Introduction

Psychiatry is concerned with the study and management of disorders of mental function: primarily thoughts, perceptions, emotions and purposeful behaviours. Psychological medicine, or liaison psychiatry, is the discipline within psychiatry that is concerned with psychiatric and psychological disorders in patients who have physical complaints or conditions. This chapter will primarily concern itself with this particular branch of psychiatry.

The long-held belief that diseases are either physical or psychological has been replaced by the accumulated evidence that the brain is functionally or anatomically abnormal in most if not all psychiatric disorders. Physical, psychological and social factors, and their interactions must be looked into, in order to understand psychiatric conditions. This philosophical change of approach rejects the Cartesian dualistic approach of the mind/body biomedical model and replaces it with the more integrated biopsychosocial model.

The psychiatric history

As in any medical specialty, the history is essential in making a diagnosis. It is similar to that used in all specialties but tailored to help to make a psychiatric diagnosis, determine possible aetiology, and estimate prognosis. Data may be taken from several sources, including interviewing the patient, a friend or relative (usually with the patient’s permission), or the patient’s general practitioner. The patient interview also enables a doctor to establish a therapeutic relationship with the patient. Box 23.2 gives essential guidance on how to safely conduct such an interview, although it is unlikely that a patient will physically harm a healthcare professional. When interviewing a patient for the first time, follow the guidance outlined in Chapter 1 (see pp. 10–12).

Components of the history are summarized in Table 23.1.

Table 23.1 Summary of the components of the psychiatric history

The mental state examination (MSE)

The history will already have assessed several aspects of the MSE, but the interviewer will need to expand several areas as well as test specific areas, such as cognition. The MSE is typically followed by a physical examination and is concluded with an assessment of insight, risk and a formulation that takes into account a differential diagnosis and aetiology. Each domain of the MSE is given below; abnormalities that might be detected and the disorders in which they are found are summarized in Table 23.2. The major subheadings are listed below.

Classification of psychiatric disorders

The classification of psychiatric disorders into categories is mainly based on symptoms and behaviours, since there are currently few diagnostic tests for psychiatric disorders. There currently exists an unhelpful dualistic division of psychiatric disorders from neurological diseases, since the pathologies of at least the majority of each group of conditions are located in the brain, e.g. Alzheimer’s disease causing dementia and a pseudobulbar palsy causing emotional lability.

Psychiatric classifications have traditionally divided up disorders into neuroses and psychoses.

There are several problems with a neurotic-psychotic dichotomy. First, neuroses may be as severe in their effects as psychoses. Second, neuroses may cause symptoms that fulfil the definition of psychotic symptoms. For instance, someone with anorexia nervosa may be convinced that they are fat when they are thin, and this belief would meet all the criteria for a delusional belief. Yet we would traditionally classify the illness as a neurosis.

The ICD-10 Classification of Mental and Behavioural Disorders published by the World Health Organization has largely abandoned the traditional division between neurosis and psychosis, although the terms are still used. The disorders are now arranged in groups according to major common themes (e.g. mood disorders and delusional disorders). A classification of psychiatric disorders derived from ICD-10 is shown in Table 23.5, and this is the classification mainly used in this chapter (ICD-11 will be available in 2014).

Table 23.5 International classification of psychiatric disorders (ICD-10)

The Diagnostic and Statistical Manual of the American Psychiatric Association (DSM-IV-TR) is an alternative classification system (DSM-V in 2013).

Causes of a psychiatric disorder

A psychiatric disorder may result from several causes which may interact. It is most helpful to divide causes into the three ‘Ps’: predisposing, precipitating and perpetuating factors.

Psychiatric aspects of physical disease

People with non-psychiatric, ‘physical’ diseases are more likely to suffer from psychiatric disorders than those who are well. The most common psychiatric disorders in physically unwell patients are mood or adjustment disorders and acute organic brain disorders (delirium). The relationship between psychological and physical symptoms may be understood in one of four ways:

Table 23.6 Psychiatric conditions sometimes caused by physical diseases

Factors that increase the risk of a psychiatric disorder in someone with a physical disease are shown in Table 23.7.

Table 23.7 Factors increasing the risk of psychiatric disorders in the general hospital

Setting

Physical conditions

Treatment

The sick role and illness behaviour

The sick role describes behaviour usually adopted by ill people. Such people are not expected to fulfil their normal social obligations. They are treated with sympathy by others and are only obliged to see their doctor and take medical advice or treatments.

Illness behaviour is the way in which given symptoms may be differentially perceived, evaluated and acted (or not acted) upon by different kinds of persons. We all have illness behaviour when we choose what to do about a symptom. Going to see a doctor is generally more likely with more severe, distressing and numerous symptoms. It is also more likely in introspective individuals who focus on their health.

Abnormal illness behaviour occurs when there is a discrepancy between the objective pathology present and the patient’s response to it, in spite of adequate medical investigation and explanation.

Functional or psychosomatic disorders

So-called functional (in contrast to ‘organic’) disorders are illnesses in which there is no obvious pathology or anatomical change in an organ and there is a presumed dysfunction of an organ or system. Examples are given in Table 23.8. The psychiatric classification of these disorders would be somatoform disorders, but they do not fit easily within either medical or psychiatric classification systems, since they occupy the borderland between them. This classification also implies a dualistic ‘mind or body’ dichotomy, which is not supported by neuroscience. Since current classifications still support this outmoded understanding, this chapter will address these conditions in this way.

Table 23.8 ‘Functional’ somatic syndromes

The word psychosomatic has had several meanings, including psychogenic, ‘all in the mind’, imaginary and malingering. The modern meaning is that psychosomatic disorders are syndromes in which both physical and psychological factors are likely to be causative. So-called medically unexplained symptoms and syndromes are very common in both primary care and the general hospital (over half the outpatients in gastroenterology and neurology clinics have these syndromes). Because orthodox medicine has not been particularly effective in treating or understanding these disorders, many patients perceive their doctors as unsympathetic and seek out complementary or even alternative treatments of uncertain efficacy.

Because epidemiological studies suggest that having one of these syndromes significantly increases the risk of having another, some doctors believe that these syndromes represent different manifestations of a single ‘functional syndrome’, indicating a global somatization process. Functional disorders also have a significant association with depressive and anxiety disorders. Against this view is the evidence that the majority of primary care people with most of these disorders do not have either a mood or other functional disorder. It also seems that it requires a major stress or the development of a co-morbid psychiatric disorder in order for such sufferers to see their doctor, which might explain why doctors are so impressed with the associations with both stress and psychiatric disorders. Doctors have historically tended to diagnose ‘stress’ or ‘psychosomatic disorders’ in people with symptoms that they cannot explain. History is full of such disorders being reclassified as research clarifies the pathology. An example is writer’s cramp (p. 1122) which most neurologists now agree is a dystonia rather than a neurosis.

The likelihood is that these functional disorders will be reclassified as their causes and pathophysiology are revealed. Functional brain scans suggest enhancement of brain activity during interoception in more than one syndrome. Interoception is the perception of internal (visceral) phenomena, such as a rapid heartbeat.

Fibromyalgia (chronic widespread pain, CWP)

This controversial condition of unknown aetiology overlaps with chronic fatigue syndrome, with both conditions causing fatigue and sleep disturbance (see p. 1162). Diffuse muscle and joint pains are more constant and severe in CWP, although the ‘tender points’, previously thought to be pathognomonic, are now known to be of no diagnostic importance (see p. 509.) Different specialists have different views.

CWP occurs most commonly in women aged 40–65 years, with a prevalence in the community of between 1% and 11%. There are associations with depressive and anxiety disorders, other functional disorders, physical deconditioning and a possibly characteristic sleep disturbance (Table 23.9). Functional brain scans suggest that patients actually perceive greater pain, supporting the idea of abnormal sensory processing, and this may be in part related to abnormalities in the regulation of central opioidergic mechanisms.

Other chronic pain syndromes

A chronic pain syndrome is a condition of chronic disabling pain for which no medical cause can be found. The psychiatric classification would be a persistent somatoform pain disorder, but this is unsatisfactory since the criteria include the stipulation that emotional factors must be the main cause, and it is clinically difficult to be that certain.

The main sites of chronic pain syndromes are the head, face, neck, lower back, abdomen, genitalia or all over (CWP, fibromyalgia). ‘Functional’ low back pain is the commonest ‘physical’ reason for being off sick long term in the UK (p. 503). Quite often, a minor abnormality will be found on investigation (such as mild cervical spondylosis on the neck X-ray), but this will not be severe enough to explain the severity of the pain and resultant disability. These pains are often unremitting and respond poorly to analgesics. Sleep disturbance is almost universal and co-morbid psychiatric disorders are commonly found.

Irritable bowel syndrome

This is one of the commonest functional syndromes, affecting some 10–30% of the population worldwide. The clinical features and management of the syndrome and the related functional bowel disorders are described in more detail in chapter 6. Although the majority of sufferers with the irritable bowel syndrome (IBS) do not have a psychiatric disorder, depressive illness should be excluded in people with constipation and a poor appetite. Anxiety disorders should be excluded in people with nausea and diarrhoea. Persistent abdominal pain or a feeling of emptiness may occasionally be the presenting symptom of a severe depressive illness, particularly in the elderly, with a nihilistic delusion that the body is empty or dead inside (see p. 1168).

Multiple chemical sensitivity, Candida hypersensitivity, and food allergies

Some complementary health practitioners, doctors and patients themselves make diagnoses of multiple chemical sensitivities (MCS) (e.g. to foods, smoking, perfumes, petrol), Candida hypersensitivity and allergies (to food, tap-water and even electricity). Symptoms and syndromes attributed to these putative disorders are numerous and variable and include all the functional disorders, mood disorders and arthritis. Scientific support for the existence of these disorders is weak, particularly when double-blind methodologies have been used.

Type 1 hypersensitivities to foods such as nuts certainly exist, although they are fortunately uncommon (approximately 3/1000) (see pp. 68, 69). Direct specific food intolerances also occur (e.g. chocolate with migraine, caffeine with IBS).

Candidiasis can occur in the gastrointestinal tract in immunocompromised individuals, such as those with AIDS. Vaginal candidiasis can occur after antibiotic treatment in otherwise healthy women. A double-blind and controlled study of nystatin in women diagnosed as having candidiasis hypersensitivity syndrome showed that vaginitis was the only condition relieved more by nystatin than placebo. There is little evidence of Candida having a systemic role in other symptoms.

In spite of this evidence, the patient is often convinced of the legitimacy and usefulness of these diagnoses and their treatments.

Premenstrual syndrome

The premenstrual syndrome (PMS) consists of both physical and psychological symptoms that regularly occur during the premenstrual phase and substantially diminish or disappear soon after the period starts.

The premenstrual (late luteal) dysphoric disorder (PMDD) is a severe form of PMS with marked mood swings, irritability, depression and anxiety accompanying the physical symptoms. Women who generally suffer from mood disorders may be more prone to experience this disorder. The prevalence of PMS does not vary between cultures and is reported by the majority (75%) of women at some time in their lives. Severely disabling PMS (PMDD) occurs in about 3–8% of women.

The cause of the premenstrual syndrome remains unclear, although exacerbating factors include some of those outlined in Table 23.9. Research suggests that abnormalities of reproductive hormone receptors may play a role.

The menopause

The clinical features and management of the menopause are described on page 973. A prospective study has shown that there is no increased incidence of depressive disorders at this time. Such a significant bodily change, sometimes occurring at the same time as children leaving home, is naturally accompanied by an emotional adjustment that does not normally amount to a pathological state.

Somatoform disorders

As explained in the section on functional disorders (p. 1162), the classification of somatoform disorders is unsatisfactory because of the uncertain nature and aetiology of these disorders. However, there are certain disorders, beyond those described in ‘functional disorders’, that present frequently and coherently enough to be usefully recognized.

Dissociative/conversion disorders

A dissociative disorder is a condition in which there is a profound loss of awareness or cognitive ability without medical explanation. The term dissociative indicates the disintegration of different mental activities, and covers such phenomena as amnesia, fugues and pseudoseizures (non-epileptic attacks).

Conversion disorder occurs when an unresolved conflict is converted into usually symbolic physical symptoms as a defence against it. Such symptoms commonly include paralysis, abnormal movements, sensory loss, aphonia, disorders of gait and pseudocyesis (false pregnancy). The lifetime prevalence has been estimated at 3–6 per 1000 in women, with a lower prevalence in men. Most cases begin before the age of 35 years. Dissociation is unusual in the elderly.

Clinical features

The various symptoms are usually divided into dissociative and conversion categories (Table 23.10). Dissociative disorders have the following three characteristics that are necessary in order to make the diagnosis:

They occur in the absence of physical pathology that would fully explain the symptoms.

They are produced unconsciously.

Symptoms are not caused by overactivity of the sympathetic nervous system.

Table 23.10 Common dissociative/conversion symptoms

Dissociative (mental)	Conversion (physical)
Amnesia Fugue Pseudodementia Dissociative identity disorder Psychosis

Paralysis

Disorders of gait

Tremor

Aphonia

Mutism

Sensory symptoms

Globus hystericus

Hysterical fits

Blindness

Other characteristics include:

Symptoms and signs often reflect a patient’s ideas about illness.

There is usually abnormal illness behaviour, with exaggeration of disability.

There may have been significant childhood traumas.

Primary gain is the immediate relief from the emotional conflict.

Secondary gain refers to the social advantage gained by the patient by being ill and disabled (sympathy of family and friends, being off work, disability pension).

Physical disease is not uncommonly also present (e.g. pseudoseizures are more common in someone with epilepsy).

Dissociative amnesia commences suddenly. Patients are unable to recall long periods of their lives and may even deny any knowledge of their previous life or personal identity. In a dissociative fugue, patients not only lose their memory but wander away from their usual surroundings, and, when found, deny all memory of their whereabouts during this wandering. The differential diagnosis of a fugue state includes post-epileptic automatism, depressive illness and alcohol misuse.

Multiple personality disorder is rare, but dramatic, and may be triggered by suggestion on the part of a psychotherapist. There are rapid alterations between two or more ‘personalities’ in the same person, each of which is repressed and dissociated from the other ‘personalities’. A differential diagnosis is rapid cycling manic depressive disorder which would explain sudden apparent changes in personality.

Differential diagnosis

Dissociation is usually a stable and reliable diagnosis over time, although high rates of co-morbid mood and personality disorders are found in chronic sufferers. Particular care should be taken to make the diagnosis on positive grounds, and not simply on the basis of an absence of a medical diagnosis. Care should also be taken to exclude or treat co-morbid psychiatric disorders.

Aetiology

Functional brain scans differ between healthy controls feigning a motor abnormality and people with a similar conversion motor symptom, which suggests that dissociation involves different areas of the brain from simulation (Fig. 23.1). Functional brain scanning of a patient with conversion paralysis has shown that recalling a past trauma not only activated the emotional areas, such as the amygdala, but also reduced motor cortex activity. This would suggest that conversion involves a disinhibition of voluntary will at an unconscious level, so that the patient can no longer will something to happen.

Figure 23.1 Statistical parametric maps superimposed on an MRI scan of the anterior surface of the brain, orientated as though looking at a person head on. Red region shows hypofunction of people with conversion motor symptoms. The green region shows hypofunction of healthy controls feigning the same motor abnormality.

(From Spence SA, Crimlisk HL, Cope H et al. Discrete neurophysiological correlates in prefrontal cortex during hysterical and feigned disorders of movement. Lancet 355:1243–1244, with permission.)

The psychoanalytical theory of dissociation is that it is the result of emotionally charged memories that are repressed into the unconscious at some point in the past. Symptoms are explained as the combined effects of repression and the symbolic conversion of this emotional energy into physical symptoms. This hypothesis is difficult to test, although there is some evidence that people with dissociative disorders are more likely to have suffered childhood abuse, particularly when the abuse was both sexual and physical and started early in childhood. Caution should be taken with such a history obtained by therapies that ‘recover’ childhood memories that were previously completely unknown to the patient.

People with dissociative disorders by definition adopt both the sick role and abnormal illness behaviour, with consequent secondary gains that help to maintain the illness.

Management

The treatment of dissociation is similar to the treatment of somatoform disorders in general, outlined above and in Box 23.4. The first task is to engage the patient and their family with an explanation of the illness that makes sense to them, is acceptable, and leads to the appropriate management. An invented example of a suitable explanation is given in Box 23.5. Such an explanation would be modified by mutual discussion until an agreed understanding was achieved, which would serve as a working model for the illness. Provision of a rehabilitation programme that addresses both the physical and psychological needs and problems of the patient would then be planned.

A graded and mutually agreed plan for a return to normal function can usually be led by the appropriate therapist (e.g. speech therapist for dysphonia, physiotherapist for paralysis).

At the same time, a psychotherapeutic assessment should be made in order to determine the appropriate form of psychotherapy. For instance, couple therapy will address a significant relationship difficulty; individual psychotherapy could ease an unresolved conflict from childhood.

 Box 23.5

An example of an explanation that might be given for a dissociative disorder

You told me about the tremendous shock you felt when your mother suddenly died. This was particularly the case since you hadn’t spoken to her for so long beforehand, after that big disagreement with her over your wedding to John. You weren’t able to say good-bye before she died. Your brain was overloaded with grief, guilt and anger all at once. I wonder whether that is why you aren’t able to speak now. I wonder whether it’s difficult to think of anything to say that would make things right, particularly since you can’t speak with your mother now.

Abreaction brought about by hypnosis or by intravenous injections of small amounts of midazolam may produce a dramatic, if sometimes short-lived, recovery. In the abreactive state, the patient is encouraged to relive the stressful events that provoked the disorder and to express the accompanying emotions, i.e. to abreact. Such an approach has been useful in the treatment of acute dissociative states in wartime, but appears to be of less value in civilian life. It should only be contemplated in the presence of an anaesthetist with suitable resuscitation equipment to hand.

Hypnotherapy is psychotherapy while the patient is in a hypnotic trance, the idea being that therapy is more possible because the patient is relaxed and not using repression. This may allow the therapist access to the previously unconscious emotional conflicts or memories. There are no published trials of this technique in dissociation, which Freud gave up as unsuccessful in order to found psychoanalysis, but some hypnotherapists claim good results. Care should be taken to avoid a catastrophic emotional reaction when the patient is suddenly faced with the previously repressed memories.

Prognosis

Most cases of recent onset recover quickly with treatment, which is why a positive diagnosis should be made early. Those cases that last longer than a year are likely to persist, with entrenched abnormal illness behaviour patterns. One study found that 83% were still unwell at 12 years’ follow-up.

FURTHER READING

olde Hartman TC, Borghuis MS, Lucassen PL et al. Medically unexplained symptoms, somatisation disorder and hypochondriasis: course and prognosis. A systematic review. J Psychosom Res 2009; 66:363–377.

Stone J, Smyth R, Carson A et al. Systematic review of misdiagnosis of conversion symptoms and “hysteria”. BMJ 2005; 331:989.

Sleep difficulties

Sleep is divided into rapid eye movement (REM) and non-REM sleep:

Figure 23.2 Sleep architecture, showing cycles of slow wave sleep interspersed with rapid eye movement (REM) sleep. Sleep is ‘staged’ by the electroencephalogram (EEG). Deeper sleep (stages 3 and 4) is demonstrated by slow waves on the EEG. Sleep occurs in cycles, light sleep (accompanied by REM sleep and dreaming) to deep sleep and back again, and these cycles last about 90 min.

REM sleep is accompanied by dreaming and physiological arousal. Slow wave sleep is associated with release of anabolic hormones and cytokines, with an increased cellular mitotic rate. It helps to maintain host defences, metabolism and repair of cells. For this reason slow wave sleep is increased in those conditions where growth or conservation is required (e.g. adolescence, pregnancy, thyrotoxicosis).

Insomnia is difficulty in sleeping; a third of adults complain of insomnia and in a third of these, it can be severe.

Primary sleep disorders include sleep apnoea (p. 818), narcolepsy which responds to tigotine, the restless legs syndrome (Ekbom’s) (see p. 616) and the related periodic leg movement disorder, in which the legs (and sometimes the arms) jerk while asleep.

Delayed sleep phase syndrome occurs when the circadian pattern of sleep is delayed so that the patient sleeps from the early hours until mid-day or later, and is most common in young people. Night terrors, sleep-walking and sleep-talking are non-REM phenomena, called parasomnias, are most commonly found in children, but can recur in adults when under stress or suffering from a mood disorder. Sleep disorders secondary to another medical diagnosis will not be discussed here.

Psychophysiological insomnia commonly occurs secondary to functional, mood and substance misuse disorders, and when under stress (Box 23.6). It can often be triggered by one of these factors, but then become a habit on its own, driven by anticipation of insomnia and day-time naps.

Insomnia causes day-time sleepiness and fatigue, with consequences such as road traffic accidents. Assessment should pay particular attention to mood, life difficulties and drug intake (especially alcohol, nicotine and caffeine), and the timing of the insomnia should be ascertained:

Habitual alcohol consumption should be carefully estimated since even a small excess can be a potent cause of insomnia, as well as recent withdrawal. Caffeine is perhaps the most commonly taken drug in the UK, and its effects are easily underestimated. Six cups of real coffee a day are likely to cause insomnia in the average healthy adult. Caffeine is not only found in tea and coffee, but is also found in chocolate, cola drinks and some analgesics. Prescription drugs that can either disturb sleep or cause vivid dreams include most appetite suppressants, glucocorticoids, dopamine agonists, lipid-soluble beta-blockers (e.g. propranolol) and certain psychotropic drugs (especially when first prescribed, e.g. fluoxetine, reboxetine, risperidone).

Hypersomnia is not uncommon in adolescents with depressive illness, occurs in narcolepsy, and may temporarily follow infections such as infectious mononucleosis.

Mood (affective) disorders

Depressive disorders

Depressive disorders or ‘episodes’ are classified by the ICD-10 as mild, moderate or severe, with or without somatic symptoms. Severe depressive episodes are divided according to the presence or absence of psychotic symptoms.

Clinical features of depressive disorder

Whereas everyone will at some time or other feel ‘fed up’ or ‘down in the dumps’, it is when such symptoms become qualitatively different, pervasive or interfere with normal functioning that a depressive illness has occurred. Depressive disorder, clinical or ‘major’ depression, is characterized by disturbances of mood, speech, energy and ideas (Table 23.11). Patients often describe their symptoms in physical terms. Marked fatigue and headache are the two most common physical symptoms in depressive illness and may be the first symptoms to appear. Patients describe the world as looking grey, themselves as lacking a zest for living, being devoid of pleasure and interest in life (anhedonia). Anxiety and panic attacks are common; secondary obsessional and phobic symptoms may emerge. Symptoms should last for at least 2 weeks and should cause significant incapacity (e.g. trouble working or relating to others) in order to be dealt with as an illness.

Table 23.11 Characteristic features of depressive illness

Characteristic	Clinical features
Mood	Depressed, miserable or irritable
Talk	Impoverished, slow, monotonous
Energy	Reduced, lethargic, lacking motivation
Ideas	Feelings of futility, guilt, self-reproach, unworthiness, hypochondriacal preoccupations, worrying, suicidal thoughts, delusions of guilt, nihilism and persecution
Cognition	Impaired learning, pseudodementia in elderly patients
Physical	Insomnia (especially early waking), poor appetite and weight loss, constipation, loss of libido, erectile dysfunction, bodily pains
Behaviour	Retardation or agitation, poverty of movement and expression
Hallucinations	Auditory – often hostile, critical

In the more severe forms, diurnal variation in mood can occur, feeling worse in the morning, after waking in the early hours with apprehension. Suicidal ideas are more frequent, intrusive and prolonged. Delusions of guilt, persecution and bodily disease are not uncommon, along with second-person auditory hallucinations insulting the patient or suggesting suicide. In severe depressive illness, particularly in the elderly, concentration and memory can be so badly affected that the patient appears to have dementia (pseudodementia). Delusions of poverty and non-existence (nihilism) occur particularly in this age group. Suicide is a real risk, with the lifetime risk being approximately 5% in primary care patients, but 15% in those with depressive illness severe enough to warrant admission to hospital. People with bipolar disorder are also at greater risk of suicide. Screening questions for depressive illness are shown in Box 23.7.

 Box 23.7

Screening questions for depressive illness

During the last month, have you often been bothered by feeling down, depressed or hopeless?

During the last month, have you often been bothered by having little interest or pleasure in doing things?

If one or both of the answers is ‘yes’, assess further for depressive illness.

Epidemiology

About a third of the population will feel unhappy at any one time, but this is not the same as depressive illness; the middle-aged feel least happy compared to the young and elderly. The point prevalence of depressive illness is 5% in the community, with a further 3% having dysthymia (see below). It is more common in women, but there is no increase with age, and no difference by ethnic group or socioeconomic class (apart from a clear association with unemployment). Married and never married people have similar prevalence rates, with separated and divorced people having two to three times the prevalence. Some studies have suggested that depressive illness is becoming more common.

Depressive illnesses are more common in the presence of:

physical diseases, particularly if chronic, stigmatizing or painful

excessive and chronic alcohol use (probably the most depressing drug humans use)

social stresses, particularly loss events, such as separation, redundancy and bereavement

interpersonal difficulties with those close to the patient, especially when socially humiliated

lack of social support, with no confiding relationship.

Depressed people with another physical disorder view themselves as more sick and disabled, visit their doctors almost four times as often as the non-depressed physically ill, stay in hospital longer, adhere less to medical advice and medication, and undergo more medical and surgical procedures. Depressive illness may be associated with increased mortality (excluding suicide) in people with physical illness, such as myocardial infarct.

Dysthymia

Dysthymia is a mild or moderate depressive illness that lasts intermittently for 2 years or more and is characterized by tiredness and low mood, lack of pleasure, low self-esteem and a feeling of discouragement. The mood relapses and remits, with several weeks of feeling well, soon followed by longer periods of being unwell. It can be punctuated by depressive episodes of greater severity, so-called ‘double depression’.

Seasonal affective disorder

Seasonal affective disorder is characterized by recurrent episodes of depressive illness occurring during the winter months in the northern hemisphere. Symptoms are similar to those found with atypical depressive illness, in that patients complain of hypersomnia, increased appetite (with carbohydrate craving) and weight gain, with profound fatigue. Such patients have a higher prevalence of bipolar affective disorder, and some doctors are uncertain whether the condition is different from normal depressive illness, with the accentuation of mood that naturally occurs by season. However, there is evidence that seasonal depressive illness can be successfully treated with bright light therapy given in the early morning, which causes a phase advance in the circadian rhythm of melatonin. In contrast, the same treatment given in the early evening, with consequent phase delay of melatonin secretion, is less antidepressant. Selective serotonin reuptake inhibitors (SSRIs) are alternative treatments

FURTHER READING

Hickie IB, Ragers NL. Novel melatonin-based therapies. Lancet 2011; 378:621–631.

Puerperal affective disorders

Affective illnesses and distress are common in women soon after they have given birth.

‘Maternity blues’ describe the brief episodes of emotional lability, irritability and tearfulness that occur in about 50% of women 2–3 days postpartum and resolve spontaneously in a few days.

Postpartum psychosis occurs once in every 500–1000 births. Over 80% of cases are affective in type and the onset is usually within the first 2 weeks following delivery. In addition to the classical features of an affective psychosis, disorientation and confusion are often noted. Severely depressed patients may have delusional ideas that the child is deformed, evil or otherwise affected in some way, and such false ideas may lead to either attempts to kill the child or suicide. The response to speedy treatment is generally good. The recurrence rate for a psychosis in a subsequent puerperium is 20–30%.

Non-psychotic postnatal depressive disorders occur during the first postpartum year in 10% of mothers, especially in the first 3 months. Risk factors are first pregnancy, poor relationship with the partner, ambivalence about the pregnancy, and emotional personality traits. The Edinburgh Postnatal Depression Scale (EPDS) is a 10-item questionnaire and can be used as an effective screening tool. Depressive illness after childbirth is clinically similar to other depressive illnesses, but lack of emotional bonding with the baby is common.

Differential diagnosis

The differential diagnoses of depressive illness are shown in Table 23.12. Other psychiatric disorders are the most common misdiagnoses. Some 90% of patients presenting with a depressive illness while misusing alcohol will no longer be depressed 2 weeks after their last drink.

Table 23.12 Common differential diagnoses of depressive illness

Other psychiatric disorders

Alcohol misuse Amfetamine (and derivatives) misuse and withdrawal Borderline personality disorder Dementia Delirium Schizophrenia Normal and pathological grief

Organic (secondary) affective illness

Physical causes which are both necessary and sufficient as a cause

Cushing’s syndrome

Thyroid disease (although sometimes depression persists after treatment)

Hyperparathyroidism

Corticosteroid treatment

Brain tumour (rarely without other neurological signs)

Pathological (abnormal) and normal grief are described on page 1180. Pathological grief is closely associated with depressive illness.

Investigations

A corroborative history can be valuable in helping to exclude differential diagnoses such as alcohol misuse and elucidating maintaining factors such as a poor relationship with a partner. Physical investigations should be guided by the history and examination:

They will often include measurement of free T₄ and TSH (particularly in women), calcium, sodium, potassium, mean corpuscular volume, γ-glutamyl transpeptidase, haemoglobin, white cell count, ESR or plasma viscosity

Less commonly a chest X-ray, anti-nuclear antibody, morning and evening cortisols, electroencephalogram or a brain scan are indicated.

The aetiology of unipolar depressive disorders

The aetiology of unipolar depressive disorders is multifactorial and a mixture of genetic and environmental factors.

Genetic

Unipolar depression is probably polygenic, but no linkage has been firmly identified. The risk of unipolar depression in a first-degree relative of a patient is approximately three times the risk of the non-affected. The concordance of unipolar depression in monozygotic twins is between 30% and 60%, the concordance increasing with more recurrent illnesses. Polymorphisms that increase the risk of depression involve monoamines and their receptors, but studies are inconclusive. Serotonin transporter genes (5-HTTLPR) have provided a particular focus of interest in recent years, although a recent meta-analysis failed to demonstrate a putative gene-environment interaction with stressful life events and the subsequent development of a depressive episode. The issue is complicated by the genetic influence on sleep habits, emotional personality, and even life events, which are all involved in the genesis of depressive illness.

Neuroimaging changes

The use of functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) has revealed a number of abnormalities in the brains of people with major depression. These changes are non-specific but involve regions that are associated with both the emotional and cognitive abnormalities seen in depression. Increased brain ventricle volume, orbitofrontal, dorsolateral frontal and anterior cingulate cortex altered activation have all been implicated. The hippocampus is smaller in several stress-related neuropsychiatric disorders, including recurrent depression.

Biochemical changes

Monoamines

The monoamine deficiency theory of depressive illness is supported by the efficacy of monoamine reuptake inhibitors and the depressive effect of dietary tryptophan depletion.

Neuroimaging studies have recently revealed a raised density of monoamine oxidase A (MAO-A) receptors. It is proposed that depression is related to a chronic and ongoing depletion of these neurotransmitters as a result of this enzyme’s increased activity and its interaction with region-specific monoamine transporter densities. The relative transporter densities in particular regions and how they are affected by the global reduction in monoamine levels are then thought to determine the particular expression of the depressive illness and which symptoms predominate.

Neuroendocrine tests also suggest that the serotonin neurotransmitter system is downregulated. 5HT_1a and 5HT₂ receptor subtypes are thought most likely to be involved. Receptor-labelled functional brain scans suggest that dopamine underactivity is related to psychomotor retardation.

Hypothalamo-pituitary-adrenal axis

The administration of exogenous steroids is associated with the onset of depressive symptoms and people with Cushing’s syndrome often demonstrate depressive episodes. Acute stress, whether physical or psychological, is associated with a rise in serum glucocorticoids. Severe depressive episodes have been associated with hypercortisolaemia (of note cortisol is low in ‘atypical’ depression with hyperphagia and hypersomnia). This cortisol dysregulation has been associated with impaired glucocorticoid negative feedback, adrenal hyper-responsiveness to ACTH and hypersecretion of CRH.

Exposure to the high levels of cortisol is thought to directly affect neuronal plasticity and lower resistance to neuronal damage. The hippocampus seems especially susceptible to this, resulting in atrophic changes. This in turn has further deleterious effects on wider neuroendocrine function, resulting in a self-perpetuating dysregulation that may serve to maintain and/or worsen the illness.

The interplay at this level becomes more complicated, with reduced central and peripheral glucocorticoid receptor sensitivity, hypothalamo-pituitary-adrenal (HPA) axis upregulation and the release of pro-inflammatory cytokines that may in turn explain changes in mood, fatigue, appetite, sensitivity to pain and libido (note that depression is a potentially dangerous side-effect of interferon treatment). Additionally at the cellular level, this affects monoamine transport, causes neuronal apoptosis and dysfunction of glial cells normally responsible for maintaining neuronal homeostasis.

Brain-derived neurotrophic factor

Healthy interactions between neurones and glial cells are maintained by brain-derived neurotrophic factor (BDNF), which is found in its greatest concentration in the hippocampus and cerebral cortex. It promotes cell growth and long-term potentiation (the enhancement of synchronous firing between two neurones). Pro-BDNF, its precursor, promotes the reduction of dendritic spines and apoptosis. BDNF is then involved in the growth and activity of neural networks.

Animal studies show it is reduced under stressful conditions.

Postmortem studies show reduced concentrations in suicide compared with non-suicide deaths.

Adult humans with untreated depressive illness have three times lower concentrations when compared with both healthy controls or those that have received antidepressant treatment.

Low levels normalize with antidepressant treatment.

BDNF therefore has potential as an objective marker of depression and its response to treatment as well as being a potential target for treatment of the disorder itself.

Sleep

A reduced time between onset of sleep and REM sleep (shortened REM latency) and reduced slow wave sleep both occur in depressive illness. These abnormalities persist in some patients when they are not depressed. Families with several sufferers of depressive illness can share these traits, suggesting that sleep patterns may be inherited and predispose to depression.

Childhood traumas and personality

Physical, sexual and emotional abuse or neglect in childhood all predispose adults to depressive illness, but the effect is non-specific. Both ‘neurotic’ (emotional) and perfectionist personality traits are risks for depressive illness, and these may be determined as much by genetic factors as childhood environment.

Social factors

Some 30% of women will develop a depressive illness after a severe life event or difficulty, such as a divorce, and this is compounded by low self-esteem and a lack of a confiding relationship. Unemployment is a significant risk factor in men.

An integrated model of aetiology

Stress is more likely to trigger depressive illness in a person predisposed by lack of social support and/or certain personality traits. Stress in turn triggers various brain changes in both stress hormones (such as the release of corticotrophin-releasing hormone) and neurotransmitters (e.g. serotonin) that are both known to be altered in depressive illness. We can thus start to glimpse the model of an integrated biopsychosocial model of depressive illness. This model challenges dualistic ideas that depressive illnesses are either psychological or physical; depressive illnesses involve both the mind and the body, which are themselves indivisible.

FURTHER READING

Belmaker RH, Agam G. Major depressive disorder. N Engl J Med 2008; 358:55–68.

Frye MA. Bipolar disorder – a focus on depression. N Engl J Med 2011; 364:51–59.

Krishnan V, Nestler EJ. The molecular neurobiology of depression. Nature 2008; 455:894–902.

Moussavi S, Chatterii S, Verdes E et al. Depression, chronic diseases, and decrements in health: results from the World Health Surveys. Lancet 2007; 370:851–858.

Treatment of depressive illness

The patient needs to know the diagnosis to provide understanding and rationalization of the overwhelming distress inherent in depressive illness. Knowing that self-loathing, guilt and suicidal thoughts are caused by the illness may have an ‘antidepressant’ effect. The further treatment of depressive disorders involves physical, psychological and social interventions (Box 23.8).

 Box 23.8

Management of depressive illness

Physical

Stop depressing drugs (alcohol, steroids)

Regular exercise (good for mild to moderate depression)

Antidepressants (choice determined by side-effects, co-morbid illnesses and interactions)

Adjunctive drugs (e.g. lithium; if no response to two different antidepressants)

Electroconvulsive therapy (ECT) (if life-threatening or non-responsive)

Psychological

Education and regular follow-up by same professional

Cognitive behaviour therapy (CBT)

Other indicated psychotherapies (couple, family, interpersonal)

Social

Financial: eligible benefits, debt counselling

Employment: acquire or change job or career

Housing: adequate, secure tenancy, safe, social neighbours

Young children: child-care support

Treatments combined

The most effective treatment is a mixture of CBT and an antidepressant

Patients who are actively suicidal, severely depressed (with or without psychotic symptoms) should be admitted to hospital. Admission is necessary for perhaps 1 in 1000 people with clinical depression in primary care. This provides the patient a break from self-care, and allows support, listening, observation, the close monitoring of treatments and the prevention of suicide. Avoid the pitfall of not treating a depressive illness just because it seems an ‘understandable’ reaction to serious illness or difficult circumstances. This is particularly likely to happen if the patient is elderly, severely or even terminally ill.

Exercise

There is good evidence that regular exercise, particularly involving other people, can help relieve depressive illness of mild or moderate severity. The benefit is independent of a physical training effect.

Drugs used in the treatment of clinical depression

Today’s antidepressants are designed to provide an acute increase in monoamine activity. They do this either through preventing reuptake or enzymatic degradation. Note that this occurs acutely and that although an equally rapid depletion of monoamines has an acute mood lowering effect, the mood elevating benefits of these drugs require weeks of continuous administration. The benefits are therefore unlikely to be due to this mechanism alone.

The effects of chronic administration of monoamine reuptake inhibitors are various. Examples include an increase in the synthesis of binding proteins necessary for serotonin receptor activity and increases in cyclic AMP activation which in turn increases BDNF synthesis, enhances glucocorticoid receptor sensitivity and inhibits cytokine signalling. These may be secondary to the acute restoration of monoamine levels but rely upon transcriptional and translational changes that alter neuronal plasticity. It is this protein synthesis-dependent process that is currently thought to be the final pathway responsible for the clinical effect of the drugs.

As the neurobiology for depressive illness becomes clearer, so too are novel approaches to its treatment; some of the novel targets under active investigation are listed in Box 23.9.

 Box 23.9

Potential targets for novel antidepressant agents

Brain-derived neurotrophic factor (BDNF) Tumour necrosis factor-alpha (TNF-α) Interleukin-1 beta (IL-1β) Glucocorticoid receptors Corticotrophin-releasing hormone Melanin concentrating hormone

Alpha-melanocyte stimulating hormone Ghrelin Leptin Orexins Neuropeptide Y Nesfatin-1

General approach to drug treatment of depression

Recreational drugs such as alcohol should be stopped. Prescribed medicines suspected of exacerbating depression, such as corticosteroids, should be gradually stopped or reduced to a safe minimum.

Treatment with antidepressants is more successful when accompanied by sufficient patient education and regular follow-up, particularly a week after starting treatment and throughout the following 6 weeks. Dysthymia responds less well to antidepressants than does a depressive episode.

The commonest two pharmacological types of antidepressants are selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs). All antidepressants have similar efficacy and speed of onset. Choice depends on their side-effects, which can be used to positive effect (sedating drugs given at night to enhance sleep), and their safety. A course of antidepressants should be given until 6 months after recovery from a first episode to prevent relapse. Stopping antidepressants immediately upon recovery leads to a 50% relapse rate within 6 months. The two greatest problems with these drugs are persuading the patient to take them and adherence, since 80% of the UK public wrongly believe that they are addictive.

Drug choices in specific circumstances

Recurrent episodes: maintenance treatment with the antidepressant at the dose that obtained remission should be continued for at least 2 years. Maintenance treatment beyond this point should be re-evaluated, taking into account age, co-morbidities and risk factors.

Refractory depressive illness: whilst 50% may show response, as few as 30% of individuals (outpatients) get a complete remission with the first choice of antidepressant agent. Strategies available at this point are switching drug classes or augmentation with other agents. This should be overseen by a specialist.

Psychotic depression needs either a combination of an antidepressant and an antipsychotic drug or electroconvulsive therapy.

Bipolar depressive illness: monotherapy with quetiapine has been proposed as the drug of choice. Other drugs include mood stabilizers or olanzapine, either alone or in combination with an antidepressant (see p. 1176).

Selective serotonin reuptake inhibitors (SSRIs)

SSRIs selectively inhibit the reuptake of the monoamine serotonin (5HT) within the synapse, and are thus termed ‘selective serotonin reuptake inhibitors’ or SSRIs. Citalopram and its laevo isomer escitalopram, fluvoxamine, fluoxetine, paroxetine and sertraline have the advantage of causing less serious or disabling side-effects than tricyclics. For instance, SSRIs do not usually cause significant weight gain. Because of their long half-lives they can also be given just once a day, normally in the morning after breakfast. For these reasons patients adhere more to treatment and therefore SSRIs are now first-line treatments for depressive disorders. Vilazodone is a recently introduced SSRI and serotonin 1A receptor partial agonist. It also has partial agonist effects and does not cause weight gain.

The most common side-effects of SSRIs resemble a ‘hangover’ and include nausea, vomiting, headache, diarrhoea and dry mouth. Insomnia and paradoxical agitation can occur when first starting the drugs. Adolescents, in particular, may develop suicidal thoughts with SSRIs; only fluoxetine is licensed in the UK for adolescents for this reason. Further studies suggest that this is a small risk, if present, and no study has shown a significant increased risk of suicide itself. One in five patients also has sexual side-effects, such as erectile dysfunction and loss of libido. Uncommon side-effects include restless legs syndrome (see p. 616) and hyponatraemia.

A risk of bleeding is associated with SSRIs and is thought to be due to the inhibition of serotonin uptake by platelets as part of normal aggregation in response to vascular injury. To date, much of the reported incidence relates to gastrointestinal bleeding and any patient with one or more risk factors for upper gastrointestinal bleeding, such as taking a non-steroidal anti-inflammatory, should be given gastro-protection with a proton pump inhibitor. A risk of intra- and postoperative bleeding has also been reported. Whilst antidepressants are certainly not a reason to withhold a surgical intervention, it is certainly an added risk of which the surgeon and anaesthetist should be aware.

’Serotonin syndrome’. This is a toxic hyperserotonergic state, which can be caused by the ingestion of two or more drugs that increase serotonin levels, e.g. an SSRI combined with a monoamine oxidase inhibitor (MAOI), dopaminergic drugs (e.g. selegiline) or a tricyclic antidepressant. Symptoms include agitation, confusion, tremor, diarrhoea, tachycardia and hypertension; hyperthermia is characteristic. This is a potential medical emergency and treatment needs to begin with admission to hospital under a medical team.

Discontinuity syndrome, a specific withdrawal syndrome, has also been reported with SSRIs. This is characterized by shivering, anxiety, dizziness, ‘electric shocks’, headache and nausea. Patients should be warned not to omit a dose and to gradually reduce SSRIs when stopping them. All antidepressants have the potential to cause a discontinuity syndrome if suddenly stopped.

Tricyclic antidepressants (TCAs)

These drugs potentiate the action of the monoamines, noradrenaline (norepinephrine) and serotonin, by inhibiting their reuptake into nerve terminals (Fig. 23.3). Other tricyclics in common use include nortriptyline, doxepin and clomipramine. Dosulepin, imipramine and amitriptyline are the three most commonly used in the UK, but many related compounds have been introduced, some having fewer autonomic and cardiotoxic effects (e.g. lofepramine).

Figure 23.3 The effect of drugs used in the treatment of depression on CNS serotonergic and adrenergic functioning. The majority of released serotonin and noradrenaline is rapidly removed from the synapse by reuptake into the neurone (yellow circles). There is a range of antidepressants, which vary in their abilities to inhibit the reuptake of serotonin or noradrenaline, thus enhancing the synapse concentrations of these transmitters. Stimulation of presynaptic α₂-adrenoceptors reduces monoamine release mirtazapine block these presynaptic autoreceptors, and increases the release and transmission of noradrenaline and serotonin. Other drugs act by significantly blocking postsynaptic receptors which are upregulated in depression. MAO, monoamine oxidase; NA, noradrenaline; NRI, (selective) noradrenaline reuptake inhibitor; SNRI, serotonin and noradrenaline reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TCA, ‘classic’ tricyclic antidepressant.

(From Waller DG, Renwick A, Hillier K, eds. Medical Pharmacology and Therapeutics, Edinburgh: Saunders; 2010, with permission.)

TCAs have a number of side-effects (Table 23.13). In long-term treatment or prophylaxis, weight gain is most troublesome. Because of their toxicity in overdose, it is wisest not to prescribe them to outpatients who have suicidal thoughts without monitoring or giving the drugs to a reliable family member to look after.

Table 23.13 Side-effects of tricyclic antidepressants

Antimuscarinic effects	Convulsant activity
Dry mouth Constipation Tremor Blurred vision Urinary retention	Lowered seizure threshold
	Other effects
	Weight gain Sedation Mania (rarely)
Cardiovascular
QT prolongation
Arrhythmias
Postural hypotension

Trazodone is different from other tricyclic antidepressants and acts by blockade of 5HT₂ receptors.

SNRIs and NRIs

These antidepressants blocks a number of different neurotransmitter receptors both at the synapse and elsewhere. Their different receptor profiles cause different side-effects.

SNRIs: venlafaxine is a potent blocker of both serotonin and noradrenaline (norepinephrine) reuptake (SNRI). It has negligible affinity for other neurotransmitter receptor sites and so produces less sedation and fewer antimuscarinic effects. It can be given in slow-release form with the advantage of once-daily dosage. Nausea is the commonest side-effect and patients should be monitored for hypertension. It should not be prescribed in those with either uncontrolled hypertension or in those prone to cardiac arrhythmias; if an arrhythmia is suspected, it should be ruled out by ECG before starting treatment. Duloxetine works in a similar way to venlafaxine and has been found especially helpful with chronic pain.

NSSA: mirtazapine is a 5HT₂ and 5HT₃ receptor antagonist and a potent α₂-adrenergic blocker. The consequent effect is to increase both noradrenaline (norepinephrine) and selective serotonin transmission: an NSSA. It can be given at night to aid sleep and rarely causes sexual side-effects. Mirtazapine can be sedating in low dose and can cause weight gain. An uncommon adverse effect is agranulocytosis.

Monoamine oxidase inhibitors (MAOIs)

These act by irreversibly inhibiting the intracellular enzymes monoamine oxidase A and B, leading to an increase of noradrenaline (norepinephrine), dopamine and 5HT in the brain (see Fig. 23.3). Because of their side-effects and restrictions while taking them, they are rarely used by non-psychiatrists. MAOIs also produce a dangerous hypertensive reaction with foods containing tyramine or dopamine and therefore a restricted diet is prescribed. Tyramine is present in cheese, pickled herrings, yeast extracts, certain red wines, and any food, such as game, that has undergone partial decomposition. Dopamine is present in broad beans. MAOIs interact with drugs such as pethidine and can also occasionally cause liver damage.

Reversible inhibitors of monoamine oxidase A (RIMAs)

An example is moclobemide; usual dose 300 mg daily. These drugs appear to have fewer side-effects than the MAOIs (insomnia and headache, but some sexual problems) and constitute a low risk in overdose. Patients prescribed such antidepressants should be told that they can eat a normal diet, but should be careful to avoid excessive amounts of food rich in tyramine (see above).

Antidepressant augmentation

If two trials of antidepressants have failed, adding a second concomitant drug, e.g. lithium or tri-iodothyronine, can sometimes be helpful.

Antidepressant use in general medicine

Cardiac disease. In people with cardiac disease, SSRIs, lofepramine and trazodone are preferred over more quinidine-like compounds.

Epilepsy. MAOIs and mirtazapine do not affect epileptic thresholds.

Drug interactions. SSRIs are metabolized by the cytochrome P450 system, unlike venlafaxine, mirtazapine and reboxetine; the latter therefore have fewer drug interactions.

Herbal medicine. Care should be taken not to prescribe antidepressants while a patient is taking the herbal antidepressant St John’s wort, which interacts with serotonergic drugs in particular.

Elderly. Doses of antidepressants should initially be halved in the elderly and in people with renal or hepatic failure.

Pregnancy. Antidepressants should be avoided if possible in pregnancy and breast-feeding. If other treatments are ineffective, the risks of drug therapy should be balanced against taking no treatment, since depression can affect fetal progress and future mother-child bonding. Tricyclic antidepressants are generally believed to be safe in pregnancy, with no significant increase in congenital malformations in fetuses exposed to them. However, occasionally their antimuscarinic side-effects produce jitteriness, sucking problems and hyperexcitability in the newborn. Postpartum plasma levels of babies breast-fed by treated mothers are negligible. SSRIs do not seem to be teratogenic but manufacturers advise against their use in pregnancy until more data are available. Pulmonary hypertension in the newborn is a rare complication. MAOIs should be avoided during pregnancy because of the possibility of a hypertensive reaction in the mother.

FURTHER READING

Stewart DE. Depression during pregnancy. N Engl J Med 2011; 365:1605–1611.

FURTHER READING

Payne JL. Antidepressant use in the postpartum period: practical considerations. Am J Psychiatry 2007; 164:1329–1332.

Electroconvulsive therapy (ECT)

ECT is the treatment of choice in severe life-threatening depressive illness, particularly when psychotic symptoms are present. It is sometimes essential treatment when the patient is dangerously suicidal or refusing to eat and drink and when a rapid resolution is required such as in postpartum depressive illness, when returning baby to mother as soon as it is safe so to do forms part of the treatment.

The treatment is performed under general anaesthetic and involves the passage of an electric current across two electrodes applied to the anterior temporal areas of the scalp, in order to induce an epileptic fit. It was previously believed that the extent of the generalized seizure was proportional to its efficacy. We now know that the motoric seizure is less significant than its electrophysiological evidence (spike and wave activity on an EEG), without which benefit is unlikely to be seen. Treatments are normally given twice a week for 3–6 weeks.

ECT is a controversial treatment, yet it is free of serious side-effects. Most side-effects are due to the general anaesthetic; post-ictal confusion and headache are not uncommon, transient and short-term retrograde amnesia and a temporary defect in new learning can occur during the weeks of treatment, but these are typically short-lived effects. The frequency with which defects in autobiographical memory occur during the time of treatment should be noted. These are discrete and in most instances not recognized by the patient, unless the particular memory is actively sought. It is necessary to warn patients of its possibility beforehand.

Uncommonly used treatments

Transcranial magnetic stimulation (TMS) shows moderate efficacy, but is uncommonly used. Psychosurgery is very occasionally considered in people with severe intractable depressive illness, when all other treatments have failed (see p. 1181). A third improve remarkably, while a further third improve somewhat.

Vagal nerve and deep brain stimulation may represent major advances in the management of chronic and treatment-refractory depressive disorders, but definitive trials are not available.

Psychological treatments

Cognitive behaviour therapy (CBT)

Beck developed CBT to reverse the negative cognitive triad with which patients regarded themselves, their situation and their futures. It involves the identification of the negative automatic thoughts that maintain the negative perceptions that feed depression. They commonly include: catastrophizing (e.g. making a ‘mountain out of a mole-hill’); overgeneralizing (e.g. ‘I failed an exam; therefore I am a failure as a person.’); categorical (‘black or white’) thinking (e.g. ‘My work is either perfect or abysmal.’). CBT then involves identifying the links between these thoughts, consequent behaviour, and feeling low, and then testing their logic. This is done by looking at the evidence either in the therapy sessions (e.g. Q: ‘Did you pass the other exams you took?’; A: ‘Yes; I guess I did.’) or by behavioural ‘experiments’ (e.g. showing the ‘abysmal’ work to a colleague and asking their opinion).

There is good evidence that individual CBT is as effective as antidepressant drugs for mild and moderate depressive illness. CBT is also effective in preventing a relapse of clinical depression. Individual CBT is more effective than group delivered therapy, and there is preliminary evidence that computer delivered CBT programmes are also helpful, when used to supplement therapist involvement. Mindfulness-based CBT, based on the use of meditation, can play a helpful role in prevention of recurrence.

Interpersonal psychotherapy

This psychotherapy is probably as effective as both antidepressants and CBT in mild and moderate clinical depression. The therapist focuses on a patient’s interpersonal relationships involved in, or affected by, their illness (especially relationship changes or deficiencies), using problem-solving techniques to help the patient to find solutions.

Other psychotherapies

Couple therapy is particularly effective when a patient is in a problematic relationship that may be contributing to the perpetuation of the depressive illness; both the patient and partner attend therapy.

Family therapy is effective not only in a family with problems, but also as a way of helping the family to help the patient get better. It may involve understanding one family member’s ‘depression’ as a systemic ‘solution’ for a wider problem within the family.

Social treatments

Many people with clinical depression have associated social problems (Box 23.8). Assistance with social problems can make a significant contribution to clinical recovery. Other social interventions include the provision of group support, social clubs, occupational therapy and referral to a social worker. Educational programmes, self-help groups, and informed and supportive family members can help improve outcome.

Prognosis

Depression is one of the leading causes of disease burden worldwide. People with major depressive illness are between 1.5 and 2 times more likely to die than non-depressed people in the next 16 years, and the risks are not only suicide, but also cardiovascular. Depression produces greater disability than angina, arthritis, asthma and diabetes, which makes effective treatment and prevention imperative.

The majority of patients have recovered by 6 months in primary care and 12 months in secondary care. About a quarter of patients attending hospital with depressive illnesses will have a recurrence within 1 year, and three-quarters will have a recurrence within 10 years. People with recurrent depressive illnesses should be offered prevention. This may involve CBT that concentrates on relapse prevention, other forms of psychotherapy, antidepressant medication and advice on lifestyle activities such as regular exercise. Full-dose antidepressants are the most effective prophylaxis in recurrent depressive disorders.

Mania, hypomania and bipolar disorder

Mania and hypomania almost always occur as part of a bipolar disorder. The clinical features of mania include a marked elevation of mood, characterized by euphoria, over-activity and disinhibition (Table 23.14). Hypomania is the mild form of mania. Hypomania lasts a shorter time and is less severe, with no psychotic features and less disability. Hypomania can be distinguished from normal happiness by its persistence, non-reactivity (not provoked by good news and not affected by bad news) and social disability.

Table 23.14 Clinical features of mania

The social disability of mania can be severe, with disinhibited behaviour leading to significant debts (from overspending), lost relationships (from promiscuity or irritability), social ostracism and lost employment (from reckless or disinhibited behaviour).

Some patients have a rapid cycling illness, with frequent swings from one mood state to another. A mixed affective state occurs when features of mania and depressive illness are seen in the same episode. Cyclothymia is a personality trait with spontaneous swings in mood not sufficiently severe or persistent to warrant another diagnosis.

Differential diagnosis

Acute intoxication with recreational drugs such as amfetamines, amfetamine derivatives (MDMA: Ecstasy), and cocaine can mimic mania. Up to a quarter of people with Cushing’s syndrome develop mania. Similarly corticosteroids can induce mania less commonly than depressive illness. Dopamine agonists (e.g. bromocriptine) are also known to sometimes induce mania.

Epidemiology

The lifetime prevalence of bipolar affective disorder is 1% across the world. Unlike unipolar depressive illness, it is equally common in men and women, supporting its different aetiology. There is no variation by socioeconomic class or race. The mean age of onset is 21; earlier than unipolar depression. The higher prevalence found in divorced people is probably a consequence of the condition.

Aetiology

Genetic

There is strong evidence for a genetic aetiology in this disorder. There is a 60–80% concordance rate in monozygotic twins, compared to 15% in dizygotic twins, suggesting a high rate of heritability. Adoption studies show similar rates, so this high rate is probably genetic and not due to the family environment. Linkage studies have so far proved disappointing, with several polymorphism associations being found, and a recent large study finding similar polymorphisms to those associated with schizophrenia.

Biochemical

Brain monoamines, e.g. serotonin, seem to be increased in mania. Dexamethasone tends not to suppress cortisol levels in people with mania, suggesting a similar pattern of non-suppression to that seen in severe depressive illness.

Psychological

The effect of life events is much weaker in bipolar compared with unipolar illnesses; most effect being apparent at first onset. Similarly, personality does not seem to be a major influence, in contrast to unipolar depression, although there is some evidence of a link with the creativity and divergent thinking that is an advantage in the right occupation

FURTHER READING

Cipriani A et al. Comparative efficacy and acceptability of antimanic drugs in acute mania. Lancet 2011; 378:1306–1315.

Treatment

Acute mania or hypomania

This is summarized in Table 23.15.

Acute mania is treated with an atypical antipsychotic (neuroleptic), sodium valproate or lithium. The atypical antipsychotics aripiprazole, olanzapine, quetiapine and risperidone are particularly recommended, especially with behavioural disturbance. Doses are similar to those used in schizophrenia. The behavioural excitement and overactivity are usually reduced within days, but elation, grandiosity and associated delusions often take longer to respond.

Lithium may be used in instances where compliance is likely to be good; however, the screening necessary prior to its use (see below) may prohibit its use in these circumstances as a first-line agent.

Valproic acid is also helpful in hypomania or in rapidly cycling illnesses (see below).

Table 23.15 Treatment options for the management of acute mania or hypomania

Choice of agent is determined largely by clinical judgement, contraindications and prior response
Stop antidepressant medication
If the patient is NOT on antimanic medication, then start:	If the patient is already ON antimanic medication:
Antipsychotic, e.g. aripiprazole 15 mg daily	If taking an antipsychotic: Check dose and compliance. Increase if possible or add valproate or lithium
Or Valproate 750 mg daily	If taking valproate: Check plasma levels and increase dose aiming for a serum concentration of 125 mg/L as tolerated and/or add an antipsychotic (this should be done if mania is severe)
Or Lithium 0.4 mg daily to serum lithium of 0.4–1.0 mmol/L	If taking lithium: Check plasma levels and increase the dose to gain a level of 1.0–1.2 mmol/L if necessary (Note: higher than usual reference range) or add an antipsychotic (this should be done if mania is severe)
	If taking carbamazepine: add an antipsychotic if appropriate
If response is inadequate: Antipsychotic + valproate or lithium
A short-acting benzodiazepine may be added to assist with agitation in all patients

Prevention in bipolar disorders

Since bipolar illnesses tend to be relapsing and remitting, prevention of recurrence is the major therapeutic challenge in management. A patient who has experienced more than two episodes of affective disorder within a 5-year period is likely to benefit from preventive treatments. Recommendations include lithium, olanzapine, and valproic acid (so long as the patient is not a woman at risk of pregnancy).

Lithium

Lithium (carbonate or citrate) is one of the two main agents used for prophylaxis in people with repeated episodes of bipolar illness (the other being valproic acid). It is rapidly absorbed from the gastrointestinal tract and more than 95% is excreted by the kidneys; small amounts are found in the saliva, sweat and breast milk. Renal clearance of lithium correlates with renal creatinine clearance. Lithium is a mood-stabilizing drug that prevents mania more than depression. It reduces the frequency and severity of relapses by half and significantly reduces the likelihood of suicide. Its mode of action is unknown, but lithium is known to act on the serotonergic system. Poor response to lithium is associated with a negative family history, an unstable premorbid personality, and a rapid cycling illness. Recent pharmacogenetic work suggests that certain polymorphisms may predict response.

Plasma levels. These should be monitored weekly, with blood drawn 12 h after the last dose (a ‘trough’ level) until a steady state is reached and at 3-monthly intervals thereafter. The minimum level for prophylaxis is 0.4 mmol/L, with an optimum range of between 0.6 and 0.75 mmol. Levels higher than this may afford further protection against manic episodes but the relationship with depression is less clear. For this reason, the therapeutic range is typically quoted as 0.5–1.0 mmol/L. Fluctuations in plasma levels increase the risk of relapse.

Screening prior to starting lithium and at 6-monthly intervals thereafter includes:

Thyroid function (free T₄, TSH and thyroid autoantibodies). Lithium interferes with thyroid function and can produce frank hypothyroidism. The presence of thyroid autoantibodies increases the risk.

Parathyroid function. Serum calcium and parathyroid hormone levels higher in 10% of patients.

Renal function (serum urea and creatinine, estimated glomerular filtration rate and 24-hour urinary volume). Long-term treatment with lithium causes two renal problems: nephrogenic diabetes insipidus (DI) and reduced glomerular function (see p. 566). The best screen for DI is to ask the patient about polyuria and polydipsia.

Toxicity. Patients should carry a lithium card with them at all times, be advised to avoid dehydration, and be warned of drug interactions, such as with NSAIDs and diuretics. As with all medications, it is vital to discuss side-effects and signs of toxicity (these are listed in Box. 23.10).

 Box 23.10

Lithium

Pregnancy. As a rule, lithium is not advised during pregnancy, particularly in the 1st trimester, because of an increased risk of fetal malformation (Ebstein’s anomaly). Between 25% and 30% of women with a history of bipolar disorder relapse within 2 weeks of delivery. Restarting lithium within 24 h of delivery (if the mother is prepared to forgo breast-feeding) markedly reduces the risk of relapse.

Other mood stabilizers

Valproic acid (as the semisodium salt) is recommended both in prophylaxis and treatment of manic states. Second-line treatments include carbamazepine and lamotrigine. Some patients who do not respond to lithium may respond to these anticonvulsants or a combination of both. People with rapid cycling illnesses show a better response to anticonvulsants than to lithium. For antimanic treatment, dosage in the initial stage of treatment will be 200 mg twice daily of carbamazepine, increasing to a normal dose of 800–1000 mg.

Other drugs which appear to exercise a prophylactic mood-stabilizing effect include olanzapine and risperidone.

Both carbamazepine and valproate can be teratogenic (neural tube defects) and should be avoided in pregnancy. Other side-effects of these drugs are given in Chapter 22.

FURTHER READING

Benazzi F. Focus on bipolar disorder and mixed depression. Lancet 2007; 369:935–940.

Fountoulakis KN. Treatment of bipolar disorder: a systematic review of available data and clinical perspectives. Int J Neuropsychopharmacol 2008; 11:999–1029.

McKnight RF et al. Lithium toxicity profile: a systematic review and meta-analysis. Lancet 2012; 379:721–728.

The BALANCE Investigators. Lithium plus valproate combination versus monotherapy for relapse prevention in bipolar disorders (BALANCE). Lancet 2010; 375:365–375.

Prognosis

The average duration of a manic episode is 2 months, with 95% making a full recovery in time. Recurrence is the rule in bipolar disorders, with up to 90% relapsing within 10 years.