Article Text

Download PDFPDF

Drug treatment for depression in patients inside rehabilitation wards
  1. Unidad de Investigaciones Clinicas
  2. Antiguo Hospital Concepción Beistegui
  3. Regina 7, CP 06080, México D F, México
  4. email: svale{at}

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Editor,—Shah and colleagues report on the prevalence of psychiatric disorders affecting elderly people institutionalised in a rehabilitation unit.1 They conclude that depression is common among older adults, it is a treatable condition, and that in cases where there are cognitive impairments associated with concomitant depression, the cognitive impairments are worsened by the depressive disease. Although the high prevalence of cognitive impairments and depression in these patients is important and it is necessary to call attention to it, I believe that from the psychiatric perspective, the problem has been oversimplified. Therefore, we cannot accept the elementary proposition of Shahet al about a drug trial with selective serotonin reuptake inhibitors (SSRIs) to elucidate the correct diagnosis when diagnostic doubts remain.

    Depressive states in older patient groups have a tremendous clinical heterogeneity. Beside the depressive elderly patient without complicated somatic pathology or psychiatric co-morbidity, that will respond adequately to antidepressant drugs, a significant proportion of other symptomatic depressed patients (whose cases are precisely studied by Shah and colleagues in their article) can be considered as follows.

    (1) Elderly depressed patients who may show greater cognitive deficits compared with age similar normal subjects.2 These patients also present (subcortical) dysfunction of learning and memory, comprising the so-called “depressive pseudodementia”, and may show reversible cognitive deficits after successful somatic treatment of depression.

    (2) Patients who display cognitive deficits characterised by severe prefrontal dysfunction, with perseveration, psychomotor retardation, and long P300 latency.3

    (3) Patients who present features of depression that are related to underlying vascular disease and neurological lesions, corresponding to the hypothesised “vascular depression”.4

    (4) Patients with late life onset of cognitive deficits, psychomotor retardation and limited depressive ideation, many of whom will have the apathy syndrome5 that frequently follows brain damage in caudate, putamen and thalamus, usually secondary to cerebrovascular heterogeneous diseases (and which may overlap with the previously mentioned third subgroup).

    While SSRIs may be useful to the first subgroup, they are useless in the second and third groups, while in the fourth, dopamine agonists like bromocriptine are required. Hence, it is erroneous to overgeneralise that “depressed elderly respond well to SSRIs” like Shah et al suggest. Moreover, Shah and colleagues also state that patients with dementia may become depressed, particularly if they have insight into their condition. The frequent depressive symptoms founded in these patients may be early manifestations of Alzheimer's disease,6 in which case, cholinesterase inhibitor drugs instead of antidepressants are indicated. And what about the caution needed in the SSRI prescription because of their significant drug interactions7 resulting from interference with components of the hepatic “P-450 enzyme system”? (where Shah and colleagues state that these drugs are safe in the elderly, in spite of their frequent use of multiple medications). Drug treatments for every elderly disturbance, like for any other human complaint, must always be a careful skilled decision.

    Incidentally, Shah and colleagues state that there are no biological diagnostic tests for depression. Beside the dexamethasone cortisol test, the high prevalence of brain dysfunction in the geriatric depressed and cognitive impaired patient suggests that the computer analysed, quantitative electroencephalographic record may help not only in the brain damage differential diagnosis but also in disclosing a depressive disease by showing the characteristic increased anterior alpha power and decreased coherence.8 Also functional magnetic resonance imaging can yield valuable diagnostic information, showing abnormal activation in the left medial prefrontal cortex and in the right anterior cingulate gyrus9 in depressive patients.


    The authors respond: Depression in the elderly has enormous heterogeneity,1-1however it is very common, especially in those with concomitant physical illness. Its recognition in this area is important because of the effect it has on length of hospital stay and results of rehabilitation. Psychiatric intervention has been shown to increase recovery rate, reduce duration of stay, and reduce the need for residential care after discharge and therefore reduce costs.1-2

    First line treatment of depression in the elderly with SSRIs has been repeatedly recommended in the literature, not least because they tend to be prescribed and tolerated in therapeutic dosages.1-3Any additional drug treatment in the elderly must be considered with due risk of interactions and side effects: given appropriate care, the SSRIs have been used safely in this population for many years.1-4 1-5

    It is also accepted clinical practice to give a trial of antidepressants to patients with cognitive impairment if there is doubt as to whether the degree of depression may be causing/worsening the impairment. We know of no clinical service where functional magnetic resonance imaging or quantitative electroencephalography would be carried out as an alternative.

    The dexamethasone suppression test is not accepted as a diagnostic tool in the elderly because of the high incidence of false positives.1-6

    If depressive illness does not respond to treatment, liaison psychiatry services should be involved. However we would disagree strongly with the alternative suggestions given by Dr Vale.

    Depression accompanying cerebrovascular changes can respond to antidepressants, and again a trial should be given.1-7Depression as a prodromal feature of Alzheimer's disease and multi-infarct dementia is well recognised as adiagnosis made with hindsight.

    The cholinesterase inhibitors are not licensed as antidepressants, and given the difficulty in many areas to obtain funding for their use in established Alzheimer's disease are unlikely to become a treatment option for depression, nor is there any published evidence of efficacy in this area.

    The dopamine agonist bromocriptine is again not licensed for use as an antidepressant and, other than a few anecdotal reports in the literature, is not a recognised treatment alternative. Given the relative frailty of this population and risk of side effects including precipitating psychotic features, we would not recommended its use outside specialised neuropsychiatric facilities.

    In summary, our paper reported a study carried out in a district general hospital rehabilitation service; results and conclusions are applicable to everyday clinical practice in such areas.


    1. 1-1.
    2. 1-2.
    3. 1-3.
    4. 1-4.
    5. 1-5.
    6. 1-6.
    7. 1-7.