|
| | Authors | Country | Intervention | Study design and population | Findings and conclusions |
|
| 1 | Jonker et al. [12] | Amsterdam, Netherlands | NSAIDs (including aspirin) (n = 137) versus control (no NSAIDs) (n = 475) | Population-based study (n = 612) via the Longitudinal Aging Study Amsterdam (LASA) | For a population with an age range varying from 62 to 85, this study found no differences between users and nonusers of NSAIDs in mean MMSE scores. Also no differences were found between individual subscores on items such as recall (both immediate and delayed) and coding tasks between users and nonusers. However, with regard to a decline in memory, NSAIDs seem to lower the risk. This protective effect was more associated with the use of aspirin and no other NSAIDs. They suggest that the additional antiplatelet properties of aspirin might contribute to its protective effects. |
|
| 2 | Kang and Grodstein [13] | USA | NSAIDs (including aspirin) (n = 11,216) versus control (no NSAIDs) (n = 8,749) | Prospective cohort study (n = 19,965) via the Nurses’ Health Study cohort of women aged 70–81 | For women between the ages of 70 and 81, this study found that NSAID use has a substantial effect on cognitive function and decline. Long-term users of nonaspirin NSAIDs showed marginally better overall performance on tests measuring cognitive decline (verbal fluency, delayed recall, East Boston Memory Test, TICS, etc.) relative to those who do not use NSAIDs. The authors postulate that the protective effects of nonaspirin NSAIDs against cognitive decline might be due to the inhibition of cyclooxygenase mediated inflammation and the decrease of amyloid-beta protein, an important factor in Alzheimer’s Disease pathogenesis. They also state that the use of NSAIDs reduces the risk of cerebrovascular pathologies and vascular dementia. |
|
| 3 | Devine and Rands. [14] | London, UK | Aspirin (n = 38) 75 mg/day (n = 28) and 150 mg/day (n = 10) versus control (no aspirin) (n = 40) | Retrospective case-notes analysis (n = 78) of patients with ischemic VD | This study primarily highlights the difficulty in removing indication bias in studies involving the efficacy of aspirin in those with VD; that is, aspirin in those with VD might be indicated and prescribed for patients with other copresenting conditions; most of the time, including in this study, those in the aspirin group had significantly higher rates of vascular disease compared to the nonaspirin group (not necessarily due to the use of aspirin itself). In this study with a small sample size, it was shown that aspirin has a small protective effect (improves survival times to institutionalization and to death) in those with dementia and with those who lived with a care-taker. They argue that this is due to better compliance relative to those who live alone with dementia, where compliance with aspirin may not be guaranteed. The authors also note that this effect is not that robust when accounting for the lower mean age of the patients prescribed aspirin and that larger studies are needed. |
|
| 4 | Nilsson et al. [15] | Sweden | Aspirin (n = 175) versus paracetamol (n = 143) versus NSAID (not including aspirin or paracetamol) (n = 54) versus D-propoxyphene (n = 69) versus control (n = 261) | Population-based study (n = 702 or 351 twin pairs) via the Swedish longitudinal study “Origins of Variance in the Old-Old : Octogenarian Twins” (OCTO-Twin) | This study found that aspirin might reduce the risk of Alzheimer’s Disease (AD). Both nonaspirin NSAIDs and aspirin showed a numerical reduced risk of AD and cognitive decline, even after correcting for stroke, TIA, myocardial infarction, angina pectoris, and congestive heart failure, but this reduction was statistically insignificant. The authors also note that, in all studies concerning the efficacy of aspirin in reducing dementia risk, the use of over-the-counter aspirin and the use of prescribed aspirin must both be analyzed. They also suggest that access to pharmacotherapy in patients with dementia differs from that in those without dementia which must also be accounted for. They further support the idea that the protective effects of NSAIDs and aspirin come from COX-1 inhibition and slowing down of the inflammatory processes that contribute to AD development. |
|
| 5 | Cornelius et al. [16] | Stockholm, Sweden | Aspirin (n = 154) versus NSAIDs (not including aspirin) (n = 76) versus control (n = 1,079) | Population-based study (n = 1,299) via the longitudinal Kungsholmen Project | The apoE epsilon4 allele has been found to be a risk factor for both AD and other cerebrovascular diseases and this study aimed to find an association between dementia, the use of NSAIDs, and the apoE apsilon4 allele. This study found that those who were on aspirin and did not have the apoE epsilon4 allele had an increased risk of AD. They also note that this finding persists even after correcting for other possible underlying diseases and indications. This study additionally found that there is a small protective effect of NSAIDs against AD in general, but this was not statistically significant. The authors postulate that dementia in those who are negative for the apoE epsilon4 allele may not owe its pathogenesis to inflammatory reactions but through some other unknown mechanism. Similarly, the protective effects of NSAIDs in the elderly population might be due to protection against other comorbidities and vascular changes and not necessarily due to the anti-inflammatory effects and the prevention of AD. |
|
| 6 | Shepherd et al. [17] | Sydney, Australia | Aspirin (n = 69) versus control (no aspirin) (n = 82) | Population-based study (n = 151) via the Sydney Older Persons Study (SOPS) | This study aimed to find an association between human leukocyte antigen (HLA-DR) genotype, aspirin use, and performance in cognitive testing. This study supports the protective effect aspirin has on cognitive performance. This effect was also seen in individuals with the HLA-DRB1∗01 allele. The presence of the DRB1∗05 allele was associated with reduced cognitive performance. The study also suggests that there is no interaction between aspirin use, HLA-DR genotype, and cognitive performance. The authors postulate that the protective effects of aspirin are twofold: anti-inflammatory effects (which they consider might not be the primary mode of protection due to similar protective effects seen with aspirin at both low and high doses) and the prevention of cerebral vascular damage or increased cerebral perfusion due to aspirin’s antiplatelet effects. |
|
| 7 | Thal et al. [18] | USA | Rofecoxib 25 mg (n = 725) versus placebo (n = 732) | Randomized, double-blind, placebo-controlled study (n = 1,457) comparing rofecoxib and placebo | This study found that rofecoxib, a selective COX-2 inhibitor, does not delay a diagnosis or the progression of AD. This finding also suggests that the inhibition of COX-2 is not a useful therapeutic marker for AD and that COX-2 is not an important factor in AD pathogenesis. One nonintuitive finding from this study is that rofecoxib might lead to increased conversion from MCI to AD, but the authors suggest this is more likely a chance finding or is due to the selective NSAID’s ability to increase blood pressure (but no evidence was found for this hypothesis). |
|
| 8 | Kang et al. [19] | USA | Aspirin (n = 3,215) versus placebo (n = 3,162) | Observational cohort study (n = 6,377) via the Women’s Health Study (a randomized, double-blind, placebo-controlled study investigating the use of aspirin for the primary prevention of cardiovascular disease) | For women older than 65 years receiving treatment for nearly 10 years, this study found similar cognitive performance between those taking aspirin and those receiving placebo. This null finding was shown at all three timepoints of assessment into the study, as well as in the 3 to 6 years following the study when measuring average cognitive decline. The authors note that, in the final assessment, the aspirin group performed better in the category fluency test, a partial assessment of executive function which is a cognitive system that is heavily dependent on the presence of vascular disease, but disclaim that this was the only assessment that tested executive function changes and was not a primary outcome. This study also found that subjects who had high cholesterol or who were smokers experienced less cognitive decline than those who were receiving placebo. This protective effect was not found in those without hyperlipidemia or those who were not smokers. |
|
| 9 | AD2000 Collaborative Group et al. [20] | UK | Aspirin (n = 156) versus control (no aspirin) (n = 154) | Randomized, double-blind, placebo-controlled study (n = 310) comparing aspirin and placebo (the AD2000 trial) | This study originally aimed to find the efficacy of donepezil in the management of AD. Additionally, the authors wanted to test the protective effects of aspirin in patients with AD. This study found that, for patients with AD and with no other cardiovascular disease, low-dose aspirin does improve cognitive functioning but only marginally (half-point improvement on the MMSE). The study’s authors argue that the follow-up time was short (2 years) and if a longer time was considered, then the effect of aspirin may have been more significant. However, they note the tradeoff between longer follow-up and reduced compliance and drop-off, especially when considering subjects of elderly age with dementia. Considering the tradeoffs, in addition to the increased hemorrhagic risk of long-term aspirin use, the authors conclude that although some evidence exists that aspirin might slow cognitive decline, the risks outweigh the benefits and aspirin cannot always be recommended. |
|
| 10 | ADAPT Research Group et al. [21] | USA | Celecoxib (n = 726) versus naproxen (n = 719) versus placebo (n = 1,083) | Randomized, placebo-controlled study (n = 2,528) comparing celecoxib, naproxen, and placebo (primary prevention trial, the Alzheimer’s Disease Anti-Inflammatory Prevention Trial (ADAPT)) | The primary prevention trial ADAPT sought to evaluate the use of NSAIDs, such as naproxen (a nonselective COX-1 and COX-2 inhibitor) and celecoxib (a selective COX-2 inhibitor), for the primary prevention of AD. The motivation for this study comes from in vitro and in vivo mouse models that show that some NSAIDs (not necessarily celecoxib or naproxen) lower the level of amyloid-beta protein and the epidemiological evidence of lower incidence of dementia in people taking NSAIDs. ADAPT found that NSAIDs do not show a protective effect and may also lower the performance on cognitive testing. The scores on cognitive assessments were significantly lower for those taking naproxen (but not with celecoxib) when compared with placebo. Some treatments in ADAPT were terminated early due to the increased cardiovascular risk associated with celecoxib. They suggest that the protective anti-inflammatory effects of NSAIDs are only advantageous when given several years before the development of dementia symptoms and not when given close to diagnosis of MCI or AD. The authors conclude that naproxen and celecoxib should not be used for the prevention of AD. |
|
| 11 | Arvanitakis et al. [22] | USA | Aspirin (n = 389) versus nonaspirin NSAID (n = 215) versus control (no NSAIDs or aspirin) (n = 1,434) | Population-based study (n = 1,019) via the Religious Orders Study | This study found that NSAID use, either ibuprofen or aspirin, had no relationship with incident AD or changes in cognitive functioning. In the 325 deceased subjects that had associated neuropathology autopsies, the authors again found that NSAID use did not correlate with any measures of AD pathology. The data presented in this study can be considered to be stronger than previous investigations examining NSAID use and cognitive decline because of the extensive follow-up conducted, but the authors maintain that NSAIDs are not related to reduced cognitive decline. |
|
| 12 | Diener et al. [23] | 35 countries worldwide | Aspirin 25 mg and extended-release dipyridamole 200 mg (n = 9,382) versus clopidogrel 75 mg (n = 9,430) (each given with either telmisartan 80 mg or placebo) | Randomized, double-blind, placebo-controlled study (n = 18,812) comparing aspirin with extended-release dipyridamole and clopidogrel (the Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS) trial) | This study aimed to find a possible improvement in cognitive outcomes after stroke with either aspirin, extended-release dipyridamole (ER-DP), clopidogrel, or telmisartan. The authors found that, across all treatment arms, functional impairments were similar. The authors postulate that neither aspirin and ER-DP nor clopidogrel is neuroprotective and that both of these treatments are indeed neuroprotective but yield no changes in functional impairment after stroke. None of the treatments had an influence on cognitive decline and MMSE scores. No change in cognitive outcomes was also found in the smaller subgroup of patients with recurrent stroke. The authors note that the follow-up period of 2 to 5 years in the PRoFESS study may have been too short to properly show the effects of the treatments on cognitive outcomes, especially when juxtaposed to other trials with longer follow-ups that show a greater total decrease in blood pressure and even the reduction of cognitive decline. |
|
| 13 | Price et al. [24] | Scotland | Aspirin 100 mg (n = 850) versus placebo (n = 858) | Randomized, double-blind, placebo-controlled study (n = 1,708) comparing aspirin and placebo | This study, like many before it, investigated the effects of aspirin on cognitive function in those at risk for cardiovascular complications. The authors found that there were no significant differences between those taking aspirin and those on placebo with regard to cognitive function. They were motivated by the experimental evidence that aspirin reduces some cardiovascular disease and can thusly indirectly reduce cognitive decline and the risk of vascular dementia. However, the authors conclude that there is no cognitive benefit seen with aspirin usage in those without dementia and at moderately increased cardiovascular risk. They postulate that a population that is already presented with cognitive decline or with advanced vascular disease, or both, might benefit from such a treatment. |
|
| 14 | Small et al. [25] | California, USA | Celecoxib (n = 36) versus placebo (n = 36) | Randomized, double-blind, placebo-controlled study (n = 72) comparing celecoxib and placebo | This study supports the authors’ hypothesis that a selective COX-2 inhibitor like celecoxib can benefit cognitive performance in people with memory complaints when taken daily for 18 months. This study included more middle-aged participants than other usual studies assessing the efficacy of NSAIDs and the cognitive benefits of celecoxib were modest, mainly improving only information processing and retrieval and not episodic memory performance. They bolster these results with PET imaging of glucose metabolism that show increased PET activity in brain regions associated with semantic information processing and retrieval, such as the prefrontal cortex. They note that a number of previous studies state that COX-2 inhibitors can have a null or sometimes negative effect on cognitive performance, but the authors note that the subjects in this study were on average a decade younger than those in the ADAPT study, for example. The younger mean age also contributes to a reduced concern for cardiovascular complications associated with celecoxib and longer administration periods. |
|
| 15 | Szekely et al. [26] | USA | NSAIDs (n = 1,180) versus control (no NSAIDs) (n = 2,049) | Observational study (n = 3,229) via the prospective Cardiovascular Health Cognition Study | This study found that NSAID use was correlated with a reduced risk of dementia and AD. The authors postulate that the protective effects of NSAID use were associated with the APOE status of the subjects, as subjects with multiple APOE epsilon 4 alleles had he least incidence of dementia. Additionally, this study found that the association with AD was seen with neither acetaminophen nor aspirin but the null finding could be due to either a dosage-dependent effect or a misclassification of usage duration (no information on the subject’s NSAID usage prior to the study). |
|
| 16 | Waldstein et al. [27] | USA | NSAIDs and aspirin | Observational study (n = 2,300) via the prospective Baltimore Longitudinal Study of Aging | This study aimed to investigate the relationship between NSAID and aspirin use and performance on multiple tests of cognitive function in those without dementia. The authors found that NSAID use was related to less decline on tests that assess concentration, memory, and executive function. Aspirin was found to be related to less decline on tests that assessed memory, concentration, and visual memory. Though the actual rate of change of cognitive performance is small, some cross-sectional findings suggest that these medications are neuroprotective, but it is difficult to determine the biological mechanism responsible. |
|
| 17 | Kern et al. [28] | Gothenburg, Sweden | Aspirin (n = 129) versus control (no aspirin use) (n = 552) | Prospective, population-based cohort study (n = 681) of women in Sweden (from the Prospective Population Study of Women and the H70 Birth Cohort Study) | This study aimed to investigate the effect of low-dose aspirin on cognitive function in elderly women with differing cardiovascular risk profiles. This study found that, in women with high risk for cardiovascular disease, low-dose aspirin treatment was linked to less cognitive decline. The authors also note that the use of aspirin did not affect the incidence of dementia, but they owed this null finding to the short follow-up period. The authors also note that the observed increase in cognitive function associated with aspirin use is most likely only seen in those with preclinical dementia, and those with later-stage dementia might not yield as much of a neuroprotective effect. |
|
| 18 | Taguchi et al. [29] | Japan | Cilostazol (n = 70) versus control (no cilostazol) (n = 22) | Retrospective observational study (n = 92) via medical records at Sumotoitsuki Hospital | This study found that cilostazol may preserve cognitive function in those with mild cognitive impairment. In particular, improvements to orientation in time and delayed recall were found, which are critical areas that are first to decline in dementia. The authors postulate that the protective effects of cilostazol come from its dual effects on both AD etiopathogenesis (inhibition of amyloid-beta development) and cerebral perfusion. |
|
| 19 | Ihara et al. [30] | Japan | Donepezil (alone) (n = 87) versus combination therapy of donepezil and cilostazol (n = 69) | Retrospective observational study (n = 156) of subjects in Japan with a history of donepezil use with or without cilostazol use | This study found that cilostazol can preserve cognitive function in patients with mild dementia who are concomitantly taking donepezil for approximately two years. The study’s authors hypothesize that the benefit of cilostazol lies in its dual ability to reduce both ischemia and amyloid-beta related neurodegeneration. They also note that the synergistic effects of donepezil and cilostazol can be owed to each drug having different vascular targets: donepezil being endothelium-dependent vasodilation and cilostazol being endothelium-independent vasodilation. One limitation of this study is that it is not clear why one patient would receive cilostazol as an add-on but others would not. Regardless, the study’s findings support that cilostazol can delay cognitive decline in patients in early-stage AD or with mild dementia but not in established moderate-to-severe dementia. |
|
| 20 | Zhang et al. [31] | China | Antiplatelet agents (aspirin, clopidogrel, ozagrel) (n = 435) versus control (no antiplatelets) (n = 83) | Retrospective observational study (n = 518) of subjects in China with a diagnosis of acute ischemic stroke | This study found that antiplatelet agents such as aspirin, clopidogrel, and ozagrel reduce the risk of recurrent myocardial and cerebral infarctions and, in general, improve the prognosis of stroke survivors. The authors also found that, in acute ischemic stroke, taking antiplatelet agents shows a favorable functional and cognitive outcome. This effect is nonsignificant, however, when correcting for the use of such agents in the convalescent period of stroke (3 months after stroke). The authors note that, in comparison to other studies examining similar treatments, the population in this study had relatively minor neurologic deficits. |
|
| 21 | Pearce et al. [32] | 81 clinical centers worldwide | Aspirin and clopidogrel (n = 1,468) versus aspirin and placebo (no clopidogrel) (n = 1,448) | Randomized, double-blind, placebo-controlled study (n = 2,916) (Secondary Prevention of Small Subcortical Strokes (SPS3) trial) | This study investigated the effects of dual antiplatelet therapy with aspirin and clopidogrel in patients with cerebral small vessel disease, a history of lacunar strokes, and their effects on cognition. The study’s findings suggest that, for these patients (in the absence of a recurrent stroke), adequate management of blood pressure and a dual antiplatelet regimen leads to minimal cognitive decline in a short term (3 years). However, no clinically significant difference was found between the dual antiplatelet plus clopidogrel group and the antiplatelet plus placebo group. |
|
| 22 | Wichmann et al. [33] | Wisconsin, USA | Aspirin (n = 1,583) and nonaspirin NSAIDS (n = 1,111) versus control (no aspirin nor NSAIDs) (n = 1,965) | Observational study via the prospective Epidemiology of Hearing Loss Study of patients with mild cognitive impairment and sensory loss; participants without cognitive impairment at baseline | This study’s findings do not support NSAID’s protective effect against cognitive impairment. Aspirin use was related to increased risk of cognitive impairment, possibly due to indication bias. Those who used aspirin at baseline but not 5 years before had a higher risk of cognitive impairment over the next decade. The current study found an elevated risk of cognitive impairment among aspirin users but only among current users at baseline. This study assessed NSAID use and incident cognitive impairment over a 20-year period and found no indication of a protective effect. Debate continues over nonaspirin NSAID use and dementia risk because most observational research yields contradictory results. |
|
| 23 | Chang et al. [34] | Taiwan | Aspirin (n = 2,876) versus control (no aspirin) (n = 10,720) | Retrospective cohort study (n = 28,321) of patients with type 2 diabetes mellitus without a prior diagnosis of dementia | A daily dose of 40 mg aspirin may reduce the incidence of Alzheimer’s Disease in people with type 2 diabetes but not in people with non-Alzheimer’s dementia. The absence of statistical significance of low-dose aspirin’s protective benefits against non-Alzheimer’s dementia was likely owing to small sample size. The authors additionally found that increasing the dosage of aspirin greater than 80 mg/day increased the risks of both non-AD and AD in those with T2DM. |
|
| 24 | Oveisgharan and Hachinski [35] | Canada | Unspecified salicylate therapy | Observational study (n = 4,878) of patients with cognitive impairment via the longitudinal Canadian Study of Health and Ageing | This study suggests that salicylate use may reduce dementia risk in older people who score poorly on a basic neuropsychological test and are 2 to 3 times more likely to acquire dementia than people with normal cognition. The authors additionally found that there was no link between incident dementia amongst normal cognition patients and the use of salicylates; however, for subjects with normal copying pentagons-writing scores in the studied neuropsychological tests, salicylates enhanced dementia incidence. It is probable that some of these people had preclinical AD diseases such amyloid angiopathy, and salicylates have been linked to brain microbleeds, which are linked to dementia. |
|
| 25 | Tai et al. [36] | Taiwan | Cilostazol (n = 2,287) versus control (n = 6,861) | Prospective cohort study (n = 9,148) of patients who had ever taken cilostazol for at least 3 months via Taiwan’s National Health Research Institute | This study found a significant link between a decreased risk of dementia in those without a prior history of dementia and the use of cilostazol. The authors note that this relationship was dose-dependent (high-dose cilostazol significantly decreased the risk of dementia). They additionally find that cilostazol use was notable for reducing the risk of dementia in those with vascular pathologies like ischemic heart disease and cerebral vascular disease and that this neuroprotective effect was notably seen in men older than 65 years. |
|
| 26 | Staszewski et al. [37] | Poland | Aspirin | Observational study of patients with first-ever LS (n = 49), diagnosed VP (n = 16), and VD (n = 39) due to CSVD | This study’s main conclusion is that AR is frequent in CSVD patients and is linked to radiological progression and lacunar stroke risk. The prevalence of AR was similar in patients with acute (lacunar stroke) and chronic (VD and VP) CSVD. They found no difference in ASA treatment duration between AR and RTA patients. A similar frequency of AR was found in individuals who received ASA prior to study admission and those who started ASA after trial entry. Those with VD or VP had a higher risk of vascular events attributable to age (mean 72 years), hypertension and dyslipidemia, and diabetes than those without VD or VP. The authors also found that TG levels were linked to AR regardless of statin use (patients with AR had slightly higher TG). |
|
| 27 | Ding et al. [38] | Stockholm, Sweden | Unspecified anticoagulation and antiplatelet therapy | Population-based cohort study with patients at risk for either VD or AD due to CVD (total n = 2,685) and those specifically with AF (n = 243) | This population-based investigation found that anticoagulant but not antiplatelet therapy lowered the risk of dementia in AF patients. The study’s authors also note that AF in the elderly is associated with an increased risk of cognitive decline and dementia. Because of this somewhat novel association between these two pathologies, this study motivates the need for trials investigating the effects of using antithrombotics and anticoagulants for the management of dementia. The authors hypothesize that the link between AF and dementia is due to reduced cardiac output, increased blood stasis, blood hypercoagulability, and the increased risk of cerebrovascular lesions, factors that are often targeted by the novel direct oral anticoagulants. |
|
| 28 | Yang et al. [39] | Taiwan | Aspirin (n = 1,525) and control (n = 1,525) | Population-based study (n = 6,028) of patients with cognitive impairment and LOD via the National Health Insurance of Taiwan survey | This study shows that aspirin consumption reduces the risk of dementia and also provides evidence that LOD is a risk factor for dementia. The authors note that, in this study, like many studies before it investigating the effects of aspirin in the elderly with dementia, many of the subjects copresented with cardiovascular disease, especially those with LOD, and this association possibly explains the protective effects of aspirin. Their findings should be taken with caution in light of the ASPREE trial’s poor findings on low-dose preventative aspirin use in healthy older people; nevertheless, the authors maintain that aspirin might still be beneficial for thrombosis prevention in the elderly with dementia, cardiovascular risk, and LOD, especially in those who do not present with the increased hemorrhagic risk of aspirin. |
|
| 29 | Viticchi et al. [40] | Italy | Unspecified antiplatelet and anticoagulation therapy | Retrospective cohort analysis of hospitalized patients with preexistent AF (total n = 1,705) and those with VD (n = 193) via the Atrial Fibrillation in Critically Ill (AFICILL) study | This study aimed to evaluate the effect of compliance and adherence to antiplatelet and anticoagulant therapy in patients with dementia and atrial fibrillation. Primarily, the authors found that antiplatelets and anticoagulants are often underadministered or not given to patients where these medications are indicated. They also found that patients with other forms of dementia were also undertreated, albeit less so than VD patients. Moreover, VD patients had a higher incidence of acute cerebrovascular disorders during hospitalization, with some having a new stroke/TIA, compared to patients with other forms of dementia. VD patients had higher thromboembolic risk scores than non-VD patients. Despite this, VD patients had a lower rate of major bleeding and a greater rate of stroke. They also showed no significant difference in treatment failure compared to nondemented patients. The authors conclude that the risk of bleeding in the elderly can lead to undertreatment, especially those affected by dementia, and this leads to a source of understudied treatment effects on this patient population. |
|
| 30 | Matsumoto et al. [41] | Japan | Aspirin (n = 1,259) versus control (no aspirin) (n = 1,277) | Follow-up cohort study (n = 2,536) of patients with type 2 diabetes mellitus (Japanese Primary Prevention of Atherosclerosis with Aspirin for Diabetes (JPAD)) | This study demonstrates that low-dose aspirin may help prevent dementia in T2D women but not necessarily in men. This study is the first to use real-world data to assess the long-term relationship between low-dose aspirin and dementia risk in T2D patients. The authors postulate that aspirin’s dementia-preventive effects may be strengthened in people with cardiovascular risk. They note this as a possible source of bias especially since dementia and cardiovascular disease share common factors in their etiopathogenesis. This study also features a longer follow-up (11.4 years) than most other studies and could be a reason for the positive finding not seen in shorter aspirin related trials. |
|
| 31 | Ryan et al. [42] | USA and Australia | Aspirin (n = 9,525) versus placebo (n = 9,589) | Randomized, double-blind, placebo-controlled study (n = 19,114) of patients with no history of CVD but diagnosed with dementia (Aspirin in Reducing Events in the Elderly (ASPREE)) | In ASPREE, they found that daily low-dose aspirin did not improve disability-free survival but increased the chance of severe hemorrhage. Low-dose aspirin had no effect on dementia incidence. On average, the treatment groups did not benefit from aspirin when it came to clinically probable or plausible AD, MCI, or cognitive impairment. Aspirin had no effect on global or specific cognitive domains. Treatment results did not differ between subgroups characterized by age, sex, ethnicity, health, or prior NSAID usage. Because of the rigor of this large double-blinded placebo-controlled trial, this study provides good evidence that starting low-dose aspirin for those older than 70 years does not reduce the risk of cognitive decline, dementia, or AD in older adults. |
|
| 32 | Khezrian et al. [43] | Scotland | Aspirin (n = 68) versus control (no aspirin) (n = 250) | Longitudinal observational study (n = 318) via the Aberdeen 1936 Birth Cohort study | This study aimed to describe the relationship between white-matter hyperintensity total lesion volumes, cardiovascular risk, cognitive decline, and the administration of aspirin. This study found that aspirin is expectedly beneficial for the prevention of recurrent cardiovascular incidents such as stroke or MI but can be detrimental for information processing speed; they find that this negative effect of aspirin is exacerbated by white matter hyperintensity burden. The authors conclude that, in future analyses of the relationship between cardiovascular risk and cognitive decline, assessments of white matter hyperintensity volumes should be included, which was shown to be a significant predictor of cognitive decline and, in particular, information processing speed. |
|
| 33 | Huang et al. [44] | Southern Taiwan | Cilostazol (n = 45) versus aspirin or clopidogrel (n = 45) | Prospective open-label study (n = 90) of subjects with a diagnosis of cognitive impairment after ischemic stroke | This study describes poststroke cognitive alterations. The cilostazol and control groups had similar changes in cognitive and global status and the study concludes that cilostazol did not make a significant difference in global cognitive functioning relative to other antiplatelet drugs. After 6 months, over 60% of patients improved their cognitive function and over 70% improved their global status in both treatment arms. Cilostazol may be regarded as “noninferior” for poststroke cognitive change due to the bias of higher recruiting into the cilostazol group for the treatment of PAOD, which many of the subjects copresented with. Cilostazol is recommended by Taiwan’s national health insurance standards for treating PAOD symptoms of intermittent claudication and thus many subjects were recruited into the cilostazol treatment arm. |
|
| 34 | Lee et al. [45] | Korea | Warfarin (n = 25,948) versus DOACs (n = 46,898) (rivaroxaban, dabigatran, apixaban, edoxaban) | Observational study (n = 72,486) via the Korean nationwide claims database | Compared to warfarin, DOACs had equivalent dementia, vascular dementia, and Alzheimer’s dementia risks. For individuals aged 65 to 74 yesars and those who had previously had a stroke, DOACs were associated with a decreased incidence of dementia than warfarin. When compared to warfarin, edoxaban and rivaroxaban had a decreased risk of dementia. |
|
| 35 | Kim et al. [46] | Korea | Cilostazol (n = 127) versus aspirin (n = 129) | Randomized, double-blind, controlled study (n = 256) comparing cilostazol and aspirin | This study aimed to investigate any differences seen in white matter changes in cerebral small vessel disease, such as lacunar infarctions, when receiving cilostazol or aspirin. Both groups, over the two-year period of the trial, had increased WMH volume to total WM volume, but the treatment arms showed no significant differences. The cilostazol group had reduced ischemic vascular events relative to the aspirin group, but this was a secondary outcome. |
|
