IPA Bulletin Recent Advances - Volume 25, Number 4

By Dr. Robert Barber and Professor Robert Baldwin

Raised homocysteine has been implicated in Alzheimer’s disease (AD) and vascular disease, though the pathological and clinical relevance of this finding remains to be clarified. In terms of neuropathology, putative mechanisms linking homocysteine and neuronal damage have included potentiating oxidative neuronal injury, endothelial dysfunction, cerebral microangiography, apoptosis and betaamyloid peptide mediated toxicity. It may also exert a direct neurotoxic effect and impair production of key molecules such as neurotransmitters, phospholipids and myelin. The potential therapeutic relevance of the “homocysteine hypothesis” lies in the relationship between homocysteine and vitamins. Deficiency in folate and vitamin B12 can increase the level of homocysteine and there is evidence that even in the absence of vitamin deficiency, vitamin preparations or food enriched with folic acid, alone or in combination with vitamin B12 and B6, can reduce plasma homocysteine levels. However, whether dietary supplements of folate and vitamin B could prevent the emergence of AD or benefit patients with established AD is not clear. The Alzheimer Disease Collaboration Study group based in the United States (Aisen, et al. JAMA. 300(15): 1774-83, 2008) completed a multicenter, randomised control trial using high-dose folate, vitamin B12 and vitamin B6 supplementation in 409 (340 completers) patients with mild to moderate AD (MMSE 14-26). The trial duration was 18 months and the primary outcome measure was the ADAS-cog. Although active treatment reduced homocysteine levels, neither the primary or secondary outcome measures showed an improvement, and indeed, those treated with vitamins were more likely to become depressed. It is possible that the effects of homocysteine in AD could be independent of folate and vitamin B. Other studies have found no link between vitamin B levels (folate, vitamin B12 and B6).

New Therapeutic Option for Alzheimer’s disease?

Schilling et al (Nat Med. 14(10): 1106- 11, 2008) used transgenic mouse modeling of AD to identify an enzyme called glutaminyl cyclase that was implicated in the formation of Alzheimer-like pathology, notably pyroglutamate-modified Abeta peptides. Furthermore, a glutaminyl cyclase inhibitor reduced the formation of these peptides as well as plaque formation and gliosis. Memory performance also improved, leading the authors to conclude that inhibition of glutaminyl cyclase offers a new therapeutic option for AD.

Treating Sleep Apnoea in Alzheimer’s disease

Preliminary findings from a clinical trial suggests that treating patients with AD and obstructive sleep apnoea with continuous positive airways pressure can have some benefits on cognition, though the finding needs replicating (Ancoli-Israel, et al. J Am Geriatr Soc. 15, 2008).

Further Insights into the Link between Cerebrovascular Disease and Dementia

Autopsy prospective studies continue to be one of the most robust ways to explore the relationship between cerebrovascular and cognitive impairment. Further results from such a study - the Baltimore Longitudinal Study of Aging - were published in Annals of Neurology (Troncoso, et al. 64(2):168- 76, 2008). One hundred and seventynine subjects underwent autopsy examination, and both symptomatic and asymptomatic infarcts were found to increase the likelihood of developing dementia. This in turn was related to the number of strokes not their size. Location was also important, with only hemispherical rather than subcortical infarcts associated with dementia.

The autopsy study also found an interaction between Alzheimer-type pathology and cerebrovascular disease, a finding reinforced by a separate in vivo MRI study showing the severity of white matter lesions predicted the rate of decline in AD (Brickman, et al. Arch Neurol. 65(9):1202-8, 2008).

Using a Virtual World to Test Real-World Navigational Abilities

As the authors of this study argue, directly testing a person’s navigational ability can be both time consuming and difficult (Cushman, et al. Neurology. 71(12): 888-95, 2008). To explore the feasibility of using a simulated virtual reality test as a viable alternative, the researchers compared the performance of subjects with varying degrees of cognitive impairment using both simulated and real-world navigational tasks. They found the test performance was similar, supporting the view that simulated tests are valid. Interestingly, subjects with normal ageing and Alzheimer’s disease showed underperformance, raising concerns about potential risks in both groups.

The Burden of Alzheimertype Pathology Changes with Age

Researchers from the United States (Haroutunian, et al. Arch Neurol. 65(9): 1211-7, 2008) completed 317 autopsy examinations to determine how the burden of Alzheimer-type pathology in AD varies as a function of age. Interestingly, the density of neuritic plaques and neurofibrillary tangles increased with the severity of dementia in those aged between 60-80 years, but by the time subjects were aged 90-107, there was no difference in the burden of pathology between patients with dementia compared to aged matched non-demented controls, suggesting neuropathological mechanisms for AD vary as a function of age.

The Best Way of Predicting MCI to AD?

The prediction of subjects who will convert from MCI to AD has been approached from numerous angles. In a further study, the predictive accuracy was optimised by combining five measures: cognitive test performance, informant history of functional decline, smell identification test, and MRI hippocampal and entorhinal cortex volumes (Devanand, et al. Biol Psychiatry. Aug 22, 2008). Overall, this gave a specificity of 90% and sensitivity of 85.2%.

Healthy Body, Healthy Mind?

Interest has been growing to understand the potential value of exercise in promoting a “healthy brain”. Researchers from Australia completed a 6-month randomised trial to assess whether increasing physical activity reduces cognitive decline among adults aged 50 and over with subjective memory problems - but not dementia (Lautenschlager, et al. JAMA. 300(9): 1027-37, 2008). The activity program was home-based and the primary outcome measure, the ADAS-cog, was measured at the end of the activity program and follow-up at 18 months. Overall modest improvements were observed in favour of the activity program, with a mean difference between the active intervention group and care as the usual/plus education group of - 1.3 points on the ADAS-cog at 6 months, with differences still apparent after 18 months.

Double or Quits

Twenty years ago, the term “doubledepression” was introduced to denote major depression superimposed on chronic dysthymia. Hybels and colleagues from the United States (American Journal of Geriatric Psychiatry. 16: 300-309, 2008) explored whether having depression plus dysthymia confers a worse prognosis than major depression alone in patients aged sixty and over. Eighty-seven of 250 patients followed up over 10 years had double-depression and reported worse outcomes than those with depression alone; they took longer to achieve improvement or remission and had higher overall depression scores throughout the three-year follow-up period. Although this may seem clinically fairly obvious, it has not been demonstrated before in older patients.

An Interesting Case of …

Although much journal space is devoted to primary research, there is still room for the case discussion (‘grand round’) approach. Turvey and Klein use this format to outline the practical application of psychological interventions in a patient with chronic illness and physical impairment (American Journal of Psychiatry.165: 569-574, 2008). The patient had hypertension, diabetes, COPD, heart failure, arthritis and obesity. The authors highlight how it is physical impairment, particularly for social and recreational activities, rather than simply the presence of a medical condition, that predicts and exacerbates depression. They then describe a twelve-session psychotherapy treatment plan which incorporates elements of interpersonal psychotherapy and behavioural activation. This was comprised of mainly telephone contacts and three face-to-face interviews. The areas covered included realistic goal setting, activation and addressing the role of transition and loss. The authors provided booklets to accompany the intervention and the account is highly readable. So too is an earlier evidence-based review by Jürgen Unützer from Seattle, Washington, United States of the management of late-life depression, again centred on a case history (New England Journal of Medicine. 357: 2269-2276, 2007).

Atypicals

Atypical antipsychotics are commonly used now for bipolar disorder. Do they work in older people? Sajatovic and colleagues from the United States (Journal of Clinical Psychiatry. 69: 41- 46, 2008) conducted an open-label prospective study of aripiprazole in 20 adults with a mean age of 60 over 12 weeks, showing significant improvement in rated mania as well as some improvement in functional status. Adverse effects occurred infrequently (3 patients with restlessness, 3 with greater than 7% weight gain and 2 with sedation). This was a small study, but there is very little literature regarding older people specifically with bipolar disorder, so it is encouraging.

Relapse prevention with antidepressants works. What about atypicals? Alexopoulos and colleagues from the United States (American Journal of Geriatric Psychiatry.16: 21-30, 2008) evaluated risperidone following remission with citalopram in 63 patients with a mean age of 63.4 years. Those on the combination of risperidone plus citalopram (compared to citalopram plus placebo) had a median relapse time of 105 days compared to 57 days in those on placebo augmentation. Although this difference was not statistically significant, the authors point out that this may be an important improvement over other treatment strategies (as described later in this article). Interestingly, the dosages used were an average of 40 mg of citalopram and only 0.8 mg (mean) of risperidone. The latter fits with recent work on atypicals (not old age specific) in treatment resistant depression – low dosages seem to work.

Old and Ill

Depression is a common accompaniment of medical admission to hospital in older people. Cole and colleagues (American Journal of Geriatric Psychiatry. 16: 175-178, 2008) studied 86 older medical in-patients without depression and assessed them at 3, 6 and 12 months. Incidence of depression using DSM IV criteria for major depression was 30%. This in itself is quite an eye opener. Of almost fifty candidate risk factors, those which predicted the recurrence of depression included a prior episode of depression, less contact with children, more depressed mood with diurnal variation and inadequate emotional support. Oddly, those with low scores of medical co-morbidity at the start of the study were at greater risk of depression. The authors speculate that perhaps starting from relatively good health, they had more to lose. Yet, measures of physical illness severity also did not correlate with the risk of depression, but risk factors are additive and should not be treated in isolation. This paper also reminds us that perhaps one of the best screening questions for detecting depression in older patients admitted to medical wards is to ask whether they have ever had depression before.

Which Comes First?

Depressive symptoms are known to predate Alzheimer’s disease relatively frequently. Vascular dementia is often said to be associated with more frequent or severe depression. Luchsingar and colleagues (International Journal of Geriatric Psychiatry. 23: 922-928, 2008) from the United States studied 526 elderly people without dementia at baseline who were followed up for 5 years. The Hamilton Scale was used to assess depressive symptoms at baseline. Because they assessed subjects for diabetes, hypertension, heart disease, smoking and stroke, they were able to explore the interaction between depressive symptoms, vascular disease and later Alzheimer’s disease (one theory being that vascular risk factors are the mediating factor between depression and later dementia). Interestingly, they did not find such a link, but rather that there was a dose-response association between depressive symptoms and the later development of Alzheimer’s disease which could not be explained by vascular risks. The authors speculate that perhaps Alzheimer’s and depressive symptoms share a pathological mechanism arising from damage to structures such as the hippocampus.

Rescue Package

We know from research conducted by the Pittsburgh group in the United States, that older patients who require antidepressant augmentation because of inadequate initial response or early relapse during treatment are less likely to recover than individuals not needing augmentation (Dew, MA, et al Am J Psychiatry. 164: 892-899, 2007). The same group (Karp JF et al, Journal of Clinical Psychiatry. 69:457-463, 2008) has studied the effects of switching non-responders to escitalopram (n=40) to the dual-acting drug duloxetine in an open-label ‘rescue’ study. 50% percent showed a full response and 17.5% a partial response. Interestingly, the median dosage was 90mg, meaning that switching dosages may be higher than initial treatment dosages (which is often said to be 60mg). Although these data are perfectly consistent with the literature on augmentation, the use of a single drug rather than combinations may have benefits, both clinical (reduced side effect burden) and financial (in those countries where medication must be paid for). Given recent concerns about SSRIs and suicidality, it is reassuring to see that a large Canadian cohort study (n=128,229, mean age 75 years) (Rahme, E et al Journal of clinical Psychiatry. 69:349- 357, 2008) showed no evidence of increased risk of death by suicide in SSRI users compared to non-users. In fact, the overall rates were lower for SSRI users, probably because most suicides occur to depressed persons not receiving treatment. However, patients who switched to another antidepressant did seem to have an increased suicide rate. The database could not elaborate on this finding, but it might be that patients who require switching or augmentation are a more ‘at risk’ group for suicide.

 

Reprinted from IPA Bulletin, Volume 25, Number 4