Key Points
Anticholinergic adverse drug reactions—such as confusion, urinary retention, and constipation—are not limited to older adults and can occur even when only one drug is involved.Current anticholinergic scoring tools do not necessarily reflect the full spectrum of medications implicated in the anticholinergic events reported through pharmacovigilance channels. These findings highlight the need for updated guidelines and greater awareness about anticholinergic risk among healthcare professionals.
Introduction
As a crucial neurotransmitter in both the central and peripheral nervous systems, acetylcholine (ACh) has a key role in the regulation of a broad variety of physiological processes. Acetylcholine acts on nicotinic and muscarinic receptors. Muscarinic receptors are particularly important in many therapeutic strategies and mediate a range of processes, such as muscle contraction, heart rate regulation, and modulation of the autonomic nervous system including heart rate regulation, intestinal motility, eye functions.
Many drugs are designed to either mimic or block the actions of ACh through the muscarinic receptors and are thus used to treat conditions like overactive bladder, Parkinson's disease, and neurological and psychiatric conditions [1]. Others exhibit anticholinergic properties in addition to their primary pharmacological effects. As such, cholinergic and anticholinergic drugs are widely prescribed to older adults [1].
Although anticholinergic drugs are effective in the treatment of many conditions, their use also may lead to adverse drug reactions (ADRs). These reactions include confusion, tachycardia, constipation, intestinal occlusion, and urine retention [2, 3], and are therefore of special concern in older adults. A high “anticholinergic burden” (i.e., the cumulative effect of taking one or more drugs with anticholinergic properties) can exacerbate individual effects and increase the risk of functional decline.
Hence, the careful management of anticholinergic drugs is extremely important in clinical practice.
There are various ways of indirectly assessing the anticholinergic activity of drugs. The evaluation of serum anticholinergic activity (SAA) is the acknowledged “gold standard” laboratory technique [4]. However, this evaluation is not available on a routine basis—partly because it involves a radioreceptor assay.
As a consequence, various anticholinergic drug scales have been developed for calculation of the anticholinergic burden [5–17]. All these scales use quite the same method to score drugs. A score of 1 is assigned to drugs with a possible anticholinergic effect.
A score of 2 is given when cognitive side effects are expected, and a score of 3 or higher when such side effects are common or particularly severe. The overall anticholinergic burden for each patient prescription is determined by summing the anticholinergic score of each constituent drug. However, the levels of agreement between the various anticholinergic burden scales are only low to moderate [18].
Furthermore, these scores have been developed independently in particular countries and so cannot be easily extrapolated to other countries in which the drugs included in the original scoring system are not always marketed. Consequently, numerous published studies have sought to adapt existing anticholinergic burden scales to their respective national pharmacopoeias, incorporating local drug availability and expert assessment. As a result, the scales show significant discrepancies in drug selection: some drugs are given a score of 3 on one scale and a score of 1 on another.
Next, some anticholinergic drugs are not referenced by any of the scales. Lastly, the scales currently in use consider the anticholinergic effects of each individual drug without considering the overall effects associated with an entire pharmacological drug class (e.g., antihistamines, tricyclic antidepressants, etc.). This can lead to drugs in the same class being given different weights on a given scale.
In addition, most of these scales are based on the drugs’ in vitro activity and expert opinion on selected molecules, while the potential acute clinical consequences of these exposures in real-life settings are often overlooked.
We hypothesised that these scales may not capture all the relevant ADRs. The overall aim of the present study was therefore to determine the extent to which currently available anticholinergic burden scales capture the ADRs reported in the French Pharmacovigilance Database (Base Nationale de Pharmacovigilance) and provide an in-depth analysis of all reported anticholinergic ADRs associated with medications previously identified as having a high anticholinergic risk. More specifically, our objectives (based on a selection of drug classes known to induce central and peripheral anticholinergic ADRs) were to describe (i) the number and types of anticholinergic ADRs notified in the French pharmacovigilance database (FPVD), and (ii) the types of anticholinergic drug as a function of their classification by the anticholinergic burden scales.
Materials and Methods
Data Source
Individual case safety reports (ICSRs) from the FPVD were analysed. The FPVD was established in 1973 and consists of a network of 30 regional centres, to which all ADRs should be reported spontaneously by health professionals and/or patients.
Each ICSR includes anonymised administrative data (the notifier’s status), demographic variables (patient age and gender), drug-related information (the international non-proprietary name [INN] or trade name, the indication, the initiation date, the discontinuation date, the dosing regimen, and the administration route), and information on the suspected ADR (coded according to the Medical Dictionary for Regulatory Activities [MedDRA®]), and on the concomitant drugs [19]. There are five levels in the MedDRA® hierarchy, which range from very general (the system organ class [SOC]) to specific [the preferred term (PT) or the lowest-level term]). The dictionary also includes standardised MedDRA® queries (SMQs), which are collections of MedDRA® terms consistent with a description of a clinical syndrome associated with the ADR and the drug exposure.
As such, SMQs are useful for wide-ranging searches.
The regional pharmacovigilance centres, through multidisciplinary committees, evaluate the causal relationships between ADRs and drugs. For each ADR report, imputability (defined as the likelihood of a causal relationship between the suspected drug and the reported ADR) is assessed according to the FPVD’s standard procedure [20]. If a drug is considered to be at least probably responsible for the adverse event (AE), it is defined as “suspect” or “interacting” in the ICSR; if not, it is defined as “concomitant”.
An ADR is categorised as “serious” if it results (at any dose) in an untoward medical occurrence, such as (from the least serious to the most serious) important medical event, initial or prolonged hospitalisation, persistent or significant disability/incapacity, congenital anomaly, a life-threatening event, or death.
Selection of Anticholinergic ADRs
In the present study, we selected ADRs recorded in the FPVD between January 1st, 1985, and November 30th, 2024; this period was selected because many anticholinergic drugs were introduced decades ago, and underreporting tends to increase with the time since market authorisation [21]. Given that the SMQ "anticholinergic syndrome" was not suitable for our descriptive project, we refined the case definition for several reasons. First, this SMQ includes PTs that are not typically associated with anticholinergic toxicity (e.g., toxicity to various agents, thinking abnormal, and ataxia).
Second, some included PTs are overly broad or insufficiently specific for reliably capturing anticholinergic ADRs. Third, the SMQ provides an incomplete representation of gastrointestinal manifestations, as it contains only PTs restricted to the oral cavity (e.g., dry mouth, dysphagia, thirst) and lacks lower gastrointestinal tract symptoms commonly described in cases of anticholinergic toxicity. Therefore, we selected PTs corresponding to typical anticholinergic ADRs, such as confusion, angle-closure glaucoma (referred to simply as “glaucoma” hereafter), tachycardia, urinary retention, constipation, intestinal obstruction, and other peripheral anticholinergic ADRs (Table 1).
This selection was based on pharmacological knowledge [2, 3]. We excluded cases that were not confirmed by a physician, pregnancy-related cases, and cases for which the intake of the drug of interest and the ADR were chronologically incompatible (i.e., a chronological parameter reported as ‘0’). Only cases categorised as “suspect” were included.Table 1Definitions of anticholinergic adverse drug reactions (ADRs)Type of anticholinergic ADRMedDRA® preferred termAnticholinergic effectsAnticholinergic effectAnticholinergic syndromeConfusionConfusional stateDisorientationGlaucomaAcute myopiaAngle closure glaucomaBorderline glaucomaGlaucomaGlaucoma drainage device placementGlaucoma surgeryIntraocular pressure increasedIridotomyNarrow anterior chamber angleOcular hypertensionPupillary blockTrabeculectomyTachycardiaHeart rate increasedInappropriate sinus tachycardiaMaximum heart rate increasedSinus tachycardiaSupraventricular tachycardiaTachycardiaTachycardia paroxysmalUrinary retentionBladder dilatationUrinary retentionConstipationConstipationFaecalomaFaeces hardInfrequent bowel movementsObstructive defaecationIntestinal obstructionBowel obstruction surgeryColitis ischaemicColon gangreneDistal intestinal obstruction syndromeDuodenal obstructionDuodenal stasisGastric ileusGastric volvulusGastrointestinal decompressionGastrointestinal gangreneGastrointestinal ischaemiaGastrointestinal mucosal necrosisGastrointestinal necrosisGastrointestinal obstructionGastrointestinal perforationIleusIleus paralyticIleus spasticIntestinal gangreneIntestinal infarctionIntestinal ischaemiaIntestinal obstructionIntestinal perforationIntestinal pseudo-obstructionIschaemic enteritisLarge intestinal obstructionLarge intestinal obstruction reductionLarge intestine perforationLower gastrointestinal perforationMalignant gastrointestinal obstructionNecrotising colitisNecrotising gastritisNecrotising oesophagitisObstruction gastricOesophageal obstructionSmall intestinal obstructionSmall intestinal obstruction reductionSubileusVolvulusMydriasisMydriasisPupil fixedOther peripheral anticholinergic effectsAnhidrosisDry eyeDry mouthHyperhidrosisThirst
Drugs Analysed
The drugs used for case selection were (i) those with a high anticholinergic score, according to existing scales [5, 6, 10–12, 14–17, 22–34] and (ii) all the compounds in the corresponding Anatomical Therapeutic Chemical (ATC) drug class (ATC level 4) (Supplementary Table 1). Drugs containing multiple active substances within an ATC class were not included in the study.
Statistical Analyses
Descriptive statistics were used to summarise the cases’ baseline characteristics. Depending on the type and distribution of the data, we used the chi-square test for categorical variables, and Student’s t-test, Welch’s t-test or the Wilcoxon rank-sum test for continuous variables.
To explore patterns of co-implication, chord diagrams were used to visualise pairwise associations between medications identified as potentially imputable for the same type of ADR.
Results
Regarding the selected drugs, 101,365 cases had been notified in the FPVD. Of these, 9629 (9.5%) corresponded to at least one anticholinergic ADR (Supplementary Table 2). Patients who experienced at least one anticholinergic ADR had a median age of 61 years (interquartile range [IQR], 38–79), and the majority were women (58%) (Table 2).Table 2Characteristics of the study populationOverall (N = 101,365)Cases without an anticholinergic ADR* (N = 91,736)Cases with at least one anticholinergic ADR (N = 9629)p-valueWomen, n (%)MenSex unknown58,573 (58%)42,451 (42%)34153,034 (58%)38,385 (42%)3175539 (58%)4066 (42%)240.24Age, median [IQR]Missing58 [37–76]212357 [37–76]194961 [38–79]174<0.001IQR interquartile range*All these patients had an adverse drug reaction (ADR) but not an anticholinergic ADR, as defined in the Methods sectionWe considered that each case concerned a single patient; that is a strong hypothesis and explains why the present analysis is exploratory
Types of Anticholinergic ADR
Confusion was the most frequently reported anticholinergic ADR (4603 cases, of which 81% were classified as serious), followed by tachycardia (n = 1541 cases, 70% serious), urinary retention (1061 cases, 75% serious), other peripheral anticholinergic effect (866 cases—mostly dry mouth—31% serious), mydriasis (826 cases, 57% serious), constipation (762 cases, 58% serious), intestinal obstruction (561 cases, 98% serious), and, lastly, glaucoma (103 cases, 72% serious) (Table 3). Patients presenting confusion were older (median age: 75 years [IQR, 57–84]) than those reporting tachycardia (38 [22–58]) or mydriasis (24 [11–40]). Women were overrepresented for all types of anticholinergic ADRs other than intestinal obstruction.Table 3Age and sex distribution and the proportion of serious cases, by type of anticholinergic adverse drug reaction (ADR)Type of anticholinergic ADR (N patients)AgeMedian [IQR]Missing dataWNomen (%)Missing data% of cases judged to be seriousConfusion (N = 4603)75 [57–84]622675 (58%)881%Glaucoma (N = 103)63 [51–73]271 (69%)-72%Tachycardia (N = 1541)38 [22–58]37897 (58%)470%Urinary retention (N = 1061)65 [40.25–80]15533 (50%)375%Constipation (N=762)60 [41.75–77]14435 (57%)158%Intestinal obstruction (N = 561)57 [37–71]8251 (45%)198%Mydriasis (N = 826)24 [11–40]17513 (62%)257%Other peripheral anticholinergic effects (N = 866)53 [36–67]32551 (64%)631%IQR interquartile rangeA single patient may have experienced more than one type of ADR and is counted once for each ADR they experienced.
For example, a single patient can be counted under both "confusion" and "constipation"
Most cases (n = 8812; 92%) had experienced only one type of anticholinergic ADR, while 717 (7%) had experienced two types, and 100 (1%) had experienced three or more types. When considering patients having experienced two or more anticholinergic ADRs, the most frequently observed combinations were urinary retention with confusion, urinary retention with constipation, and mydriasis with tachycardia (data not shown).
Drugs Associated with Anticholinergic ADRs
In the present analysis, the FPVD was queried about 290 molecules, of which 107 had the highest possible anticholinergic score on the scales studied here (Supplementary Table 1). A total of 140 molecules were reported in the FPVD as having led to at least one anticholinergic ADR investigated.
The drug classes with the highest number of reports were (by far), anxiolytics, antidepressants, and antipsychotics (Supplementary Table 3). The pharmacological classes varied as a function of the type of ADR. The pharmacological classes most frequently associated with confusion, tachycardia, urinary retention, and constipation were anxiolytics, antidepressants, and antipsychotics (Supplementary Table 3).
In cases of intestinal obstruction, antipsychotics were implicated in half of the reports, followed by anticholinergic antiparkinsonian drugs. Laxatives were reported as concomitant drugs in 80 of the 615 cases of constipation or intestinal obstruction associated with antipsychotics, and in 48 of 273 cases involving antiparkinsonian drugs. Regarding glaucoma, an antidepressant was suspected in nearly half of the cases.
For mydriasis, around one-quarter of the reported cases involved mydriatic agents, and one-fifth occurred during treatment with an anxiolytic.
Drugs with a high anticholinergic burden (such as paroxetine, amitriptyline, tropatepine, and cyamemazine) were in the top five most frequently implicated drugs for most types of anticholinergic ADR (Table 4). Interestingly, some of the drugs associated with anticholinergic ADRs in the present study are not present in (or are assigned a low score by) the anticholinergic scales used widely in routine clinical practice (e.g., the Anticholinergic Cognitive Burden [ACB] scale) (Table 5). For example, loxapine has a moderate anticholinergic score according to several scales but was among the top five drugs associated with constipation, intestinal obstruction, tachycardia, and urinary retention.
Likewise, bromazepam has a low anticholinergic score on the ACB scale but was among the top five drugs implicated in cases of mydriasis.Table 4The five most frequently implicated drugs, by type of anticholinergic adverse drug reaction (ADR)Type of anticholinergic ADRATCINNN (%)Highest anticholinergic score reported by at least one scaleConfusion (n = 5974)N06AB05Paroxetine323 (5.4%)3 (ACB, Chew, AAS, DRS, BAA, mACB1, mACB2)N05BA08Bromazepam296 (5.0%)1 (DRS, AIS)N06AA09Amitriptyline288 (4.8%)5 (ATS), 4 (Chew, AAS, DRS), 3 (Summers, ABD, ADS, ARS, ACB, CrAS, mARS, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)C03CA01Furosemide276 (4.6%)3 (ABC)N05BA04Oxazepam269 (4.5%)1 (ADS, ALS, DRS, Ger-ACB, mACB2, CALS, AIS, Swe-ABS)Glaucoma (n=117)N06AB05Paroxetine16 (13.7%)3 (ACB, Chew, AAS, DRS, BAA, mACB1, mACB2)N06AA04Clomipramine8 (6.8%)4 (DRS), 3 (ABC, ADS, ACB, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)N06AB06Sertraline8 (6.8%)2 (mACB2)N06AB04Citalopram7 (6.0%)2 (Chew, AAS, DRS)R03BB01Ipratropium bromide7 (6.0%)4 (AAS), 3 (BAA, AIS)Tachycardia (n = 2061)N05AA06Cyamemazine151 (7.3%)3 (AIS)N06AA09Amitriptyline113 (5.5%)5 (ATS), 4 (Chew, AAS, DRS), 3 (Summers, ABD, ADS, ARS, ACB, CrAS, mARS, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)N05AH01Loxapine106 (5.1%)2 (ADS, ACB, Ger-ACB, KABS, mACB2, CALS, AIS)N05AH02Clozapine95 (4.6%)4 (Chew, AAS, DRS)N05AH03Olanzapine76 (3.7%)3 (ACB, Chew, AAS, DRS, BAA, KABS, mACB1, mACB2)Urinary retention (n = 1531)N05AA06Cyamemazine96 (6.3%)3 (AIS)N05BB01Hydroxyzine94 (6.1%)4 (DRS), 3 (ABC, ADS, ARS, ACB, mARS, Ger-ACB, BAA, KABS, mACB2, CALS, AIS, Swe-ABS)N06AA09Amitriptyline88 (5.7%)5 (ATS), 4 (Chew, AAS, DRS), 3 (Summers, ABD, ADS, ARS, ACB, CrAS, mARS, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)N05AH01Loxapine80 (5.2%)2 (ADS, ACB, Ger-ACB, KABS, mACB2, CALS, AIS)N04AA12Tropatepine75 (4.9%)3 (ABC, CALS, AIS)Constipation (n = 1223)N05AH01Loxapine96 (7.8%)2 (ADS, ACB, Ger-ACB, KABS, mACB2, CALS, AIS)N05AA06Cyamemazine79 (6.5%)3 (AIS)N04AA12Tropatepine73 (6.0%)3 (ABC, CALS, AIS)N06AA09Amitriptyline65 (5.3%)5 (ATS), 4 (Chew, AAS, DRS), 3 (Summers, ABD, ADS, ARS, ACB, CrAS, mARS, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)N05AH02Clozapine56 (4.6%)4 (Chew, AAS, DRS)Intestinal obstruction (n = 1147)N04AA12Tropatepine127 (11.1%)3 (ABC, CALS, AIS)N05AH01Loxapine123 (10.7%)2 (ADS, ACB, Ger-ACB, KABS, mACB2, CALS, AIS)N05AA06Cyamemazine103 (9.0%)3 (AIS)N05AH02Clozapine102 (8.9%)4 (Chew, AAS, DRS)R06AD01Alimemazine62 (5.4%)3 (AEC)Mydriasis (n = 1021)S01FA01Atropine112 (11.0%)4 (Chew, DRS), 3 (Summers, Peters, ADS, ARS, ACB, CrAS, ALS, mARS, AEC, ATS, BAA, KABS, mACB1, CALS, AIS, Swe-ABS)S01FA04Cyclopentolate96 (9.4%)3 (Summers)A04AD01Scopolamine75 (7.3%)4 (DRS), 3 (Summers, Peters, ADS, ACB, CrAS, AEC, Ger-ACB, KABS, CALS, AIS, Swe-ABS)N06AB05Paroxetine64 (6.3%)3 (ACB, Chew, AAS, DRS, BAA, mACB1, mACB2)N05BA08Bromazepam35 (3.4%)1 (DRS, AIS)Other peripheral anticholinergic effects (n=1056)N04AA09Amitriptyline90 (8.5%)5 (ATS), 4 (Chew, AAS, DRS), 3 (Summers, ABD, ADS, ARS, ACB, CrAS, mARS, AEC, Ger-ACB, BAA, KABS, mACB1, mACB2, CALS, AIS, Swe-ABS)M03BX01Baclofen84 (8.0%)2 (ARS, CrAS, mARS, BAA, mACB1, CALS, AIS)G04BD04Oxybutynin48 (4.5%)3 (ABC, ADS, ARS, ACB, Chew, AAS, mARS, DRS, AEC, Ger-ACB, BAA, KABS, mACB1, CALS, AIS, Swe-ABS)N05AA06Cyamemazine47 (4.5%)3 (AIS)N06AB05Paroxetine47 (4.5%)3 (ACB, Chew, AAS, DRS, BAA, mACB1, mACB2)For each type of anticholinergic ADR, the number (n =) corresponds to the total number of drug-ADR pairs. For example, 4603 patients experienced confusion, but two or more drugs were implicated in many cases, resulting in a total of 5974 drug-confusion pairs.Anticholinergic scales: AAS, Anticholinergic Activity Scale [11]; ABC, Anticholinergic Burden Scale [10]; ACB, Anticholinergic Cognitive Burden [5]; ADS, Anticholinergic Drug Scale [22]; AEC, Anticholinergic Effect on Cognition Scale [12]; AIS, Anticholinergic Impregnation Scale [23]; ALS, Anticholinergic Load Scale [14]; ARS, Anticholinergic Rating Scale [6]; ATS, Anticholinergic Toxicity Scale [24]; BAA, Brazilian Anticholinergic Activity Drug Scale [16]; CALS, CRIDECO Anticholinergic Load Scale [25]; Chew, Chew's list [26]; CrAS, Clinician-rated Anticholinergic Scale [27]; DRS, Delirogenic Risk Scale [34]; Ger-ACB, German Anticholinergic Cognitive Burden [15]; KABS, Korean Anticholinergic Burden Scale [17]; mACB1, modified Anticholinergic Burden Scale [28]; mACB2, modified ACB [29]; mARS, modified ARS [30]; Peters, Peters’ Scale [31]; Summers, Summers’ Scale [32]; Swe-ABS, Swedish Anticholinergic Burden Scale [33] ATC Anatomical Therapeutic Chemical, INN International Nonproprietary NamesTable 5The 10 drugs with the highest number of reported anticholinergic adverse drug reactions5th level ATC classINNAbsolute no. reported anticholinergic ADRsNo. anticholinergic scales citing the drug (out of 22)N06AA09Amitriptyline72721N05AA06Cyamemazine6941N06AB05Paroxetine63118N05AH01Loxapine5067N05BB01Hydroxyzine50615N04AA12Tropatepine4963N06AA04Clomipramine46414N05BA12Alprazolam45017N05BA08Bromazepam4482N05BA04Oxazepam43110The drug with the highest number of reported anticholinergic adverse drug reactions is derived from Supplementary Table 3ADRs adverse drug reactions, ATC Anatomical Therapeutic Chemical, INN International Nonproprietary Names
Of the 9629 cases featuring at least one anticholinergic ADR, 7324 (76%) involved a single drug, while 2305 (24%) involved two or more drugs with anticholinergic properties. In these 2305 cases, the three most common combinations were loxapine-tropatepine, cyamemazine-tropatepine, and olanzapine-cyamemazine (Fig. 1 and Supplementary Figs 1 to 7).Fig.
1Pairwise associations in cases (n = 2305) in which two or more drugs were implicated. Note: only pairs that occurred at least 20 times are shown here. The broader the link between two drugs, the more frequently the association was found to be involved in anticholinergic adverse drug reactions (ADRs).
The most frequently reported drug pairs were loxapine-tropatepine, cyamemazine-tropatepine, and olanzapine-cyamemazine
Discussion
Using data from the FPVD, we conducted an in-depth analysis of all reported anticholinergic ADRs associated with medications previously identified as having a high anticholinergic risk and other medications in the same pharmacological classes. When considering nearly 10,000 reports, confusion was the most frequently reported anticholinergic ADR, followed by two peripheral ADRs (tachycardia and urinary retention). Intestinal obstruction was the ADR with the highest proportion of serious cases: 98%.
Anxiolytics, antidepressants and antipsychotics were the three drug classes most frequently implicated in the anticholinergic ADRs analysed here. It is noteworthy that some of the individual drugs implicated in the anticholinergic ADRs studied here are not included in some existing anticholinergic risk scales (such as alizapride, aprepitant, thiocolchicoside, rupatadine, bilastine, among others) or are only assigned low risk scores by these scales (such as bromazepam, levocetirizine, trimebutine, among others). Furthermore, our results highlighted a major weakness of current anticholinergic scoring tools in that they do not necessarily reflect the full spectrum of medications implicated in the anticholinergic events reported through pharmacovigilance channels.
Anticholinergic ADRs have been particularly well characterised in older adults, who are especially vulnerable to central anticholinergic effects [6, 35, 36]. However, the median patient age observed in the present study (61 years) shows that anticholinergic ADRs (and particularly peripheral anticholinergic ADRs) do not occur solely in older adults. Anticholinergic ADRs are therefore not limited to geriatric or poly-medicated patients, and greater awareness among prescribers is warranted.
Notably, the median age was 57 among patients having experienced intestinal obstruction (classified as serious in 98% of cases) and 24 among those having experienced mydriasis (likely related to excessively intense and unintended mydriasis occurring in the context of cycloplegia induction for ophthalmologic examination in young children).
Confusion was reported most frequently amongst the anticholinergic ADRs in this large pharmacovigilance dataset, followed by urinary retention and tachycardia. Central anticholinergic ADRs (such as confusion, delirium, hallucinations, sedation, and cognitive impairment) are well documented—particularly in older adults. These reactions are primarily attributed to the antagonism of muscarinic receptors in the central nervous system, and especially the M1 receptors involved in cognitive function.
Epidemiological studies have evidenced an association between the cumulative anticholinergic burden on one hand and cognitive decline, an elevated risk of delirium, and even long-term dementia on the other—particularly in frail or older adult patients [37, 38] However, it is increasingly acknowledged that central side effects may not be solely caused by a drug’s anticholinergic activity. Pharmacokinetic factors (such as high lipophilicity and the ability to cross the blood-brain barrier) can amplify central ADRs. Moreover, pharmacodynamic interactions involving other neurotransmitter systems (dopaminergic, serotonergic, or histaminergic pathways) might also contribute to central symptoms – especially in the context of polypharmacy.
This involvement might explain why drugs with a low, moderate, or zero assigned anticholinergic score can be associated with central nervous system effects in clinical practice [39, 40].
Peripheral anticholinergic ADRs were frequently reported in the FPVD. The two peripheral ADRs reported most were tachycardia and urinary retention. It is noteworthy that glaucoma was the least frequently reported peripheral anticholinergic ADR.
The peripheral ADRs varied in terms of seriousness; indeed, the proportion of serious cases ranged from 57% for mydriasis to 98% for intestinal obstruction. However, differences in seriousness might reflect confounding by indication, patient characteristics, or differential reporting patterns.
We found that, in most cases, only a single anticholinergic ADR was reported. Pharmacologists in French pharmacovigilance centres are trained to actively investigate and document all potential anticholinergic side effects when assessing cases flagged for an anticholinergic ADR. This suggests that cholinergic imbalance rarely manifests as a full-blown anticholinergic syndrome (defined by the presence of two or more symptoms) but more often as a single, isolated anticholinergic symptom.
Notably, in approximately three-quarters of the reported cases, only one drug with anticholinergic potential was suspected to be responsible for the adverse reaction. This finding contrasts with the concept of anticholinergic burden as defined by scoring systems, which generally assumes that anticholinergic adverse reactions result from the cumulative anticholinergic properties of multiple drugs. This does not preclude the occurrence of ADRs in patients exposed to several anticholinergic medications; however, multidrug situations were not predominant in the pharmacovigilance reports analysed in the present study.
Nevertheless, clinicians should remain vigilant when prescribing combinations of drugs with anticholinergic properties, as these might still potentiate the risk of anticholinergic AEs.
We also demonstrate that the most frequently reported ADRs varied as a function of the drug class. For example, urinary retention was most often linked to drugs used to treat urinary incontinence, whereas gastrointestinal disorders were the predominant ADRs for anxiolytics, antidepressants, and antipsychotics. These findings might foster the implementation of strategies for mitigating anticholinergic ADRs.
In line with our observation of a high proportion of gastrointestinal ADRs linked to antipsychotics without systematic use of laxatives, these effects warrant greater awareness among prescribers because some might be preventable. For example, the systematic use of laxatives could be considered when initiating antipsychotic therapy—even in younger patients. Constipation is a well-known, prevalent reaction to antipsychotics in both older adults and younger adults [41, 42].
In comparison, laxatives are routinely prescribed for the prevention of opioid-induced constipation [43]. Notably, the systematic use of laxatives in patients treated with antipsychotics or other drugs with anticholinergic properties does not feature in the current guidelines. However, one study has shown that the routine prescription of laxatives to inpatients receiving antipsychotics was associated with a lower frequency of cases of severe constipation [44].
The present study focused initially on anticholinergic ADRs associated with a selection of drugs that have been assigned high anticholinergic scores on existing scales and was subsequently extended to include drugs from the same pharmacological class. Indeed, several anticholinergic burden scales have been developed to estimate the cumulative effect of medications with anticholinergic properties—particularly in older or frail populations. These scales have been designed to identify patients at high risk of adverse outcomes, including cognitive impairment, falls, delirium, and other anticholinergic-related ADRs.
Each scale assigns a score to an individual drug, based on the known anticholinergic potency. The ACB scale is among the tools most widely applied in clinical research [5], and has been used to assess the potential cognitive effects of polypharmacy involving anticholinergic agents [45]. However, these scales differ in their drug coverage, scoring methods, and clinical validation, which can limit their consistency and sensitivity.
Indeed, two or three of the top five drugs most frequently implicated in the anticholinergic ADRs in the present study (i.e., sertraline, citalopram, bromazepam) were not listed in the commonly used ACB scale. This finding (i) highlights an important limitation of existing scoring systems, which may not fully capture the range of medications involved in real-world anticholinergic-related events, and (ii) underscores the need for ongoing refinement and expansion of these tools, to ensure that emerging evidence is adequately reflected in clinical risk assessments. Adapting existing anticholinergic burden scales (such as the ACB) to a given national context would require a dedicated evidence-based procedure.
This procedure should involve mapping the scale’s items to the drugs authorised and used in the country of interest, reviewing scores using national pharmacovigilance and clinical data, and applying structured expert consensus methods (e.g., Delphi panels) to adjudicate where there is uncertainty. The symptoms belonging to the anticholinergic syndrome should also be clearly defined. Lastly, to ensure the modified scale’s clinical and regulatory relevance, validated against country-specific, real-world data on outcomes would be required.
When prescribing a new medication or re-evaluating an existing treatment, careful attention should be paid to the pharmacological class in general and those with anticholinergic properties in particular. Within a pharmacological class, drugs often share mechanisms of action, receptor targets, and structural features and are likely to have similar anticholinergic potential. Moreover, the patient’s personal medical history should be reviewed, to minimise the risk of AEs.
Treatment with a combination of drugs with anticholinergic properties might increase the risk of anticholinergic ADRs through synergistic or cumulative pharmacodynamic effects. In high-risk patients, treatments lacking anticholinergic properties should (if available) be preferred. This approach helps to optimise safety while maintaining treatment efficacy.
The present study had several strengths. First, we analysed a comprehensive, nationwide database that included all ADRs (nearly 10,000) reported in France during a defined period. Second, and in contrast to studies that focused on a single type of effect or drug class, our analysis provided a broad, comprehensive overview of all anticholinergic ADRs.
Last, by directly comparing anticholinergic risk scales with real-world safety data, our approach offered a more relevant, practical assessment of anticholinergic risks.
However, our study had several limitations. First, underreporting might have affected the completeness and representativeness of the pharmacovigilance data—particularly for well-known ADRs. Indeed, one study showed that only 6% of serious ADRs are spontaneously reported by healthcare professionals [46].
Underreporting impedes the accurate assessment of the true incidences and risk profiles of these ADRs. Furthermore, the data collected over our long study period might have been influenced by changes over time in pharmacovigilance practices and case assessment procedures, which might have contributed to heterogeneity in reporting. Second, we were unable to calculate precise incidence rates for individual drugs, as the analysis relied solely on reported cases and lacked comprehensive data on drug prescription rates in France.
Therefore, the frequencies of ADRs in the database may not fully reflect the real-world rates of occurrence and should be interpreted with caution. Finally, it should be noted that we only analysed suspected drugs that were part of the predefined list of anticholinergic agents, along with molecules belonging to the same pharmacological classes; other suspected drugs reported in the ICSRs were not included in our analysis.
Conclusion
Overall, currently available anticholinergic scoring tools have various limitations and fail to capture the full range of medications involved in anticholinergic events reported through pharmacovigilance channels. Our present findings highlighted the significant burden of anticholinergic ADRs in younger patients, not only older adults or poly-medicated patients. Furthermore, we identified certain anticholinergic ADRs (including constipation and intestinal obstruction) associated with antipsychotic use; these are important targets for prevention and management.
Our findings underscore the need for greater awareness among clinicians and the development of tailored strategies for reducing the occurrence of these potentially serious ADRs. Finally, although current anticholinergic burden scales are valuable guides to the risk of anticholinergic ADRs, it is crucial to consider a drug’s complete pharmacological profile regardless of its assigned score, particularly when an anticholinergic ADR arises.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (PDF 1877 KB)