Dement Geriatr Cogn Disord 1999;10:342344

Acetylcholine and Delirium

Larry E. Tune Susan Egeli
Division of Geriatric Psychiatry, Wesley Woods Center at Emory University, Atlanta, Ga., USA

Key Words Acetylcholine W Delirium W Polypharmacy

Abstract The neurotransmitter acetylcholine has been implicated in animal and human studies of delirium. This chapter will briefly review the clinical studies focussing on measurement of serum levels of anticholinergic activity in delirious states. Three approaches have been taken. First, to identify medications currently prescribed that have subtle anticholinergic effects. The current list includes 48 commonly prescribed medications. Second, to associate serum anticholinergic activity with delirium in various clinical states including postcardiotomy delirium, postelectroconvulsive delirium, delirious elderly medical inpatients, and nursing home patients. Third, to intervene in patients with elevated anticholinergic activity by reducing known anticholinergics and correlating this reduction with clinical measures of cognition and delirium. Our most recent data investigates the impact of anticholinergics on demented patients. Rates of delirium were significantly higher in patients receiving larger numbers of anticholinergics.

Data from a variety of animal and clinical studies support the hypothesis that acetylcholine is one of the cricital neurotransmitters in the pathogenesis of delirium. Support for this comes from a wide variety of sources. From animal studies, cholinergic neurotransmission is impaired in the following models of encephalopathy/delirium: hypoxia, nitrite poisoning, thiamine deficiency, he-

patic failure, carbon monoxide poisoning, and hypoglycemia. Some of these have been briefly reviewed by other participants in this symposium and will not be reviewed here [14]. It is interesting that encephalopathies with diverse etiologies all have disturbances of cholinergic neurotransmission as one central abiding feature. Acetylcholine may serve as final common neurotransmitter pathway. Clinically, there are good reasons to focus on the importance of acetylcholine, especially in the aged, and particularly in dementia. First, both animal and human studies show that acetylcholine neurotransmission decreases with age. It is likely that the age-related loss of cholinergic reserve is a major reason that age is a consistent risk factor for delirium. Second, for dementia, there is the dual effect of age-associated decline in acetylcholine as well as the focal loss of acetylcholine associated with dementia. This is best shown in Alzheimers disease where cholinergic cell bodies in the nucleus basalis of Meynert die as a result of the disease process. Similar lesions have been found for other dementias. Clinical studies manipulating acetylcholine support the role of anticholinergics in contributing to cognitive impairments. Challenge studies with antagonists in normal elderly adults have consistently produced deficits in information processing leading to a reduced ability to focus, or sustain attention as well as compromised short-term memory [reviewed in 5]. In cognitively impaired patients, challenge studies with cholinergic antagonists consistently exacerbate preexisting cognitive impairments. Cholinomimetic agents (physostigmine, tacrine, donepezil) enhance memory function in both dementia and delirium. Our laboratory has focused on serum anticholinergic levels as a simple, reliable predictor of cognitive toxicity

ABC
Fax + 41 61 306 12 34 E-Mail karger@karger.ch www.karger.com

1999 S. Karger AG, Basel 14208008/99/01050342$17.50/0 Accessible online at: http://BioMedNet.com/karger

Larry E. Tune, MD Division of Geriatric Psychiatry Wesley Woods Center at Emory University 1841 Clifton Road NE, Atlanta, GA 30329 (USA) Tel. +1 404 728 4960, Fax +1 404 728 4963

Table 1. Anticholinergic drug levels in 25 medications ranked by the frequency of their prescription for elderly patients

Medicationa

Anticholinergic drug level ng/ml of atropine equivalents 0.22 0.25 0.08 0.25 0.00 0.00 0.00 0.11 0.44 0.00 0.00 0.12 0.55 0.00 0.22 0.15 0.00 0.11 0.86 0.00 0.02 0.00 0.00 0.00 0.22

Furosemide Digoxin Dyazide (hydrochlorothiazide and triamterene) Lanoxinb Hydrochlorothiazide Propranolol Salicylic acid Dipyridamole Theophylline anhydrous Nitroglycerin Insulin Warfarin Prednisolone Methyldopa Nifedipine Isosorbide dinitrate Ibuprofen Codeine Cimetidine Diltiazem hydrochloride Captopril Atenolol Metoprolol Timolol Ranitidine
a

Drug concentration: 10 8 M. b A digoxin compound.

from anticholinergic compounds. This has been tested in a variety of clinical settings. In 1979, we developed a simple radioreceptor assay technique, in which serum from patients was added to a tissue homogenate containing the muscarinic antagonist, 3H-quinuclidinyl benzilate. In this assay, serum samples with varying amounts of anticholinergic medications then competed with 3H-quinuclidinyl benzilate for muscarinic receptors. Samples were then compared to known amounts of atropine, and expressed as atropine equivalents. The advantage of this approach is its simplicity and its ability to assess the cumulative effects of multiple medications as well as pharmacologically active metabolites. Some of these investigations antedated the development and use of more reliable and valid assessment instruments (e.g. Confusion Assessment Method, DRS) and so correlations were, for the most part, made among clinical diagnosis (based on DSM-III, DMSIII-R criteria), clinical change in brief clinical assessment of cognition, and serum anticholinergic level. With these tools, serum anticholinergic levels have been significantly

associated with the presence and severity of delirium in a number of settings: postcardiotomy delirium, ICU delirium, postelectroconvulsive delirium, and the nursing home [for full review, see 4]. Since serum anticholinergic levels presumably reflect the net effect of medications (both parent compounds and active metabolites), our group and others have pursued three experimental approaches. These approaches were (1) to determine the anticholinergic properties of all medications involved in these investigations of delirious patients, (2) to assess in blinded, naturalistic studies preand postdelirium measures of anticholinergic levels in at risk populations, and (3) intervention studies to attempt to lower anticholinergic medications in patients with cognitive impairment. The hypothesis for this last strategy is that lowering the cumulative anticholinergic effect will have a positive effect on measures of cognition. Table 1 is a previously published list of medications with modest anticholinergic effects. Its importance lies in the fact that the list represents the 25 most commonly prescribed medications in the elderly in 1992. Many of these medications cannot be discontinued simply because of a modest anticholinergic effect. The list does underscore the point that there are over 600 drugs with anticholinergic effects [1600; 5]. A review of any geriatric clinic will support the view that anticholinergics are disproportionately administered to the elderly. Tables 2 and 3 provide an expanded list of parent compounds with modest anticholinergic effect. These medications were identified in two studies (unpubl. results) detailed below. Both studies addressed the issue of cumulative anticholinergic toxicity in patients with dementia. In these studies, as patients were identified, medications were recorded and subsequently assayed using the radioreceptor assay described above. This list is by no means comprehensive. It does reflect medications taken by elderly patients in studies of acute and subacute delirium. Two studies conducted by unrelated research groups support the generalizability of our principal finding that the serum anticholinergic level is associated with the presence and severity of delirium. Mach et al. [4] found that delirious, medically ill patients had significantly elevated serum anticholinergic levels compared to nondelirious, medically ill patients. Resolution of delirium in these patients was associated with lower serum anticholinergic levels. One central question is whether or not anticholinergic toxicity is independently associated with delirium or merely indicates that the patients are sicker and require more medications. Flacker and Lipsitz [6] addressed this in a cross-sectional study of 67 acutely ill elderly inpatients.

Acetylcholine and Delirium

Dement Geriatr Cogn Disord 1999;10:342344

343

Table 2. Additional medications with anticholinergic effects as

Table 3. Medications for delirium versus dementia with behavioral

determined by radioreceptor assay Atropine Tobramycin Clindamycin Corticosterone Ampicillin Hydralazine Diazepam Captopril Dexamethasone Desipramine Gentamycin Piperacillin Imipramine Cefmandole Imuran Cycloserine Pancuronium Prednisolone Amitriptyline Phenelzine Oxycodone Chlordiazepoxide Diphenhydramine Cefoxitin Furosemide Keflin Clindamycin Chlorazepate Oxazepam Vancomycin Hydroxyzine Cyclosporin Flurazepam Hydrocortisone Tobramycin Phenobarbital Chlorthalidonc Alprazolam Diltiazem Valproic acid Amantidine Ticrocillin Clindamycin Thioridazine Flunitrazepam Desipramine Diltiazem Dipyrimadole

symptom groups Delirium Dementia with behavioral symptoms 3.7B0.39 1.61B0.17 1.70B0.19

Admission medications Anticholinergic medications Psychotropic medications

4.68B0.33 2.34B0.195* 2.1B0.15

Multivariate logistic analysis of variables associated with delirium included significant impairment in ADL scores, admission diagnosis of infection, white blood cell count elevation, and serum anticholinergic activity. The serum anticholinergic level was independently associated with delirium, and a greater number of symptoms of delirium was associated with higher serum anticholinergic activity. The next strategy is an intervention strategy. Tolefson et al. [5] identified 34 nursing home residents, all of whom had received more than one anticholinergic medication for more than 2 weeks. All patients were then randomly assigned to an intervention or nonintervention group. The goal of the intervention group was a 25% reduction in an anticholinergic index; no reduction was attempted for the nonintervention control group. Psychometric tests done before and after the intervention included the Buschke selective reminding test, geriatric depression scale, Wechsler Memory scale-digits foreward (WMSF), the Saskatoon Delirium Checklist (SDC), and the Psychogeriatric Dependency Rating Scale (PGDRS), and the
References
1 Gibson G, Blass JP, Huang H-M, et al: The cellular basis of delirium and its relevance to age related disorders including Alzheimers disease. Int Psychogeriatr 1991;3:373395. 2 Gjedde A, Lockwood AH, Duffy TE, et al: Cerebral blood flow and metabolism in chronically hyperammonemic rats: Effect of an acute ammonia challenge. Ann Neurol 1978;3:325 330. 3 Posner JB, Plum F: The toxic effects of carbon dioxide and acetazolamide in hepatic encephalopathy. J Clin Invest 1960;39:12461258.

Mini-Mental State Exam (MMSE). Prior to the intervention, significant correlations were found between the serum anticholinergic level and impairments identified by the SDC, PGDRS, and WMSF. Following the intervention (n = 15) both the SDC and WMSF improved significantly. The MMSE improved in the anticipated direction, though the level of significance was p = 0.07. To test the hypothesis that cumulative anticholinergic toxicity from commonly prescribed medications is an important, independent etiology for delirium in elderly demented patients, we conducted the following retrospective chart review investigation. Inpatient records from 91 patients who satisfied diagnostic criteria for delirium (n = 47) or dementia behavioral disturbance (severe agitation; n = 44) were investigated. Average age was 79.3 years, and 60% of patients were female. Clinical assessments included the Confusion Assessment Method [7, 8], and MMSE [9]. Presence or absence of anticholinergic medications was based on a comparison of the patients medication list with medications listed in tables 1 and 2. Table 3 summarizes the significant differences (medications only) between the two groups. As a group, delirious patients took more anticholinergic medications (p = 0.007). Significantly more delirious patients took 5 or more prescription medications compared to the nondelirious demented patients (p = 0.002). Not surprisingly, the delirious patients had significantly more medical comorbidity. * p = 0.007.

4 Tune L, Ross CA: Delirium; Coffey CE, Cummings JL (eds): Textbook of Geriatric Neuropsychiatry, Washington, APA Press, 1994, pp 35365. 5 Tollefson GD, Montague-Clouse J, Lancaster SP: The relationship of serum anticholinergic activity to mental status performance in an elderly nursing home population. J Neuropsychiatry Clin Neurosci 1991;3:314319. 6 Flacker J, Lipsitz LA: Serum anticholinergic activity changes with acute illness in elderly medical patients. Am J Geriatr Psychiatry 1998;6:4754.

7 Inouye SK, van Dyck C, Alessi CH, et al: Clarifying confusion: The confusion assessment method. Ann Intern Med 1990;113:941 948. 8 Inouye SK, Charpentier PA: Precipitating factors for delirium in hospitalized elderly persons: Predictive model and interrelationship with baseline vulnerability. JAMA 1996;275: 852857. 9 Folstein MF, Folstein SE, McHugh PR: MiniMental State: A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189198.

344

Dement Geriatr Cogn Disord 1999;10:342344

Tune/Egeli