IPA Bulletin Recent Advances - Volume 23, Number 2

By Dr. Robert Barber, and Professor Robert Baldwin

What’s in a brain?

Is the vascular depression hypothesis backed up by neuropathological data? Rajkowska and colleagues (Biological Psychiatry, 2005; 58: 297-306) from the United States examined post-mortem samples from the orbito-frontal cortex in 15 older patients with major depression compared to 11 control subjects (mean age 75 ±9 years). They also compared their results with earlier findings from younger depressed patients. The density of pyramidal neurones in the orbito-frontal cortex was reduced by 30% in elderly depressed subjects with a particular emphasis on layers VIII and IVc. Glial cells were not affected. The effects were more pronounced in older than younger people but in the older sample findings were equally split between early vs late onset. The authors propose that because pyramidal neurones in layers V and III have been linked to glutamatergic pre-frontal projections to the striatum, a reduction in the packing density of these neurones could suggest disruption to pre-fronto stiatal fibres as a result of white matter lesions. Indirectly, then, these data do support the vascular depression hypothesis although other causes of degeneration besides vascular disease are possible.

Predicting response – executive function

 Turning from post-mortem data to living patients, in an antidepressant trial with Citalopram Alexopoulos and colleagues (Biological Psychiatry 2005; 58: 201-204) studied 112 older patients with major depression (average age 73) looking particularly for treatment response in relation to executive dysfunction as measured by the Mattis Dementia Rating Scale (initiation and perseveration score) and the Stroop Color-Word Test. Response to treatment was assessed by a 50% reduction on the Hamilton Scale. An abnormal score on the initiation/perseveration test or on response inhibition tasks with the Stroop predicted a limited response to anti-depressants. Since executive dysfunction is one of the expressions of altered fronto-striatal limbic circuitry, the results are consistent with previous research suggesting that vascular damage leading to brain pathology is both an aetiological determinant of depression but also affects treatment response. The authors highlight the need for a different treatment approach for such patients who may require additional interventions with behavioural or psychological approaches.

Vascular depression

Two small studies from the International Journal of Geriatric Psychiatry add further support to the close links between depression and vascular disease. In the first, from Tai Wan, Cheen and colleagues (International Journal of Geriatric Psychiatry, 2006; 21: 36-42) studied 14 patients with depression with an onset after 50 with 11 non-depressed controls. The depressed patients were moderately ill, having a mean Hamilton score of just 22. They found a high correlation between carotid intimamedia thickness (IMT) and visually-rated white matter lesions in depressed but not control subjects. The fact that they were late onset rather than early onset depressions suggest that the artherosclerosis, of which IMT is a marker, may have contributed to the depression rather than the other way round, although this certainly cannot be ruled out. In the second study Baldwin and colleagues (International Journal of Geriatric Psychiatry, 2006; 21: 57-63) from the UK followed up a cohort of depressed patients and control subjects over roughly 3 years. It was possible to re-interview 30 control subjects and 32 depressed ones. A clinical measure of outcome was developed and baseline prognostic variables were entered into multi-variate equations. Mortality was significantly higher in the depressed groups ( 7 of 48 original trace subjects compared to no controls). Six patients developed dementia, one control and five depressed subjects. This was not, however, significant. Predictors of a poor clinical course included a higher ESR, greater Hachinski index score, increased Body Mass Index (BMI) and a lower High Density Lipoprotein (HDL). This too was a late onset group and taken together the data do suggest further evidence for cerebrovascular disease as both an aetiological and prognostic factor, in keeping with current thinking.

Prognosis: don’t give up on the older patient
Whether or not depression in later life is associated with vascular disorder, is the prognosis better, the same or worse than at younger ages? This was addressed by Mitchell and Subramanian (American Journal of Psychiatry, 2005; 162: 1588- 1601). Searching the literature between 1966 and July 2004, they managed to find 24 publications which satisfied their criteria. Many of the studies gave contradictory results, but overall the authors conclude that the general prognosis of late life depression is no different from that at other times of life but with some provisos. These are that age itself is not the main determinant but age of onset and medical comorbidity. Recurrent depression from earlier life confers a worse prognosis but so too does late onset depression accompanied by medical comorbidity, although the mechanisms are probably different. Furthermore, the authors concluded that whilst remission rates might not be that different, relapse rates appear to be higher among older patients, supporting the view that continuation therapy may need to be longer in older than younger patients.

…and pay attention to medical co-morbidity

What is the influence of comorbid medical conditions on the adequacy of depression treatment? In a study from United States, Harman and colleagues (Journal of the American Geriatric Society, 2005; 53: 2178-2183) analyzed data from the Medical Expenditure Panel which collects information at an individual level on health care utilisation. The study was limited to those with self-reported depression (n=498). One hypothesis was that\18 IPA Bulletin • June 2006 chronic medical diseases might increase primary care contacts so that depression would be recognised. An alternative hypothesis is that such conditions obscure the recognition of depression. Adequate depression management meant either an appropriate antidepressant or psychological intervention. Over a year, a third of the cohort had received depression treatment. After controlling for certain baseline variables, those with hypertension and diabetes were more likely to receive adequate depression care (odds ratios of 1.81 and 1.77 respectively, p<0.05) whereas heart disease and arthritis showed no significant difference. Therefore the study supports the first hypothesis, that increased contact with primary care may increase the chances of depression being recognised and managed, but the relationship appears to be complex because it varied depending on the type of physical disorder. The authors suggest that the perception of the individual primary care physician may be an important ingredient. Depression in dementia: serious at all levels Starkstein and colleagues (American Journal of Psychiatry, 2005; 162: 286- 293) examined the frequency of major and minor depression in Alzheimer’s disease in a United States sample of 670 patients. They used the Structured Clinical Interview for DSM (SCID) and used an inclusive diagnostic approach. The results showed that even mild levels of depression are associated with significant impairment in function but that more severe depression (using DSMIV) had more neuropsychiatric dysfunction and extra-pyramidal signs. On the whole the findings fit a dimensional model of depression in Alzheimer’s disease and the authors also found that NIMH criteria, which do not require symptoms to be present for most days most of the time, tended to result in low specificity for depression, particularly in the later stages of dementia. Longitudinal studies will be helpful to disentangle cause and effect.

…with important links to functional decline

On cue then is a study of the natural history of depression in Alzheimer’s disease by Holtzer and colleagues (Journal of American Geriatric Society, 2005; 53: 2083-2089). They studied 536 patients with Alzheimer’s disease for up to 14 years in the United States and with smaller groups from France and Greece. Depressive symptoms were assessed using the Columbia University Scale for Psychopathology in Alzheimer’s Disease (CUSPAD). The percentage of patients depressed at baseline from first to fifth year remained stable during the first three years averaging 42%. This then tailed off to 28% and 24% for the fourth and fifth years respectively. Using time-dependent analysis to examine risk factors for the first episode of depression during the follow up period, the authors think their data points to lower functional activity rather than cognitive status as the main factor. Clearly the reduced function is likely to be linked to emerging poor cognition, but if the former precedes the latter it offers some scope for intervention. Longitudinal studies like this have considerable advantages over cross- sectional ones and might help explain some of the variation in prevalence rates of depression in AD.

What to do when treatment falters

Treatment resistant depression is common in older patients with major depression. A common clinical strategy is to combine anti-depressants but is this effective? Dodd and colleague from Australia (Journal Of Affective Disorders, 2005: 85: 1-11) systematically studied the literature to January 2005. They were only able to find eight randomised controlled trials, showing variable results. Sixteen ‘open-label’ trials were of varying quality and with differing anti-depressant combinations. Numerous case series and chart reviews were found. Among these were several reporting adverse events from anti-depressant combinations, including one case of Serotonin Syndrome. It seems that clinical practice has developed ahead of evidence. A number of treatment algorithms were also examined. Steps included a higher dose of a monotherapy drug, class switching of anti-depressants, lithium augmentation, ECT and, of course, combining anti-depressants, which is now a common and popular strategy. Nevertheless the authors are reasonably positive about combination treatment, with uncontrolled studies showing that about 50% of resistant patients treated in this way will respond. As there is very little data specifically for elderly we have to extrapolate from broader research like this.

Why we should be bothered about apathy

Apathy is a distinct clinical entity and should be distinguished from depression. It is often associated with stroke but are there specific lesions more likely to generate it? Brodaty and colleagues, also from Australia (Psychological Medicine, 2005: 35:1706-1716) studied 167 patients admitted to hospital stroke units assessed at three to six months. They used a validated measure of apathy and had 109 controlled subjects. All received the Hamilton Rating Depression Scale and the 15 Item Geriatric Depression Scale. Apathy was five times more common in stroke patients than controls, with a prevalence of about one in five cases of stroke. It was associated with older age, worsened functional level and poorer global cognitions. It was not, however, correlated with stroke severity or stroke volume. There was a trend towards rightsided lesions and imaging hyperintensities. The study also supported the independence of depression and apathy as distinct although overlapping concepts. Of 35 patients with apathy, 17% were depressed: conversely of 14 depressed patients 43% had high apathy scores.
For those interested in an overview of executive dysfunction in late life mood and anxiety disorders, there is a comprehensive review by Mohlman (Journal of Geriatric Psychiatry and Neurology, 2005;18:97-108).

Innovative technologies

Diffusion Tensor Imaging (DTI) can give an index of microstructural integrity of white matter. This is of interest in the vascular depression hypothesis. Nobuhara and colleagues from Japan (Journal of Neurology, Neurosurgery and Psychiatry, 2006; 77: 120-122) studied 13 patients with a mean age of 63 who met DSM-IV criteria for major depression. They were compared with 13 age matched controls. Using DTI, white matter and anisotropy was reduced and suggested possible loss of integrity within frontal and temporal white matter fibre tracks. Since diffusion and anisotropy assesses tissue water mobility, a number of explanations are possible for this. As there was some evidence to suggest an inverse relation between white matter DTI values in the inferior frontal brain region and severity of depression, the authors speculate that possibly the neural pathway to the caudate and other limbic regions are damaged. The data were not corrected for multiple comparisons and numbers are small but they do add to the intriguing picture of damage to affective regulation circuits within the brain in late life depression.

…and lastly:

Themed issues can be quite useful. Supplement 1 of Volume 25, 2005, of the Journal of Clinical Psychopharmacology is devoted to late life depression including these topics: the efficacy of antidepressants in treatment; trajectories of treatment response, pharmokinetic imperatives; the role of placebo control groups, the optimum length of antidepressant trials in older people and improving treatment trials with innovative statistical designs. It is all informative.

 

Reprinted from IPA Bulletin, Volume 23, Number 2

Previous Recent Advances