IPA Bulletin Recent Advances - Volume 24, Number 2

By Dr. Robert Barber, and Professor Robert Baldwin

Treatments in Alzheimer’s Disease

Antipsychotic medication – An important study published in the New England Journal of Medicine (Schneider et al 2006 355(15):1525-37) assessed the effectiveness of atypical antipsychotics in the management of psychosis, aggression and agitation in patients with Alzheimer’s disease (AD). This USA based, 42-site study followed the trial design and methodology of the NIMH clinical antipsychotic trials of intervention effectiveness (in this case CATIE-AD) which aims to assess outcomes in a way that reflects “real life” clinical practice. Unlike most previous clinical trials in AD, outpatients were recruited (n=421) and the follow-up period was longer (up to 36 weeks). In phase 1 of the study, subjects were randomised to olanzapine, quetiapine, risperidone or placebo.

The main outcome measure was “time to discontinuation of treatment for any reason”. This measure reflects the composite outcome of efficacy, safety and tolerability, i.e. how the balance of benefits and adverse events plays out for each individual patient as viewed by the patient, carers and the clinician. Using this measure, there was no difference between any of the drugs and placebo with both being discontinued relatively soon (median 5-8 weeks) and in most subjects (77 to 85%).

Furthermore, “discontinuation due to adverse events or intolerability” was more common in patients receiving an antipsychotic agent (16-24%) compared with placebo (5%). However, olanzapine (mean dose 5.5 mg per day) and risperidone (mean dose 1.0mg per day), but not quetiapine (mean dose 56.5mg), were more effective than placebo as measured by the “time of discontinuation of treatment due to lack of efficacy” (22.1; 26.7, 9.1; 9.0 weeks respectively). Overall, the authors conclude that the advantages in efficacy were offset by the adverse effects, and the use of antipsychotics to manage psychosis, aggression or agitation in AD may be restricted to patients who have discernible benefits in the absence of side effects. Phase 2 study involves allocating subjects who have discontinued treatment to citalopram – so hopefully this will further inform the clinical management of patients who are experiencing distressing and challenging symptoms.

Exercise
Rolland et al (J Am Geriatr Soc 2007 55(2):158-65) used a randomised controlled trial design to evaluate the effect of exercise on nursing home residents with Alzheimer’s disease. The study, conducted over one year and involving 134 residents, used an exercise program consisting of 2 hours per week of walking, strength, balance and flexibility training. Over the year, residents receiving the program experienced a slower decline in activities of daily living and better walking speed, although there was no obvious impact on nutrition, behaviour and mood.

Donepezil
Wallin et al (Dement Geriatr Cogn Discord 2007;23(3):150-60) reported on the outcome of the Swedish Alzheimer Treatment study – a descriptive and prospective study (n=435 outpatients) to evaluate the long- term benefits of donepezil in routine clinical settings. After 3 years, the mean decline in MMSE scores was 3.8 points (95% CI 3.0-4.7). This appears to be better than predictions from historical untreated cohorts, with the authors concluding that the beneficial effects of treatment with donepezil in routine clinical practice continues to be observed 3 years after initiating treatment.

Memantine
Pomara et al (Alzheimer Dis Assoc Discord 2007 21(1):60-64) conducted a post hoc analysis of a 24-week randomised, double- blind, placebo controlled trial to evaluate the effects of memantine on cognition in patients with mild to moderate AD. Individuals receiving memantine showed less decline than those on placebo on 5 out of 11 ADAS-cog subscales, including orientation, language and memory. The authors suggest these findings are consistent with trials conducted in moderate to severe AD, for which the treatment is currently licensed in the USA and Europe.

Celecoxib
Epidemiological studies suggest that nonsteroidal anti-inflammatory drugs may delay or prevent the onset of AD, perhaps by lowering the levels of amyloid ß peptide. To evaluate whether there are treatment benefits in established mild to moderate Alzheimer’s disease, Soininen et al (Dement Geriatr Cogn Discord 2007;23(1):8-21) conducted a 52-week randomised, double-blind placebo controlled study using celecoxib, a cylcooxygenase-2 selective inhibitor. Overall, there were no differences between placebo and celecoxib (200mg bd) in the primary outcome measures for cognition (ADAScog) and global effect (CIBIC+).

Statins
Reducing cholesterol using statins could in turn reduce the amount of Abeta peptides, raising the possibility that statins could have preventive or treatment benefits in AD. However, in a recent meta-analysis (Zhou et al Dement Geriatr Cogn Discord 2007;23(3):194-201) statins failed to show a beneficial effect on the risk of AD.

“Cognitive Reserve” Hypothesis in Alzheimer’s Disease

Katzman and colleagues in 1980s recognised that there was a discrepancy between the degree of Alzheimer’s disease identified at autopsy and the clinical manifestations of the disease in life. This led them to speculate certain patients may have greater brain or cognitive “reserve” and show greater resilience to the underlying AD pathology. Such patients may go undiagnosed in life or only develop symptoms when the illness is more biologically advanced. Various factors have been suggested to promote or be associated with professional attainment, leisure activities, familial antecedents, IQ and brain size.

To further explore this hypothesis, Roe et al (Neurology 2007 68(3):223-8) investigated the association between years in education and the presence or otherwise of dementia within 1 year of death among a large cohort of individuals with neuropathological diagnosis of AD. In line with the hypothesis, individuals with greater years of education were indeed less likely to have a clinical diagnosis of dementia in the presence of AD pathology. The authors speculate that individuals with greater educational attainment are more able to cope with AD pathology, though currently there is no evidence that education protects from Alzheimer’s disease pathology per se.

Day Hospitals

Day hospitals are a cornerstone of many services but do they work? There is remarkably little evidence but a study by McKenzie and colleagues (International Psychogeriatrics, 2006; 18: 631-641) is illuminating. This study covered over 700 patients with depression admitted to a day hospital in Toronto, Canada, over a 16 year period, representing about 70% of all referrals from its opening . All were given the Geriatric Depression Scale (GDS) and the Hamilton Rating Scale for Depression (HRS-D), along with other clinical scales. In terms of reduction in the GDS and the HRSD scores, there were large standardised effect sizes for those admitted. Between 30% and 40% of those attending experienced at least a 50% improvement in mood outcome. Interestingly, the authors compare their data to a reasonably recent placebo-controlled trial of Sertraline showing that the improvement they observed on the Hamilton score was equivalent to that seen in the RCT. Large naturalistic studies have been out of favour in recent years but this study reminds us how valuable they can be.

Bipolar Disorder

Bipolar disorder in later life is associated with much morbidity and health care usage. Lehmann and Rabins (International Journal of Geriatric Psychiatry, 2006; 21: 1060-1064) conducted a retrospective review of in-patients in a major US hospital over a 5 year period. Of 73 evaluable records, the distribution of age of onset suggested a twin peak with cut off of approximately 45 years. Early-onset bipolar patients were more likely to have been aggressive or threatening prior to admission and were more likely than late onset patients to have poor adherence to prescribe medication. However relapse and re-hospitalisation was common in both types, as was medical comorbidity. In 11% of patients the admission appeared to have been precipitated by a planned reduction in medication. It is unwise then to assume that bipolar disorder tends to “burn out” with advancing age.

Urinary Symptoms in Depression

Wong and colleagues (Journal of Affective Disorders, 2006; 96: 83- 88) from Hong Kong studied depression (Chinese version of the GDS) in 1979 participants who had completed a lower urinary tract screening questionnaire, all male. After controlling for other factors likely to be important in depression, severe lower urinary tract symptoms were associated with a 3.36 increased odds of having depressive symptoms. Although it is known that incontinence is associated with depression, this study adds the finding that there was a dose-response relationship between increasing non-cancer related urinary symptoms and depression.

Help-Seeking in Different Ethnic Groups

In a qualitative research study Lawrence and colleagues (Psychological Medicine, 2006; 36: 1375-1383) studied 110 older adults with and without depression (Black Caribbean 32, South Asian 33, White British 45) from the London, UK. The aim was to explore attitudes towards self-help, social support and health care beliefs in depression. Whilst the use of medication was endorsed by all three groups, all valued the opportunity to talk and felt there was insufficient time to do this with the general practitioners. There was also a tendency for participants to view depression as something which had to be combated from within although this may reflect a very limited exposure to any sort of counselling services. Black Caribbean participants viewed “talking with God” as a positive strategy and those from a South Asian background oriented themselves more towards the family. The participants were from primary and not specialist care and only about 10% thought seeing a psychiatrist was appropriate for depression. There was concern about stigma. The paper is rich with quotes from participants.

Why Worry?

Lindesay from the UK (Psychological Medicine, 2006; 36: 1625-1633) analysed data from the Second National Survey of Psychiatric Morbidity (2000), with the hypotheses that worry content changes with age and that the likelihood of worry declines with age, even in those with mental disorder. An interview schedule for worry was used along with one for mental disorder, life events and functional limitations. Of this large sample (8580) 1442 were 65-64 and 1268 65-74. After adjustment for confounding variables, older adults reported significantly fewer worries overall than younger age groups. Possibly older adults use more passive and emotion-focused strategies and younger people more problem focused ones. Interestingly, worry about health was a feature of the young and middle aged rather than old age. There was also some evidence that specific worries in later life were more indicative of mental disorder than in younger adults. There was a reasonably constant association between worry and evidence of mental disorder. The authors speculate that increasing age and experience may lead to “wise detachment” but a problem with the survey is a lack of participants above the age of 74.

Brain Again

Can brain imaging changes predict outcome in late-life depression? Patankar and colleagues from Manchester, England (Journal of Affective Disorders, 2007; 97: 265-270) examined the predictive value of Virchow-Robin spaces (VRS). VRS are perivascular spaces in the subcortical brain regions with a characteristic appearance on T1-weighted inversion recovery sequences on MR scanning and may be sensitive indicators of cerebral microvascular disease. In a sample of 50 depressed patients (29 monotherapy responders and 21 more resistant cases) compared to 35 controls, white matter hyperintensities were higher in the resistant group but not statistically so. For basal ganglia VRS measures there was a highly significant group difference between both resistant vs control subjects and resistant vs responder patients. VRS scores were 80% sensitive and 62% specific for poor outcome. Vascular scores did not differ between the groups so this is an interesting visual measure that could be incorporated in future scanning studies where microvascular disorder is suspected.

Not Hip-Hip Hooray

Two studies in the Journal of the American Geriatric Society (JAGS) have researched depression and hip fracture. Lenze and colleagues (JAGS, 2007; 55: 81-86) from Pittsburgh, US, examined factors predictive of the incidence of major depression after hip fracture. In a sample of 126 patients, 14.3% developed major depression following fracture (none had depression at baseline) with the main risk in the first 10 weeks, especially the first two. Of predictors, none of the characteristics of the hip fracture itself were predictive but 46% of patients with high apathy scores developed depression compared to only 11% of those with lower scores. The authors suggest that signs of disinterest and amotivation early after recovery from hip fracture indicate incipient depression. Burns and colleagues from Manchester, England (JAGS, 2007; 55: 75-80) explored treatment, both preventative and intervention, in 293 older patients posthip fracture surgery. 121 depressed subjects were allocated to either a multi-faceted nurse intervention (N=61) or usual care (N=60). There was a very slight reduction of uncertain clinical significance in the intervention group. 172 participants with no depression were allocated to a CBT intervention or usual care in equal numbers but there was no significant difference after 6 weeks in the numbers who developed depression. Although patients with a hemiarthroplasty or total hip replacement were more likely to develop depression, there were no differences in physical outcomes (eg mobility or pain) in either the treatment or prevention arms. This is disappointing, but the study may have been under-powered. The accompanying editorial in JAGS highlights the high prevalence and complexities of treating depression in this setting.

Suicidality

Szanto and colleagues from Pittsburgh, US (Journal of Affective Disorders, 2007; 98: 153-161) examined 437 patients (234 treated with Paroxetine, 203 with Nortriptyline) for suicidal ideation (on the Hamilton Rating Scale for Depression sub-item). This they correlated with chronicity of depression, anxiety symptoms and akathesia during 12 week treatment periods. 12.6% reported suicidality for longer than the first 4 weeks whilst 15.6% had reported suicidality but this had resolved. A final 7.8% reported suicidality later during the treatment (‘emergent’). Patients with emergent and persistent suicidality had higher levels of unresolved depression whilst those with resolved suicidal thoughts had a more favourable treatment response. Importantly, there was no difference between Nortriptyline and Paroxetine in terms of emergent suicidality and no link to akathesia, although akathesia was not marked in this study. As the authors comment, emergent or chronic suicidal ideation, even though infrequent, may be a marker for difficult-to-treat depression.

Conner and colleagues from the US (Journal of Affective Disorders, 2007; 97: 123-128) used the Suicide Intention Scale (SIS) to study 117 depressed patients, minimum age 50 (mean 61 years) who had attempted suicide, three quarters by overdosing. The level of planning of the suicide attempt was assessed in relation to cognitive impairment, physical function (including ADL), subjective hopelessness and living alone, factors suggested by previous research. Lower cognitive function was associated with lower levels of planning self-harm but poor physical function, contrary to expectation, was associated with lower rather than higher level of planning. Hopelessness was not linked to level of planning. Older age was associated with higher planned suicide attempts. These findings to do not mean there is no association between poor physical state and/or cognitive function and the planning of suicide but rather that these factors may predispose to more impulsive suicidal behaviour. Different patterns of self-harm in later life may therefore be linked to different risk factors.  

Reprinted from IPA Bulletin, Volume 24, Number 2

Previous Recent Advances