HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 60:375-379 (2005)
© 2005 The Gerontological Society of America

Apolipoprotein {varepsilon}4 Allele and Problems With Orientation Are Associated With a Persistent Decline in Cognition in Community-Dwelling Elderly Persons

Sharon J. Wayne1,, Bruno J. Vellas2, Steven G. Brodie3, Philip J. Garry3 and Richard N. Baumgartner1

1 Aging and Genetic Epidemiology Program, Department of Internal Medicine, University of New Mexico School of Medicine, Albuquerque.
2 Departments of Internal Medicine and Clinical Gerontology, Acute Unit for Alzheimer's Patients, Toulouse, France.
3 Department of Pathology, University of New Mexico School of Medicine, Albuquerque.

Address correspondence to Sharon J. Wayne, MPH, Aging and Genetic Epidemiology Program, Department of Internal Medicine, University of New Mexico School of Medicine, Albuquerque, NM 87131-5666. E-mail: swayne{at}salud.unm.edu


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. A decline in cognitive test scores in elderly persons can signal the beginning of a descent into dementia or may indicate only a short-term cognitive disturbance. It would be clinically useful to distinguish between the two outcomes and to identify characteristics of each.

Methods. Four hundred thirty-seven community-dwelling elderly persons were given the Mini-Mental State Examination (MMSE) annually for an average of 7 years. A low score between baseline and final MMSE was identified. A low score 3 or more points lower than baseline score indicated cognitive decline. This decline was called persistent if the final MMSE score was also at least 3 points lower than baseline MMSE score; otherwise, the decline was considered transient.

Results. Twenty participants (4.6%) experienced a persistent cognitive decline, 67 participants (15.3%) experienced a transient cognitive decline. Presence of the apolipoprotein {epsilon}4 allele was significantly associated with persistent cognitive decline (age-adjusted odd ratio [OR] = 11.46, p <.0001) but not with transient cognitive decline (age-adjusted OR = 1.53, p =.219). Incorrect answers on the orientation part of the MMSE at the time of cognitive decline was associated with persistent decline compared to transient decline (age-adjusted OR = 3.58, p =.058).

Conclusions. Persistent cognitive decline is an infrequent occurrence in community-dwelling elderly persons. Presence of the {epsilon}4 allele and errors made by the subject on questions of orientation may be useful in determining whether a cognitive decline is likely to be persistent.

IN clinical practice, physicians often see evidence of cognitive decline in their elderly patients. For some patients, this is the beginning of a process that leads to severe dementia or Alzheimer's disease; for others, it is a transient state, perhaps brought on by a passing illness or temporary bereavement, which does not indicate that the decline will be permanent. It would be valuable for physicians to be able to distinguish between persistent and transient cognitive decline; however, few studies have investigated this difference.

The strong and consistent association between the apolipoprotein {epsilon}4 allele and the development of Alzheimer's disease (1,2) leads us to suspect that the {epsilon}4 allele will be associated with persistent, but not transient, declines in cognition. Associations between the {epsilon}4 allele and measures of cognitive impairment other than Alzheimer's disease have been found (3–6), but these associations are not as consistent as those between the {epsilon}4 allele and Alzheimer's disease. We suspect that the true association between the {epsilon}4 allele and non-Alzheimer's cognitive decline can be weakened by the inclusion of declines that are transient.

Because it may not always be practical to test people for presence of the {epsilon}4 allele, other indicators of persistent cognitive decline would be useful. Of particular interest is whether there are specific questions on a cognitive examination that may indicate which participants will have sustained as opposed to transient declines.

The purpose of this analysis is to describe the incidence of persistent cognitive decline in a sample of community-dwelling elderly persons and to measure the association between this decline and the {epsilon}4 allele. We will also distinguish between persistent and transient cognitive decline and evaluate whether specific questions used to measure cognitive status can help predict which decline is likely to be persistent.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Participants were members of the New Mexico Aging Process Study (NMAPS), a prospective cohort study of the factors influencing successful aging that began in 1979. The study is described in detail elsewhere (7); a brief summary will be given here. To be admitted to the study, persons had to be at least 60 years of age, live independently in the community, and be free of major medical conditions that would inhibit their ability to participate in a long-term longitudinal study. Participants were seen annually at which time they gave a blood sample, underwent multiple body measurements, and completed various interviews and tests. They were not required to maintain good health to continue in the study. Participants who dropped out or died were replaced to keep the study size at approximately 400 persons.

Cognitive status was measured using the Mini-Mental State Examination (MMSE) (8), standard version (maximum score 30). This examination was given annually to all participants beginning in 1989. The present analysis includes MMSE data through 2001.

To define a decline in cognition, we identified the participants' lowest MMSE score between their first and last MMSEs. If this low score was 3 or more points lower than their initial score, the participant was said to have had a decline in cognition. The decline was defined as persistent if their last MMSE score was at least 3 points lower than their first MMSE score. If the last MMSE score was less than 3 points lower than the first, participants were defined as having a transient cognitive decline.

Apolipoprotein E genotypes were assessed by restriction fragment length polymorphism analysis of polymerase chain reaction products, according to the procedure of Hixson and Vernier (9). The assay was conducted in the NMAPS laboratory and was begun in 1996. For analysis purposes, participants were classified as to presence or absence of the {epsilon}4 allele; {epsilon}4 was considered present if participants were homozygous ({epsilon}4/{epsilon}4) or heterozygous ({epsilon}2/{epsilon}4 or {epsilon}3/{epsilon}4) for {epsilon}4.

Other variables of interest included sex, age, education, activity level, body mass index (BMI), and the occurrence of new medical conditions possibly related to a decline in cognition. Activity level was assessed using a modification of the instrument first described by Shapiro and colleagues (10). A higher score indicates greater activity. BMI was calculated in the standard way: weight in kilograms divided by height in meters, squared. Information on incident morbid conditions was obtained from self-report of the participants at their annual visit. The following conditions were selected because of their possible association with subsequent cognitive decline: stroke, diabetes, heart attack, congestive heart failure, and cancer. Any of the above conditions which were newly diagnosed after the participants' initial MMSE were counted as incident conditions.

Tests of statistical significance include chi-square tests to evaluate differences in proportions and analysis of variance or Wilcoxon signed-rank tests to make comparisons for continuous variables. Logistic regression analysis was used to assess the effect of multiple risk factors on the outcome of sustained cognitive decline.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
To be included in this analysis, participants were required to have at least three MMSE scores during the study period (1989–2001) so that a drop in score between first and last MMSE could be calculated; 525 participants met this criterion. Of these, 88 did not have apolipoprotein E genotyping completed, leaving a final sample size of 437. Table 1 shows the characteristics of these 437 participants at baseline. The sample was 65.5% female and well-educated, with 83.6% having at least some postsecondary schooling. The mean age was 72.5 years (SD = 5.8), the mean BMI was 25.3 k/m2 (SD = 3.8), and the mean activity score was 18.4 (SD = 6.2). The average baseline MMSE score for these participants was 29 (SD = 1.4). The percentage of participants with at least one {epsilon}4 allele was 23.3%.


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics of Sample at Baseline (N = 437).

 
Table 2 shows that 350 participants (80.1% of the sample) had no decline in cognition during the follow-up period; 67 participants (15.3% of the sample) had a decline in cognition which was transient; and persistent cognitive decline occurred in 20 participants (4.6% of the sample). Mean follow-up time differed significantly by group, with the "No Decline" group having the shortest follow-up (6.6 years) and the group with a transient decline having the longest follow-up (8.8 years). Participants with a persistent cognitive decline were followed for an average of 7.7 years. The average time to decline was the same for those whose decline was persistent (4.5 years) as it was for those whose decline was transient (4.4 years).


View this table:
[in this window]
[in a new window]
  		Table 2. Sample Characteristics by MMSE Change.

 
The distribution of the {epsilon}4 allele differs significantly by outcome category. Sixty-five percent of participants who had a persistent cognitive decline had at least one {epsilon}4 allele compared to 23.9% of those with a transient drop in cognition and 20.9% of those who did not experience a drop in MMSE score (p <.0001). Tests of significance also indicated differences between the three groups for age, BMI, and the development of congestive heart failure. Participants with sustained cognitive decline were older than the other two groups (78.1 vs 75.2 and 71.7 years), had lower BMI values (23.3 vs 25.3 and 25.4), and were more likely to develop congestive heart failure (10.0% vs 1.5% and 1.4%) during the follow-up period. There were no statistically significant differences between the groups for the other characteristics presented. In general, participants with a transient drop in cognition were more similar to those who did not have a drop in cognition than to those with a sustained cognitive decline.

Figure 1 shows the mean MMSE scores for each of the three groups for their initial MMSE, final MMSE, and the low score between these two. Participants with cognitive decline were required to have a drop of at least 3 points, and those with a persistent decline were required to have a final MMSE score that was also at least 3 points lower than their first MMSE score. The figure shows that participants with persistent decline continue to drop after their low score, whereas those whose drop is transient rebound to nearly their baseline score.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Mean Mini-Mental State Examination (MMSE) score at baseline, end of follow-up, and low score in between baseline and end of follow-up for participants with persistent cognitive decline, transient cognitive decline, and no cognitive decline

 
The results of logistic regression analysis are shown in Table 3. The top half of the table shows the odds ratios for persistent cognitive decline using those with no decline in cognition as the reference group. The unadjusted odds ratio describing the effect of the apolipoprotein {epsilon}4 allele on sustained cognitive decline is 7.05, a statistically significant association. The full logistic model contains the five variables that were significantly associated with sustained decline in cognition in Table 2. This full model shows that the presence of the apolipoprotein {epsilon}4 allele remains a strong and significant predictor of sustained cognitive decline even after controlling for age, follow-up time, BMI, and incident congestive heart failure. Increasing age is also significantly associated with this decline. There is considerable confounding by age of the association between the {epsilon}4 allele and sustained cognitive decline—those with sustained cognitive decline are older than those with no change in cognition (Table 2), and participants with an {epsilon}4 allele are younger, in this sample, than those without an {epsilon}4 allele (71.7 vs 72.8 years, p =.10). The result of this confounding is that when age is added to the logistic model, the odds ratio increases substantially, to 11.46. Follow-up time, BMI, and the development of congestive heart failure were not associated with persistent cognitive decline in a model that also included the {epsilon}4 allele and age.


View this table:
[in this window]
[in a new window]
  		Table 3. OR, 95% CIs, and p Values for Logistic Regression Models Evaluating the Effect of the APOE4 Allele on Persistent MMSE Decline and on Transient MMSE Decline.

 
The bottom half of Table 3 shows the odds ratios for the effect of the {epsilon}4 allele, follow-up time, and age on transient decline in cognition. The reference group is again those participants who had no drop in cognition. The {epsilon}4 allele is not significantly associated with the development of transient cognitive decline, but increasing age and follow-up time are significantly associated with this outcome.

Table 4 compares answers to specific questions on the MMSE of those with persistent and transient cognitive decline. The percentages indicate which questions were answered incorrectly at the time of the drop in MMSE score compared to the baseline MMSE score. Participants with persistent decline were significantly more likely to give incorrect answers to questions on the current date and identifying the clinic by name than were participants with transient decline. They were also significantly more likely to give an incorrect answer to any one of the orientation questions (61.1% vs 17.6%). Participants with transient decline were more likely to make errors on the Attention & Calculation section (e.g., spelling the word "WORLD" backwards) than were participants with persistent decline (60.8% vs 38.9%), although this difference is not statistically significant.


View this table:
[in this window]
[in a new window]
  		Table 4. Comparing Subjects With Persistent and Nonpersistent Cognitive Decline on Answers to Individual Questions on the Mini-Mental State Examination (MMSE).

 
Table 5 shows the results of logistic regression analysis used to distinguish between participants with persistent and those with transient cognitive decline. An incorrect answer to any question on the orientation section of the MMSE is associated with a sevenfold increased risk of persistent decline, a statistically significant association. The addition of age and the presence of the {epsilon}4 allele to the model shows {epsilon}4 to be significantly associated with risk of persistent decline and increasing age to be of borderline significance (p =.081). With age and {epsilon}4 in the model, an incorrect answer to an orientation question is no longer significantly associated with persistent decline in cognition (odds ratio = 3.58, p =.058), but, given the small number of participants who experience this outcome, the association between an incorrect answer to questions of orientation and persistent decline is certainly suggested by these data.


View this table:
[in this window]
[in a new window]
  		Table 5. ORs, 95% CIs, and p Values for Logistic Regression Models Evaluating Risk Factors for Persistent Decline Compared to Transient Decline.

 

   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
This study finds that the association between the apolipoprotein {epsilon}4 allele and cognitive decline is very strong when cognitive decline is defined as a persistent, as opposed to transient, drop in MMSE score. We also find compelling evidence that a persistent decline in cognition is more likely to be associated with orientation problems on the MMSE than is a transient decline. These findings, if supported by other studies, would be important for clinical practice, providing information that may help distinguish patients whose newly identified cognitive decline may or may not be transient.

We find that a persistent 3-point decline in MMSE score occurs infrequently in a healthy, community-dwelling elderly population, at a rate of less than 5% over a follow-up period averaging 7 years. This definition of cognitive decline, which is very specific, may explain why the association between the {epsilon}4 allele and cognitive decline found in this study is stronger than associations reported in other studies (3,6,11–13).

A transient cognitive decline was more frequent than a persistent one, occurring in 15% of participants. This is an interesting group, and one that should be studied in more detail. Some studies (14–17) have found that depression or an incident morbid condition such as a stroke can diminish cognition for a limited period of time. While comorbidity was infrequent in this sample, we do find that a somewhat higher percentage of participants with a transient decline had at least one of our preselected morbid conditions than did participants without decline or with persistent decline. This association was not statistically significant, but, given the lack of precision in our measurement of comorbidity, may warrant further investigation.

The association between subsets of questions from the MMSE and cognitive decline has been studied in Alzheimer's disease (18–22). The results are mixed, as might be expected because the methodologies and follow-up times differ greatly among studies. Some studies are cross-sectional, whereas others consider only baseline MMSE scores as opposed to change in scores over time. However, most studies report an association between memory and/or orientation problems and Alzheimer's disease. Our study does not find an association between memory problems (recall) and persistent cognitive decline, but does find a link between this decline and problems with orientation.

Several weaknesses of the study need to be considered. The study sample size is small, and the study outcome is relatively rare. As a result, our statistical power to detect weak associations is low. Nonetheless, the associations detected were strong. Another potential problem is that of drop-out bias. Participants tend to drop out of the NMAPS when they become demented; however, we are describing an early decline in MMSE, one that occurs prior to a diagnosis of dementia. Serious morbidity other than dementia has been the main cause for drop-out from the NMAPS (35%), followed by movement from the Albuquerque area (27%). Over the past 18 years, 23 participants developed Alzheimer's disease or other dementia, and follow-up was discontinued because it was no longer possible to obtain competent informed consent. There was no significant difference among groups for mean duration of follow-up. For these reasons we do not believe our results are seriously biased by differential drop-out. Lastly, our sample is a well-educated one, and therefore not representative of the general community-dwelling population. It has been suggested that the MMSE has an education bias (23), although a recent study (24) refutes this suggestion. In any case, we do not believe that our results are confounded by education because there was no significant difference between groups in percentage with post-high school education.

In conclusion, this study presents a clinically important and novel way of defining cognitive decline—one that incorporates a measure of whether the decline is likely to be persistent. The presence of the {epsilon}4 allele is strongly associated with a cognitive decline that is defined as persistent. Confusion by the subject regarding orientation to time or place may be a clinically useful indicator for this persistent cognitive decline. More studies are needed to better understand the distinction between persistent cognitive decline and declines that are likely to be transient.


   Acknowledgments
 
Supported by National Institutes of Health grants R01 AG02049 and AG10149.

The findings reported in this article were previously presented in poster form at the June 2003 meeting of the Society for Epidemiologic Research (SER) in Atlanta, Georgia, and as a presented paper at the July 2003 International Academy of Nutrition & Aging (IANA) 2nd Congress in Albuquerque, New Mexico. The findings have not been disclosed in any other manner.


   Footnotes
 
Decision Editor: John E. Morley, MB, BCh

Received September 15, 2003

Accepted September 16, 2003


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Corder EH, Saunders AM, Strittmatter WJ, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science. 1993;261:921-923.[Abstract/Free Full Text]
  2. Farrer LA, Cupples LA, Haines JL, et al. Effects of age, sex, and ethnicity on the association between apolioprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA. 1997;278:1349-1356.[Abstract/Free Full Text]
  3. Bretsky P, Guralnik JM, Launer L, Albert M, Seeman TE. The role of APOE-E4 in longitudinal cognitive decline: MacArthur studies of successful aging. Neurology. 2003;60:1077-1081.[Abstract/Free Full Text]
  4. Dik MG, Jonker C, Comijs HC, et al. Memory complaints and APOE-E4 accelerate cognitive decline in cognitively normal elderly. Neurology. 2001;57:2217-2222.[Abstract/Free Full Text]
  5. Howieson DB, Camicioli R, Quinn J, et al. Natural history of cognitive decline in the old old. Neurology. 2003;60:1489-1494.[Abstract/Free Full Text]
  6. Hyman BT, Gomez-Isla T, Briggs M, et al. Apolipoprotein E and cognitive change in an elderly population. Ann Neurol. 1996;40:55-66.[Medline]
  7. Garry PJ, Goodwin JS, Hunt WC, Hooper EM, Leonard AG. Nutritional status in a healthy elderly population: dietary and supplemental intakes. Am J Clin Nutr. 1982;36:319-331.[Abstract/Free Full Text]
  8. Folstein MF, Folstein SE, McHugh PR. Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.[Medline]
  9. Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with Hha1. J Lipid Res. 1990;31:545-548.[Abstract]
  10. Shapiro S, Weinblatt E, Frank CW, Sager RV. The HIP study of incidence and prognosis of coronary heart disease. J Chron Dis. 1965;18:527-558.[Medline]
  11. Kim KW, Youn JC, Jhoo JH, et al. Apolipoprotein E epsilon 4 allele is not associated with the cognitive impairment in community-dwelling normal elderly individuals. Int J Geriatr Psychiatry. 2002;17:635-640.[Medline]
  12. Staehelin HB, Perrig-Chiello P, Mitrache C, Miserez AR, Perrig WJ. Apolipoprotein E genotypes and cognitive functions in healthy elderly persons. Acta Neurol Scand. 1999;100:53-60.[Medline]
  13. Yaffe K, Cauley J, Sands L, Browner W. Apolipoprotein E phenotype and cognitive decline in a prospective study of elderly community women. Arch Neurol. 1997;54:1110-1114.[Abstract/Free Full Text]
  14. Bowler JV, Hadar U, Wade JP. Cognition in stroke. Acta Neurol Scand. 1994;90:424-429.[Medline]
  15. Narushima K, Chan KL, Kosier JT, Robinson RG. Does cognitive recovery after treatment of poststroke depression last? A 2-year follow-up of cognitive function associated with poststroke depression. Am J Psychiatry. 2003;160:1157-1162.[Abstract/Free Full Text]
  16. Rabins PV. The prevalence of reversible dementia in a psychiatric hospital. Hosp Community Psychiatry. 1981;32:490-492.[Abstract/Free Full Text]
  17. Rabins PV, Merchant A, Nestadt G. Criteria for diagnosing reversible dementia caused by depression: validation by 2-year follow-up. Br J Psychiatry. 1984;144:488-492.[Abstract/Free Full Text]
  18. Small BJ, Fratiglioni L, Viitanen M, Winblad B, Backman L. The course of cognitive impairment in preclinical Alzheimer disease. Arch Neurol. 2000;57:839-844.[Abstract/Free Full Text]
  19. Ercoli LM, Siddarth P, Dunkin JJ, Bramen J, Small GW. MMSE items predict cognitive decline in persons with genetic risk for Alzheimer's disease. J Geriatr Psychiatry Neurol. 2003;16:67-73.[Abstract]
  20. Small BJ, Viitanen M, Backman L. Mini-mental state examination item scores as predictors of Alzheimer's disease: incidence data from the Kungsholmen Project, Stockholm. J Gerontol. 1997;52A:M299-M304.
  21. Fillenbaum GG, Wilkinson WE, Welsh KA, Mohs RC. Discrimination between stages of Alzheimer's disease with subsets of mini-mental state examination items. Arch Neurol. 1994;51:916-921.[Abstract/Free Full Text]
  22. Galasko D, Klauber MR, Hofstetter R, Salmon DP, Lasker B, Thal LJ. The mini-mental state examination in the early diagnosis of Alzheimer's disease. Arch Neurol. 1990;47:49-52.[Abstract/Free Full Text]
  23. Fillenbaum GG, Hughes DC, Heyman A, George LK, Blazer DG. Relationship of health and demographic characteristics to Mini-Mental State examination score among community residents. Psychol Med. 1988;18:719-726.[Medline]
  24. Jones RN, Gallo JJ. Education bias in the mini-mental state examination. Int Psychogeriatr. 2001;13:299-310.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS