe shortness of breath and chest pain. He also reported increasing fatigue, anxiety, and stress regarding work, graduate school, and his wife’s recent miscarriage. After negative cardiac workups, his primary care physician diagnosed panic attacks. He referred Mr. J to stress management classes and prescribed clonazepam, 1.5 mg/d, which was discontinued after 2 months.

One week after discontinuing clonazepam, Mr. J experienced chest pain, shortness of breath, and anxiety while awake. A cardiologist ruled out cardiac pathology. Mr. J’s primary care physician prescribed sertraline, 25 mg/d, and propranolol, 60 mg/d and 10 mg as needed, for anxiety.

Shortly after, Mr. J moved to a different city. His new primary care physician discontinued sertraline and propranolol and started paroxetine, titrated to 20 mg/d. A psychiatrist diagnosed Mr. J with panic disorder without agoraphobia, continued paroxetine, and added alprazolam, 1 mg/d as needed. Mr. J’s anxiety symptoms were moderately controlled for several years.

After his son was diagnosed with attention-deficit/hyperactivity disorder (ADHD), Mr. J also was evaluated and found to have ADHD and major depressive disorder, single episode. Mr. J received methylphenidate, 54 mg/d, and paroxetine was titrated to 40 mg/d, with moderate results.

Approximately 6 years before presenting to our clinic, Mr. J reported worsening daytime fatigue, which was treated with modafinil, 200 mg/d. He experienced significant improvement. The next year methylphenidate was switched to amphetamine/dextroamphetamine, then discontinued because of side effects. His physician started Mr. J on atomoxetine, which also was discontinued because of side effects.

Two years later, Mr. J complained of gradual worsening daytime sleepiness. Polysomnography revealed that Mr. J had severe OSA and periodic limb movement disorder. After he began nighttime CPAP and pramipexole, 0.25 to 0.5 mg/d, and continued modafinil, 200 mg/d, his anxiety symptoms completely resolved. Several months later Mr. J’s physician discontinued paroxetine because Mr. J reported it caused mildly decreased concentration.

The authors’ observations

The etiology of Mr. J’s anxiety is unclear; however, he does not meet criteria for:

• panic disorder, because he denies persistent concern about having more attacks or the implications or consequences of panic attacks, or significant change in behavior related to panic attacks (Table 2)⁸
• generalized anxiety disorder, because between panic attacks Mr. J’s baseline anxiety related to “real-world” stressors is mild, intermittent, and easily controllable⁸
• substance-induced anxiety disorder, because Mr. J denies using caffeine, tobacco, alcohol, or illicit drugs. Also, for many years he worked for a company that performed random drug screening.

Table 2

Diagnostic criteria for panic disorder without agoraphobia

Although it is difficult to draw a conclusion from a single case, Mr. J’s dramatic improvement with CPAP warrants speculation about possible etiologic relationships among daytime panic attacks, nighttime panic attacks, and OSA.

According to DSM-IV-TR, a panic attack has a distinct period of intense fear or discomfort (Table 3).⁸ Recurrent panic attacks can lead to anticipatory anxiety, which is an intense fear and/or dread of having another panic attack.⁹ According to Steven Reiss’ expectancy theory, anxiety sensitivity—ie, the fear of anxiety or fear of fear—may be a risk factor for panic disorder.¹⁰ Therefore, past panic attacks may increase the likelihood of future panic attacks.

Table 3

Diagnostic criteria for panic attack*

Mr. J’s panic symptoms may be caused by multiple OSA-induced nocturnal panic attacks. These nighttime panic attacks may predispose him to daytime attacks. It is possible that Mr. J had subclinical panic disorder before developing OSA. In this scenario, his OSA-induced nocturnal panic attacks may have worsened his panic disorder. Unfortunately, we do not know precisely how long Mr. J has had OSA—only that he was diagnosed with the condition 4 years before presenting to our clinic.

Mr. J responded moderately to paroxetine monotherapy but experienced rapid resolution of his panic attacks with a combination of paroxetine and CPAP. CPAP monotherapy sufficiently prevented panic attacks for 4 years. Finally, when Mr. J experienced a single panic attack several months before presenting to our clinic—at the end of a very stressful year—reintroducing paroxetine prevented subsequent attacks. This supports our hypothesis that OSA may predispose or indirectly cause patients to develop daytime panic attacks. Alternately, this case suggests that OSA may cause subclinical panic disorder to present as an acute condition.

We rule out anxiety disorder secondary to a general medical condition (OSA) and diagnose Mr. J with anxiety disorder not otherwise specified.

The authors’ observations

We continue paroxetine at 20 mg/d because it was working fairly well with minimal side effects. The sleep medicine specialist maintained modafinil, 200 mg/d. Laboratory studies—including a comprehensive metabolic panel, complete blood count with differential, and thyroid stimulating hormone—were within normal limits except a fasting blood glucose of 123 mg/dL, for which we referred Mr. J to his primary care physician.

OUTCOME: Discontinue paroxetine?

One month later, Mr. J denies panic attacks, other anxiety symptoms, or other psychiatric symptoms and is sleeping well. However, he reports that his mildly decreased concentration persists and he wants to stop paroxetine.

After discussing the risks and benefits, Mr. J and the treatment team decide to continue paroxetine at 20 mg/d. We cite peer-reviewed literature that recommends continuing antidepressants for at least 1 year and possibly indefinitely after symptom resolution to control panic disorder symptoms.⁹ In addition, we discuss the lack of studies comparing different lengths of treatment with SSRIs for apparent OSA-induced panic attacks that respond to SSRI/CPAP therapy. Because Mr. J was doing well and experiencing minimal side effects, he feels he would be better served with a longer period of psychopharmacologic treatment.

Six months later, Mr. J says his anxiety symptoms are well controlled and generally unchanged except for an occasional “little flutter” of anxiety every 3 or 4 days that lasts several seconds. For 1 year, he reports no recurrence of panic attacks, compliance with CPAP, and stable OSA.

Related resource

• Saunamäki T, Jehkonen M. Depression and anxiety in obstructive sleep apnea syndrome: a review. Acta Neurol Scand. 2007;116(5):277-288.

Drug brand names

• Alprazolam • Xanax
• Amphetamine/dextroamphetamine • Adderall
• Atomoxetine • Strattera
• Clonazepam • Klonopin
• Ergocalciferol • Calciferol
• Modafinil • Provigil
• Methylphenidate extended release • Concerta
• Paroxetine • Paxil
• Pramipexole • Mirapex
• Propranolol • Inderal
• Sertraline • Zoloft

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.