IPA Bulletin
Recent Advances

By Dr. Robert Barber and Professor Robert Baldwin

Ways to reduce the risk of developing dementia?
Does marital status affect risk of dementia? A 21-year longitudinal study from Finland suggests this could be the case! Intriguingly, individuals who cohabited with a partner in mid-life were less likely than all other categories (single, separated, or widowed) to show cognitive impairment later in life (Håkansson et al BMJ. 2009;339:b2462). Overall, the greatest risk for Alzheimer's disease (AD) was observed in individuals carrying the apolipoprotein E e4 allele who lost their partner before mid-life and who were still widowed or divorced at follow-up. The authors speculated this association could result from the interplay of social and genetic factors.

Further evidence from recent prospective studies (Scarmeas et al JAMA. 2009;302(6):627-37) supports previous findings that higher levels of physical activity and greater adherence to a Mediterranean-type diet are (independently) associated with reduced risk for AD. Indeed they may also be associated with a reduced risk of developing mild cognitive impairment (MCI) and subsequent conversion to Alzheimer's disease (Scarmeas et al Arch Neurol. 2009;66(2):216-25).

Findings from Solomon and colleagues also added to growing body of literature that higher levels of serum cholesterol in mid-life carry an increased risk of AD and VaD in later-life (Dement Geriatr Cogn Disord. 2009;28(1):75-80).

Neuroimaging in Alzheimer's disease.
The United States-based $60 million Alzheimer’s disease Neuroimaging Initiative (www.adni-info.org/) reflects the growing expectations of neuroimaging in the study of dementias. As results from this longitudinal study emerge, they are likely to have a major influence on understanding the clinical contribution of MRI to the assessment and management of AD, as well as our understanding about the underlying neurobiology of AD. The primary MRI data set is based on 472 individuals diagnosed as normal, MCI, or AD. Recently published findings suggest that across a spectrum from normal ageing to early Alzheimer's disease the emerging atrophy is not uniform across regions, nor does it follow a linear trajectory (McDonald et al Neurology. 2009;73(6):457-65). For example, increases in atrophy rates were greater in early disease within medial temporal cortex, whereas increases in atrophy rates were greater at later stages in prefrontal, parietal, posterior temporal, and cingulate cortex.

CSF biomarkers in Alzheimer's disease.
An international, multicentre study evaluated CSF biomarkers in the early identification of AD in patients with MCI (Mattsson et al JAMA. 2009;302(4):436-7). Overall, the best results in terms of sensitivity (83%) and specificity (72%) were obtained by a composite measure of the ratio of beta-amyloid(1-42)/total tau protein, and tau phosphorylated at position threonine 181. However, doubts about the consistency of the analytical techniques and clinical procedures between sites were raised, as the findings were less accurate than those reported from single-center studies.

But how do CSF markers compare to other in vivo investigations? Data from the Alzheimer’s disease Neuroimaging Initiative (Vemuri et al Neurology. 2009;73(4):294-301) found combining MRI and CSF measures yielded a better prediction of those individuals at risk of converting from MCI to AD than either source alone. However, MRI measures were more closely linked to future clinical/functional decline than the CSF biomarkers.

Seizures in Alzheimer disease: Who, when, and how common?
Researchers from New York completed a longitudinal study of patients with AD to estimate the incidence of seizures and any predisposing risk factors (Scarmeas et al Arch Neurol. 2009;66(8):992-7). From a sample of around 450 patients, only a relatively small proportion (1.5%) developed seizures. But nevertheless this represented an increased risk compared to the general population risk, with an estimated hazard ratio of around eight. Younger age was the only identified risk factor.

Neuropathology of Alzheimer’s disease and mild cognitive impairment.
Autopsy findings are crucial to our understanding of the relationship between common age-related neuropathies. One such significant study was published by Schneider and colleagues in the Annals of Neurology (2009;66(2):200-8). Their findings, based on 483 autopsied participants from the Religious Orders Study and the Rush Memory and Aging Project with probable AD, MCI or no cognitive impairment, found mixed pathologies were common. These findings, in conjunction with other studies, support the notion that the underlying pathology of clinical disorders such as AD and MCI is inherently varied. For example, of 179 persons with probable AD, 87.7% had pathologically confirmed AD, and 45.8% had mixed pathologies, most commonly AD with macroscopic infarcts (n = 54), followed by AD with neocortical LB disease (n = 19) and both (n = 8). Just over half the individuals with a clinical diagnosis of MCI had pathologically diagnosed AD and nearly 20% had mixed pathologies.

Treatment of vascular risk factors is associated with slower decline in Alzheimer’s disease.
In recognition, as shown in the previous report, that vascular disease and Alzheimer's disease often co-exist, a study from France explored the impact of treating vascular risk factors (VRF) on the progression of cognitive impairment in Alzheimer's disease (Deschaintre et al Neurology. 2009;73(9):674-80). Around 300 patients with Alzheimer's disease without clinical cerebrovascular disease (mean age approximately 72 years, mean MMSE score 21.6) were followed up for an average of 2.3 years. VRF included hypertension, dyslipidemia, diabetes mellitus, tobacco smoking, and atherosclerotic disease. Overall, the more VRF treated the better the outcome: patients with all their VRF treated (n=31.8%) declined less than those with none of their VRF treated (n=25.7%). Those with some VRF treated (n=42.5%) tended to have an intermediate decline. As the authors concluded, randomized controlled trials are needed to confirm this association, but the data suggests treatment of VRF in AD can make an important contribution.

Neuropsychological decline in frontotemporal lobar degeneration: A longitudinal analysis.
A longitudinal study found the key clinical phenotype of the different subtypes of frontotemporal lobar degeneration (FTLD) remained distinct over time, and different from the neuropsychological decline observed in Alzheimer's disease (Libon et al Neuropsychology. 2009;23(3):337-46.). The authors suggested this could reflect the unique anatomic distribution of relative disease burden in FTLD and Alzheimer's disease.

Intervention studies in dementia – Thumbnail reports.
A small, randomised study from France reported some benefits using music therapy in patients with anxiety and depressive symptoms on Alzheimer's disease (Dement Geriatr Cogn Disord. 2009;28(1):36-46).

A randomized trial based in three countries found a psychosocial intervention targeted at the caregivers whose spouse had Alzheimer's disease had no difference in terms of outcome: delay in nursing home placement or mortality (Brodaty et al Am J Geriatr Psychiatry. 2009;17(9):734-43). But the study revealed Australian patients were more likely to be admitted to a nursing home earlier than patients in the United States or United Kingdom.

Atomoxetine is a norepinephrine reuptake inhibitor used in the treatment of attention-deficit hyperactivity disorder. Findings from a recent 6-month RCT in Alzheimer's disease found the addition of atomoxetine to ongoing treatment with a cholinesterase-inhibitor did not significantly improve cognitive function (Mohs et al Am J Geriatr Psychiatry 2009;17(9):752-9.).

An 8-week pilot study using prazosin reported some improvements in agitation/aggression in patients’ Alzheimer's disease with a mean change in NPI score of -19 points (Wang et al Am J Geriatr Psychiatry 2009;17(9):744-51). Prazosin antagonizes norepinephrine at postsynaptic alpha-1 adrenoreceptors, and has been used in the treatment of hypertension and benign prostatic hypertrophy.

Inflammation and Depression.
In the InChiant study of 991 people aged over 65, baseline depression (using CES-D) along with inflammatory markers (C-reactive protein, interleukins and Tumour Necrosis Factor) were measured with follow up at 3 years and 6 years (Milaneschi et al, Biological Psychiatry, 2009; 65: 973-978). After adjusting for confounding variables, there was a 2.32 fold increase in the risk of depression amongst those with the highest quartiles of IL-1ra. Besides showing a cross-sectional association, this study benefits from follow up, showing that IL-1 ra levels predicted incident depression at 6 years (but not 3 years). IL-1ra is considered a reliable marker of inflammation. The study therefore supports the “cytokine hypothesis of depression,” although translating this from the laboratory to clinical practice requires more research.

Prevention was a major topic of the IPA 2009 Montréal Congress. Identifying at-risk patients is therefore important. Jinenez et al from Spain (Psychological Medicine, 2009; 39: 1201-1209) explored the role of possible biomarkers in identifying people at risk of post-stroke depression. In particular they propose that stroke may alter inflammatory responses. 134 patients with stroke were evaluated for depression at the time of discharge. Of 104 who completed follow up, 22% were depressed at one month. Amongst a range of inflammatory markers there were no differences between those with depression and those without, but serum leptin was markedly higher in those with depression (p <0.0001). The role of leptin in depression is controversial. Linked to obesity, leptin is important in energy homeostasis and in the inflammatory response. HPA suppression may be one mechanism whereby leptin synthesis increases and in this study the leptin increase was independent of other inflammatory markers. This study did not report body mass index or metabolic syndrome parameters but apparently in other studies these factors alone do not explain a link between depression and leptin.

Antidepressants.
Mukai and Tampi (Clinical Therapeutics. 2009; 31(5):945-61) reviewed the literature (January 03 to January 09) on antidepressants questioning whether single action antidepressants in later-life are as efficacious as dual acting ones. From a long list of 407 studies, 18 met the criteria for inclusion and were RCTs with either a placebo or active control group. The paper goes into some detail on these 18 studies. One tends to think that studies of tricyclic antidepressants finished in the 1980s but two of these studies were recent and involved comparing SSRIs with tricyclics (amitriptyline and trimipramine). There were no significant differences between these two classes for outcome. Surprisingly, there was no evidence that dual acting drugs were any better than single action ones but there was a caution that the mean dose of venlafaxine in the selected studies suggesting that only serotonin reuptake was being inhibited; or put another way that the dosages were too low. There were no studies of duloxetine vs SSRIs but the single study of this agent showed significant advantage over placebo. As with the paper discussed in the last IPA Bulletin (Nelson et al, American Journal of Geriatric Psychiatry 2008; 16: 558-567) longer duration of treatment was associated with better outcomes.

Bipolar disorder and psychotic depression.
Some differences have been reported in the symptomatology and outcome of bipolar disorder in older vs younger adults. Whether these are due to age of onset is not clear. The European Mania In Bipolar Longitudinal Evaluation of Medication (EMBLEM) is a 2-year observational study (Oostervink et al, Journal of Affective Disorders, 2009; 116: 176-183) of 3459 patients, 475 were aged over 60, 323 with an onset before the age of 50 and 141 beyond that. Symptoms were similar, save for psychosis which was significantly more common in younger patients. Suicide was more common in younger patients but rapid cycling more so in older patients, especially those with an early onset. Antidepressants were prescribed more often in the late onset elderly group suggesting that perhaps these cases were initially diagnosed as unipolar depression. The use of lithium and typical antipsychotics was lower in the late onset elderly compared to early onset. Overall there was no difference in recovery rates between the younger and older patients but within the elderly group the late onset patients recovered faster. Amongst the elderly group then, early onset bipolar disorder in later-life and late onset bipolar disorder are at two ends of a spectrum in terms of severity, complexity and recovery. Late onset cases appear to have the best prognosis.

Recommendations for psychotic depression usually involve either starting with an antidepressant and adding an antipsychotic if no response, starting with an atypical antipsychotic (since some have antidepressant properties) or combination treatment from the outset. Which is best? Meyers and colleagues (Archives of General Psychiatry, 2009; 66: 838-847) studied the combination of sertraline and olanzapine vs olanzapine alone in 259 patients, half of whom were aged over 60. The combination was more effective than olanzapine alone and the efficacy was as good in the older group as the younger group. The study suffered from attrition rates of more than 40%, showing how difficult it is to carry out a study of this type. Side-effects were no worse in the older group and in fact weight gain, a problem across all patients, was worse in the younger age groups. The results are also concordant with a study by Wijstra and colleagues (August number of Acta Psychiatrica Scandivica 2009) in which patients aged under 65 were given a combination of venlafaxine and quetiapine vs quetiapine alone vs imipramine alone, with the combination again proving to be superior to either single modality treatment.

Prognosis.
Keeping people well after depression is an important goal. Alexopoulous and colleagues (American Journal of Psychiatry 2009; 166: 882-890) followed up an earlier cohort as part of the “PROSPECT” (Prevention of Suicide in Primary Care Elderly: Collaborative Trial). This was an intervention study using antidepressants and/or psychotherapy and depression care management in patients with minor and major depression in the community. Compared to usual care, the intervention group patients did significantly better if they had major depression, with a greater decline in suicidal ideation and better response rate over 24 months. Amongst patients with minor depression the intervention was not particularly effective. The rates of depression treatment were between 84.9%-89% among those in the intervention group compared to 49-62% in the usual care group, meaning that greater access to and adherence to treatment may have been crucial to better outcomes.

Last word.
Post mortem studies of late-life depression are rare. Khundakar et al (British Journal of Psychiatry, 2009; 195: 163-169) examined the Newcastle brain tissue bank identifying 17 patients with major depression and 10 control subjects. The hypothesis was that in depression there will be reduced neuronal density and volume due to ischaemia, in accordance with the vascular depression hypothesis. They did identify reduced pyramidal cell volume in the dorsolateral pre-frontal cortex but not reduced pyramidal neurone density, which had been reported earlier by the only other comparable study. These changes affected particularly layer 5 pyramidal neurones which are thought to be quite vulnerable to ischaemia. Although there was no difference in late versus early onset cases, this data supports the vascular depression hypothesis.




 

Reprinted from IPA Bulletin, Volume 26, Number 4