Hostname: page-component-8448b6f56d-mp689 Total loading time: 0 Render date: 2024-04-24T19:24:29.469Z Has data issue: false hasContentIssue false

Managing depression in physical illness

Published online by Cambridge University Press:  02 January 2018

Rights & Permissions [Opens in a new window]

Extract

Depressive disorders are more common in patients with physical illness than in those without, with up to one-third of medical in-patients reporting mild to moderate symptoms of depression (Rodin & Voshart, 1986). Some medical conditions have a stronger association with psychiatric illness than others, for example the prevalence rates of depressive illness in patients with diabetes, cardiac or neurological disease is about 25%, but not much more than the general population in those with hypertension. Medical in-patients are more likely to have depression than are out-patients. There are a number of potential factors that may contribute to this increased risk of depression in people with physical illness, as outlined in Box 1.

Type
Research Article
Copyright
Copyright © The Royal College of Psychiatrists 2002 

Depressive disorders are more common in patients with physical illness than in those without, with up to one-third of medical in-patients reporting mild to moderate symptoms of depression (Reference Rodin and VoshartRodin & Voshart, 1986). Some medical conditions have a stronger association with psychiatric illness than others, for example the prevalence rates of depressive illness in patients with diabetes, cardiac or neurological disease is about 25%, but not much more than the general population in those with hypertension. Medical in-patients are more likely to have depression than are out-patients. There are a number of potential factors that may contribute to this increased risk of depression in people with physical illness, as outlined in Box 1.

Box 1 Factors contributing to the increased risk of depression in physical illness

Biological

Hormonal, nutritional, electrolyte or endocrine abnormalities

Effects of medication

Physical consequences of systemic and/or intracerebral disease

Psychological

Sense of loss associated with serious medical illness

Effects on body image, self-esteem, sense of identity

Impaired capacity to work and maintain relationships

Despite this increased prevalence, there has been widespread neglect of depression in patients with medical disease (Royal College of Physicians & Royal College of Psychiatrists, 1995), contributing to a number of poor outcomes. Physical problems are underdiagnosed in psychiatric patients. Patients with physical illness and depression have poorer levels of functioning and higher levels of mortality and morbidity compared with patients with similar illnesses but without depression (Reference Wulsin, Vaillant and WellsWulsin et al, 1999). Physicians may assume that the medical illness caused the depression, or is a contraindication to antidepressant drug treatment. Medical patients with depression use more medical health care resources than those without, both for assessment and treatment – for example, the frequency of a patient's attendance at a medical out-patient clinic is better predicted by how depressed he or she is than by his or her underlying medical condition. This contrasts with evidence of good outcomes in patients who do receive appropriate treatment, including relief of psychological distress, improved rehabilitation potential (following a stroke or myocardial infarct) and reduced health care costs.

This article outlines the assessment of depression in patients with physical illness, explores some of the diagnostic issues involved and reviews the treatment options available.

Relationship between depression and physical illness

The comorbidity of depressive and physical illnesses may be understood in a number of different, potentially overlapping, ways. First, and most commonly, the physical condition causes the depressive illness. This may be owing to a presumed direct biological mechanism, especially if the disease involves the endocrine or central nervous systems. Box 2 gives examples of some of the many drug treatments for physical illness that may have depressive side-effects. In a more comprehensive list, Bazire (2001) notes well over 100 drug treatments reported as causing depressive illness.

Box 2 Examples of drug treatments with depressive side-effects

Steroids

Beta-blockers, digoxin, calcium channel blockers

Aminophylline, theophylline

Non-steroidal anti-inflammatory drugs

Cimetidine

Metoclopramide

Levodopa, methyldopa

Isotretinoin

Interferon α

The physical condition may also trigger depression via a psychological response, especially if the condition is particularly disfiguring, disabling or potentially fatal. Overall, two-thirds of depressive illness in general medical wards is a direct result of the physical illness or its treatment. Box 3 outlines the factors that increase the likelihood that patients who are physically ill will develop a depressive illness. Of these, the single most important predictor is a previous episode of depression.

Box 3 Risk factors for the development of depression in physical illness

Past history of psychiatric disorder

Illnesses or treatments affecting the central nervous system (such as stroke)

Chronic painful disabling or disfiguring illness hindering self-care (such as rheumatoid arthritis)

Life-threatening illness (such as cancer, myocardial infarction)

Major and unpleasant treatments

Lack of social support

Second, the physical problems develop or increase owing to depressive illness. This may range from acute liver failure following paracetamol overdose in a patient with depression to the physical complications of chronic substance misuse. Depressive illness contributes to other alterations in health-related behaviours, including poor compliance with medications, diet, exercise and utilisation of health care services. Dinan (1999) has suggested that the increased risk of coronary artery disease and reduced bone mineral density in patients with depression is caused by increased activation of the hypothalamic–pituitary–adrenal axis causing hypercortisolism.

Third, the depressive and physical illnesses may have a common cause, for example bereavement or stress triggering stroke and depression.

Finally, the depressive illness may be coincidental and unrelated to physical problems, as both physical and psychiatric conditions are common in the general population. Depression pre-dates the medical illness in up to 25% of patients with comorbid depression, and it is associated with an increase in somatic complaints.

Assessment

The basic principles of biopsychosocial assessment and management of any patient with a psychiatric disorder apply, with some modification. Patients within the general medical system may be angered or distressed by the recommendation for psychiatric referral. A joint Colleges report on the psychological care of surgical patients (Royal College of Surgeons & Royal College of Psychiatrists, 1997) offers guidance on how to instigate such a referral. Seeing the patient in the setting of the general practitioner's surgery or a medical hospital is more likely to encourage attendance. Checking that an in-patient has been told of the psychiatric referral before you arrive on the ward can prevent embarrassment for you both.

It is essential to have some understanding of the patient's physical problems, treatment and prognosis. The relevant issues are better clarified by direct liaison with the medical team, rather than relying on medical documentation alone.

During the introduction, it is important to clarify that you are a psychiatrist and to explore at this point what the patient feels about the referral. This may vary from concerns that they are mad, to relief at having their emotional distress recognised, and the discussion encourages rapport. The initial focus of the assessment from this point should concentrate on the general medical history, before gradually introducing and elaborating on psychological symptoms, as tolerated by the patient.

Diagnosis of depression in physical illness

Depression is more difficult to diagnose in patients with physical illness. The biological symptoms of depression include anorexia, weight loss, sleep disturbance, lethargy and psychomotor retardation, all of which may equally be due to the physical illness. However, the consensus is that it is probably best to apply psychiatric diagnostic criteria without modification, while being aware of the risk of overdiagnosis, so that all symptoms are considered whatever their cause may be. It is particularly important that the assessment focuses on the cognitive symptoms of depression. Somatic symptoms are used to support the diagnosis of major depression if they are severe, disproportionate to the medical illness and temporally related to the affective cognitive symptoms of depression. Hawton et al (1990) reported three cardinal affective symptoms that help to differentiate depression from non-depression in medical patients – depressed mood, morning depression and hopelessness. Inappropriate guilt, feeling punished, lowered self-esteem and suicidal ideation are reported less frequently by patients with depression in a medical rather than a psychiatric setting. However, patients with depression who are also medically ill are more likely to feel anxious, pessimistic and helpless than those who have depression but are physically well.

Covert manifestations of depression include poor compliance or refusal of essential medical treatment. The number of somatic complaints is directly proportional to the likelihood of depressive illness in the medical as well as the psychiatric patient and they may present with abnormal or exaggerated illness behaviour, pain or other somatic symptoms rather than low mood.

Differential diagnosis

It is also important to consider other potential causes of low mood in patients with physical illness. These include the normal and understandable emotional reactions to physical ill health. Adjustment reactions, for example those triggered by the stress of physical illness, are the most common psychiatric diagnosis, affecting approximately one-quarter of general medical patients. Psychological symptoms of depression are common in delirium, particularly quiet or hypoactive delirium, which is often referred as depressive illness for psychiatric assessment (Reference MeagherMeagher, 2001). Differentiating between the two conditions is particularly relevant to allow early treatment and avoid potential increased confusion resulting from the anticholinergic side-effects of antidepressant therapy. Substance misuse also needs to be considered as relevant to lowered mood in a number of ways, such as in a patient whose increasing alcohol misuse has resulted in a severe exacerbation of psoriasis. In organic affective disorders such as hypothyroidism, it is important to treat the underlying physical condition appropriately and then to reassess mood before considering antidepressant medication. Finally, a trial of antidepressant therapy may be useful in diagnostic dilemmas.

Rating scales

A number of screening tools, such as the Hospital Anxiety and Depression Scale (Reference Zigmond and SnaithZigmond & Snaith, 1983), have been specifically developed for use in the general hospital setting. The Beck Depression Inventory (Reference Beck, Ward and MendelsonBeck et al, 1961) has been helpful, given its emphasis on cognitive symptoms, for example in assessing depression in patients with multiple sclerosis. However, a recent systematic review looking at the usefulness of routinely administered psychiatric questionnaires in non-psychiatric settings concluded that it was a costly exercise that did not influence clinicians' behaviour (Reference Gilbody, House and SheldonGilbody et al, 2001). Such screening tools cannot replace thorough clinical examination of individual patients.

Assessment of capacity and the Mental Health Act

Occasionally the psychiatrist will be called to assess capacity in a patient with depression who is refusing treatment. All adults are presumed to have capacity unless it is proved otherwise. Capacity is considered present if the patient can comprehend, retain and believe the information about the treatment, and if he or she weighs that information in arriving at his or her decision. It is considered to be task-specific, with greater capacity required for more important decisions, such as discontinuation of renal dialysis. In Scotland, the medical treatment section of the Adults with Incapacity (Scotland) Act awaits imminent introduction. This will be relevant for any non-emergency medical intervention. All emergency treatments will still be covered by common law.

The Mental Health Act 1983 allows the legal detention of patients who meet the necessary criteria of mental disorder, require hopsital admission and are unwilling to enter or stay in hospital. These issues often require clarification with medical and general nursing colleagues, who may erroneously believe that the Mental Health Act is required to ensure treatment of a non-compliant patient. Although it is not relevant to the treatment of physical disorder as such, the Mental Health Act may apply where the physical disorder contributes to the mental disorder, such as when using thyroxine to treat mental disorder resulting from hypothyroidism.

Treatment of depression in physical illness

The question of where a patient who has severe depression should be treated depends essentially on the severity of his or her medical condition. If his or her physical condition can only be treated in a medical hospital setting then the medical and nursing staff need to be involved in discussions regarding whereabouts in the hospital (consider transfer to a ground floor ward, for example, balanced against the risk of losing touch with the core medical team), suicide risk, levels of observation required, whether a psychiatric nurse is required and if the patient is detainable. There is an increased risk of suicide in patients who have chronic physical illness and pain. Particular caution is required with elderly patients in this category, who may deny significant suicidal ideation. When writing in the medical records it is best to highlight relevant aetiological factors, diagnosis and the management plan, for purposes of confidentiality. The full psychiatric history can then be confined to psychiatric records.

Again, the treatment approaches that apply are the same as in the case of any patient with depression, but with some modification. This will depend on the cause and severity of the patient's physical illness and the treatment setting. The biopsychosocial approach is particularly relevant here. As with any patient with depression, treatment should always involve supportive and problem-solving strategies, education, support for family members and opportunities to discuss social difficulties. For hospital in-patients, regular physiotherapy and occupational therapy are essential to help patients to cope with boredom, pain and isolation on the ward. Address any contributory factors, such as treatment of any medical cause of the depression. If it is not possible to withdraw a medication contributing to the depressed mood then antidepressant medication should be considered.

Psychological treatments

The main potential benefit of considering a purely psychotherapeutic approach to treatment is the avoidance of drug interactions or side-effects exacerbating the physical problems. However, it is often impracticable owing to expense, lack of availability of skilled therapists, the difficulty in its application in the ward environment or for a patient who has difficulty in travelling, and the fact that it is less acceptable to some patients and their physicians than medication. Guthrie & Creed (1996) note the paucity of evidence for the efficacy of psychological treatment in chronic physical disease. Psychological therapy has been particularly neglected in older patients, who often require longer periods of treatment.

Specific psychological therapies such as problem-solving, cognitive–behavioural therapy (CBT) or interpersonal therapy (IPT) may be used as an alternative or adjunct to antidepressant medication for mild depressive disorders in those with physical illness. Of these, CBT has been the most frequently applied. The intrinsic acknowledgement of physical symptoms within the theoretical framework is particularly useful, and CBT is now well incorporated by health psychologists into cardiac rehabilitation and chronic pain programmes.

There has been recent evidence of the usefulness of psychotherapy for depression in patients with a range of physical illnesses. This includes the study by Lustman et al (1998) of 10 weeks of individual CBT along with supportive diabetic education. The authors concluded that this was an effective treatment for major depression in patients with non-insulin dependent diabetes, and also noted improved hypoglycaemic control. Kunik et al (2001) reported that a brief 2-hour CBT session resulted in a significant reduction of depressive and anxiety symptoms (but no change in physical functioning) in patients with chronic obstructive pulmonary disease, compared with a group that received 2 hours of simple education. In a prospective study, CBT was also reported to have an impact on levels of depression in an unselected group of out-patients with rheumatoid arthritis (Reference Leibing, Pfingsten and BartmannLeibing et al, 1999). The beneficial effects of various psychosocial interventions on anxiety and depression in oncology patients have been well described. There have been recent developments in psychooncology, ranging from cognitive–existential group therapy to family grief therapy (Reference Bloch and KissaneBloch & Kissane, 2000).

White (2001) has recently published a guide on the application of CBT in the treatment of chronic medical problems such as cancer and diabetes and in cardiology and dermatology.

Medication

General issues

Antidepressant medication is cheap and readily available. It should be considered whenever symptoms of depression persist, despite treatment of the underlying medical condition or brief psychotherapeutic interventions. The threshold for prescribing antidepressants is generally higher in this patient group, however, because of the risk of metabolic complications and/or drug interactions. The choice of drug depends on the physical illness of the patient and medications already prescribed, influenced as in usual practice by response to previous treatments. Because of increased sensitivity to side-effects in those with medical illness, treatment may need to be introduced at a lower dose, such as paroxetine 5–10 mg, before cautiously increasing this. Dividing out the dose may also improve tolerance.

Essential prescribing information is published in the British National Fomulary (BNF; British Medical Association & Royal Pharmaceutical Society, 2002), which should always be consulted about precautions needed and potential drug interactions. Some important examples of drug interactions are given in Box 4. Other useful references are the Psychotropic Drug Directory (Reference BazireBazire, 2001) and the Maudsley Prescribing Guidelines (Reference Taylor, McConnell, Duncan-McConnell, Taylor, McConnell and Duncan-McConnellTaylor et al, 2001), which provide more detailed information on psychotropic prescribing in medical conditions. Particular care is required in patients with renal, hepatic, diabetic, neurological and cardiovascular disease, as outlined below. In general, the most important maxims in prescribing for this population are (a) use your BNF and (b) start low and go slow.

Box 4 Important drug interactions

Reduce and monitor lithium levels if taking non-steroidal anti-inflammatory drugs

Reduce propranolol dose if selective serotonin reuptake inhibitors (SSRIs) have been prescribed

Reduce doses of tricyclic antidepressants (TCAs) or moclobemide if taking cimetidine

Increase dose of TCAs by up to one-third if on a high-fibre diet

Reduce and monitor warfarin dose if fluoxetine or fluvoxamine have been prescribed

Reduce dose of donepezil if an SSRI is added

Avoid monoamine oxidase inhibitors with opioids

Dextropropoxyphene increases carbamazepine (CBZ) levels

Avoid TCAs and CBZ if bone marrow suppression occurs due to chemotherapy

SSRIs/clomipramine and selegiline (Parkinson's) may trigger acute confusion

Lithium and sumatriptan (migraine) may trigger central nervous system toxicity

TCAs have reduced efficacy and increased toxicity with the oral contraceptive pill

Tricyclic antidepressants v. selective serotonin reuptake inhibitors

We have evidence from the Cochrane Collaboration that antidepressants are both effective and acceptable for treatment of depression in patients with physical illness (Reference Gill and HatcherGill & Hatcher, 2001). The authors conclude that antidepressant treatment is significantly more likely than placebo to result in clinical improvement in patients with a wide range of physical illness. They also comment on the acceptability of such treatments to patients, with a drop-out rate of about 1 in 10. Trends were noted of tricyclic antidepressants (TCAs) being more effective than selective serotonin reuptake inhibitors (SSRIs), but also more likely to produce drop-outs. The anticholinergic side-effects of the TCAs have been a particular problem for patients with physical illness. However, SSRIs and the other newer antidepressants are not without their disadvantages. Parker (2001) has argued convincingly that SSRIs are less effective in the treatment of melancholic depression then TCAs, with the two groups appearing equally effective in the treatment of non-melancholic depression. A judicious balance of informed clinical judgement, patients' symptoms, side-effects (wanted and unwanted) and cost helps to direct the choice of antidepressant.

Given the increasing first-line use of SSRIs in this population, it is important to be aware of the differences within groups. These are largely based on their differences in half-life and potential for cytochrome P450 (CP450) inhibition. Anderson & Edwards (2001) have produced useful guidance on how to choose between the individual drugs. For example, compared with other SSRIs, fluoxetine may be more likely to cause dermatological reactions, agitation and weight loss. Its long half-life may also be a disadvantage because of more prolonged effects if an adverse reaction did occur. Fluvoxamine appears most likely to trigger gastrointestinal disturbance but is associated with the least sexual side-effects. Paroxetine causes more severe discontinuation effects and sexual side-effects as well as a higher incidence of sedation and weight gain. Citalopram and sertraline have the lowest potential for drug interactions involving hepatic metabolism and so may be preferred in patients prescribed these. SSRIs have been shown to increase the risk of gastrointestinal bleeding in the elderly or those who have a history of gastrointestinal bleeds (Reference Van Walraven, Mamdani and Wellsvan Walraven et al, 2001). As the risk of bleeding is proportionate to the degree of serotonin blockade, citalopram in particular should be avoided in this group of patients.

Elderly patients

Caution is advised for all aspects of prescribing for elderly people. For example, they are particularly sensitive to postural hypotension and anticholinergic side-effects from TCAs. SSRIs, moclobemide, trazodone and nefazodone are better options owing to lower cardiac and anticholinergic effects, although half-lives of sertraline and citalopram are prolonged in elderly patients.

Specific disorders

Cardiovascular disease

Myocardial infarction is associated with an increased incidence of depression, which in turn is associated with increased mortality post-myocardial infarction. It is therefore important to actively instigate treatment, bearing in mind that all antidepressants are best avoided for 2 months after a myocardial infarction, if possible. Usually, the first choice is an SSRI medication, avoiding fluvoxamine and citalopram if there is a risk of overdose, which has been linked to cardiotoxicity. SSRIs may trigger an initial modest slowing of the heart rate, to which the patient usually adapts. TCAs cause orthostatic hypotension and increased heart rate, which may exacerbate angina. Although antiarrhythmic in high doses, their anticholinergic and quinidine-like effects can trigger cardiac arrhythmias, and are only prescribed in patients with severe conduction defects if they have pacemaker cover. A recent case–control study has reported a dose-related increased risk of ischaemic heart disease in patients who had ever taken dothiepin but not other TCAs (Reference Hippisley-Cox, Pringle and HammersleyHippisley-Cox et al, 2001). Of the newer antidepressants, mirtazapine seems safe for patients with arrhythmias. Venlafaxine may cause hypertension, with blood pressure monitoring recommended at doses above 200 mg daily. Paroxetine, trazodone or nefazodone may be considered if sedation is required, but the patient must be monitored for hypotension. Lithium is known to cause electrocardiogram (ECG) changes, such as flat or inverted T-waves (generally benign), and widened QRS complexes. Hence it should be prescribed cautiously in patients with pre-existing conduction abnormalities. It is contraindicated in cardiac failure and sick sinus syndrome, but otherwise is not a significant risk. A baseline ECG and yearly repeats, as well as routine urea and electrolyte monitoring is essential and, of course, particular caution is required when prescribed along with diuretics.

Hepatic disease

All antidepressants are predominantly metabolised by the liver and so have increased half-lives with reduced clearance. The severity of impairment rather than the underlying aetiology is the most important factor to consider in prescribing for this group. Renal function may also be affected. As the risk of drug toxicity increases with disease severity, lower starting and total doses of medication are recommended. Of the SSRIs, paroxetine prescribed at the lower end of the dose range seems to be the safest option. Citalopram in the lower dose range and fluoxetine at half the usual dose, or given on alternate days, will have less sedative effects. Sertraline and lofepramine are to be avoided.

The sedative and constipating side-effects of the TCAs may precipitate or unmask subclinical hepatic encephalopathy. Of the group, imipramine is probably the best, commencing at 10 mg three times daily for 2 weeks, then increased by 10 mg weekly until it is effective. Venlafaxine, nefazodone, mirtazapine and reboxetine should all be started at a low dose with cautious titration. Monoamine oxidase inhibitors (MAOIs) are hepatotoxic and may precipitate coma. If essential, the starting dose should be very low, with moclobemide being the safest option. As lithium undergoes minimal hepatic metabolism, it is the mood stabiliser of choice in liver disease.

Renal disease

In this group of patients, TCAs are probably safer than SSRIs. As with hepatic disease, the degree of renal impairment rather than the cause is most important. Renal impairment may be present in the elderly without a raised creatinine level. TCA metabolites are excreted by the kidneys, hence accumulation may occur. Again, start low and go slowly; for example commence amitriptyline at 25 mg/day in divided doses and increase by 25 mg weekly as tolerated. Avoid lofepramine because much of it is excreted, unchanged, by the kidneys.

Of the SSRIs, sertraline is not recommended by its manufacturers in renal failure. Fluoxetine, citalopram and paroxetine should be started at 10 mg daily (or 20 mg alternate days in the case of fluoxetine) in patients with a glomerulofiltration rate of at least >10 ml/min. Carbamazepine and valproate are the preferred mood stabilisers in renal failure. Lithium should only be prescribed if absolutely necessary, at low doses, on alternate days, with frequent checking of serum levels.

Diabetes mellitus

Selective serotonin reuptake inhibitors may reduce serum glucose by up to 30% and cause appetite suppression, resulting in weight loss. Fluoxetine should be avoided, owing to its increased potential for hypoglycaemia, particularly in non-insulin dependent diabetes. Its side-effects of tremor, nausea, sweating and anxiety may also be misinterpreted as due to hypoglycaemia (Reference BazireBazire, 2001). If fluoxetine is prescribed, the patient should be advised of the need to monitor serum glucose levels regularly. TCAs are more likely to impair diabetic control as they increase serum glucose levels by up to 150%, increase appetite (particularly carbohydrate craving) and reduce the metabolic rate. They are generally considered safe unless the diabetes is very poorly controlled or is associated with significant cardiac or renal disease. Antidepressants such as amitriptyline, imipramine and citalopram are also used to treat painful diabetic neuropathy. Of the mood stabilisers, lithium can be used safely in patients without renal disease. Sodium valproate may give false positive urine tests (for glucose) in patients with diabetes.

Neurological disorders

In patients with epilepsy, all non-MAOIs may lower the seizure threshold. This applies to the TCAs more so than the SSRIs, which are generally the first choice. A slow rate of introduction reduces the risk of a seizure. Of the TCAs, doxepin is considered less and amitriptyline more epileptogenic. The manufacturer recommends avoidance of all MAOIs other than moclobemide while on carbamazepine. Lithium is very epileptogenic in overdose but otherwise this is not a problem at treatment doses.

Depression is common in patients with multiple sclerosis. It may be triggered by drug treatments such as interferon β, adrenocorticotrophic hormone or steroid therapy, in which case a reduction or withdrawal of the medication should be considered. Although the TCAs are effective, they are poorly tolerated owing to side-effects, including increased hypertonicity and spasms. The SSRIs are therefore the first choice for these patients.

Depression has been reported as more common in vascular than Alzheimer's dementia, unrelated to the level of cognitive impairment, and occurring at any stage of dementia. The presentation will often be atypical and should be considered for patients showing a sustained change of behaviour. Tune (1998) recommends the use of SSRIs with a potentially stimulant effect such as fluoxetine and sertraline. There is some evidence of donepezil having an antidepressant effect. As donepezil is metabolised by CP450 isoenzyme inhibitors, it should be used with caution in patients on SSRIs.

The amotivational syndrome of Parkinson's disease may be difficult to distinguish from depressive illness, and the additional symptom of anhedonia is helpful in differentiating the two. TCAs are effective and their anticholinergic effects may help extrapyramidal symptoms (EPS), but potentially increase confusion. SSRIs may exacerbate tremor and should not be prescribed for patients on selegeline (an MAO-B inhibitor), owing to serotinergic reactions and increased EPS. Clomipramine and MAOIs should similarly be avoided with selegiline. Carbamazepine is the preferred mood stabiliser as lithium and valproate may exacerbate tremor.

Antidepressant choice after stroke depends on the patient's age, epileptic potential and medications, which will often include anticoagulants or cardiac treatments. SSRIs are the drugs of choice. TCAs again are poorly tolerated owing to their anticholinergic side-effects, epileptogenicity and cardiac side-effects. Of the group, nortriptyline seems to be the best tolerated and has been used effectively to treat post-stroke depression. Trazodone and nefazodone have less cardiac and anticholinergic side-effects and may be useful alternatives. Potential interactions may occur with aspirin, which increases the free plasma concentration of highly protein-bound drugs such as fluoxetine, paroxetine, sertraline and valproate. Citalopram is the least likely of the SSRIs to interact with warfarin.

Electroconvulsive therapy

The American Psychiatric Association Task Force's report on electroconvulsive therapy (ECT) (American Psychiatric Association, 1990) cites no absolute contraindications to ECT. With continuing advances in the use of ECT it is now used relatively safely in cardiac, pregnant and elderly patients. However, it is wise to proceed particularly cautiously in patients with increased intracranial pressure, a recent intracranial bleed or myocardial infarction, cerebral or aortic aneurysms, acute respiratory tract infection and patients at risk of complications from a general anaesthetic.

Conclusion

As so many of the patients we see have a comorbid physical illness, all psychiatrists need to be proficient in the management of depression in physical illness. Recent advances in psychiatric treatments, including brief focused psychological therapies and increasingly selective medications, have now given us the opportunity to make a significant impact on the health of this previously neglected patient group.

Multiple choice questions

  1. 1. The following may be signs of depressive illness in a medical patient:

    1. a refusal of renal dialysis

    2. b abnormal illness behaviour

    3. c poor compliance with medical treatment

    4. d hypoactive delirium

    5. e multiple somatic symptoms.

  2. 2. Depressive illness is more likely in:

    1. a medical out-patients than in-patients

    2. b patients with a past psychiatric history

    3. c patients with disfiguring illnesses

    4. d patients with hypertension

    5. e patients who have multiple sclerosis.

  3. 3. The following symptoms are particularly useful in helping to diagnose depression in physical illness:

    1. a depressed mood

    2. b sleep disturbance

    3. c hopelessness

    4. d psychomotor retardation

    5. e morning depression.

  4. 4. The SSRIs are usually the drugs of first choice in the following conditions:

    1. a cardiovascular disease

    2. b hepatic disease

    3. c renal disease

    4. d neurological disorders

    5. e patients with a history of gastrointestinal bleeding.

  5. 5. The following statements regarding the treatment of depression in patients with diabetes are correct:

    1. a fluoxetine is the preferred drug treatment

    2. b TCAs may cause hypoglycaemic episodes

    3. c TCAs should be avoided even if the diabetes is well controlled

    4. d sodium valproate is the mood stabiliser of choice

    5. e citalopram may also be used to treat diabetic neuropathy.

MCQ answers

1 2 3 4 5
a T a F a T a T a F
b T b T b F b T b F
c T c T c T c F c F
d F d F d F d T d F
e T e T e T e F e T

References

American Psychiatric Association (1990) The Practice of ECT: Recommendations for Treatment, Training and Privileging. Washington, DC: American Psychiatric Press.Google Scholar
Anderson, I. M. & Edwards, J. G. (2001) Guidelines for choice of selective serotonin reuptake inhibitor in depressive illness. Advances in Psychiatric Treatment, 7, 170180.Google Scholar
Bazire, S. (2001) Psychotropic Drug Directory: The Professionals' Pocket Handbook and Aide Memoire. Lancaster: Quay Publishing Ltd.Google Scholar
Beck, A. T., Ward, C. H., Mendelson, M. et al (1961) An inventory for measuring depression. Archives of General Psychiatry, 4, 561571.Google Scholar
Bloch, S. & Kissane, D. (2000) Psychotherapies in psycho-oncology: an exciting new challenge. British Journal of Psychiatry, 177, 112116.Google Scholar
British Medical Association & Royal Pharmaceutical Society (2002) British National Formulary. London & Wallingford: BMJ Books & Pharmaceutical Press.Google Scholar
Dinan, T. G. (1999) The physical consequences of depressive illness. BMJ, 318, 826.Google Scholar
Gill, D. & Hatcher, S. (2001) Antidepressants for depression in people with physical illness. In: Cochrane Database of Systematic Review, 2. Oxford: Update Software.Google Scholar
Gilbody, S., House, A. & Sheldon, T. (2001) Routinely administered questionnaires for depression and anxiety: a systematic review. BMJ, 322, 406409.Google Scholar
Guthrie, E. & Creed, F. (1996) Treatment methods and their effectiveness. In Seminars in Liaison Psychiatry (eds Guthrie, E. & Creed, F.) pp. 238273. London: Gaskell.Google Scholar
Hawton, K., Mayou, R. & Feldman, E. (1990) Significance of psychiatric symptoms in the general medical patient with mood disorder. General Hospital Psychiatry, 12, 296302.CrossRefGoogle Scholar
Hippisley-Cox, J., Pringle, M., Hammersley, V. et al (2001) Antidepressants as a risk factor for ischaemic heart disease: case-control study in primary care. BMJ, 323, 666669.Google Scholar
Kunik, M., Braun, U., Stanley, M. et al (2001) One session cognitive behavioural therapy for elderly patients with chronic obstructive pulmonary disease. Psychological Medicine, 31, 717723.Google Scholar
Leibing, E., Pfingsten, M., Bartmann, U. et al (1999) Cognitive–behavioural treatment in unselected rheumatoid arthritis outpatients. Clinical Journal of Pain, 15, 5866.Google Scholar
Lustman, P., Griffith, L., Freedland, K. et al (1998) Cognitive behaviour therapy for depression in type 2 diabetes mellitus. A randomised, controlled trial. Annals of Internal Medicine, 129, 613621.Google Scholar
Meagher, D. (2001) Delirium: optimising management. BMJ, 322, 144149.Google Scholar
Parker, G. (2001) ‘New’ and ‘old’ antidepressants: all equal in the eyes of the lore? British Journal of Psychiatry, 179, 9596.Google Scholar
Royal College of Physicians & Royal College of Psychiatrists (1995) The Psychological Care of Medical Patients. Recognition of Need and Service Provision. London: Royal College of Physicians & Royal College of Psychiatrists.Google Scholar
Royal College of Surgeons of England & Royal College of Psychiatrists (1997) Report of the Working Party on the Psychological Care of Surgical Patients. London: Royal College of Surgeons of England and Royal College of Psychiatrists.Google Scholar
Rodin, G. & Voshart, K. (1986) Depression in the medically ill: an overview. American Journal of Psychiatry, 143, 696705.Google Scholar
Taylor, D., McConnell, H. & Duncan-McConnell, D. (2001) The Maudsley Prescribing Guidelines (6th edn) (eds Taylor, D., McConnell, H., Duncan-McConnell, D. et al). London: Martin Dunitz Ltd. Google Scholar
Tune, L. (1998) Depression and Alzheimer's disease. Depression and Anxiety, 8 (suppl. 1), 9195.Google Scholar
Van Walraven, C., Mamdani, M., Wells, P. et al (2001) Inhibition of serotonin reuptake by antidepressants and upper gastrointestinal bleeding in elderly patient: retrospective cohort study. BMJ, 323, 655658.Google Scholar
White, C. A. (2001) Cognitive Behaviour Therapy for Chronic Medical Problems: A Guide to Assessment and Treatment in Practice. West Sussex: John Wiley & Sons Ltd.Google Scholar
Wulsin, L., Vaillant, G. & Wells, V. (1999) A systematic review of the mortality of depression. Psychosomatic Medicine, 61, 617.Google Scholar
Zigmond, A. S. & Snaith, R. P. (1983) The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica, 67, 361370.Google Scholar
Submit a response

eLetters

No eLetters have been published for this article.