Bradycardia Due to Donepezil in Adults: Systematic Analysis of FDA Adverse Event Reporting System

Abstract

Background:

Bradycardia is a physiological condition characterized by a decrease in heart rate and is a side effect of many drug classes. Bradycardia has been reported as an adverse event for patients receiving donepezil for Alzheimer’s disease (AD) treatment.

Objective:

The purpose of the paper is to systematically investigate the association between the occurrence of bradycardia in adults and the usage of donepezil using clinical data derived from the FDA Adverse Event Reporting System (FAERS) database.

Methods:

The risk of bradycardia in patients who only took donepezil was compared with those of patients who only took over-the-counter medications, multiple arrhythmia drugs, or other medications for AD treatment. In addition, this study sought to determine if this heightened bradycardia risk was influenced by sex, age, and dosage.

Results:

The results indicated that there was a significant greater likelihood of reporting bradycardia in patients administered donepezil than most of the drugs investigated. There was no significant association between age or the dosage of donepezil and the likelihood of reporting bradycardia. However, males were found to be more likely than females to report bradycardia as an adverse event. Tumor necrosis factor inhibition and the stimulation of endothelial nitric oxide synthase were proposed to be the primary mechanism of actions which confer elevated bradycardia risk when using donepezil.

Conclusion:

These findings identified strong association between the usage of donepezil and bradycardia in adults as well as provided insight into the underlying molecular mechanisms that induce bradycardia by donepezil.

Keywords

Acetylcholinesterase inhibitors AChE Alzheimer’s disease bradycardia donepezil endothelial nitric oxide synthase

INTRODUCTION

Alzheimer’s disease (AD) is a neurological disorder characterized by progressive decline in cognitive function including memory processing, of which typically manifests as the first pathological sign of the early stages of cognitive decline as many AD cases originate in the hippocampus, spatial perception, and critical thinking [1, 2]. As a result of the age distributions of many countries skewing older due to improved health care and an increase in life expectancy, the number of diagnosed cases, and subsequent deaths, attributed to AD has risen substantially since 2000 and is expected to continue to rise over the next few decades. Rates of AD cases have increased approximately 146% from 2000 to 2018, of which approximately 122,000 deaths were attributed to AD in 2018 alone [3]. Because of this high mortality, AD is currently the 6th highest cause of mortality among all age groups as well as the 5th leading cause of death in individuals 65 and older [3].

Four drugs including donepezil (Aricept), rivastigmine (Exelon), galantamine (Razadyne), and memantine (Namenda) are currently administered to AD patients as either part of a single regimen or as a combination therapy [4]. Donepezil is an acetylcholinesterase (AChE) inhibitor. Acetylcholine is a chemical messenger that modulates long-term and immediate memory and facilitates the activation of cholinergic neurons through stimulation of action potentials [5]. Acetylcholine is generated in the cytoplasm by the enzymatic action of choline acetyltransferase, of which converts acetyl-CoA and choline into acetylcholine that is subsequently transported into synaptic vesicles by the vesicular acetylcholine transporter [6]. Calcium-mediated depolarization of these synaptic vesicles stimulates exocytosis of acetylcholine into the synaptic cleft, at which point acetylcholine can bind to various receptors including muscarinic and nicotinic receptors to innervate neuronal function [7]. Electrical signaling is subsequently terminated through AChE-mediated hydrolysis of the bound neurotransmitter into acetate and choline, of which are subsequently recycled into pre-synaptic cells and reused in the synthesis of new acetylcholine [6].

Through alternative splicing and the use of different promoters, AChE is present in several variants including monomeric AChE-R, dimeric AChE-E, and tetrameric AChE-S, the latter of which is predominantly expressed in the brain [8]. Tetrameric AChE-S associates with proline-rich membrane anchor protein and anchors to neuromuscular junctions, the site of which allows for rapid hydrolysis of receptor-associated acetylcholine [9]. Although acetylcholine modulates innervation of neurons comprising the central nervous system and peripheral nervous system, one of the primary sites of acetylcholine production and stimulation is the hippocampus, the site of memory formation and one of the earliest brain regions that experiences AD-related degradation [6]. In AD, forebrain neurons are largely degenerated, and acetylcholine levels decrease significantly. In addition, regulation of the cholinergic system is disrupted, and AChE begins to associate with phosphorylated tau proteins in order to promote amyloid-β plaques, thus driving the characteristic pathology of the disease [10]. Donepezil, along with other AChE inhibitors, stabilizes acetylcholine levels and improves neuronal function in patients with early stages of AD [4]. Thus, these AChE inhibitors seek to restore the cholinergic stimulation of these neurons involved in learning and memory to slow the progression of cognitive decline.

In adults, a typical heart rate is between 60 and 100 beats per minute with the maximum exercise-induced heart rate decreasing with respect to age [11]. Bradycardia is a clinical condition defined by a reduction in heart rate and may result from infectious agents that cause damage to heart muscles (myocarditis) as well as an adverse side effect of some medications [12]. Although some cases of bradycardia may be benign, severe instances of bradycardia can result in chest pain, loss of consciousness, extended periods of unexplained low heart rate, and eventual heart failure [11].

Multiple studies have previously shown an association between the use of donepezil or other AChE inhibitors and subsequent bradycardia diagnosis in AD patients. For example, a case report details a 72-year-old man who was diagnosed as having cognitive decline, as measured by Mini-Mental State Examination, and was given a progressively increasing dose of donepezil to treat his declining cognition [13]. When administered a 10 mg, the patient reported brain fog and was subsequently reduced to a 5 mg dose, at which point he was reporting improved cognitive function. However, an 8-month follow-up revealed that he was suffering from heart palpitations and was diagnosed with sinus bradycardia due to sinoatrial exit block [13]. The patient discontinued use for approximately 1 month and the heart palpitations were resolved. However, when he resumed usage of the drug to treat his cognitive decline and was administered a 7.5 mg dose, an electrocardiogram revealed that the patient now had an asymptomatic irregular sinus bradycardia. When his dose was reduced to 5 mg again, the patient’s arrhythmias subsided, and his heartbeat returned to normal [13].

Additional studies have shown an association between AChE inhibitors use and an increase in reported bradycardia cases. A study analyzing the prevalence of bradycardia diagnoses in AD patients administered an AChE inhibitor regimen between 1999 and 2007 from the New England Veterans Affairs Healthcare System found an adjusted hazard ratio of 1.4 between the treated and untreated groups [14]. This indicates that bradycardia was approximately 40% more likely to be diagnosed in individuals in the treatment group relative to the untreated group. A population study analyzing individuals older than 65 years of age from Ontario, Canada who were hospitalized for bradycardia within 9 months of being administered an AChE inhibitor generated an odds ratio of 2.13, indicating a strong association between recent initiation of an AChE inhibitor regimen and hospitalization due to bradycardia symptoms [12].

Despite these cases, there are no current systematic analyses of the association between donepezil and bradycardia in AD patients. FDA Adverse Event Reporting System (FAERS) is the largest repository of passively reported adverse drug events (ADEs) in the world. Through voluntary reporting by consumers, healthcare professionals, and manufacturers, more than 10 million ADEs from patients receiving complex combinations of medications have been collected in FAERS. The availability of real-world adverse drug effect data from FAERS proves to be an essential resource in the discovery of possible adverse drug effects.

In this study, we performed a comprehensive analysis on the association between the usage of donepezil and the occurrences of bradycardia. The frequency of reported bradycardia cases in the FAERS database was compared between donepezil and other drugs including over the counter (OTC) drugs, other AD drugs, and arrhythmia drugs that may cause bradycardia. In addition, confounding factors such as age, sex, and dosage were investigated.

MATERIAL AND METHODS

FAERS records from 2012 Quarter 4 to 2020 Quarter 4 were downloaded from the FDA website. The FARES raw data contains seven tables including drug information (DRUG), drug reaction (REAC), patient outcome for the event (OUTC), patient demographic information (DEMO), report sources for the event (RPSR), drug therapy start dates and end dates (THER), and the indications for use (INDI). The DRUG, REAC, OUTC, and DEMO tables were selected. The people who take a drug were extracted by searching the DRUG table using both generic name (for example donepezil) and brand name (such as Aricept, Eranz, Ziledon, and Donila) of the drug. The adverse drug events of these patients who take the drug were identified from the REAC table. The patient information including age, sex, reporting date, and country of origin for reported cases were extracted from the DEMO table. The serious outcomes of events were extracted from the OUTC table. The records of individuals younger than 18 years old were excluded in this study. To avoid the inclusion of duplicate values, each caseID (individual) that reported bradycardia as an adverse event was counted as one case, regardless of the number of follow-ups and subsequent bradycardia reports.

A disproportionality analysis was performed to compare different groups in the risk of bradycardia. The reporting odds ratio (ROR) and proportional reporting ratio (PRR) [15, 16] were calculated. These two values evaluate the proportion of reports containing bradycardia in one group versus the proportion of reports of bradycardia in the control group. The 95% bound of the confidence intervals of ROR or PRR do not include 1.0, indicating that the association between that drug and bradycardia is significant. Data analysis was performed using R statistical software.

RESULTS

The association between donepezil and bradycardia in adult patients who take donepezil only

In the FAERS database, many individuals are prescribed more than one drug. As a result, it is difficult to determine whether bradycardia was attributed to donepezil, another drug that is being taken, or is due to drug-drug interactions between the therapeutics administered. Thus, patients that were only on a donepezil regimen were investigated in this project. Figure 1 represents the top 10 reported adverse events among adult patients taking donepezil exclusively. The most reported adverse event was bradycardia (124 cases). Other frequently reported adverse events included vomiting, diarrhea, confused state, and nausea. As shown in Table 1, total cases of bradycardia cases in patients who were taking donepezil (including patients that were only on a donepezil regimen and patients prescribed multiple drugs including donepezil) were 652. That is to say, bradycardia (528 cases) was also seen in adult patients who were taking donepezil in combination with other drugs. This indicates that bradycardia is a common adverse event of donepezil.

Fig. 1

This figure shows the frequency of the top ten most commonly reported adverse side effects by adult patients prescribed donepezil for Alzheimer’s disease (AD) treatment. Of the ten commonly reported symptoms, bradycardia was the most prevalent side effect in patients prescribed exclusively donepezil.

Table 1

Comparison of the frequency of reported bradycardia cases in adult patients prescribed donepezil and patients prescribed a donepezil-comprised combination therapy. Of the 1625 patients that were only prescribed donepezil, 124 people reported having bradycardia. When assessing the group in which donepezil was used in combination with other drugs, 528 people reported bradycardia as an adverse event

	With Bradycardia	Without Bradycardia
Patients taking donepezil only	124	1501
Patients taking donepezil with other drugs	528	17658
Total	652	19159

Figures 2 and 3 describe the temporal and geo-graphic distribution respectively of reported bradycardia cases by patients prescribed donepezil. Based on the dataset utilized, from 2012 to 2020, the number of bradycardia cases reported by AD patients prescribed donepezil exhibited an overall positive trend with each year observing a higher frequency of reported cases than the prior year, with the exception of the years 2014 and 2015. The highest number of bradycardia cases (approximately 30) was reported in 2020. In addition, patients from many different countries including Japan, US, UK, France, Italy, and Canada that received donepezil for AD treatment reported bradycardia as an adverse event. Thus, this suggests that there is a lack of bias present in the temporal and geographic distribution of reported bradycardia cases by patients administered donepezil for AD treatment, and bradycardia is a globally observed complication.

Fig. 2

This figure shows the temporal distribution of reported cases of bradycardia by adult patients prescribed donepezil from 2012 to 2020. Generally, the number of reported bradycardia cases increased in each subsequent year except 2014-2015.

Fig. 3

This figure shows the geographic distribution of reported cases of bradycardia by patients prescribed donepezil from 2012 to 2020. Cases were reported by adult patients from many different countries with the greatest number of cases being reported by the United States, Great Britain, Italy, and Japan.

In order to validate the observed frequency of bra-dycardia cases from the FAERS database, positive and negative controls were utilized. The positive control included a class of therapeutics used in the treatment of hypertension and arrhythmias including propranolol (Inderal), atenolol (Tenormin), metoprolol (Toprol-XL), sotalol (Betapace), verapamil (Isoptin), or diltiazem (Cardizem). Those drugs are known to cause bradycardia in some patients. In contrast, the negative controls included most common OTC drugs reported by Amazon.com that are generally safe and have very few reports of bradycardia as an adverse event. As shown in Table 2, 86 patients only taking propranolol reported bradycardia as a symptom with a ROR of 0.609 (p = 0.0006 compared with donepezil) while taking propranolol. 90 patients only prescribed atenolol (ROR = 0.986; p = 0.922) reported bradycardia as an adverse event while 242 patients only administered metoprolol (ROR = 0.764; p = 0.019) reported having bradycardia as a side effect. With regards to patients prescribed sotalol (ROR = 0.517; p = 0.0221), verapamil (ROR = 0.838; p = 0.210), and diltiazem (ROR = 0.782; p = 0.0836), 14, 95, and 92 patients respectively reported bradycardia as an adverse event to the FAERS database. Thus, the likelihood of patients exclusively administered donepezil reporting bradycardia is significantly higher than the likelihood of subsequent bradycardia diagnosis after being prescribed either propranolol, metoprolol, or sotalol, of which are known to cause bradycardia in some patients. However, based on p-values, the likelihood of bradycardia diagnosis in patients exclusively administered donepezil is not significantly higher than in patients that are only prescribed atenolol, diltiazem or verapamil, although the ROR values are less than 1.

Table 2

This table shows the frequencies, reporting odds ratio (ROR), proportional reporting ratio (PRR), and corresponding p-values (compared with donepezil) of reported bradycardia cases in adult patients administered or taking one of a series of drugs. Drug classes include over the counter (OTCs) drugs, arrhythmia therapeutics, and other AD drugs. With the exception of atenolol, diltiazem, and verapamil, all associations were found to be statistically significant using p = 0.05 as the cutoff of significance

	With Bradycardia	Without Bradycardia	ROR	PRR
Donepezil	124	1501	1.000
OTC drugs
Acetaminophen	49	24512	0.0242 (0.0173∼0.0338)	0.0261 (0.0189∼0.0363)
			p < 0.0001	p < 0.0001
Ibuprofen	16	22679	0.0085 (0.0051∼0.0144)	0.0092 (0.0055∼0.0155)
			p < 0.0001	p < 0.0001
Omeprazole	4	21031	0.0023 (0.0008∼0.0062)	0.0025 (0.0009∼0.0067)
			p < 0.0001	p < 0.0001
Cetirizine	4	15042	0.0032 (0.0012∼0.0087)	0.0035 (0.0013∼0.0094)
			p < 0.0001	p < 0.0001
Aspirin	7	4771	0.0178 (0.0083∼0.0381)	0.0192(0.0090∼0.0410)
			p < 0.0001	p < 0.0001
Arrhythmia treatment
Propranolol	86	1710	0.609 (0.458∼0.808)	0.628 (0.481∼0.819)
			p = 0.0006	p = 0.0006
Atenolol	90	1105	0.986 (0.743∼1.308)	0.987 (0.760∼1.281)
			p = 0.922	p = 0.922
Metoprolol	242	3832	0.764 (0.611∼0.957)	0.778 (0.632∼0.959)
			p = 0.019	p = 0.019
Sotalol	14	328	0.517 (0.294∼0.909)	0.537 (0.313∼0.921)
			p = 0.0221	p = 0.0221
Verapamil	95	1373	0.838 (0.635∼1.105)	0.848 (0.655∼1.097)
			p = 0.210	p = 0.210
Diltiazem	92	1425	0.782 (0.591∼1.033)	0.795 (0.612∼1.031)
			p = 0.0836	p = 0.0839
AD drugs
Memantine	26	1383	0.228 (0.148∼0.350)	0.242 (0.159∼0.367)
			p < 0.0001	p < 0.0001
Galantamine	11	764	0.174 (0.0935∼0.325)	0.186 (0.101∼0.342)
			p < 0.0001	p < 0.0001
Rivastigmine	38	5270	0.0873 (0.0604∼0.126)	0.0938 (0.0655∼0.134)
			p < 0.0001	p < 0.0001

The frequency of bradycardia was also compared between the negative control (OTC drugs) and donepezil. Of the 24,561 users of acetaminophen only, 49 reported bradycardia as an adverse side effect (ROR = 0.0242; p < 0.0001) while 16 of the 22,695 users of ibuprofen reported bradycardia as a side effect (ROR = 0.0085; p < 0.0001). For individuals that utilized omeprazole or cetirizine, 4 individuals (ROR = 0.0023; p < 0.0001) and 4 individuals (ROR =0.0032; p < 0.0001) reported bradycardia respectively. Finally, 7 patients who utilized aspirin reported bradycardia as a subsequent adverse event (ROR =0.0178; p < 0.0001). When compared to the negative controls, bradycardia was 50–400 times more likely to be reported by patients administered donepezil than people that used an OTC medication.

The frequency of reported bradycardia was then compared in patients that were administered exclusively one of four commonly medications for AD treatment: donepezil, memantine, galantamine, and rivastigmine. As shown in Table 2, only 26 patients that were exclusively administered memantine reported bradycardia as an adverse consequence of taking the drug (ROR = 0.228; p < 0.0001 compared with donepezil). In addition, 11 patients that were only administered galantamine exhibited bradycardia (ROR = 0.174; p < 0.0001) and 38 patients administered exclusively rivastigmine reported bradycardia (ROR = 0.0873; p < 0.0001). Thus, patients who are prescribed donepezil for AD treatment are more likely (4–11 times more likely) to report bradycardia as an adverse health event than patients who were prescribed memantine, galantamine, and rivastigmine. Because galantamine and rivastigmine are also AChE inhibitors, it is suggested that the mechanism of action that confers this observed higher ROR in patients administered donepezil for AD treatment is a trait unique to donepezil and may work in concert with AChE inhibition.

Association between age and reported bradycardia cases in patients prescribed donepezil

The association between age and the frequency of reported bradycardia cases in patients prescribed donepezil was then analyzed (Table 3). 16 AD patients aged 50–65 that were prescribed done-pezil reported bradycardia. 35 patients aged 66–80 (ROR = 0.722; p = 0.306, compared with the age 50–65 group) reported bradycardia after being prescribed donepezil while 51 patients older than 80 years of age (ROR = 0.921; p = 0.787) reported having bradycardia after being prescribed donepezil. Thus, there was no statistically significant association between the age of the patient and their subsequent reporting of bradycardia as an adverse event after being treated with donepezil.

Table 3

The frequencies of reported bradycardia cases were compared among the following age groups in order to determine if age influenced the risk of bradycardia diagnosed: 50–65, 66–80, and older than 80. Odds ratios were then calculated by comparing the number of reported bradycardia cases in each age group with the bradycardia frequency for the 50–65 age demographic. No statistically significant association was found between the age of a patient prescribed donepezil and the likelihood that they report bradycardia as an adverse event

	With Bradycardia	Without Bradycardia	ROR	PRR
50–65	16	126
66–80	35	382	0.722 (0.386∼1.348)	0.745 (0.426∼1.304)
			p = 0.306	p = 0.303
>80	51	436	0.921 (0.508∼1.671)	0.929 (0.547∼1.578)
			p = 0.787	p = 0.787

Association between sex and bradycardia in AD patients prescribed donepezil

The association between sex and frequency of bradycardia diagnosis was then assessed in patients currently prescribed donepezil. As shown in Table 4, of the 589 male patients analyzed in the study, 56 reported bradycardia as a side effect of donepezil usage. In contrast, 52 females that were administered donepezil for AD treatment (ROR = 0.653; p = 0.0337 compared with male) reported having bradycardia as a consequence of donepezil usage. Relative to males, females were about two thirds less likely to report bradycardia as an adverse event when taking donepezil for AD treatment. Thus, there was a statistically significant association between the biological gender of the patient and the likelihood of reporting bradycardia as a side effect after taking donepezil.

Table 4

The role of sex was also assessed with respect to bradycardia risk in patients prescribed donepezil for AD treatment. For males, 56 patients reported bradycardia as a side effect upon being placed on a donepezil regimen while 533 patients were not diagnosed with bradycardia. In contrast, 52 females were reported as being diagnosed with bradycardia after taking donepezil while 758 females were not diagnosed with bradycardia. Thus, when comparing the ROR value between males and females, females were approximately two thirds less likely to report bradycardia as a side effect than females when taking donepezil for AD treatment

	With Bradycardia	Without Bradycardia	ROR	PRR
Male	56	533
Female	52	758	0.653 (0.441∼0.967)	0.675 (0.470∼0.970)
			p = 0.0337	p = 0.0336

Dosage effect on bradycardia

The association between dosage of donepezil and the occurrence of bradycardia was also evaluated. As shown in Table 5, when comparing the dose-response (5 mg versus 10 mg), no significant difference in reported bradycardia cases was found (p = 0.483).

Table 5

The association between dosage of donepezil and the occurrence of bradycardia. No significant difference in reported bradycardia cases was found (p = 0.483) between two dose groups (5 mg versus 10 mg)

	With Bradycardia	Without Bradycardia	ROR	PRR
Dose (5 mg)	23	357
Dose (10 mg)	22	275	1.242 (0.678 ∼2.275)	1.242 (0.678 ∼2.275)
			p = 0.483	p = 0.483

Severity of bradycardia cases

To assess the seriousness of the reported bradycardia cases, the difference in mortality rates and hospitalization rates was analyzed between people that reported bradycardia as an adverse event and those that did not (Table 6). Although there was no death case reported in those that reported bradycardia, 80 people (64.5%) were subsequently hospitalized in the bradycardia group while 431 people (28.7%) that did not report bradycardia as an adverse event were hospitalized. When comparing the frequency of additional serious medical events, 39 people (31.4%) who reported bradycardia as an adverse event were classified as such relative to 618 people (41.2%) who did not report bradycardia.

Table 6

The seriousness of the reported bradycardia cases. Of those that reported bradycardia, 80 people (64.5%) were subsequently hospitalized while 431 people (28.7%) that did not report bradycardia as an adverse event were hospitalized

	With bradycardia	Without bradycardia
Hospitalization	80	431
Other serious (Important Medical Event)	39	618
Death	0	59
Disability	1	38
Life threatening	0	42
Required intervention	0	4
Congenital Anomaly	0	1
NOT AVAILABLE	4	308
Total cases	124	1501

DISCUSSION

Although some studies have demonstrated the association between reported bradycardia and AChE inhibitors as a drug class, this is the first systematic analysis of the association between bradycardia and donepezil usage. This is also one of the first studies to compare bradycardia risk for patients administered donepezil with the bradycardia risk associated with OTCs, other commonly used AChE inhibitors, and arrhythmia drugs that are known to cause bradycardia in some patients.

It has been shown that an odds ratio can be converted into an effect size by first taking the natural log of the odds ratio and then dividing the resulting number by 1.81 [17]. When compared with OTC drugs, the ROR values for developing bradycardia in patients taking donepezil was 50–400 times higher. The corresponding effect size was approximately 2.16 to 3.31. The ROR for bradycardia in patients administered donepezil relative to patients taking another drug with bradycardia as a known side effect was approximately 1.01∼1.93, resulting in an effect size of 0 to 0.36. Finally, the comparison of other drugs used in AD treatment with donepezil yielded an effect size of 0.82 to 1.35. Based on Cohen’s w criteria for effect size, there was a large, small, and medium effect size when comparing bradycardia frequency in recipients of donepezil with OTCs, arrhythmia drugs, and other commonly used in AD treatment respectively.

We observed 528 bradycardia cases out of 18,186 individuals when taking into account patients that were administered donepezil while taking other drugs. The percentage of bradycardia was slightly lower than it is in single use (donepezil only) patients (124 bradycardia cases out of 1,625 individuals). One possible reason for this is that a greater number of adverse events will be reported as the number of drugs administered increases. That is to say, the total case number is high for individuals taking more drugs. Therefore, the percentage of bradycardia is a little bit lower for those people.

It was observed that the number of bradycardia cases reported by AD patients prescribed donepezil generally increased from 2012 to 2020. It can be explained by two reasons. First, due to the development of the FAERS database and the Medwatch reporting system, the number of total annual FAERS records will rise over time. Therefore, more annual bradycardia cases will increase during time. In addition, label changes or news reports about donepezil and bradycardia may prompt an increase in reported adverse events in FAERS system.

The observed sex differences may be a result of physiological differences between males and females, most notably due to the reduced heart rate in males [18]. As a result of a decreased heart rate, males may be more susceptible to dysregulation in heartbeat control and are more prone to developing bradycardia [18]. When comparing the dose-response (5 mg versus 10 mg), no significant difference in reported bradycardia cases was found (p = 0.483). However, more cases are needed to validate the conclusions about these factors.

In addition, the hospitalization rate was higher in the bradycardia group (64.5%) than in the without bradycardia group (28.7%). However, the difference in death, life threatening cases, and cases that resulted in disability were higher in the group that did not report bradycardia. These results may be a product of an increased likelihood of detecting bradycardia when in a hospital setting. Thus, we cannot conclusively prove a causal relationship between bradycardia and subsequent medical outcome.

In multiple studies, AChE inhibitors have been linked to a variety of adverse events including bradycardia. Inhibition of AChE stifles the recycling of acetylcholine and allows for prolonged innervation of neurons in the hippocampus and other regions of the brain [6]. In particular, acetylcholine stimulates neurons comprising the parasympathetic nervous system, resulting in a decrease in cardiac output, decreased heart rate, and a return to a physiological state of rest [19]. Thus, the typical mechanism of action that confers bradycardia as a known side effect of AChE inhibitors is overstimulation of the parasympathetic nervous system, thus demonstrating why AChE inhibitors such as donepezil, rivastigmine, and galantamine are more likely to cause bradycardia in patients than OTCs.

However, two additional mechanisms of action are responsible for the higher frequency of reported bradycardia cases in patients that are administered exclusively donepezil for treatment of cognitive decline. In the first mechanism, donepezil serves as an agonist of endothelial nitric oxide synthase (eNOS), of which subsequently generates the potent neurotransmitter nitric oxide [20, 21]. The source of nitric oxide generation frequently dictates whether or not the parasympathetic nervous system, which primarily reduces heart rate, or the sympathetic nervous system is stimulated, which elevates heart rate [21]. In the case of donepezil-mediated stimulation of eNOS activity, the resulting NO produced innervates neurons comprising the parasympathetic nervous system through downstream signaling. Nitric oxide binds to soluble guanylate cyclase and generates a positive feedback loop that strengthens cGMP-mediated downstream signaling [22]. An increase in cGMP production subsequently promotes protein kinase G (PKG) activity and stimulates downstream activity of myosin phosphatase, of which targets smooth muscle cells and facilitates smooth muscle relaxation through the removal of a phosphate group on the regulatory chain of myosin II [22]. However, upregulation of PKG activity is associated with bradycardia and is a product of extensive relaxation of the smooth muscles of the heart [22].

The second proposed mechanism that contributes to the elevated risk of bradycardia in patients administered donepezil is the donepezil-mediated inhibition of TNF activity. Under normal physiological conditions, activation of TNF leads to downstream activation of NF-κB through the recruitment of a member of the TNF receptor associated factor (TRAF) family of proteins [23]. TRAF mediates downstream activation of NF-κB inducing kinase and subsequent phosphorylation of IκB kinase (IKK), an event which releases NF-κB from its sequestered state and at which point enables migration of NF-κB to the nucleus [23]. In the nucleus, NF-κB functions as a transcription factor and stimulates the expression of many different downstream genes, of which includes inducible nitric oxide synthase (iNOS) [24]. Nitric oxide generated by iNOS serves as a compensatory mechanism to eNOS-mediated lowering of heart rate and has been shown to elevate heart rate (tachycardia) through stimulation of neurons comprising the sympathetic nervous system [21]. However, disruption of this signaling pathway through donepezil-mediated inhibition of TNF signaling stifles the restoration of a low heart rate to a physiologically standard heart rate, leaving a patient in a state of perpetual lowered heart rate. Thus, the coupling of overstimulation of the parasympathetic nervous system with the inhibition of the sympathetic nervous system could be a possible means in which donepezil imposes a greater risk of developing bradycardia than other AChEs. Figure 4 summarizes the proposed mechanisms for elevated bradycardia risk in patients taking donepezil for AD treatment.

Fig. 4

The proposed mechanisms to explain the elevated risk of bradycardia in patients prescribed donepezil relative to the risk of bradycardia. The first mechanism is through stimulation of endothelial nitric oxide synthase and the overproduction of nitric oxide, leading to an increase in muscle relaxation and lowered heart rate (bradycardia). The second mechanism functions through the inhibition of TNF-mediated overexpression of inducible nitric oxide synthase (INOS) through suppression of the transcription factor NF-κB. This may prevent the compensatory increase in heart rate facilitated by INOS and maintain a physiological state of bradycardia.

As a result of the relative novelty of this study, some limitations exist. Because this study focused on bradycardia in patients prescribed only one drug to avoid the effects of other medications, some drugs may either not be widely prescribed or may primarily be used in conjunction with other AD therapeutics, thus leading to a possibly low sample size for these drugs. In addition, OTC drugs are not perfect negative controls as underreporting is somewhat high. Therefore, the odds ratio may not be accurate for these drugs. However, with the development of MedWatch program, approximately 1 million new reports are added into the FAERS database every year so that reliability of results should continue to increase over the next several years.

Using known and postulated drug targets for donepezil, a mechanism was proposed to explain the disparity in reported bradycardia cases between donepezil, other AChE inhibitors, OTCs, and arrhythmia therapeutics. Additional biological verification and analysis of transcriptome data would be necessary to further elucidate the underlying mechanisms that drive the greater bradycardia risk associated with donepezil usage. Further studies will seek to explain this observed discrepancy and identify additional variables that may be associated with an increased risk of bradycardia. In addition, future studies will extrapolate on how various drug-drug interactions [15, 25] may affect bradycardia risk and how different combination therapies may be associated with an increase in bradycardia cases.

CONCLUSION

The most frequently reported adverse event for adult patients administered donepezil was bradycardia, a condition defined as an abnormally low heart rate. Utilizing clinical data derived from the FAERS database, the frequency of reported bradycardia cases in patients taking different medications was compared. Patients administered donepezil were 4–11 times more likely to report bradycardia as an adverse event than patients administered memantine, galantamine, and rivastigmine. When compared to a series of drugs that are commonly used to treat arrhythmias and have bradycardia as a known side effect, bradycardia was found to be a more commonly observed adverse event in patients administered donepezil. Sex differences were observed between men and women and may be a result of men having lower heart rates on average than their female counterparts, thus making them more susceptible to abnormally low heart rates. However, no statistically significant association was present between age or dosage of donepezil and the likelihood of reporting bradycardia as an adverse event upon donepezil treatment. Despite these promising results, further biological validation is required to strengthen the demonstrated association between donepezil usage and increased risk of bradycardia. Future directions will focus on transcriptomic data and the validation of nitric oxide synthase dysregulation as the source of the disparity in bradycardia risk between donepezil usage and the use of other AChE inhibitors as well as possible drug-drug interactions.

Footnotes

ACKNOWLEDGMENTS

This research was funded by a National Institutes of Health grant (1R03AG070536-01) to Feng Cheng.

Authors’ disclosures available online ().

References

Setti

, Hunsberger

, Reed

(2017) Alterations in hippocampal activity and Alzheimer’s disease. Transl Issues Psychol Sci 3, 348–356.

Morris

, Armbruster

, Silva

, Widell

, Cheng

(2020) The association between the usage of non-steroidal anti-inflammatory drugs and cognitive status: Analysis of longitudinal and cross-sectional studies from the global Alzheimer’s Association interactive network and transcriptomic data. Brain Sci 10, 961–973.

(2020) 2020 Alzheimer’s disease facts and figures.heimers Dement}. Alz{- 16, 391–460.

Shaikh

, Verma

, Siddiqui

, Ahmad

, Rizvi

, Shakil

, Biswas

, Singh

, Siddiqui

, Shakil

, Tabrez

, Kamal

(2014) Current acetylcholinesterase-inhibitors: A neuroinformatics perspective. CNS Neurol Disord Drug Targets 13, 391–401.

Haam

, Yakel

(2017) Cholinergic modulation of the hippocampal region and memory function. J Neurochem 142(Suppl 2), 111–121.

Ferreira-Vieira

, Guimaraes

, Silva

, Ribeiro

(2016) Alzheimer’s disease: Targeting the Cholinergic System. Curr Neuropharmacol 14, 101–115.

Abdrakhmanov

, Petrov

, Grigoryev

, Zefirov

(2013) Depolarization-induced calcium-independent synaptic vesicle exo- and endocytosis at frog motor nerve terminals. Acta Naturae 5, 77–82.

Zimmermann

(2013) Neuronal AChE splice variants and their non-hydrolytic functions: Redefining a target of AChE inhibitors?. Br J Pharmacol 170, 953–967.

Garcia-Ayllon

, Small

, Avila

, Saez-Valero

(2011) Revisiting the role of acetylcholinesterase in Alzheimer’s disease: Cross-talk with P-tau and beta-amyloid. Front Mol Neurosci 4, 22.

10.

Alvarez

, Opazo

, Alarcon

, Garrido

, Inestrosa

(1997) Acetylcholinesterase promotes the aggregation of amyloid-beta-peptide fragments by forming a complex with the growing fibrils. J Mol Biol 272, 348–361.

11.

Christou

, Seals

(2008) Decreased maximal heart rate with aging is related to reduced beta-adrenergic responsiveness but is largely explained by a reduction in intrinsic heart rate. J Appl Physiol (1985) 105, 24–29.

12.

Park-Wyllie

, Mamdani

, Li

, Gill

, Laupacis

, Juurlink

(2009) Cholinesterase inhibitors and hospitalization for bradycardia: A population-based study. PLoS Med 6, e1000157.

13.

Rosenbloom

, Finley

, Scheinman

, Feldman

, Miller

, Rabinovici

(2010) Donepezil-associated bradyarrhythmia in a patient with dementia with Lewy bodies (DLB). Alzheimer Dis Assoc Disord 24, 209–211.

14.

Hernandez

, Farwell

, Cantor

, Lawler

(2009) Cholinesterase inhibitors and incidence of bradycardia in patients with dementia in the veterans affairs new England healthcare system. J Am Geriatr Soc 57, 1997–2003.

15.

Pham

, Cheng

, Ramachandran

(2019) A comon study of algorithms to detect drug-adverse event associations: Frequentist, Bayesian, and machine-learning approaches. Drug Saf 42, 743–750 paris.

16.

Evans

, Waller

, Davis

(2001) Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 10, 483–486.

17.

Chinn

(2000) A simple method for converting an odds ratio to effect size for use in meta-analysis. Stat Med 19, 3127–3131.

18.

Casin

, Kohr

(2020) An emerging perspective on sex differences: Intersecting S-nitrosothiol and aldehyde signaling in the heart. Redox Biol 31, 101441.

19.

Harvey

(2012) Muscarinic receptor agonists and antagonists: Effects on cardiovascular function. Handb Exp Pharmacol, 299-316.

20.

Constantinescu

, Constantinescu

, Andercou

, Mironiuc

(2013) Influence of sildenafil and donepezil administration on the serum redox balance in experimentally induced lower limb critical ischemia. Clujul Med 86, 250–258.

21.

Almeida

, Oliveira

, Benini

, Crestani

(2019) Differential roles of hippocampal nNOS and iNOS in the control of baroreflex function in conscious rats. Brain Res 1710, 109–116.

22.

Stasch

, Pacher

, Evgenov

(2011) Soluble guanylate cyclase as an emerging therapeutic target in cardiopulmonary disease. Circulation 123, 2263–2273.

23.

Shi

, Sun

(2018) Tumor necrosis factor receptor-associated factor regulation of nuclear factor kappaB and mitogen-activated protein kinase pathways. Front Immunol 9, 1849.

24.

Katsuyama

, Shichiri

, Marumo

, Hirata

(1998) NO inhibits cytokine-induced iNOS expression and NF-kappaB activation by interfering with phosphorylation and degradation of IkappaB-alpha. Arterioscler Thromb Vasc Biol 18, 1796–1802.

25.

, Shen

, Pietsch

, Nagar

, Fadli

, Huang

, Tu

, Cheng

(2015) A novel algorithm for analyzing drug-drug interactions from MEDLINE literature. Sci Rep 5, 17357.