IPA Bulletin Recent Advances - Volume 25, Number 2

By Dr. Robert Barber and Professor Robert Baldwin

Anxious elders
Not much is known about the treatment of anxiety disorders in older adults. Pinquart and Duberstin (American Journal of Geriatric Psychiatry, 2007; 15: 639-651) conducted a meta-analysis, including both controlled and uncontrolled studies, of anxiety in patients aged at least 60 years. A total of 19 studies investigated pharmacotherapy and 12 reviewed the effects of behavioural intervention and one study examined both kinds. Treatment effects were impressive, somewhat higher than reported in younger adults and included some reduction in co-morbid depressive symptoms. Pharmacological interventions were somewhat more effective than behavioural ones, although comparing studies is difficult. Drop outs rates were similar and no one drug stood out. Interestingly, studies of older people which included PTSD were absent – perhaps it is overlooked.

It is often said that anxiety disorders decline with later life and this is supported by Corna and colleagues (International Psychogeriatrics, 2007; 19: 1084-1096) in a Canadian survey (N = 12 792) of over 55 year olds. About 1% reported 12 months prevalence and 2.5% a lifetime prevalence of anxiety disorder (one-fifth occurring after the age of 55). The highest prevalence seemed to be in the 55 to 64 year old group and the disorders were disabling, so certainly worth detecting and, as the previous paper shows, treating.

Prevalence of depressive symptoms and syndromes
EURO-D is a 12-item self-report depression questionnaire which has identified “affective suffering” and “motivation” as two strong constructs in late life depression. Castro-Costa and colleagues (British Journal of Psychiatry, 2007; 191: 393-401), studied a representative sample of people aged above 50 from 10 European countries (N=22, 777). Individual EURO-D symptoms consistently varied between countries with a higher prevalence in France, Italy, and Spain (“Latin” countries). What accounts for these differences is not quite clear as the motivation factor appeared to be mediated by cognitive impairment (verbal fluency rather than memory) and could reflect subcortical brain damage, i.e. apathy rather than depression. Cultural factors may very well be underlying some differences in prevalence and the way terms were phrased could have been important: motivational items were all positively worded whereas affective suffering negatively so. These nuances may be susceptible to cultural influence.

How well do antidepressants work?
Placebo controlled trial of antidepressants are not common these days (usually trials are “head-to-head”). Raskin and colleagues from the US (American Journal of Psychiatry, 2007; 164: 900-909) conducted an RCT of Duloxetine (N=207) and placebo (N=104) over eight weeks in elderly patients. Duloxetine was significantly more effective and also was associated with a reduction in back pain scores and, interestingly, some improvement in verbal learning and memory, but not attention and executive functioning. As with many recent placebo RCTs though, the proportion achieving remission was low, at only 27.4% (vs 14.7% placebo). The figures for improvement and remission translate into NNTs of 5.3 and 7.9 respectively.

Rehabilitation of geriatric drugs?
A useful review of monoaminoxidase inhibitors has been compiled by Krishnan (Journal of Clinical Psychiatry, 2007; (suppl 8: 35-41), with some data coming from later life patients. MAO-I’s have fallen out of favour in recent years but the availability of trans-dermal delivery reducing side-effects and dietary requirements, may rehabilitate them. That said, Kok et al (Journal of Clinical Psychiatry, 2007; 68: 1177-1185) from Utrecht, Holland, found that compared to lithium treatment, Phenelzine was inferior in the augmentation of 28 depressed subjects with the mean age of 73. These patients had not responded to standard antidepressants, two-thirds were psychotic and the outcome criterion was remission. Short-term outcome (six weeks) was poor in both groups, but rose considerably to over 80% with long term follow up. Tolerance was good in both groups. The message seems to be: persistence pays and do not discount lithium.

Mood disorder and vascular risk factors
Van Der Kooy and colleagues from the Netherlands (International Journal of Geriatric Psychiatry, 2007; 22: 613-66) conducted meta-analyses of 28 studies showing that depressed mood increased the risk of a range of vascular diseases, notably myocardial infarct (for which there was best evidence) with an odds ratio of 1.43-1.63. The strongest association was for major depression with marked heterogeneity for studies involving depressive symptoms only, but suggesting a dose response relationship. Various pathways are suggested. Ahto and colleagues from Finland (International Journal of Geriatric Psychiatry 2007; 22: 757-763) used a population based study (N=1196 aged over 64) and also found the strongest predictor of coronary heart disease was among those with the most severe depression. Lastly, Subramaniam and colleagues from the UK (International Journal of Geriatric Psychiatry 2007; 22: 733-737) examined patients with bipolar disorder both over and under the age of 60. Using the Framingham stroke risk score, those in the late onset group had a higher score, although the groups did not differ significantly on overall health or cognitive status. So, more evidence that cerebrovascular disease and affective disorder are closely linked.

Some compli(e)ments …
There is much interest in complementary medicine as alternatives to medicines and Meeks et al (Journal of Clinical Psychiatry, 2007; 68: 1461- 1471) reviewed the evidence for them in older patients with depression, anxiety, or sleep disturbance (but not dementia) using PUB MED between 1966 and 2006. Thirty-three studies met their inclusion criteria and with a bewildering wide range of treatments (yoga, meditation, relaxation, music, qigong, tai-chi, vitamins, melatonin, St John’s Wort, light therapy and acupuncture). Two-thirds of the studies were positive and despite methodological limitations, mind-body interventions, e.g. tai-chi and body-based therapies (e.g. acupuncture), were modestly effective for anxiety/ depression and sleep disturbance. On the whole, they were slightly more effective than biologically based therapies (e.g. St John’s Wort). Some commonly- used treatments in the East based on energy manipulation (e.g. qigong) are less known in the West, and there is much scope for large organised trials.

A good read
The December 2007 edition of Psychological Medicine is almost a themed issue for old age psychiatry. Herrmann et al (37: 1693-1702) reviewed cognitive function in ten studies of geriatric depressive disorder, concluding that late onset depression was characterised by greater impairments of executive functioning and processing speed than early onset depression. Taylor and colleagues (37: 1763-1773) explored the relationship between orbito-frontal cortex (OFC) volumes, depression and the genetic effects of the serotonin transporter gene and APOE. OFC was smaller in depressed subjects and associated with more visually- rated white matter lesions but there was no association with the genetic polymorphisms. Both studies are consistent with “vascular depression”. Almeida et al (37: 1775-1786) measured depressive symptoms and C-reactive protein (CRP) in a community sample of men and found that although there was a significant correlation between CRP and depression, this appeared to be mediated via poor physical health. Spek et al (37: 1797-1806) explored the efficacy of an internet based psychological intervention in 300 subjects with subthreshold depression aged over 50. The internet therapy was as effective as a comparison group-delivered course (Coping with Depression (CWD)) and more effective than a waiting list group. Dougall et al (37: 1787-1795) assessed 2640 over 65 year olds for depression finding a prevalence of 8.7% (9.7% if dementia included): 10.4% women, 6.5% men. Severe depression was rare (2.7%). Disability was the strongest association. Importantly, depression did not “go away” with advancing age. Lastly, religious practice may buffer depression, but King et al (37: 1807- 1815) found that the relationships were very complex.

A gloomy end?
There is debate as to whether depression in later life increases mortality. From France, Ryan et al (British Journal of Psychiatry, 2008; 192: 12-18) studied the four year survival of 7363 older people from the community. Twice as many men (7.5%) compared to women (3.7%) died but with differing predictors. For women, the highest mortality was among those with severe depression who were not receiving treatment; whereas in men, the risk of dying increased with depression severity and was highest in those with severe depression who had been prescribed antidepressants. Type of antidepressant was not a factor. There were few deaths from suicide. Non-depressed men who were on antidepressants had no increased risk of death, arguing against antidepressants being potentially harmful. Rather, the opposite seems to be true. The authors speculate that perhaps depressed older men simply die whereas older women first become disabled. However, this study paid close attention to physical co-morbidity, so it seems unlikely that undisclosed physical illness could explain these findings. Depressed men are known to be reluctant to come forward for help but in men, all levels of depression severity are potentially harmful.

Clinical trials in dementia

Use of NSAID and aspirin in dementia.
Inflammatory mechanisms have been hypothesised to have an important role in the pathogenesis in Alzheimer’s disease (AD). Linked to this, laboratory and epidemiological findings have suggested that non-steroidal antiinflammatory drugs may reduce the risk of AD, although clinical trials for both the prevention and treatment of AD with these medications have todate been inconsistent. Szekely and colleagues from the U.S.A. (Neurology. 2008 Jan 1;70(1):17-24) conducted a further epidemiological study using participants from the cardiovascular health cognition study which included over 3,000 individuals over the age of 65 who were free of dementia at baseline. They found that the use of NSAIDs was associated with lower risk of dementia in particular AD (adjusted hazard ratio of 0.63) but not vascular dementia. However, this association was only found in those patients who had an APOE 4 allele.

However, findings from a recent clinical trial raised concerns about the use of aspirin in patient with dementia. The AD2000 collaborative group from the UK completed a randomised openlabel study of low dose in patients (n=310, mean age 75 years) with mild to moderate Alzheimer’s disease (Lancet Neurology 2008 Jan;7(1):41-9). Half the group were assigned to one 75-mg enteric-coated tablet per day whilst the remaining patients avoided its use. After 2 years, no benefits were found in favour of aspirin but the risk of bleeding was increased with 2% patients in the aspirin group having fatal cerebral bleeds.

RCT of donepezil in subcortical vascular dementia (caused by CADASIL)
The role of cholinesterase inhibitors (CHI) in non-Alzheimer’s disease patient groups remains unclear, and in particular, there have been inconsistent findings regarding their role cognitive impairment secondary to vascular brain disease. Dichgans et al (Lancet Neurol. 2008 Apr;7(4):310-8) report their findings from a 18-week randomised study of donepezil in patients (mean age 54.8 years. MMSE 10-27) with cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoencephalopathy (CADASIL), a genetic form of subcortical ischaemic vascular dementia. The onset of CADASIL is usually about 45 years of age and the potential advantage is studying this patient group is that they will have a relatively pure form of vascular dementia against which to assess the efficacy or otherwise of CHI as they are unlikely to have confounding co-morbidities such as Alzheimer type pathology. Eighty-six patients received donepezil (10mg/daily) but compared to the placebo group (n=82) there were no differences in outcome on the primary endpoint - change from baseline in the score on the vascular AD assessment scale cognitive subscale.

Can antidepressants help cognitive function in dementia?
The complex, possibly two-way, nature of the relationship between depression and dementia was further scrutinised in a study by Mossello et al from Italy (Dement Geriatr Cogn Disord. 2008 Mar 20;25(4):372-379). Drawing on the analogy that cholinesterase inhibitors can help both cognitive and non-cognitive symptoms in dementia, the authors posed the same question for antidepressants. Overall, they found antidepressant use was associated with less cognitive decline in older depressed patients with Alzheimer’s disease – an effect that was in part independent of the effect on depressive symptoms.

Systematic review of the effectiveness of cholinesterase inhibitors and memantine in dementia.
Raina et al from Canada published a systematic review of the effectiveness of cholinesterase inhibitors and memantine (Ann Intern Med. 2008 Mar 4;148(5):379-97). Ninety-six published studies, representing 59 unique studies, were reviewed with the conclusion that both cholinesterase inhibitors and memantine had consistent effects on cognition and global assessment but the overall effect size was modest. Outcomes on the domains of behaviour and quality of life were however generally less consistent.

Using memantine for non-cognitive symptoms in Alzheimer’s disease.
Wilcock et al (J Clin Psychiatry. 2008 Feb 20;e1-e8) conducted a post hoc analysis of three memantine studies focusing on patients with moderately severe to severe Alzheimer’s disease who were also either agitated or aggressive or experiencing psychosis. In summary, they found treatment with memantine provided modest benefits in neuropsychiatric symptoms specifically for the treatment of agitation and aggression.

Practice Guidelines
A clinical practice guideline on the state of current pharmacologic treatments in dementia was published by the American College of Physicians and the American Academy of Family Physicians in the Ann Intern Med. (2008 Mar 4;148(5):370-8).

Imaging in dementia

Is what we see on imaging what we think it is?
The imaging technologies that have evolved from “x-ray vision” have allowed us to look inside the human brain during life in ever increasing detail, but being able to make the link between what we see and the underlying pathology stills presents a key challenge. To try and improve our understanding of this issue, Jagust and colleagues from the U.S.A. completed a long-term clinic-pathological study allowing comparison between MRI imaging findings in life with findings from post-mortem examination (Ann Neurol. 2008 Jan;63(1):72-80). The intuitively reassuring finding that vascular changes seen in life (white matter change and lacunar infarct) are indeed secondary to vascular disease was perhaps not surprising, but the underlying pathology associated with atrophy appears to be more complex with “several pathological processes converging on similar brain structures”. In particular, hippocampal volume loss was associated with both Alzheimer-type pathology and hippocampal sclerosis, whereas the pathology of grey matter volume loss was more widespread and included Alzheimer type pathology, subcortical vascular pathology and atherosclerosis.

Intriguingly, findings from a separate clinic-pathological study raise further questions about the amyloid hypothesis of Alzheimer’s disease. Josephs et al (Ann Neurol. 2008 Feb;63(2):204-12) found the rate of brain volume loss as measured by serial MRI in life was not determined by the amount of insoluble Abeta in the gray matter at postmortem. However, significant correlations were observed between rates of brain loss and age, Braak NFT stage, and change over time in cognitive measures.

Changes in atrophy as patients progress from MCI to Alzheimer’s disease.
In a way consistent with the neuropathological models of Alzheimer’s disease progression, Whitwell et al (Neurology. 2008 Feb 12;70(7):512-20) tracked subjects who converted from MCI to Alzheimer’s Disease with serial MRI scans and showed that these patients showed greater loss in the medial and inferior temporal lobes, the temporo-parietal cortex, the anterior and posterior cingulate and frontal lobes compared with those patients who remained stable.

Review article on neuroimaging.
An interesting review by Small et on the “current and future uses of neuroimaging for cognitively impaired patients” can be found in Lancet Neurology (2008 Feb;7(2):161-72).

Genetics and biomarkers in dementia

Risk of developing Alzheimer’s disease when both parents have the disease.
Recognising the lack of evidence to address this question, Javadev and colleagues (Arch Neurol. 2008 Mar;65(3):373-8) used information from a database of 111 families in which both parents were affected by Alzheimer’s disease (AD) to determine the risk of AD in family members. Overall, they found that if both parents developed Alzheimer’s disease there was indeed an increased risk of AD in the children compared to that of the general population: 31% of those older than 60 years and 42 % of those older than 70 years had AD. The overall risk may even be higher because not all probands had reached the “at-risk” age.

Impact of APOE genotype disclosure.
Guidelines currently advise against direct testing of APOE status in Alzheimer’s disease (AD), in part because the presence of the APOE 4 allele is neither sufficient nor necessary to develop AD and quantifying risk is therefore problematic, and disease modifying treatments are not yet available. However, how would individuals react to the knowledge of their APOE genotype? Chao and colleagues from the U.S.A. published findings from the first randomised clinical trial (the REVEAL Study) for the disclosure of APOE genotype (Alzheimer Dis Assoc Disord. 2008 January/March;22(1):94- 97). Participants (N = 162) were randomised to the intervention (disclosure or control group no disclosure). All subjects received a numerical life-time risk estimate of future AD generated from knowledge of their family history and gender. Interestingly, participants who learnt they were APOE 4 positive were more likely to undertake changes in their behaviour which putatively may modify their risk of developing the condition. In a separate study, Jarvik et al review the state of knowledge with respect to risk and protective factors for AD with particular emphasis on children from persons with AD.

Can measuring plasma beta amyloid help detect the risk of developing Alzheimer’s disease?
Translating laboratory scientific findings into meaningful clinical biological markers of disease prediction or diagnosis continues to be a fascinating challenge. Extrapolating from the amyloid hypothesis of Alzheimer’s disease (AD) that as Abeta accumulates in the brain in patients with AD these changes can be detected in the plasma, Sundelöf and colleagues from Sweden (Arch Neurol. 2008 Feb;65(2):256-63) complete a large perspective study measuring plasma Abeta(40) and Abeta(42) levels in 1,045 men aged 70 years and 680 men aged 77 years to see whether the levels predicted in the development of AD. They found low plasma Abeta(40) levels were associated with the onset of AD in men aged 77 years but there was no clear relationship in the younger age group, nor for Abeta(42) in either age group. This led the authors to conclude that “the clinical value of Abeta measurements in plasma remains to be established”.

Assessing capacity in dementia
Assessing capacity in clinical practice can present considerable challenges. Martin and colleagues from the U.S.A. completed a one-year longitudinal study to assess the differences in performance of capacity in patients with mild Alzheimer’s disease and healthy adults (Am J Geriatr Psychiatry. 2008 Mar;16(3):209-19). They also tracked changes in the capacity of patients over the duration of this study as the condition progresses. Overall, they found using standardised measures of financial capacity that patients with mild Alzheimer’s disease were already substantially impaired at baseline and also experienced additional decline over the course of the next year. The authors concluded that “the study supports the importance of prompt financial supervision and planning for patients duly diagnosed with Alzheimer’s disease”.

Risk of developing dementia in Parkinson’s disease
The close link between Parkinson’s disease (PD) and the development of neuropsychiatric symptoms is wellrecognised, and a study from Buter and colleagues reinforces this association (Neurology. 2008 Mar 25;70(13): 1017-22). Reporting on 12-year study to explore the risk of developing dementia in patients with PD (n=233), they found 140 patients developed dementia during this period with the cumulative incidence of dementia steadily increasing with age and duration of PD: conditional on survival this increases to 80- 90 per cent by the age of 90 years. This led the authors to conclude that “dementia is a key part of surviving PD and must be planned for services for this condition”.

Reprinted from IPA Bulletin, Volume 25, Number 2

Previous Recent Advances