IPA Bulletin
Recent Advances - Volume 20, Number 3
By Professor John O'Brien, Dr. Robert Barber, and
Professor Robert Baldwin
This month we are delighted to announce
that Bob Baldwin from Manchester,
UK, has joined the editorial
team. He is, of course, well known to IPA
members, being a board member and having
contributed greatly to the literature (and
IPA meetings) in several areas over recent
years, most particularly in the field of late
life affective disorders. We are very pleased
that he has agreed to contribute to this column
on a regular basis
ALZHEIMER’S DISEASE AND DEMENTIA
Memantine for the treatment of
moderate to severe dementia
Following close on the heels of cholinesterase inhibitors, another class of compound has recently been licensed in
many countries for the treatment of Alzheimer’s disease and results of a pivotal
study has been published by Reisberg and
colleagues (New England Journal of Medicine
2003: 348; 1333-1341). They studied 250
patients with moderate to severe Alzheimer’s
disease (MMSE between 3 and 14) treated
for 28 weeks with 20 mg of memantine or
placebo. Primary outcome measures were
the clinicans interview based impression of
change with caregiver input (CIBIC+) and
an ADL measure. Significant improvements
on memantine seen on the CIBIC+ on observed
case analysis (p=0.06 on intention to
treat analysis) and on ADL were seen. There
was evidence of less cognitive deterioration
(mean 7 points) on the SIB (severe impairment
battery) but no effects on non-cognitive
symptoms. Memantine, a non-competitive
NMDA antagonist proposed to have
neuroprotective properties, produced a different
pattern of response from that previously
seen in trials of cholinesterase inhibitors
in that no improvement on the drug was
seen, rather less deterioration than placebo.
As such, measures of who “responds” to
treatment and who does not would be more
of a challenge than the already difficult situation
with cholinesterase inhibitors.
Does immunotherapy produce cognitive
benefits in Alzheimer’s disease?
The potential dangers of immunization
with amyloid peptide are well recognized
and have been reported in this column
previously. In a recent multi-centre study,
6% of cases developed an aseptic meningoencephalitis,
leading to cessation of the study.
However, was there any clinical evidence of
benefit? Hock et al (Neuron 2003, 38:547-
554) report cognitive results from their centre
which took part in the study. They report
24 patients who had received active immunization
compared to 6 receiving placebo in
a double-blind randomized study design. 19
of the immunized patients generated antibodies
against Abeta amyloid, these patients
performed markedly better on MMSE at one
year compared to control subjects. Those
with antibodies declined by a mean of 1.2
points compared to control group deterioration
by 6.3 points, a significant difference.
Significant differences in favour of immunization
also seen in Activities of Daily Living
(DAD score) and also on a Memory Test.
There appeared to be a correlation between
antibody response and clinical response. Although
highly preliminary, this data suggested
that immunotherapy, if it can be achieved
without serious risk to patients, might well
produce cognitive benefits. Results from all
subjects enrolled in the study would clearly
be of great interest.
Is obesity a risk factor for
Alzheimer’s disease?
Gustafson et al (Archives of Internal Medicine
2003, 163:1524-1528) investigate
this in a Swedish cohort of 392 non-demented
adults followed up from age 70 until
88. They found 93 participants developed
dementia, with women who developed dementia
having higher body mass index at age
70 and a greater degree of being overweight
than those who did not develop dementia.
For every increase in BMI of one, AD risk
increased by a striking 36%. No such risk
was seen in men. Given the frightening levels
of obese and overweight individuals in
Western society (currently estimated as over
50% of the adult population), these findings,
if confirmed, would have major public
health implications regarding future prevalence
of dementia. Mechanisms mediating
this apparently increased risk remain unclear,
though insulin resistance and diabetes
remain possibilities (see below).
Insulin and Alzheimer’s?
Watson et al (Neurology 2003, 60:
1899-1903) examined the effect of
insulin infusions of placebo (saline) or insulin
on CSF Abeta 42 levels in healthy older subjects.
Insulin infusion produced an increase
in CSF insulin and also increased levels of
CSF Abeta 42 levels. The authors hypothesize
that insulin may impede degradation of
Abeta 42 and it may be that insulin resistance
(either sub-clinical or associated with
diabetes) which has been described in adults
with AD might elevate Abeta levels and promote
Alzheimer pathology. While this raises
prospects for prevention, results also indicate
that insulin therapy itself may potentially
accelerate Alzheimer pathology.
Non-steroidal anti-inflammatories and
Alzheimer’s disease: The debate continues
There continues to be an interesting contrast
between studies suggesting that
prior use of non-steroidals may protect
against Alzheimer’s disease with efficacy
studies suggesting no benefit of prescription
in those with established disease. A metaanalysis
of observational studies is provided
by Etminan et al (British Medical Journal
2003, 327:128-131) who identify 9 studies
(6 cohorts and 3 case controls) of almost
15,000 participants which examined risk of
Alzheimer’s disease amongst users of nonsteroidal
inflammatory drugs. Pooled relative
risk was 0.72, a significant reduction, and
was most prominent in long term (mostly
greater than 24 months) users where relative
risk over AD was 0.27 (confidence intervals
0.13 to 0.58). Interestingly, pooled relative
risk in 8 studies of aspirin was 0.87 which
was non-significant. The authors concludethat non-steroidal anti-inflammatory drugs,
particularly when used for one or two years,
appear to offer some protection against the
development of Alzheimer’s disease which is
not seen with aspirin.
But…non-steroidals have no effect on
Alzheimer’s disease progression
In view of the above there is somewhat disappointing
news from a study by Aisen et
al (Journal of the American Medical Association
2003, 289:2819-2826) who report a large
multi-centre study of 351 subjects with mild
to moderate Alzheimer’s disease treated with
the non-steroidal anti-inflammatory naproxen,
the selective COX-2 inhibitor rofecoxib
or placebo for one year. There were no significant
differences between groups on either
the primary outcome measure (ADAS-COG)
or secondary measures which include Activities
of Daily Living measures and noncognitive
symptoms. As expected, both side
effects and serious adverse effects were more
commonly found in the active treatment
group than the placebo group.
Randomized controlled trial of estrogen
plus progestin raises concern about its
benefits in dementia
More disappointing news on the therapeutic front from results from the largest (n > 4000) multicentre placebo-controlled
trial of estrogen plus progestin (Shumaker
et al. and Rapp et al. Journal of the
American Medical Association 2003; 289
pages 2651-2662 and 2663-2672) which
found treatment increased the risk of dementia
in postmenopausal women aged 65 years
or older. Overall, twice as many women developed
dementia on treatment compared
with placebo (40 vs 21) over the four years
of the trial. The increased risk was seen in
Alzheimer’s disease and vascular dementia,
and differences appeared as early as 1 year into
the trial. In addition, there was no effect on
the incidence of mild cognitive impairment.
The study, which formed part of the Women’s
Health Initiative, was discontinued prematurely
in 2002 because of concern about increased
health risks associated with the combined therapy.
The authors conclude that estrogen plus
progestin should not be prescribed with the
expectation that it will enhance cognitive
performance in postmenopausal women,
and taking into consideration other findings
(such as increased coronary heart disease,
breast cancer, stroke and pulmonary embolism),
the risks of treatment outweighs the
potential benefit (such as reduced colorectal
cancer and hip fracture).
In a separate study, Thal and colleagues
found no association between estradiol and
estrone levels and cognitive performance in
patients with Alzheimer’s disease treated
with exogenous estrogen (Archives of Neurology
2003; 60:209-12).
Lithium protective against
Alzheimer’s disease?
The notion that lithium may be neuroprotective
in Alzheimer’s disease is not
entirely new but further evidence supporting
this is provided Phil et al (Nature 2003, 423:
435-439) who examined the effects of lithium,
an inhibitor of GSK3, on production of
Abeta peptide. Although it is known that
GSK3 phosphorylates tau protein (and so
lithium may prevent tau deposition), this
study showed that lithium also blocked accumulation
of Abeta peptides in the brains
of mice overproducing APP, the target for
this being GSK3 alpha. This adds to growing
evidence suggesting that lithium might
be useful in preventing neurodegenerative
pathology and supports calls for further
clinical trials.
How “silent” are silent brain infarcts?
Silent infarcts appear to increase with age
and are probably associated with the
same cardiovascular risk factors as stroke disease
(Vermeer et al. Stroke 2003; 34; 392-6).
But what is their clinical significance? This
question was addressed in a large (n > 1000)
prospective, population-based study of elderly
subjects–the Rotterdam Scan Study.
Subjects were enrolled in 1995-96 and completed
base line cognitive testing and MRI,
which was repeated in 1999 to 2000. The
researcher found base line silent infarcts
doubled the risk of dementia and were associated
with steeper cognitive decline (Vermeer
et al. New England Journal of Medicine
2003; 348:1215-22). They also were associated
with a more than 3-fold increase in the
risk of a subsequent stroke (Vermeer et al.
Stroke 2003; 34:1126-9).
DEPRESSION
Does long term treatment with sertraline
prevent recurrence of depressive illness
in the elderly?
Most clinicians might be somewhat surprised
to learn the answer is no, at
least as judged by a paper from Wilson and
colleagues (British Journal of Psychiatry 2003,
182:492-497). They undertook a large randomized
controlled trial initially involving
318 elderly subjects with depression of who
113 reached a maintenance phase to be randomized
to sertraline (50 to 150 mg) or
placebo and then followed for two years.
Recurrence rates over the two year period
were non-significantly reduced (by 8%) in
sertraline compared to placebo groups and
rates of recurrence were strikingly high
(around 50%) in both groups. The authors
conclude that negative results may be due to
lack of power in the study or the fact that
other risk factors or lack of compliance may
be responsible. The authors emphasize the
need for further studies of prophylaxis in
particular subgroups, such as those with old
age depression, and the study re-opens debate
on the degree of treatment response and
rates of relapse in older compared to
younger patients.
Changing prescription from Lithium
to Valproate: Shifting practice without
evidence?
This is somewhat the provocative title of
an interesting study by Schulman et al
(British Medical Journal 2003, 326:960-961)
who compare changing prescription habits,
specifically the use of lithium and valproate
in non-demented subjects, in a population
in Ontario over the age of 65. Over an 8
year period almost 4,000 patients started
lithium and over 5,000 started valproic acid.
More interesting was the change in pattern
of drug use over the years with new lithium
use around four times as common as new
valproic acid use in 1993 with this pattern
almost reversing by 2000. The authors challenged
this change as being the absence of
evidence based data, as there is no clear evidence
that valproic acid offers comparable or
superior efficacy to lithium in the over 65’s.
Their call, therefore, is for systematic evaluation
before whole scale change in practice.
Is depression more common after stroke
than myocardial infarction?
Aben and colleagues (Journal of Neurology
Neurosurgery and Psychiatry 2003;
74:581-5) examined whether stroke disease
is more closely linked to the onset of depression
than other acute, life-threatening and
disabling diseases that do not involve ischemic
brain damage. They followed-up 190 patients
with their first ever stroke and 200 with first
ever myocardial infarction for one year to
determine the comparative incidence of depression.
Overall, depression was equally
common in both groups, leading the authors
to conclude that the relatively high incidence
of post-stroke depression does not seem to
reflect a specific pathogenic mechanism.
Prevention of depression?
Can anything be done to prevent depressive
episode in later life? A systematic
review (Cole & Dendukuri, American Journal
of Psychiatry 2003; 160:1147-56) highlights
that sleep disturbance and disability
are two potentially modifiable risk factors,
while the effects of another, bereavement,
can be helped by timely intervention.
Life events differ between early
and late-onset depression
However, vulnerability to depression
may differ depending on age at onset.
In a study of 66 older patients and 38
controls (Grace & O’Brien, International
Journal of Geriatric Psychiatry 2003; 18:
473-478), the late-onset group experienced
fewer bereavements and more often had a
confidante, suggesting perhaps a more neurobiological
causation.
Prior depression a risk factor for Alzheimer’s
Depression is a risk factor for Alzheimer’s
disease (AD), even when the onset of
depressive symptoms occurred 25 years earlier.
The MIRAGE study of 2000 subjects with
AD and their unaffected relatives showed that
the strongest association was in families
where depression onset was in the year prior
to the onset of AD symptoms, even when
other risk factors were controlled for (Green
et al Archives of Neurology 2003; 60:753-9)
Caregivers Caregiving is a recognized risk for depression. Prigerson et al (American Journal
of Geriatric Psychiatry 2003. 11:309-319)
have evaluated a new scale designed to evaluate
stress in those caring for a person with
terminal illness. Called ‘SCARED’ high
scores were strongly correlated with major
depression with as many a 30% qualifying as
cases. But can intervention make a difference
to caregivers? In a meta-analysis, Brodaty
et al (Journal of the American Geriatric
Society 2003: 51:657-664) show that in dementia
psychosocial interventions for caregivers
are effective but that programs which
involve both patient and family are more
likely to be successful.
Prevalence of anxiety
Anxiety receives less attention than
depression or dementia. Kala et al
(Journal of the American Geriatric Society
2003: 51: 499-506) interviewed over 3000
people and found, as expected, that almost
half those who were depressed had anxiety
symptoms but 15% were anxious in the absence
of depression. The main correlates were
female gender, chronic incontinence, hearing
impairment, hypertension and poor sleep.
Detecting elder abuse
Sadly, abuse of older adults occurs.
Richardson et al (Age and Ageing 2003;
32:286-291) have demonstrated the clinical
utility of a new instrument designed to
measure knowledge and practice in this
difficult area. The instrument, ‘KAMA’
(knowledge and management of abuse),
utilizes case vignettes and could be a useful
non-threatening vehicle to train staff.
John T. O’Brien, Robert Barber, and
Robert Baldwin are the
Research Editors
of the IPA Bulletin. They welcome readers’ comments via email
(J.T.O'Brien@ncl.ac.uk).
