IPA Bulletin Recent Advances - Volume 24, Number 2

By Dr. Robert Barber, and Professor Robert Baldwin

Interface Between Vascular Disease and Alzheimer’s disease

Controlling vascular risk factors to prevent cognitive decline in the elderly was one of the key messages of a further study from the Leukoaraiosis and Disability in the Elderly (LADIS) study group (Verdelho et al J Neurol Neurosurg Psychiatry April 2007). The authors examined the impact of diabetes, hypertension and previous stroke on cognition in a sample of people aged 65 to 84 years who had varying degrees of agerelated white matter changes. They found that not only did the presence of severe white matter change increase the risk of having cognitive deficits, but the vascular risk factors also appeared to have their own independent impact on cognition, hence the authors’ conclusions. Furthermore individuals with more extensive white matter changes were found to have a greater risk of both motor and cognitive decline (Inzitari et al Arch Intern Med. 2007; 167(1):81-8).

The intriguing relationship between vascular risk factors and Alzheimer’s disease was also the focus of a further study from the same group (Korf et al Diabetic Medicine 2007 Feb; 24(2):166-71). This study found visually rated medial temporal lobe atrophy, which can be an in vivo marker of ADpathology, was associated with a history of diabetes, though not hypertension.

The importance of interplay between Alzheimer-type and vascular-type pathology was reinforced by the findings from the Rush Memory and Aging Project (Schneider et al Ann Neurol. 2007 May 14; [Epub ahead of print]). The group completed autopsy analysis of nearly 150 subjects as part of a longitudinal clinicopathological study exploring the relationship between plaques and tangles on the one hand and infarcts on the other. They found subcortical infarcts were associated with reduced memory performance and when occurring with AD pathology interacted to further worsen performance. These findings add to growing clinical literature of the importance of mixed dementias.

Clinical Predictors of Developing MCI and Alzheimer’s disease

Structural neuroimaging in life offers the opportunity to detect changes in brain morphology before symptoms emerges. Researchers from the USA (Smith et al Neurology. 2007;68(16):1268-73) acquired serial MRI scans over a five-year period in order to track changes in brain morphology in normal elderly individuals (n=136) to see whether there were changes that may predate the onset of mild cognitive impairment (MCI). They found subjects who eventually developed MCI (n=23) demonstrated decreased gray matter volumes in the anteromedial temporal lobes bilaterally and left angular gyrus before the onset of any changes in cognitive performance. Ultimately the atrophic changes seen on MRI represent the visual end-point of extensive cell death, and this technique offers the opportunity to detect changes that are “biologically” active before they are “clinically” active.

In a separate study from Sweden, Palmer et al (Neurology 2007; 68(19):1596-602) assessed whether the presence of certain neuropsychiatric symptoms could help to identify individuals who are more likely to develop AD in the future. In subjects with MCI they found the chance of developing AD was much greater in those subjects experiencing anxiety symptoms: 83% of subjects with anxiety developed AD over three years compared to around 41% without anxiety. In subjects without cognitive impairment, depressive symptoms were associated with an increased risk of developing AD. The authors speculate that certain neuropsychiatric symptoms may be related to specific neuropathologic mechanisms.

In a separate study by Tun et al (Am J Geriatric Psychiatry 2007 15(4) 314-27) patients with AD who over a three-year period were classified as being “highly symptomatic” in terms of the severity and frequency of neuropsychiatric symptoms had a worse prognosis in terms of an increased risk of moving in to nursing home care and a lower rate of survival.

Spirituality and Dementia

The relationship between spirituality and mental health is potentially complex. Intuitively, a person’s religious and spiritual beliefs and practices might have a generic positive effect on certain parameters such as quality of life but the researchers from Canada were interested to know whether they could have any impact on the outcome of organic disease, such as dementia? (Kaufman et al Neurology 2007; 68(19):1596-602). Intriguingly, they found that subjects with AD who had higher levels of spirituality and private religious practice had a slower rate of cognitive decline. No effect however was observed between the rate of cognitive decline and quality of life.

Headline Findings from Clinical Trials in Dementia

Patients with AD who received galantamine for four months had a greater reduction in verbal repetition compared to placebo-treated patients (58% vs. 24%) (Rockwood et al Neurology 2007 68(14):1116-21).

In patients with Parkinson’s disease with dementia and dementia with Lewy bodies, donepezil used for 20 weeks had beneficial effects on computerized measures of attention (Rowan et al Dement Geriatr Cogn Disord 2007; 23(3):161-7.

A 90-day pilot study involving patients with AD (n=27, 19 women, mean age 81 years) compared the effects of cyproterone (100mg /d) versus haloperidol (2mg/d) on levels of (mild) aggression. Overall, the study found cyproterone had a better efficacy and safety profile. The study was conducted between 1993 and 1998 but published in 2007 (Huertas et al J Clin Psychiatry 2007 68(3):439-44).

Adding to the view that valproate preparations cannot be recommended for the (Longergan et al Cochrane Database Syst Rev. 2004;(2):CD003945.), Herrmann et al (Dement Geriatr Cogn Disord 2007; 23(2):116-9) found valproate was neither effective nor well-tolerated when used to treat aggression in patients with Alzheimer’s disease, though again this was a small study (n=14).

A 12-week open-label pilot study involving 90 patients with mild to moderate AD reported significant benefits in cognitive performance when memantine therapy was added to existing treatment with rivastigmine (Riepe et al Dement Geriatr Cogn Disord 2007;23(5):301-6).

Lithium and Alzheimer’s Disease

Although the status of lithium as either a neuroprotective or neurotoxic agent is largely unresolved ( Fountoulakis et al Int J Neuropsychopharmacol. 2007 May 17; 1- 19), in theory lithium could have a neuroprotective benefit in Alzheimer’s disease. This hypothesis is based on the premise that activation of glycogen synthase kinase 3beta (Gsk3beta) pathway has an important role in neurodegenerative disease, and that lithium in turn can modulate GSK3beta activity. In support of this hypothesis, Rockenstein et al (J Neurosci. 2007;27(8): 1981-91) found lithium could modulate GSK3beta activity in a way that reduced amyloid production and tau phosphorylation in amyloid precursor protein (APP) transgenic mice. In a clinical setting, Nunes et al (Br J Psychiatry 2007190:359-60) found AD was less common in patients (with bipolar affective disorder) who had been on long-term lithium treatment compared to those on no treatment (5% vs. 33%).

Antidepressants – Good and Not so Good News

Despite all the advances in antidepressants over the past 20 years, have they had any impact on outcome? Montagnier and colleagues from France (Journal of the American Geriatric Society, 2006; 54: 1839- 1845) examined the PAQUID database between 1988/9 and 1999 (aged 75 at entry) with 2-3 yearly review. Antidepressant use rose from 5.2% to 11.9% over the ten-year period, mainly due to SSRIs. Alongside this was a decline in depressive symptomatology (assessed on the CES-D) from 13.8% to 8.3%, even after adjustment for co-variates. The intriguing possibility is that increased antidepressant use has led to an improvement in the outcome of depression.

However, the all-age STAR*D study, commented upon in earlier IPA Bulletins, makes more sobering reading. Rush (Am J Psychiatry, 2007; 164: 201-204) summarises the data. The higher the level of antidepressant resistance the higher the chance of relapse, so early combination or augmentation treatment is preferred to several monotherapy trials. ‘Better but not in remission’ is often recorded in patients’ notes as acceptable but actually predicts poor outcomes. Remission is the goal but was disappointingly low in the STAR*D studies. As Rush comments, would physicians accept ‘less hypertensive’ as an acceptable outcome?

More positively, Narushima and colleagues from Iowa, US (British Journal of Psychiatry, 2007; 190: 260-265) randomised 47 stroke patients to receive either an antidepressant (Fluoxetine or Nortriptyline) or placebo over a 12-week trial with neuropsychological assessment of executive function. At 21 months, those on active treatment showed improvement on executive performance irrespective of depression status, whereas in the placebo group there was decline, this holding true after adjustment for confounders. The intriguing speculation is that antidepressant use may have led to neurogenesis, which is theoretically possible.

Affect, Arteries, Anger

As part of the US Pittsburgh Healthy Heart Project, (Stewart, J et al, Arch Gen Psychiatry. 2007; 64:225-233) 360 adults, mean age 60.6, completed two measures of carotid Intima Media Thickness (IMT) to detect subclinical atheroma along with mood rating (on the Beck scale) and selfreported negative emotions. There was no link between IMT and anger, hostility or anxiety but mild to moderate depressive symptoms were associated with greater three-year change in carotid IMT. Further, somatic-vegetative depressive items on the Beck but not cognitive-affective ones were most predictive. This is similar to data from the Dutch LASA study (Bremmer et al, American Journal of Geriatric Psychiatry 2006; 14(6):523-530) which showed that major (but not sub-syndromal) depression is an independent predictor of first cardiac events, especially ischaemic ones. There seems little doubt that depression is a risk factor for vascular disease. Studies of treatments aimed at altering the impact of vascular disease and its antecedents are clearly needed. In one report, which included some young-elderly, cholesterol actually increased with antidepressant treatment; why this should be so is unclear (Gabriel, A Journal of Affective Disorders 2007; 99(1-3):273-278). Trials of vasoprotective medication in vascular depression are needed.

Two papers cast doubt on whether there is a specific clinical profile to vascular depression. Naarding and colleagues, using two databases from the Netherlands, (Psychological Medicine; 2007; 37(3):382-392) rated mood and motivational symptoms in subjects of average age 76 years with DSMIV and atherosclerosis with a mixture of medical history and physiological measures. Those with vascular risk factors reported more loss of energy but not the ‘classic’ vascular depression profile of anhedonia and psychomotor disturbance. This is similar to the community case-control Vienna Transdanube Aging (VITA) Study (Rainer et al, American Journal of Geriatric Psychiatry 2006; 14(6):531-37) which used neuroimaging to rate white matter hyperintensities and volumetric data. A third study, from the Dutch National Survey of General Practice (2001) (Nuyen et al, Journal of Affective Disorders 2007; 99(1-3):73-81) suggested a possible link between depression and vascular risk factors in the age group 50 and 69 years but, paradoxically, not among older subjects. These data do not necessarily refute the vascular depression hypothesis which may be more typical of more severe secondary care cases, but raises questions.

Perhaps these will be answered by newer imaging techniques. Nam Bae and colleagues (Biological Psychiatry 2006; 60(12):1356-1363) used diffusion tensor structural abnormalities in late-life depression (106 depressed subjects and 84 elderly nondepressed controls). Fractional anisotropy values were lower in areas connecting to the frontal regions and anterior cingulate cortex among depressed subjects. These findings, the largest to date using DTI in a psychiatric disorder, generally support the theory that micro structural alterations in white matter may reflect disconnection of cortical and sub cortical regions.

Urban and Rural Depression

Adult suicide is higher in rural than urban areas, particularly for elderly men. Are there differences in the prevalence and risk factors for depression in urban versus rural districts? Friedman and colleagues from the US (American Journal of Geriatric Psychiatry, 2007; 15: 28-41) used data from an RCT of Medicare primary care consumers, average age 80, all with some functional deficits. 650 were urban dwelling and 276 rural. 14.8% of the urban patients were depressed compared to 8.3% of rural-dwellers. For rural-dwellers there were significant associations between financial strains, lack of supports and visits to the local emergency department. Relationships to physical health were more complicated. Two of these factors are potentially modifiable whilst multiple trips to the emergency room should trigger a screen for depression.

Bipolar Disorder

Brain changes may underlie bipolar disorder (BPD) as well as depression. Using a UK lithium clinic database (Subramaniam et al International Journal of Geriatric Psychiatry 2006, 5 Dec 2006 Early View), 50 BPD patients, 30 with an early onset (under 60 years) and 20 late-onset were compared for cognitive function, physical health and vascular risk factors. Using the Framingham stroke risk prediction score later-onset cases had significantly higher scores, even after controlling for other variables such as age. Cognitive scores did not differ. Zanetti et al (Progress in Neuro-Psychopharmacology and Biological Psychiatry, 2007; 31(2):551-556) from Brazil provide an overview of the theoretical basis for a vascular mechanism in late-life BPD using a case discussion of a 72 year woman with late-onset depression complicated by severe manic episodes and transient ischaemic attacks. As the authors admit, antidepressant use may have caused the manic switch although a Canadian report of bipolar patients (mean age 76) in receipt of antidepressants (mainly SSRIs) (Schaffer et al International Journal of Geriatric Psychiatry, 2006 21(3): 275-280) could find no evidence for a manic ‘switch’. In another case history (Donovan, & Freudenreich, Journal of Clinical Psychiatry 2007; 68(5):798) a 67 year male with late-onset BPD and carotid stenosis appeared to be cured by endarterectomy, with the suggestion that cerebral perfusion might have a part to play aetiologically.
 

Reprinted from IPA Bulletin, Volume 24, Number 3

Previous Recent Advances