Psychosomatics 43:31-35, February 2002
© 2002 The Academy of Psychosomatic Medicine
Tricyclic Antidepressants and Cognitive Decline
Laura Jean Podewils, and
Constantine G. Lyketsos, M.D., M.H.S.
Received June 15, 2001; revised October 10, 2001; accepted October 18, 2001. From the Departments of Epidemiology and Mental Hygiene, Johns Hopkins Bloomberg School of Public Health, and the Neuropsychiatry Service, Department of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine, Baltimore, MD. Address correspondence and reprint requests to Dr. Lyketsos, Johns Hopkins Medical Institutions, 600 N. Wolfe St, Osler 320, Baltimore, MD 21287. E-mail: kostas{at}jhmi.edu
|
ABSTRACT
|
In this study, we sought to determine the cross-sectional and longitudinal relations of tricyclic antidepressant (TCA) use to cognitive function and cognitive change in a population-based sample of adults (n=1,488). Sociodemographic information, TCA use, and baseline scores on the Mini-Mental State Exam (MMSE) were determined in the initial two waves of the study. At wave 3, participants repeated the MMSE; the prospective relation was assessed for change between waves 2 and 3 (median 11.5 years). These findings failed to support the concept that TCA use is related to concurrent measurable cognitive deficits, and TCA use does not appear to significantly compromise memory over a substantial time span.
Key Words: Depression • Cognition
|
INTRODUCTION
|
In recent years, there has been a shift from the use of tricyclic antidepressants (TCAs) to new antidepressant classes, such as selective serotonin reuptake inhibitors (SSRIs), as first-line treatment for depression. Hesitancy with TCA administration is mainly due to their side-effect profile,1 because their efficacy remains at least comparable to that of newer therapies. TCAs, given their anticholinergic properties, may affect cognitive function in the short term. The possibility also exists that use of these agents may cause declines in cognition and memory in the long term through anticholinergic or other mechanisms. TCA use has been a particular concern among the elderly, who are at an increased risk for cognitive effects of the TCAs.
The evidence indicating that TCAs pose long-term risks to cognition is not conclusive. The preference for other agents is mainly attributable to the short-term anticholinergic effects of TCAs.1 In several literature reviews,2,3 the authors have concluded that SSRIs and TCAs have comparable efficacy and side effects, and they cited the uncertainty of any cost benefit favoring the newer agents. Other studies have found few differences in cognition after treatment with TCAs or other agents.4–6 Among special populations, such as the severely depressed or elderly patients, a meta-analysis of direct comparisons between TCAs and SSRIs7 and a more recent clinical trial8 suggested that TCAs may actually be more effective than SSRIs, although this is controversial.9 In contrast, research has demonstrated alterations in psychomotor function, reaction time, and short-term memory attributable to recent TCA use.10 It remains unclear exactly how these effects might extend to daily functioning or long-term outcomes.9,11,12
To our knowledge, no studies to date have investigated the impact of TCA use on cognitive functioning after an extended follow-up. Previous work from our group has shown that, after 11.5 years of follow-up, a population sample in East Baltimore exhibited a measurable decline on the Mini-Mental State Exam (MMSE).13 The purpose of the current investigation was to determine whether use of TCAs, compared with non-use of TCAs, is associated with poorer concurrent performance on global cognitive functioning or with greater cognitive decline after 11.5 years.
|
METHODS
|
Study Population
The Baltimore Epidemiologic Catchment Area (ECA) study is a prospective study designed to achieve a better understanding of the prevalence, patterns, and impact of psychiatric disorders in a population-based community sample of adults aged 18–80 years. Sampling methods employed for recruitment into the ECA Baltimore follow-up study have been detailed elsewhere.13 Briefly, the 3,481 individuals who participated in the initial wave in 1981 were traced for targeting in the wave-3 follow-up in 1993–1996; 1,920 of the participants completed a household survey at this time. As in recent studies from our group, only individuals who were present for all three study waves, namely 1981, 1982, and 1993–1996, were included in the present analysis. This yielded a sample of 1,488 persons.
Classification of TCA Use (Exposure)
At ECA waves 1 and 2, participants were asked if they had ever used antidepressant medications. Participants were given an exhaustive list including individual names of medications available at that time, and they responded whether they had ever used each particular medication. If participants reported current use or history of use at either wave 1 or wave 2, they were classified into the exposed group. If they did not report any use, they were considered unexposed.
Measurement of Global Cognitive Function
Folstein et al.14 developed the MMSE as a tool for measurement of global cognitive function. The MMSE has wide application as a standard in clinical practice, and age-adjusted norms have been well established.15,16 The MMSE was administered to the entire cohort in all three study waves by a trained interviewer, and the summary score on the MMSE provided a measurable assessment of current memory performance.
Measurement of Cognitive Decline
Our measure of cognitive decline was an individual MMSE change score14 for each study participant, calculated by subtracting the wave-2 score from the wave-3 score. Although a single test score on the MMSE is not particularly useful in distinguishing pathologic cognitive dysfunction,17,18 performance on the MMSE is generally stable, and changes in its value may have practical implications. The MMSE has been used previously by our team in the ECA to evaluate general risk factors for long-term changes in cognition.13
Other Covariates
Age at wave 3, highest level of education achieved (by category), race, and gender were considered as potential confounding variables in the regression models on the basis of our previous work.13
Additionally, smoking status and alcohol consumption were evaluated because they may modify the relation between TCA use and cognition. Both smoking and alcohol use were classified by reported use at the initial two waves. Preliminary exploratory analysis failed to distinguish between never and former smokers, so individuals who had quit smoking by 1982 were considered non-smokers for the follow-up time. Those who reported smoking were categorized as smokers. Alcohol status was categorized as teetotalers (never drinkers), former drinkers, and drinkers. The latter category was divided into light (1 drink/day), moderate (2–4/day), and heavy ( 5/day) drinking.
Statistical Analysis
Means and standard deviations for continuous variables, as well as frequencies and percentages for each category, were tabulated for the sociodemographic descriptors. We used t tests for continuous variables and  2 tests for categorical variables to determine significant differences for age, gender, race, education, smoking status, alcohol intake, and baseline MMSE scores.
Univariate and multiple linear regression models were used to investigate the relation between TCA use and MMSE scores at wave 1 and wave 2 cross-sectionally, and as a predictor for changes in MMSE scores between waves 2 and 3. All analyses were performed using Stata 6.0 (Stata, Galveston, TX).
|
RESULTS
|
Ninety-seven percent of the original cohort participants who reported use of antidepressants at the initial two waves (1981–1982) indicated that the agents were of the TCA type. Table 1 provides an overview of the analytic sample, comparing the TCA users (exposed; n=65) and non-users (unexposed; n=1,423) on a series of sociodemographic variables. Figure 1 compares the two groups graphically on mean MMSE scores at each wave. Mean scores and standard deviations for cognitive decline (with higher positive numbers reflecting greater decrements) in TCA users and non-users were 1.6±1.9 and 1.2±2.9, respectively. Results of univariate linear regression analyses did not find any significant associations between current or past TCA use and either wave-1 (baseline) MMSE scores (ß=-0.042, SE=0.244; P=0.864) or wave-2 MMSE scores (ß=-0.251, SE=0.238; P=0.292). Table 2 provides the results of our final multiple regression model for cognitive change ( MMSE), including age, education, race, and smoking status as significant predictors and explaining approximately 10% (R2=0.102) of the variance in cognitive change over time. TCA use was not associated with greater cognitive decline after 11.5 years (ß=-0.197, SE=0.288; P=0.484). Older age, African American race, and smoking were all associated with greater decrements in cognitive function during follow-up. Additionally, education beyond eighth grade was associated with a slower rate of decline.
|
DISCUSSION
|
Our study provides no evidence to support either short-term or long-term effects of TCAs on cognitive functioning as measured by the MMSE. This work suggests that some of the concerns regarding the long-term cognitive effects of TCAs may be unfounded. The associations between cognitive decline and both advancing age and lower education are generally accepted, but racial differences and a relation with smoking have been less consistent. On average, persons of African American race have fewer years of formal education19; the reasoning behind an independent negative association with cognitive function is less clear. Factors influencing early development that are differential by income and race, including nutritional intake, lifestyle factors, or environmental exposures, may be partially accountable. Although nicotine use may tend to increase attention and stimulus discrimination20,21 and potentially enhance cognitive performance, the deleterious effect of smoking on longitudinal changes in cognitive functioning found in our study is in line with findings of others.22–24
Our study is not without limitations, and the results should be interpreted with caution and in the context of other evidence with regard to anticholinergic agents. It is possible that selection bias may have affected our sample because attrition rates are generally higher in persons of older age and poorer health. There was substantial loss to follow-up among the TCA users. Only 24.3% of the original cohort using TCAs in one or both of the former waves were available at the wave-3 follow-up, compared with 44.2% of the persons who did not report any previous TCA use. Unfortunately, those individuals who were lost to follow-up were unable to contribute a final score on the MMSE at wave 3 and were excluded from the analysis. Such differential losses to follow-up may have inflated the type II error rate and compromised the ability to detect a relation between TCA use and cognitive decline. To address this concern directly, we compared baseline MMSE scores between those who were available at wave 3 and contributed a wave-3 MMSE score and those who were not, according to category of TCA use. Among TCA never users, those who contributed wave-3 MMSE scores performed significantly better on the MMSE at baseline than those who were lost to follow-up (26.7±5.1 vs. 24.4±9.0; t=-8.64, P<0.0001). The TCA users in the present analysis (n=65), however, were representative of the original group of TCA users (n=268), demonstrating similar mean MMSE scores at baseline (26.8±5.9 vs. 27.3±5.5; t=0.631, P=0.528). If TCA use did compromise long-term cognitive functioning, we would have anticipated that the differences between these TCA users and the TCA non-users available at the follow-up would have been even more marked (e.g., biased away from null) because the non-users at follow-up were more cognitively elite than the original group of non-users. Therefore, we do not believe that any excessive threats are posed to the internal consistency of our findings, and the substantial sample size still allows ample power upon which to draw conclusions.
Left untreated, the complex pathology of depression may interfere with cognitive functioning.9 Depression has also been reported as a key predictor of mortality.25,26 It is possible that our control group was contaminated by individuals with undiagnosed clinical depression or dysthymia, thus affecting cognition. On the other hand, some of those undergoing treatment with TCAs could be therapy resistant; either could explain the similarity of performance between the groups. No information was elicited during interviews on the indication for or previous duration of TCA use, or on whether exposure status changed during the follow-up period. If TCA treatment was relatively new therapy for an individual's depression at the time of assessment, the concurrent measurement of cognition could have been suppressed by depressive symptoms. Effective treatment for depression has the potential to restore or improve cognitive function, and therefore differences between a baseline MMSE score suppressed by depression and a decline in MMSE after 11.5 years could be essentially undetectable.
Although antidepressant medication may be a surrogate for the presence of depression, depression often goes undiagnosed and untreated.27,28 Therefore, we would expect negligible discrepancies between the proportion of undiagnosed depressive patients and the proportion of treated depressive patients unavailable for the 11.5-year follow-up.
More subtle effects on cognitive function may have occurred that were not detectable by the MMSE. However, the reliability of the MMSE and the extensive follow-up time were adequate to detect any clinically relevant differences between the groups if any did exist.
Further, skepticism may surround self-reports of medication use. However, previous work by Foy et al.29 showed high concordance rates between physician records and patient reports of use. We believe that self-report would introduce only minimal error, if any, and is a minor concern.
Despite these limitations, this paper is the first, to our knowledge, to investigate both concurrent and longitudinal cognitive effects of TCA use in a population-based sample after a substantial follow-up period. In the current epidemiologic investigation of 1,488 adults, we found no association between TCA use and current or future objective cognitive performance. We emphasize the critical consideration of each patient's best interest for both immediate and long-term outcomes when selecting the proper pharmacologic intervention.
|
ACKNOWLEDGMENTS
|
This study was supported by NIMH (MH-47447) for the Baltimore ECA follow-up.
|
REFERENCES
|
-
Hirschfield RM: Efficacy of SSRIs and newer antidepressants in severe depression: comparison with TCAs. J Clin Psychiatry 1999; 60:326-335[Medline]
-
Settle EC: Antidepressant drugs: disturbing and potentially dangerous adverse effects. J Clin Psychiatry 1998; 59(suppl 16):25-30
-
Garattini S, Barbui C, Saraceno B: Antidepressant agents: from tricyclics to serotonin uptake inhibitors. Psychol Med 1998; 28:1169-1178[CrossRef][Medline]
-
Kuny S, Stassen HH: Cognitive performance in patients recovering from depression. Psychopathology 1995; 28:190-207[Medline]
-
Allain H, Lieury A, Brunet-Bourgin F, et al: Antidepressants and cognition: comparative effects of moclobemide, viloxazine and maprotiline. Psychopharmacology (Berl) 1992; 106:S56-S61
-
Cole JO, Branconnier R, Solomon M, et al: Tricyclic use in the cognitively impaired elderly. J Clin Psychiatry 1983; 44:14-19
-
Anderson IM, Tomenson BM: The efficacy of selective serotonin re-uptake inhibitors in depression: a meta-analysis of studies against tricyclic antidepressants. J Psychopharmacol 1994; 8:238-249
-
Perry PJ: Pharmacotherapy for major depression with melancholic features: relative efficacy of tricyclic versus selective serotonin reuptake inhibitor antidepressants. J Affect Disord 1996; 39:1-6[CrossRef][Medline]
-
Oxman TE: Antidepressants and cognitive impairment in the elderly. J Clin Psychiatry 1996; 57(suppl 5):38-44
-
van Laar MW, van Willigenburg APP, Volkerts ER: Acute and subchronic effects of nefazodone and imipramine on highway driving, cognitive functions, and daytime sleepiness in healthy adult and elderly subjects. J Clin Psychopharmacol 1995; 15:30-40[CrossRef][Medline]
-
Gray SL, Lai KV, Larson EB: Drug-induced cognition disorders in the elderly. Drug Saf 1999; 21:101-122[CrossRef][Medline]
-
Moore AR, O'Keeffe ST: Drug-induced cognitive impairment in the elderly. Drugs Aging 1999; 15:15-28[CrossRef][Medline]
-
Lyketsos CG, Chen L, Anthony JC: Cognitive decline in adulthood: an 11.5 year follow-up of the Baltimore Epidemiologic Catchment Area Study. Am J Psychiatry 1999; 156:58-65[Abstract/Free Full Text]
-
Folstein M, Anthony JC, Parhad I, et al: The meaning of cognitive impairment in the elderly. J Am Geriatr Soc 1985; 33:228-235[Medline]
-
Bleecker ML, Bolla-Wilson K, Kawas C, et al: Age-specific norms for the Mini-Mental State Exam. Neurology 1988; 38:1565-1568[Abstract/Free Full Text]
-
Crum RM, Anthony JC, Bassett SS, et al: Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA 1993; 269:2386-2391[Abstract/Free Full Text]
-
Tangalos EG, Smith GE, Ivnik RJ, et al: The Mini-Mental State Examination in general medical practice: clinical utility and acceptance. Mayo Clin Proc 1996; 71:829-837[Abstract]
-
Tombaugh TN, McIntryre NJ: The Mini-Mental State Examination: a comprehensive review. J Am Geriatr Soc 1992; 40:922-935[Medline]
-
U.S. Department of Commerce, Bureau of the Census: Current Population Survey (CPS). Washington, DC, US Government Printing Office, 1997
-
Le Houezec J, Benowitz NL: Basic and clinical psychopharmacology of nicotine. Clin Chest Med 1991; 12:681-699[Medline]
-
Rezvani AH, Levin ED: Cognitive effects of nicotine. Biol Psychiatry 2001; 49:258-267[CrossRef][Medline]
-
Launer LJ, Feskens EJ, Kalmijn S, et al: Smoking, drinking and thinking. The Zutphen Elderly Study. Am J Epidemiol 1996; 143:219-227[Abstract/Free Full Text]
-
Leibovici D, Ritchie K, Ledesert B, et al: The effects of wine and tobacco consumption on cognitive performance in the elderly: a longitudinal study of relative risk. Int J Epidemiol 1999; 28:77-81[Abstract/Free Full Text]
-
Spilich CJ, June L, Renner J: Cigarette smoking and cognitive performance. Br J Addict 1992; 87:1313-1326[CrossRef][Medline]
-
Barefoot JC, Schroll M: Symptoms of depression, acute myocardial infarction, and total mortality in a community sample. Circulation 1996; 93:1976-1980[Abstract/Free Full Text]
-
Penninx BW, Geerlings SW, Deeg DJ, et al: Minor and major depression and the risk of death in older persons. Arch Gen Psychiatry 1999; 56:889-895[Abstract/Free Full Text]
-
Davidson JR, Meltzer-Brody SE: The underrecognition and undertreatment of depression: what is the breadth and depth of the problem? J Clin Psychiatry 1999; 60(suppl 7):4-9
-
Hirschfeld RM, Keller MB, Panico S, et al: The National Depressive and Manic-Depressive Association consensus statement on the undertreatment of depression. JAMA 1997; 277:333-340[Abstract/Free Full Text]
-
Foy A, O'Connell D, Henry D, et al: Benzodiazepine use as a cause of cognitive impairment in elderly hospital inpatients. J Gerontol A Biol Sci Med Sci 1995; 50:M99-M106
This article has been cited by other articles:
|
|
|
|
 
J. O. Brooks and J. C. Hoblyn
Neurocognitive Costs and Benefits of Psychotropic Medications in Older Adults
J Geriatr Psychiatry Neurol,
December 1, 2007;
20(4):
199 - 214.
[Abstract]
[PDF]
|
|
|
|
|
|
|
P. M. Doraiswamy, K. R. R. Krishnan, T. Oxman, L. R. Jenkyn, D. J. Coffey, T. Burt, and C. M. Clary
Does Antidepressant Therapy Improve Cognition in Elderly Depressed Patients?
J. Gerontol. A Biol. Sci. Med. Sci.,
December 1, 2003;
58(12):
M1137 - 1144.
[Abstract]
[Full Text]
[PDF]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 2002
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
