The association between anticholinergic load and cognitive performance, elevated blood pressure and the presence of peripheral anticholinergic side effects

Background: To quantify the association between anticholinergic drug use (burden or load) and cognitive function (MMSE), blood pressure and the presence of anticholinergic side effects.


Introduction
Older people have an increased likelihood of illness and as a result receive a disproportionate amount of over-the-counter and prescribed medications. 1 In the USA people over 65 used 30% of prescription and 40% of over-the-counter medicines, despite making up only 13% of population. 1 In the UK, the old aged comprised only 18% of population but consumed 45% of all prescription drugs.Some of these medications are prescribed inappropriately and without attention to side effects.Elderly people living in aged care facilities are even more likely to receive medications and to be exposed to side effects. 1 Blazer et al reported in one year, 97% of elderly patients in aged care facilities obtained a prescribed drug, compared with 71% of patients living in the community. 2Longitudinal data from Australia (1981 through 2002) has shown the rate of adverse drug events to be on the rise. 3The largest increase manifested in those persons aged over 80 years; a tenfold increase in men and a sevenfold increase in women. 3reover, older people frequently have low health expectations and are less likely to recognize medication adverse effects.In addition, patients with cognitive decline have difficulties in communication of their discomfort.Consequently, those living in aged care homes may rely on carers to prompt the physician to realize possible adverse drug events.This is particularly the case for anticholinergic side effects, which are among the most common drug-related problems experienced by elderly people living in nursing homes. 4,5 ccording to the Beer et al criteria 6 aimed at identifying drugs deemed inappropriate for use in the elderly (aged ≥65 years), drugs with anticholinergic properties are frequently regarded as examples of inappropriate medications to prescribe because they constitute an increased risk for many adverse drug effects, such as dry mouth, constipation, urinary retention, visual impairment, falls, confusion, drowsiness, delirium and cognitive decline. 7There are many common anticholinergic side effects which may contribute to significant clinical problems.For instance, dry mouth can lead to speech difficulties, dental decay or trouble with dentures. 2,8 hewing and swallowing may become painful and can result in refusing of solid food.Pupillary dilation and the inability to accommodate will affect near vision, thereby increasing the risk of accidents, including falls and may worsen narrow angle glaucoma in predisposed patients.In the case of constipation, anticholinergic effect may lead to stomatitis, paralytic ileus or faecal impaction. 2,8 rug-induced increases in heart rate may precipitate angina symptoms and inhibition of sweating may cause life-threatening hyperthermia.Urinary retention can cause urinary tract infection, a common cause of delirium.Men with benign prostatic hypertrophy are at high increased risk of acute urinary retention from an anticholinergic side effect. 2,8 icholinergic medications are the most common cause of drug-induced delirium.In patients with dementia, anticholinergic drugs can inhibit cognitive performance and compromise the beneficial effect of cholinergic enhancers used to treat cognitive impairment. 9,10 dditionally, polypharmacy and the concurrent chronic use of multiple drugs are a reality for older patients. 11Thus, anticholinergic adverse effects may result from the cumulative anticholinergic burden of multiple drugs.Consequently, the combination of drugs can contribute clinically important adverse drug reactions, even if the individual drugs do not cause obvious side effects. 12number of studies have reported on the common use of anticholinergic medications in elderly people.
Blazer et al 2 reported that during one year, nearly 60% of aged care facility residents had received drugs with anticholinergic activity, compared with 23% of elderly people living in the community.The Established Population of Epidemiologic Studies of the elderly study, a US prospective cohort study, showed that chlorpheniramine and diphenhydramine were among the 10 most commonly identified inappropriate drugs used in elderly patients. 13A prospective cohort study in 532 patients with intact cognitive function conducted in the primary care clinics at the Veterans Affairs Medical Center (VAMC), Iowa City, demonstrated that 27.1% of patients used at least one medication with anticholinergic action. 14 a longitudinal study in 9294 community-based elderly patients in France 15 , 6.4% of patients used drugs with anticholinergic properties.In a retrospective cohort study involving 765,423 elderly patients covered by a pharmaceutical benefit manager and who filed ≥ 1 prescription drug claim in 1999 in US 16 , amitriptyline and doxepin were reported to account for almost one quarter of all inappropriate drugs prescribed.
In addition, there are several studies that demonstrated the association between the use of anticholinergic drugs (anticholinergic load) and anticholinergic side effects.A prospective cohort study in 364 elderly patients in Italy 17 found that anticholinergic drug use was associated with impaired physical performance and muscle strength, irrespective of potential confounders and that this association correlated with the increasing number of anticholinergic drugs used.Additionally, a number of other studies have reported the association of anticholinergic drug use with cognitive impairment. 12, 18-23A retrospective, medical record review at two urban tertiary care teaching hospitals, found that the most common side-effects from anticholinergic medications were decreased secretions, tachycardia, confusion and drowsiness. 24 our knowledge, few studies have investigated the association between anticholinergic load and overall anticholinergic side-effects.Therefore, identifying the association of anticholinergic load and anticholinergic drug-related problems could contribute to the better understanding, assessment, prevention and management of anticholinergic side effects in elderly people.This study was therefore undertaken to quantify the association between the anticholinergic load and medication-related problems amongst residents of aged care homes.

SETTING
A retrospective cohort study was conducted at three aged care homes in the Perth metropolitan area, Western Australia.It was undertaken as a quality use of medicines (QUM) activity as part of the residents' annual medication management reviews (MMR), which are undertaken by an accredited pharmacist and funded by the Australian Government.

PARTICIPANTS
Participants in the study were aged care home residents aged over 65 years old.The participants were recruited through the medication management review program conducted in their aged care homes as part of standard care.All residents in aged care homes in Australia are entitled to an annual medication management review.As part of the medication management review process, the review pharmacists are required to undertake a range of quality use of medicines (QUM) activities and this study was undertaken as one of these activities.Aged care home residents at the time of their admission are required to provide written consent for medication management reviews to be undertaken.Therefore, individual patient consent for the study was not required.This study was approved by the Curtin University of Technology Human Research Ethics Committee and Ethics Committees of the aged care homes.
The anticholinergic load was derived from the total number of anticholinergic medications taken by the patient.The prevalence of cognitive impairment, hypertension and anticholinergic adverse events was evaluated based on the patient's anticholinergic load taking into consideration any co-morbidities.

STATISTICAL ANALYSIS
Descriptive statistics were complied in the sample according to age, gender, co-morbidities, the use and the number of medications with anticholinergic properties (anticholinergic load), anticholinergic side effects, and the Mini Mental State Examination scores.The association between anticholinergic load and cognitive impairment (the Mini Mental State Examination scores) was assessed using stepwise multiple regression analysis adjusted for all related co-morbidities.To assess the extent to which the presence of peripheral anticholinergic side effects was related to the anticholinergic load, a multiple logistic regression model was created to predict such a relationship taking into account the anticholinergic load and all identified co-morbidities related to each anticholinergic side effect.A p value ≤ 0.05 was considered statistically significant.All analyses were conducted using SPSS program version 15.0.

Results
A total of 182 patients' medical records were reviewed for the study.Of these, seven patients were excluded as they were found to be less than 65 years old.Therefore, 175 patients' data were included in the data analysis.For the cognitive function, data on the Mini Mental State Examination scores was obtained from 161 patients; in 12 of the remaining 14 patients Mini Mental State Examination was not able to be performed due to severe cognitive impairment and, whilst the other two patients declined to be tested.These 14 patients consisted of three males and 11 females who had a mean (SD) anticholinergic load 3.2 (1.8).
The distribution of age, sex, marital status, the use and number of anticholinergic medications, the Mini Mental State Examination scores, anticholinergic side effects, and co-morbidities are summarized in Table 1.The mean age was 87.0 (8.0) years and 71.4% of the subjects were women.Of the 175 patients, 89.6% were receiving at least one anticholinergic medication; with the mean number of medications being of 2.9 (1.9).The mean Mini Mental State Examination score of the 161 evaluable patients was 20.2 (7.0).The three most common potential anticholinergic side effects identified were hypertension (66.3%), constipation (63.4%) and blurred vision (40.0%).The most frequently prescribed medications with anticholinergic activity were temazepam (35.4%), frusemide (31.4%) and loperamide (14.3%).The most common co-morbidities were hypertension (66.3%), dementia (64.6%) and psychological/behavioral disorders (58.3%).
The association between the anticholinergic load (the number of anticholinergic medication used) and cognitive impairment was assessed using the stepwise multiple logistic regression (Table 2) after adjusting for age and the following co-morbidities: urinary tract infection, hypothyroidism, renal failure, thiamine deficiency, vitamin B12 deficiency, diabetes mellitus, dementia, psychotic disorders and Korsakoff's syndrome.Among all co-morbidities, only dementia (p=0.000) and psychotic disorders (p=0.021) were included in the regression model as they had a significant negative impact on the Mini Mental State Examination scores.There was an inverse correlation between anticholinergic load and Mini Mental State Examination scores but this was not statistically significant (beta = -0.114,p=0.142) after adjusting for these co-morbidities.Table 3 presents the factors significantly associated with each of the targeted anticholinergic side effects.
The effect of age and the various co-morbidities (diabetes mellitus, chronic renal failure, ischaemic heart disease, hypercholesterolemia and hypothyroidism) and anticholinergic load on hypertension was assessed using the multiple logistic regression.From this analysis, hypercholesterolemia (p=0.005) and the anticholinergic load (p=0.000) were the only variables that had a significant impact on increasing the likelihood of patients experiencing hypertension.Each increase in the number of anticholinergic medications prescribed was associated with a 1.53-fold increase in the likelihood of hypertension (Adjusted Odds Ratio [AOR] = 1.53, 95% CI: 1.24-1.88,p= 0.000).
For dry mouth, the related factors included in the analysis were diabetes mellitus, anaemia, dementia, stroke, psychological disorders, postmenopausal status and age.After adjusting for all these factors, only the anticholinergic load demonstrated a significant impact on the likelihood on dry mouth.The increase in the number of anticholinergic medications accounted for an increased risk in the presence of dry mouth.
Similar to dry mouth, after adjusting other contributing factors (postmenopausal status, diabetes mellitus, stroke, asthma, glaucoma and age), only the anticholinergic load had a positive association with dry eyes.
For each increase in the number of anticholinergic medications received there was an associated 1.30-fold increase in the likelihood of dry eyes (AOR = 1.30, 95%CI: 1.08-1.56,p= 0.007).
A multiple logistic regression analysis failed to show the anticholinergic load to be a predictive factor for blurred vision (p=0.084).After adjusting all related co-morbidities (macular degeneration, diabetes mellitus, cataract, stroke and glaucoma) and age, the number of anticholinergic medications did not have a significant impact on blurred vision, whereas, the presence of glaucoma (p= 0.000) and macular degeneration (p=0.005)increased the likelihood of blurred vision significantly.The anticholinergic load did not have an impact on likelihood of patients suffering from glaucoma after adjusting for age, hypertension, diabetes mellitus and the use of corticosteroids (p = 0.702).
In terms of constipation, after adjustment for other possible causes of constipation including hypothyroidism, diabetes mellitus, psychological disorders, stroke and age, the only significant predictors of constipation were the anticholinergic load (p=0.031) and psychological disorders (p=0.025).For each increase in the number of anticholinergic medications received there was an associated 1.24-fold increase in the likelihood of constipation (AOR 1.2495%CI: 1.02-1.50,p= 0.031).
Lastly, urinary hesitancy was affected by the presence of benign prostatic hyperplasia (p=0.020)prostate cancer (p=0.045)anticholinergic load (p=0.050) after adjustment of all possible risk factors (benign prostatic hyperplasia, prostate cancer, diabetes mellitus, chronic renal failure, urinary incontinence, recurrent urinary tract infection and age).For each increase in the number of anticholinergic medications received there was an associated 1.21-fold increase in the likelihood of urinary hesitancy (AOR = 1.21.95% CI: 1.00-1.45,p= 0.050).

Discussion
The results of this study demonstrated that the anticholinergic load measured by the number of anticholinergic drugs received was significantly associated with the presence of all peripheral anticholinergic side effects with the exception of blurred vision, suggesting that these side effects are clinically significant.This association was not attributable to co-morbidities or age.In the case of blurred vision, its occurrence was correlated with glaucoma and macular degeneration, but not anticholinergic load.Given that anticholinergic agents should be used with caution in patients with glaucoma a subanalysis was undertaken to determine if there was a relationship between anticholinergic load and glaucoma, and hence anticholinergic load may be secondary contributor to blurred vision.This was not found to be the case, which validated the initial finding that anticholinergic load was not associated with blurred vision.
Whilst anticholinergic load was found to be inversely correlated with decline in the Mini Mental State Examination scores, the association was not statistically significant (p=0.142).37][38] One study showed that even a low anticholinergic serum level can cause mild but measurable cognitive impairment in the presurgical elderly 39 and memory deficit in geriatric depressed subjects.Mulsant et al 20 measured serum anticholinergic activity in a sample of 201 elderly subjects living in the community and examined the relationship between serum anticholinergic activity and cognitive performance.They found that subjects with serum anticholinergic activity at or above the sample's 90 th percentile point were significantly more likely than subjects with undetectable serum anticholinergic activity to show a low Mini Mental State Examination score.However, all the studies mentioned above used serum anticholinergic activity as the anticholinergic load.Serum anticholinergic activity, despite a number of limitations, is still considered the current gold standard in quantifying the anticholinergic burden. 40ome patients and found a strong association in institutionalized elderly between the use of specific anticholinergic medications and constipation as reflected in the increased use of laxatives.The result of dry mouth and the anticholinergic load was congruent with Thomson et al's 47 study showing the higher prevalence of xerostomia in patients taking anticholinergic medications.
In a recent study by Rudolph et al 48 , the Anticholinergic Risk Scale, a ranked categorical list of commonly prescribed medications with anticholinergic potential was developed.The study then evaluated Anticholinergic Risk Scale scores as the risk factor of anticholinergic adverse effects in two cohorts of patients (a geriatric evaluation and management (GEM) cohort and in primary care cohort).Higher Anticholinergic Risk Scale scores were linked to an increased risk for central anticholinergic adverse effects in GEM group and peripheral anticholinergic side effects in both groups.
4][15][16] This finding supports the growing concern about the prescribing of drugs with anticholinergic activity deemed as an inappropriate in this age group according to the Beer et al criteria. 6Concerning was the fact that the majority of medications with anticholinergic activity being consumed by the patients in this study were available only on prescription.Therefore, it was assumed a physician prescribed the medications thereby suggesting the need of education program for prescribers aimed at producing a substantial reduction in the use of these medications. 49The high frequency of benzodiazepine use, in this case, temazepam, can increase the risk of delirium and falls. 36Therefore, healthcare providers should offer other options to such patients (i.e.nonphamacological) rather than prescribing the medication as the first choice. 50Kay et al 12 suggested that prescribers should review patient's medical history, total anticholinergic burden and level of cognitive impairment prior to prescribing anticholinergic medications.
providing medication management review services to a number of residential aged care facilities.Ms Jiraporn Puangsombat (B Pharm) is a pharmacist completing her Master of Clinical Pharmacy Degree through Curtin University of Technology.Mr Malcolm Roberts (MRPharmS, Grad Dip Pharm, Grad Dip Bus & Admin, MHA) is an Adjunct Associate Professor with the School of Pharmacy at Curtin University, he is also an accredited pharmacist providing medication management review services to numerous residential aged care facilities.

Table 1 .
Baseline demographic and clinical characteristic of study participants (N = 175)

Table 2 .
Stepwise multiple logistic regression analysis of variables associated with cognitive impairment (MMSE scores)

Table 3 .
Multiple logistic regression analysis of variables associated with anticholinergic side effects