The cardiovascular safety of dementia medications: a cross national study

Dice with broken heart

Two of the most widely available medicines for dementia in the UK are cholinesterase inhibitors and a N-methyl-D-aspartate (NMDA) receptor antagonists (NHS, 2013).  Both drugs are popular because they have been found to lead to improvements in dementia symptoms (Raina 2008). However, some have raised concerns about the safety for patients. These concerns are partly grounded in claims that whilst these medications may ‘treat’ dementia, they carry risks to patients’ cardiovascular health (McCain 2007, Haibara 1995). Some researchers have attempted to explore if these risks exist, but the clarity of their findings are marred by different methodological problems (Fosbol 2012).

Using cross-national patient records from the USA and Denmark, a new study by Fobsol et al hoped to clarify the cardiovascular risks of the dementia medication.


The researchers used a nationally representative sample of patient records from Medicare beneficiaries from 2006 through 2009 in the USA and nationwide Danish administrative registries from 1997 through to 2007.

The authors explored the link between the prescribed dementia drugs and subsequent cardiovascular health. Cox proportional hazards regressions were performed to do this.

Medicines included in the research were:

  • Three types of cholinesterase inhibitors:
    • Donepezil
    • Rivastigmine
    • Galantamine
  • One type of NMDA receptor antagonist:
    • Memantine

In the USA sample they looked for any associations between these medications and myocardial infarction (MI), heart failure, and syncope or atrioventricular block. In the Danish cohort they looked for these conditions as well as fatal and non-fatal MI and cardiac death.

Individuals over 65 treated for dementia were included. 46,737 USA beneficiaries and 29,496 Danish patients were included.


  • The MI risks in the Danish cohort were:
    • Memantine (HR 1.31)
    • Rivastigmine (HR 0.98)
    • Galantamine (HR 0.92)
  • Risks for syncope or atrioventricular block did not differ between either type of medication in the USA sample.  In the Danish cohort the risk of these events was lower for Memantine (HR = 0.63) than in the USA sample (HR = 0.97).
  • In the Danish cohort, Memantine was also associated with a greater risk of cardic death (HR = 1.31) than in the USA sample (HR = 0.97)


The author’s concluded:

There was no greater risk of cardiovascular events associated with rivastigmine than with other cholinesterase inhibitors. The greater risk of cardiovascular events in the Danish cohort users of memantine and combination therapy is probably related to selection of sicker participants, because these therapies are reserved for individuals with more-severe dementia in Denmark.


It is unclear from this research whether being on dementia medication can increase the risks of cardiovascular events

The authors only compared the risk of adverse cardiovascular effects across the different medications, and did not compare their results to individuals not on any medication. So, based on this study it is unclear whether just being on the medication can increase the risks of adverse cardiovascular health.

This study also used retrospective cross-sectional data and therefore cause and effect cannot be inferred.  A controlled experiment would be needed to establish a cause and effect relationship. However, it would be incredibly difficult to carry out a controlled experiment on the same cross-national scale as in this study.

A key strength of this study lies in a large representative sample was used.  This suggests that the effects of the drugs were the same, regardless of different socio-economic factors. The main strength of this study is that it showed that most of the results replicated cross nationally. This is particularly important  because it suggests that the effects of the drug results are similar in different environment, meaning, that it is quite likely that the risks of cardiovascular effects found were not likely to stem from environmental factors.


In summary, within the USA and Denmark cohort neither cholinesterase inhibitors or NMDA receptor antagonists were associated with greater risks to cardiovascular health. NMDA receptor antagonists may have carried a greater risk in the Danish cohort, but this was unclear. Future work is needed to explore these potential risks further.


Fosbol E, Peterson ED, Holm E. et al. Comparative Cardiovascular Safety of Dementia Medications: A Cross-National Study. The American Geriatrics Society 2012 1-7. [PubMed abstract] [PDF PowerPoint presentation]

Haibara AS, Colombari E, Chianca DA, et al. NMDA receptors in NTS are involved in bradycardiac but not in pressor response of chemoreflex. Am J Physiol 1995;269: 421-427. [PubMed abstract]

McCain KR, Sawyer TS, Spiller HA. Evaluation of centrally acting cholinesterase inhibitor exposures in adults. Ann Pharmacother 2007;41:1632-1637. [PubMed abstract]

How is Dementia Treated? NHS Choices. Accessed online (1 April 2013).

Raina P, Santaguida P, Ismaila A. et al. Effectiveness of cholinesterase inhibitors and memantine for treating dementia: Evidence review for a clinical practice guideline. Ann Intern Med 2008;148:379-397.

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+