| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
1 Department of Internal Medicine and Clinical Gerontology, Centre Hospitalier Universitaire Purpan-Casselardit, Toulouse, France.
2 Department of Epidemiology and Public Health, University of Toulouse, Toulouse, France.
3 INSERM U558, Toulouse, France.
Address correspondence to S. Gillette-Guyonnet, PhD, Service de Médecine Interne et de Gérontologie Clinique, Pavillon J. P. Junod, Centre Hospitalier Universitaire La Grave-Casselardit, 170 avenue de Casselardit, TSA40031, 31059 Toulouse cedex 9, France. E-mail: gillette.s{at}chu-toulouse.fr/sophie.gillette{at}free.fr
|
A |
|---|
|
|
|---|
Methods. A prospective observational study was performed in which a population of 455 Alzheimer's disease patients were recruited and followed up for at least 1 year between 1994 and 2002. Patients were reevaluated at 6 monthly intervals using standardized neurocognitive and geriatric evaluations in addition to complete clinical examination, standard paraclinical investigations, and recording of treatment received.
Results. The risk of rapid cognitive deterioration was significantly decreased in patients taking CEIs for at least 1 year compared to untreated patients (odds ratio [OR] = 0.56, 95% confidence interval [CI], 0.340.93; p =.025). The potential benefit of CEI use was also found on institutionalization (OR = 0.2, 95% CI, 0.080.48; p <.001) and weight loss (OR = 0.56, 95% CI, 0.320.97; p =.039) after 1 year of follow-up.
Conclusion. The special interest of this study is that all patients were recruited and followed in the same center with the same management care plan and the same medical team. This follow-up offers us a unique opportunity to compare the 1-year evolution of the disease in clinical practice before and after the marketing of CEIs and allows us to demonstrate a clinically significant improvement in patient outcome over time.
The purpose of the present study is to investigate the effect of CEI use on the AD patient's outcome. We based our study on a population enrolled between 1994 and 2002 (before and after the advent of CEIs) in our Alzheimer Clinical Research Centre, and followed up using a similar standardized management care plan, to test the hypothesis that CEI use may positively influence the natural history of the disease. The two primary end points for evaluating outcome were rapid deterioration of cognitive function and institutionalization. Secondary outcome measure was weight loss, which is one of the main complications of the disease (17). These milestones were chosen because of their clinical, social, and economic impact.
| METHODS |
|---|
|
|
|---|
From this cohort, we excluded 130 patients who had a discontinued follow-up (deaths n = 25, loss to follow-up n = 7, wishes of the family or the patient n = 36, moved to another area n = 21, missed appointments n = 35, incomplete evaluation n = 6). Analyses were conducted among 455 patients (mean age 75.4 ± 6.7 years) who achieved a complete evaluation during the 1 year of follow-up after their enrollment. Forty-three percent (n = 192) of these patients took CEIs at each evaluation during their follow-up. The percentage of participants who discontinued follow-up did not differ significantly according to the year of inclusion. At inclusion, these participants had poorer cognitive function and were more dependent than those who completed follow-up (mean MMSE score = 17.2 ± 6.1 vs 18.9 ± 5.4, p =.007; mean Instrumental Activities of Daily Living (IADL) score = 3.40 ± 2.25 vs 4.56 ± 2.15, p <.0001). However, the number of patients who discontinued follow-up was not higher among participants not treated with CEIs than among treated participants (17.5% vs 13%, p =.71).
Data Collection
Each participant in this observational study received repeated evaluations at 6 monthly intervals which were carried out by multidimensional and pluridisciplinary methods. Investigations included recording of sociodemographic data (age, sex, educational level, and living arrangements) and duration of the problem and time since diagnosis of dementia as declared by the caregiver (initial evaluation only), and cognitive evaluation by a neuropsychologist using the MMSE (20). The patient's physical disability was quantified using the Activities of Daily Living (ADL) scale (21) and the IADL scale (22). Quantitative assessment of nutritional status was carried out with the Mini Nutritional Assessment (MNA) (23). Treatments (in particular, the use of CEIs [donepezil, rivastigmine, or galantamine for patients included since 1997]), were carefully recorded at each evaluation. Lastly, the caregiver's subjective burden was evaluated using Zarit's Burden Inventory (ZBI) (24), which was completed by the caregiver at the time of the patient's visit. For each of the 22 items, the caregiver indicated with what frequency he or she experienced a given emotion during the care relationship. The possible frequency ranged from 0 (never) to 4 (nearly always). The global score, obtained by adding the scores for each item, has a theoretical range of 088. The higher the score, the greater the subjective burden. In Zarit's proposed classification, burden is considered "absent to mild" if the score is less than 21, "mild to moderate" between 21 and 40, "moderate to severe" between 41 and 60, and "severe" if the score is greater than 60. During follow-up, the following events were recorded: hospital admissions, admissions to institutions, use of new support services, and changes occurring among the patient's relatives.
Outcome Measures
Rapid cognitive deterioration.--
According to O'Hara and colleagues (25), we considered that a decrease of 3 or more MMSE points in 1 year corresponded to clinically significant worsening of the disease.
Institutionalization.-- We considered as "institutionalized" those patients who left their homes during follow-up and permanently entered retirement home, nursing home, a long-stay hospital unit, or the home of a host family.
Weight loss.-- We considered as clinically significant any weight loss of 4% or more in 1 year as considered by Wallace and colleagues (26), who have shown that it was a risk factor of increased mortality in healthy elderly persons.
Statistical Analysis
To study the potential effect of CEIs, we took into consideration treatment at 6 months and defined the treatment variable as follows: "0 exposure = not treated," "1 exposure = treated at one time point during the whole year of follow-up," "2 exposures = treated at 2 time points," "3 exposures = treated at 3 time points." Bivariate analysis was based on classic tests for means comparison in the case of quantitative variables (using analysis of variance) and for comparison of distribution frequency in the case of qualitative variables (chi-square test). Multivariate logistic regression analysis was used to study the specific effect of factors associated with rapid cognitive deterioration, institutionalization, and weight loss. Treatment and all variables related to the dependent variable with a p value less than or equal to 0.25 after bivariate analysis and considered pertinent according to the literature were included in the model (see Table 1). According to the model, the dependent variable was a loss of 3 or more MMSE points (coded yes or no), weight loss of at least 4% (coded yes or no), or institutionalization (coded yes or no). In any case, analysis was adjusted on the initial MMSE score. The year of inclusion was not used in the initial model because it was collinear with the treatment. Analysis was carried out using stepwise descending regression. Goodness-of-fit was tested using the test of Hosmer and Lemeshow. First-order interactions were tested on the final models, and no significant interaction was found. All statistical analysis was done using SAS software (version 8.02; SAS Institute, Cary, NC).
|
| RESULTS |
|---|
|
|
|---|
.0001)]. The mean rate of cognitive decline was 2.48 ± 3.78 points over the 1-year period. The rate of change in the MMSE score was not significantly different according to the initial MMSE score: It was 2.93 ± 3.32 for patients with an initial MMSE score <15, 2.70 ± 4.21 for patients with an initial MMSE score between 15 and 19, 2.18 ± 3.87 for patients with an initial MMSE score between 20 and 24, and 2.16 ± 3.21 for patients with an initial MMSE score
25 (p KruskalWallis =.2028). Forty seven and one half percent of participants showed rapid MMSE loss (
3 points in 1 year). We observed a significant decrease in the risk of cognitive deterioration by time of inclusion notably marked after the advent of CEIs in 1997 (p trend =.002). The number of participants with a rapid cognitive deterioration was significantly lower among patients who had 3 consecutive exposures to CEIs compared with those who had never been exposed (39.3% vs 58.3%; OR = 0.46, 95% CI, 0.290.73; p =.001) (Table 1). The relationship between medication use and other factors which alter the cognitive deterioration in bivariate analysis (body mass index [BMI] at entry, educational level, initial MMSE and IADL score) was next examined using a multiple regression logistic analysis. The principle finding of this analysis was that CEI use was a predictor of cognitive deterioration; the risk of rapid cognitive deterioration was decreased by 44% in patients treated with CEIs for 1 year compared to untreated patients (Table 1). A low educational level, BMI
25 kg/m2, and the number of disabilities on the IADL scale (
2) were also identified as independent protective factors against rapid cognitive deterioration.
Factors Associated With Institutionalization
Eleven percent (n = 50) of participants entered institutional care during the first year of follow-up. The simple bivariate association between CEI use and institutionalization was significant. Nearly 23% of untreated participants had entered an institution at 1 year, compared with 9% of participants with 2 consecutive exposures to CEIs (p =.011) and 3.6% of those with 3 consecutive exposures (p <.001; Table 1). Using bivariate analysis, we also found a significant decreased risk of institutionalization according to time of inclusion, initial cognitive status (MMSE score), living arrangements, use of medical support services, and caregiver burden reported at baseline. The association between CEI use and institutionalization was found by using multivariate analysis, which revealed a significant decrease in the risk of institutionalization in patients who had been treated for at least 1 year compared with untreated patients (OR = 0.20, 95% CI, 0.080.48; p
.001) (Table 1). Multivariate analysis also helped us to identify two independent risk factors for institutionalization: living arrangements (at home without spouse) and personal history of depression.
Factors Associated With Weight Loss
Mean overall weight of patients at inclusion was 62.0 ± 12.4 kg. Twenty-seven percent of participants (n = 123) had clinically significant weight loss during the follow-up period. Weight loss during follow-up was not associated with more rapid deterioration of cognitive function (
MMSE: 2.7 ± 4.0 vs 2.4 ± 3.7, p =.7495) or of independence in IADLs (
IADL: 1.3 ± 1.8 vs 1.2 ± 1.5, p =.9430). As for the two previous criteria studied, time of inclusion significantly influenced the risk of weight loss (p =.001). Simple bivariate analysis associations were found between clinically significant weight loss and living arrangements, use of medical support services, caregiver burden, disabilities (IADL score), and cognitive impairment (MMSE score) collected at baseline.
There was no significant difference in BMI between treated and nontreated groups at baseline (p =.15). Weight loss was also significantly less frequent in participants who had been exposed to CEIs throughout follow-up compared with participants who had never been treated (Table 1). The positive association with treatment was also found in multivariate analysis, which revealed a decrease of 45% of the risk of weight loss in participants treated for 1 year compared with nontreated participants (OR = 0.56, 95% CI, 0.320.97; p =.039) (Table 1).
| DISCUSSION |
|---|
|
|
|---|
Our results are consistent with those of Lopez and colleagues (13,27), who showed that there was a cognitive benefit from the use of CEIs in AD after 1 year of follow-up. The recent study of the AD2000 Collaborative Group did not find a measurable reduction of institutionalization or progress of disability in AD patients treated for 3 years (15). In this study, patients were followed up in many centers with different care plan management. In the same way, data from the REAL.FR study (15) were observed from 16 different Alzheimer's centers. Nonpharmacological management remains difficult to quantify and to assess in isolation because of the multiplicity of interventions and of the health care personnel involved.
The lower incidence of clinically significant weight loss among the treated participants is also an argument in favor of a beneficial role of CEIs. An increased incidence of weight loss is reported in patients being treated with higher doses of CEIs compared to placebo patients in several short-term clinical trials, but the proportion of patients losing weight is extremely variable from one study to another (28). Weight loss may occur only during the initiation of CEIs. Data on CEIs and weight loss are derived from short-term randomized trials. Impact of nutritional status over a longer period is unknown. Our data showed that patients being treated with CEIs for at least 1 year lost less weight. In a previous study (29), we also found that the risk of weight loss decreased with increasing length of treatment with CEIs. These data must be interpreted cautiously because it is possible that patients with a dietary problem needed to stop treatment due to the anticholinergic side effects and, then, were considered as controls.
The longitudinal follow-up which we coordinate gives us the opportunity to observe in clinical practice the evolution of patients before and after specific treatment for the disease. This study demonstrated a significantly better clinical outcome in our patients, related to these medications. The treatment of the disease remains complex, and pharmacological methods must be associated with management of the complications.
|
A |
|---|
|
|
|---|
|
F |
|---|
|
|
|---|
Received April 6, 2005
Accepted October 17, 2005
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Gillette-Guyonnet, S. Andrieu, F. Nourhashemi, E. Reynish, and B. Vellas RESPONSE TO LETTER FROM GARCIA AND COLLEAGUES ON CHOLINESTERASE INHIBITORS AND ALZHEIMER'S DISEASE OUTCOMES J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2007; 62(6): 679 - 680. [Full Text] [PDF] |
||||
![]() |
H. Y. Cheng COULD THE POTENTIAL BENEFITS OF CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE BE REDUCED AFTER ADJUSTING FOR COEXISTING DISEASES AND MULTIPLE MEDICATIONS? J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2007; 62(6): 680 - 680. [Full Text] [PDF] |
||||
![]() |
S. Gillette-Guyonnet, S. Andrieu, F. Nourhashemi, E. Reynish, and B. Vellas RESPONSE TO LETTER FROM CHENG ON CHOLINESTERASE INHIBITORS AND ALZHEIMER'S DISEASE OUTCOMES J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2007; 62(6): 680 - 680. [Full Text] [PDF] |
||||
![]() |
L. Ferrucci A Year of Excellence in Geriatric Research and Some New Dishes on the Menu J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2007; 62(5): 516 - 518. [Full Text] [PDF] |
||||
![]() |
A. Garcia, K. Thompson, K. Zanibbi, S. Geick, and R. Adams CHOLINESTERASE INHIBITORS AND ALZHEIMER'S DISEASE OUTCOMES J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2007; 62(5): 570 - 570. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|