C6.Central nervous system

Mental illness is widespread in society but remains the object of considerable stigma and a great deal of misconception. Recurrent media scares about so-called ‘maniacs’ do not help the situation. Many people suffer from anxiety or depression at some stage and the lifetime prevalence of schizophrenia is about 1 in 100. In theUK, despite the admirable motives behind the campaign to depopulate the long-stay psychiatric hospitals, the Care in the Community programme appears to have overwhelmed the community organizations that have to implement it. Increasingly, people with mental health needs are seen in primary care.

Neurological disease is assuming greater importance in an ageing population, as the preva-

lence of conditions such as Parkinson’s disease and dementia rises inexorably.

            In almost all central nervous system conditions, although psychotherapy plays a vital role, drug therapy contributes substantially to making the problems manageable and relieving an enormous amount of suffering.

Physiological principles

To understand the mechanisms, symptoms and

treatment of CNS disorders it is necessary to

review  the  functions  of  the  more  important

brain centres, their interconnections and the

transmitters that predominate in each. Only the

most simplified outline can be attempted here.

An overall view is presented in this section, but

further  specific  information  appears  in  later

sections as appropriate. More detailed accounts

may be found in the References and further

reading section (p. 454).

Brain functions can be considered in three

broad categories:

•  Input or perceptual, i.e. handling the mass of

            sensory data passed up from receptors in the

sense organs.

•  Processing that data, i.e. the cognitive func-

            tion,  which  involves  integration,  associa-

tion with stored data (memory, experience) and, especially in man, the addition of an

emotional component.

•  Output, i.e. the action decided by the cogni-

            tive function in response to input: this will

usually be either motor  (mainly voluntary

muscle) or homeostatic (mainly involuntary

muscle and glands); see Table 6.1 and Figure

6.1.

Brain centres and their disorder

The brain is conventionally considered in six

main anatomical and functional areas:

system

1. Cerebrum  -  two  hemispheres  of  cerebral

cortex,  containing  the  limbic  system  and

basal ganglia.

2. Diencephalon, containing the hypothalamus

            and thalamus.

3. Midbrain.

4. Pons.

5. Medulla oblongata.

6. Cerebellum.

Alternatively, the brain may be subdivided

into distinct regions:

•  The forebrain, which includes areas 1 and 2.

•  The  hindbrain,  which  includes  areas  4,  5

and 6.

•  The brainstem, which includes the midbrain,

            medulla and pons.

Interconnections   between   these   areas   are manifold and complex, accounting for the rich-

ness and diversity of human activity, experience and achievement.

Cerebral cortex

The  cerebral  cortex  is,  in  evolutionary  terms,

the youngest centre. It is the principal distin-

guishing feature of higher mammals. Notably

developed in man, where it contains 90% of the

total brain neurones, the cerebral cortex is the

location of abstract thought, reasoning, judge-

ment, creativity, the interpretation of sensory

input  and  also  memory.  It  functions  like  a

computer, providing an objective, logical assess-

ment of the environment as perceived via the

senses, and then producing a plan for action

depending on past experience and biological

goals. Specific areas of the cortex are dedicated to subsidiary functions, such as the speech centre and the visual, auditory and motor cortexes.

Anatomically, the cerebral cortex is subdivided into various lobes, i.e. the frontal, temporal,

parietal and occipital.

Cortical  disorders  usually  have  a  profound

effect on all CNS function. They are commonly

manifested as disorders of intellect, e.g. mental

handicap, dementia or Alzheimer’s disease, or of

movement, e.g. epilepsy. Strokes are caused by

obstruction of blood flow usually to discrete

cortical areas. The thought disorder character-

istic of schizophrenia is partly cortical, but disor-

dered  limbic  or  thalamic  influences  on  the

cortex are probably more important.

Most proven and putative neurotransmitters are  found  in  the  cortex.  Many  of  the  more recently discovered mediators, such as the endor-

phins and peptides, have yet to be definitely

linked with specific CNS functions, disorders or drug actions. They may modulate the action of the traditional transmitters.

Limbic system

This  interesting  evolutionary  development  of

the higher mammals provides mental activity

with   an   emotional   dimension.   The   limbic

system is responsible for feelings rather than

objective reasoning and is perceived consciously

as an emotional overlay, i.e. the affect or mood,

which can modify the decisions taken by the

cortex. The system may mediate rage, fear, plea-

sure and love and, by its influence on cortical

function, is responsible for beliefs as opposed to

rational thought. A materialistic interpretation

of one of the objectives of some Eastern philoso-

phies, especially meditation, would be that it

attempts to achieve control or even elimination

of limbic influences (‘the self’, ‘desire’) on the

cortex.

The contrast between limbic and cortical func-

tions is illustrated by our response to being

caught for a motoring offence. One part of us -

our limbic system - is angry, fearful or ashamed (depending on our personality): at the same

time, our cortex is calculating the effect on our insurance premium, the most effective way to appease the policeman, or perhaps even how to manage without a driving licence.

The limbic system has evolved from a struc-

ture in lower mammals concerned with olfaction

(sense of smell), and indeed it retains this func-

tion in humans. Possibly this accounts for the

emotional power that smells have on humans.

The limbic system is also involved in memory,

and we are all familiar with how strongly smells

can evoke even distant memories. The system is

structurally  complex  with  many  component

nuclei  and  important  connections  with  the

frontal and temporal lobes of the cortex, with

the reticular system and with the hypothalamus

(all  of  which  are  sometimes  considered  as

partially within the limbic system). Dopamine is

an important transmitter, as are noradrenaline

(NA, norepinephrine) and 5-hydroxytryptamine

(5-HT,  serotonin).  Gamma-aminobutyric  acid

(GABA) is an inhibitory transmitter here.

Normally in a stable personality there is a

balance between limbic and cortical influences

on   behaviour:   one   should   be   neither   too

emotional nor too unfeeling and cold. Of course,

the relative contributions in any one person,

which in part defines their personality, will be

determined  by  genetic,  nurturing,  social  and

cultural factors, providing both the diversity and

the unpredictability of human behaviour. The

affective dimension accounts for many of the

differences   between   individuals,   and   also

between man and most other animals. It is inter-

esting to speculate on the biological advantage

that the limbic system confers: possibly it is

related to the social evolution of man.

Disorders of the limbic system are likely to

cause inappropriate emotions, such as depres-

sion,  mania  or  excessive  anxiety.  Delusions

(inappropriate beliefs) may arise in the limbic

system.    The    now    discredited    prefrontal

lobotomy (leucotomy), an operation to sever the

links between the limbic system and the cortex

in severe psychiatric disorders, resulted in the

patient  becoming  emotionally  flat.  A  similar

phenomenon is sometimes seen in patients on

long-term antipsychotics.

Basal ganglia

This  group  of  interconnected  nuclei  lie  deep

within the cerebral hemispheres and are impor-

tant coordinating centres for voluntary motor

activity. If the cortex decides motor strategy and

the  cerebellum  organizes  the  main  muscular

movements, then the basal ganglia (BG) look after

the fine detail, especially of posture and tone.

system

The BG centres include the corpus striatum

(putamen  and  caudate  nucleus),  the  globus

pallidus  and the midbrain substantia nigra.

There are important two-way connections with

higher centres (especially the motor cortex), the

cerebellum  and  various  motor  nuclei  of  the

brainstem. The BG are thus vital components in

neural loops involved in the integrated control

of muscular movement. They affect muscular

activity indirectly by modulating motor cortex

output  and  also  directly  by  augmenting  or

suppressing motor neurones in the spinal cord

via the descending extrapyramidal nerve tracts

(see below). To assist them in this, the BG receive

sensory  information  ascending  from  proprio-

ceptor muscle spindles within voluntary muscle,

via the reticular system (see below).

Disorders of the BG result in tremor or inap-

propriate muscular tone, e.g. Parkinson’s disease. The three important transmitters here are acetyl-

choline, dopamine and GABA; the former two have  opposing  actions.  Over 75%  of  brain dopamine is in the BG.

Thalamus

This important relay and preliminary processing

centre for sensory data is situated in the main

afferent pathway between the sense organs and

the cortex. It may be involved, with other centres,

in conditions where there is perceptual dysfunc-

tion,  e.g.  the  hallucinations  of  schizophrenia.

Dopamine is a likely transmitter in the thalamus,

as are other catecholamines such as adrenaline

(epinephrine)  and  noradrenaline  (norepineph-

rine), and acetylcholine. The thalamus is also

involved in motor activity.

Hypothalamus

Through   its   connections   with   autonomic

centres in the medulla, the hypothalamus has

an  important  influence  on  the  output  of  the

sympathetic   and   parasympathetic   nervous

systems.  The  hypothalamus  itself  contains

centres  for  satiety,  sleep,  thermoregulation,

water  balance  and  sexual  appetite.  It  also

controls a major part of the endocrine system

through  its  connections  with  the  pituitary

gland.  Dopamine, 5-HT  and  noradrenaline

(norepinephrine) are important transmitters in

the hypothalamus. Drugs acting on dopamin-

ergic  receptors  usually  affect  hypothalamic

activity.

Brainstem

The brainstem comprises the ‘lower’, more prim-

itive  part  of  the  brain (in  effect  the  whole midbrain and hindbrain except the cerebellum). The midbrain houses visual and auditory sensory nuclei,   as   well   as   some   motor   centres. Throughout  the  brainstem  is  a  more  diffuse structure, the reticular formation.

Medulla

Many vital homeostatic centres are located here, notably   the   respiratory   and   cardiovascular centres  but  also  controls  for  the  GIT.  These centres act via the autonomic nervous system to control many essential involuntary processes. The medulla acts partly as an executive arm of the hypothalamus, which is the brain’s principal integrating centre for homeostasis.

The reticular formation

This diffuse collection of tracts and nuclei perme-

ating  the  brainstem  monitors  and  modulates

much of the brain’s input and output. Before

describing  its  two  components  it  is  necessary

briefly to consider spinal pathways in general.

Conventionally, neural pathways are classified

according to whether they carry information to

the brain or to a higher centre within it, i.e.

the ascending or afferent pathways, or signals

down from the brain to the periphery, i.e the

descending or efferent pathways. These subserve

the input and output functions identified above.

Two main pathways carry the bulk of traffic.

The afferent spinothalamic tract brings much

of the sensory information from the periphery

via the thalamus to the cortex (Figure 6.2). The

corticospinal tract (also called the pyramidal

tract from the anatomical appearance in cross-

section, where left and right tracts cross in the

medulla) carries the signals to the muscles, which

execute the decisions taken by the brain. The

corticospinal tract is routed via the cerebellum

and it receives input from the BG (Figure 6.3).

Physiological principles 369

Ascending reticular formation.   Some fibres

leave  the  spinothalamic  tract  as  it  ascends

through the medulla, and run via the reticular formation directly to other centres. For example,

the limbic system receives information directly

from the sense organs, enabling an emotional

colouring to our perceptions, e.g. the fear associ-

ated with pain. Fibres to the hypothalamus allow

it to act very quickly if necessary to maintain

homeostasis. Output from the reticular system

also bypasses the thalamus and projects directly

into the cortex: this is the so-called reticular

activating  system,  which  is  concerned  with

alertness and sleep.

This system may also have an important role

in focusing attention because it allows unneces-

sary sensory ‘noise’ to be ignored. Clearly we are

not conscious of all sensory input at all times.

Just think about the messages from the various

skin sensors in response to normal clothing: they

are entirely ignored at the conscious level most

of the time unless our attention is drawn to

them.  Conversely,  recall  how  easily  we  are

alerted, amid the noisy babble of a crowded

room,  the  moment  someone  mentions  our

name. In a similar way the reticular formation

may allow a mother to hear her baby crying in a

distant room when nobody else does.

There has been speculation on the role of the ascending   reticular   formation   in   disordered perceptions, e.g. hallucinations. Although many psychotropic drugs act here, e.g. hypnotics and antipsychotics, it is difficult to link this with

their clinical action.

Descending  reticular  formation.   The  des-

cending reticular system, a vital component of

overall  motor  function  involved  in  the  fine

control and coordination, works in parallel to

system

the corticospinal tract. The system passes signals

from various brainstem reticular nuclei via retic-

ulospinal tracts to synapses in the spinal cord

with  the  motor  neurones.  These  tracts  run

outside the main corticospinal tracts, hence the

description extrapyramidal. There are both exci-

tatory and inhibitory fibres, providing two net

effects: smoothing of movement, and control of

resting muscle tone that helps maintain posture.

This formation receives input from the BG

(themselves connected to the motor cortex), the

hypothalamus  and  cerebellum.  The  system

obtains feedback on position and tone via affer-

ents originating in the muscle spindles (proprio-

ceptors)  throughout  the  body.  Some  of  these

synapse  in  the  spinal  cord  with  the  motor

neurones,  while  others  run  directly  to  the

cerebellum.

Psychiatry and neurology

It is surprisingly difficult to make an unambiguous

distinction between these traditionally separate

specialities. Put simply, psychiatry concerns disor-

ders of thought, belief, perception and mood,

while neurology is concerned with disorders of

movement, sensation and intellect (Table 6.2).

Thus  psychiatrists  deal  with  disorders  of  the

‘mind’, e.g. depression and schizophrenia, while

neurologists are concerned with ‘brain’ disor-

ders, e.g. parkinsonism, epilepsy and migraine.

Unfortunately,  some  conditions  exhibit  both

types   of   abnormality.   For   example,   mood

changes may occur in epilepsy, and intellectual

and motor deficits may occur in schizophrenia.

Looked at another way, neurological disease is

generally felt to be caused by some organic

(anatomical) lesion, whereas psychiatric disorder

is functional - there is a problem with the way

the mind functions but no identifiable structural

defect.   But   epilepsy   often   has   no   obvious

anatomical cause, and structural defects have

been  found  in  the  brains  of  patients  with

schizophrenia.  Changes  in  neurotransmitters,

commonly found in CNS disorders, can be cited

as evidence for either model. Moreover, there is

a  school  of  thought  that  regards  psychiatric

Psychiatric disorder

Clinical aspects of psychiatric disorder

Definition

In considering psychiatry it is impossible to omit

reference to the philosophical and sociological

controversies surrounding the human mind and

its malfunction. Is the dichotomy between mind

(spirit,  soul,  etc.)  and  brain (a  collection  of

neurones and chemicals, essentially a machine

and  therefore  ultimately  predictable)  just  a

subjective artefact? Just what is meant by ‘mental

illness’? Behaviour, beliefs, personality types and

so on vary greatly from culture to culture and in

different times in history. Mental disorder might

simply be regarded as behaviour that is unusual

or unacceptable to most people in the society and

at the time in which the ‘patient’ lives.

On a more practical level, a workable defini-

tion of mental illness needs to indicate when

there is a need for intervention of some kind. It

therefore needs to answer the question, when

does a person becomes a potential patient? This

is surely when they cannot cope with everyday

life, or society cannot cope with them, because

of their mental state. This then excludes uncon-

ventional,  politically  unacceptable  or  merely

eccentric behaviour. In addition, most formal

diagnostic definitions include a frequency or

al aspects of psychiatric disorder           371

illness also as organic. Psychiatric disorders are generally treated by psychological means, e.g. psychotherapy and psychoanalysis, as well as with drugs, whereas surgery or drugs are usually appropriate for neurological disorders.

Although  the  two  remain  distinct  medical specialities, reflecting the long intellectual tradi-

tion of mind-body dualism (see next section), increasingly the theoretical approaches to under-

standing the brain and its disorders are merging both disciplines. For this book, however, the

traditional distinction will be observed.

chronicity criterion, to distinguish the condition from reversible, temporary or secondary condi-

tions; e.g. the symptoms must have been present for at least 6 months.

Classification

There have been countless attempts to classify

mental   illness   and   the   situation   changes

constantly. The most widely accepted official

classifications  currently  in  use  are  the 10th

edition  of  the  International  Classification  of

Disease (ICD-10) and the 4th edition of the

American Diagnostic and Statistical Manual of

Mental Disorders (DSM-IV). The system adopted

in this chapter broadly represents a consensus of

these  two,  with  conflation  or  simplification

where this aids understanding.

It is convenient to group psychiatric condi-

tions   into   two   broad   types,   neurosis  and

psychosis. As with most attempts to classify

biological phenomena there is much overlap,

and it must not be assumed that this classifies

individual   patients.   Only   a   minority   falls

entirely at either end of the spectrum, showing

all   the   classical   features.   Nevertheless,   this

distinction remains useful for differentiating the

various syndromes of mental illness (Table 6.3).

In general, patients with neurotic illness know

they are ill and why others consider them so,

and can see the effect it is having on themselves

and those around them, i.e. they have insight.

Although they are unable to control their symp-

toms, they retain a grasp of reality: they can

reason and be reasoned with. They do not have

delusions, i.e. fixed, false, irrational beliefs that

they hold despite evidence to the contrary, after

allowance  for  the  context  of  their  social  or

ethnic background, or hallucinations, i.e. false

perceptions that are not perceived by anyone but

them.

Thus  chronically  anxious  patients,  who  are

neurotic, might agree that it is foolish to worry so

much, but claim that they cannot help it. There

may well be something happening to them that

most people would consider distressing, but their

reaction seems excessive. They are likely to have

been an anxious, worrying type of person even

before they became ill. Such patients are often

difficult to treat; they generally respond poorly to

drugs, but perhaps better to psychotherapy.

People with schizophrenia, by contrast, are

psychotic. They usually have very poor insight

into their illness. However, it would be untrue to

suppose that they do not know they are ill. They

suffer miserably, partly from the vague feeling

that others find them or their reported experi-

ences strange. They often have beliefs or per-

ceptions  that  others  find  bizarre  or  frankly

incredible. The onset of the illness may be linked

to some life event, but this may bear no direct

relationship to the specific symptoms. Nor will

the  patient’s  prior  personality,  although  they,

or their relationships with others, may have

seemed a trifle strange, e.g. the ‘loner’ child.

Ironically, the symptoms of psychosis are often

easier to treat than those of neurosis, although it

is  unlikely  that  either  type  of  disease  can  be

fundamentally cured with current techniques.

Management

Treatment options

There are two broad categories of treatments:

•  Invasive, such as drugs and electroconvulsive

            therapy (ECT).

•  Non-invasive, including psychotherapy and

            conditioning methods.

These are compared in Table 6.4. The differ-

ences between these two approaches reflect the principal differences between two fundamental concepts of the causation of mental illness.

The functional concept assumes the problem

is with the way a person thinks or feels; it is a

matter  of  ideas  and  relationships.  It  follows

that treatment methods should employ ideas,

words,  feelings  and  relationships;  in  short, psychotherapy.

Psychoanalysis is a form of psychotherapy in

which the analyst explores the mental tensions or

conflicts (complexes)  that  may  underlie  the

psychiatric symptoms, especially anxiety. Causes

are sought in experiences during infancy or child-

hood, or in family relationships. Psychoanalysis

seeks  to  help  the  patient  understand   their

problem on the assumption that resolution of

the illness will follow. It is a lengthy, time-

consuming and not always successful process.

Simpler   psychotherapeutic   approaches   often

yield faster results, although some would argue

that this represents a less fundamental solution.

Behaviour  therapy takes a more pragmatic

approach. A patient’s illness behaviour is learned

just like any other behaviour, i.e. it is condi-

tioned. Being ill, and adopting the sick role,

brings  certain  rewards,  despite  the  suffering.

Treatment should thus be aimed at modifying

this behaviour, reducing its benefits as perceived by the patient.

The   materialistic  or   organic  concept   of

mental illness implies a distinct, and ultimately

discoverable, anatomical or biochemical lesion;

it is a matter of molecules. Thus drugs, physical

traumas   or   manipulations,   even   including,

rarely,  the  surgeon’s  knife,  are  seen  to  be

appropriate.

As usual in the great schisms in science, prob-

ably the truth will ultimately be found to lie

between the two extremes. Currently it seems that neuroses have a greater functional compo-

nent   and   respond   better   to   non-invasive methods, while psychoses are often associated with   biochemical   or   structural   defects   and respond to invasive methods.

Nevertheless, either type of treatment can be

appropriate  and  effective  for  either  group  of

illnesses, depending on clinical factors. Most

psychiatrists nowadays are eclectic. While non-

invasive   methods   may   be   theoretically   or

humanely preferable, they may frequently be

inadequate. For example, the first aim in treating severely depressed patients is to prevent them committing suicide, and psychotherapy is not rapid enough for this. Moreover, such patients may not have sufficient insight to allow this to be effective. Thus ECT has been found to be

literally life-saving in such cases.

Diagnosis and management problems

The peculiar nature of disorders of the mind, and

our   incomplete   knowledge   of   their   causes,

presents many problems in management (Table

6.5). Clearly, there is a difference between psychi-

atric diagnoses on the one hand and diagnosis of

conditions such as asthma or diabetes mellitus on

the other, because in the latter diseases there are

system

agreed  objective  diagnostic  criteria,  and  their pathology is relatively well understood. Further problems arise with pharmacotherapy, which are summarized in Table 6.6.

Anxiety

Definition

Anxiety is familiar to us all: the mouth dry, the

heart pounding, ‘butterflies in the stomach’ and

the sense of fear, panic or dread. It arises naturally

in response to an anticipated threat of some kind

and prepares us to meet that threat. The physical

symptoms are part of the sympathetic nervous

system’s ‘fright, fight or flight’ response, and the

subjective component (fright) is presumably a warning, rather like pain. Most are due to a rise in circulating adrenaline (epinephrine).

Anxiety then is a natural, healthy response of

obvious biological value. Nowadays the kind of

life-threatening dangers for which it was origi-

nally designed, like confronting wild animals,

are rarely met. Instead, it has become adapted to

modern life, helping us cope with less tangible

stresses. These may be either acute stress reac-

tions, like to public speaking, examinations or

athletic competition, or more chronic adjust-

ment   reactions,   like   to   unemployment   or

divorce. Many of us find that a certain amount

of anxiety, the feeling of being properly keyed

up, can improve our overall performance.

However, above a certain degree of stress and

its resultant anxiety, performance levels off and

then starts to decline rapidly (Figure 6.4). The

point at which this happens varies from one

individual  to  another:  it  is  part  of  our  per-

sonality.  Some  people  have  a  low  tolerance,

worrying constantly and panicking easily: even

quite a low stress level can impair their perfor-

mance. Others, perhaps more laid-back, seem to

be unmoved by normal stressors and can only

work  well  under  pressure,  e.g.  on  the  night

before an examination. We also know that some

people,  with  the  so-called  type  A  personality,

are exceptionally active and seem to thrive on

stress.

Normally the threat, actual or anticipated, that

is generating the stress is real and would be

regarded by anyone as worrying or potentially

harmful. It is important to distinguish between

the normal physiological response to this stress- anxiety  and  the  characteristics  of  an

anxiety   disorder.   Pathological   or ‘clinical’

anxiety, i.e. which is abnormal or counterpro-

ductive, can be thought of as occurring when:

•  The threat is real but the response is out of

            proportion:

-  compared with the reaction of most people

            or

-  such that it interferes with normal func-

            tioning or everyday living.

•  The threat is only imagined by the patient.

            It may take the form of a vague but oppres-

sive  feeling  of  impending  doom  that  the patient  cannot  explain  but  which  makes them anxious, worried or tense.

Thus,   a   pragmatic   definition   of   anxiety disorder would be: a prolonged or exaggerated response to a real or imagined threat, which

interferes with normal life.

The prevalence of anxiety disorder in the UK is estimated to be between 5% and 15% (depending on how it is defined), producing approximately 15% of GP consultations.

Pathophysiology

Figure 6.5 shows one way in which the mental

and physical features of anxiety may be associ-

ated with the CNS structures discussed above. The

perceived  threat,  if  real,  reaches  the  cortex

through  both  the  normal  sensory  pathways

(spinal cord-thalamus-cortex) and the ascending

reticular  system (which  runs  to  the  limbic

system).

The former path causes appropriate voluntary

muscular  action (e.g.  flight)  via  the  motor

cortex, and the required endocrine and auto-

nomic response (e.g. corticosteroid secretion and

increased heart rate and blood pressure) via the

hypothalamus and medulla. This would explain

many of the symptoms associated with stress

and anxiety. The limbic involvement accounts

for the subjective feelings of fear or panic.

This may account for the low-tolerance type of

anxiety, where the normal protective mecha-

nisms are activated at a lower threshold than

normal. But how might the anxiety caused by

an imagined threat be explained? Possibly this originates in the limbic system, perhaps as a

misinterpretation  of  sensory  input.  Impulses would then pass up to the cortex, with all the

consequences of a real threat, including what are then inappropriate autonomic responses.

This   might   explain   how   chronic   stress

becomes associated with increased heart disease

or peptic ulceration: the drive to these organ

systems   becomes   excessive,   prolonged   and

unnecessary. It also agrees with the finding that

anxiolytic drugs act on the limbic system.

Theories of anxiety

The   psychoanalytic  or   functional   view   of

anxiety disorder is that it results from internal

mental conflicts. These may arise from man’s

socialization, which demands the repression of

primitive instincts, including the reproductive

and sexual drives. This may result in inappro-

priate ideas or feelings in individuals who are

unable to repress them effectively, and these

should   be   susceptible   to   modification   by

psychotherapy.

The biological theories, on the other hand,

hold that the cause is biochemical, e.g. an imbal-

ance of amine neurotransmitters. One theory

proposes   that   potentially   anxious   impulses

arising in the limbic system are usually subject to

inhibition  via  GABA-ergic  and  monaminergic

impulses  from  the  cortex.  In  anxiety  there  is

reduced GABA-modulin activity, which allows

over-stimulation  of  the  ascending  reticular

formation  leading  to  excessive  arousal; 5-HT receptors are also involved.

The success of drugs in controlling many of

the  symptoms  of  anxiety  seems  to  support  a

biological model, but the two theories are not

necessarily  incompatible.  Functional  mental

events  would  have  to  alter  brain  activity  in

order to achieve anything, and neurochemical

changes  would  inevitably  result.  Thus,  chem-

ical  interference  would  be  expected  to  bring

about  a  symptomatic  improvement  but  not

to  affect  the  underlying  cause.  This  is  in

accord with the apparent effect of anxiolytics

in suppressing symptoms, and argues for an

eclectic psychotherapeutic-pharmacotherapeutic

approach to therapy.

Aetiology and classification

Both   genetic   and   childhood   developmental

factors  contribute  to  a  predisposition  to  the

illness, which usually starts in early adulthood.

Environmental  stressors  and  life  events  are

frequently associated with its onset and persis-

tence. In most cases the physical symptoms are

similar; however, an understanding of the aeti-

ology and also of the management is helped by

classifying it according to the nature of the

perceived threat and the severity and course of

the illness (Tables 6.7 and 6.8).

Common forms of anxiety

The two international diagnostic systems (ICD-

10 and DSM-IV) do not use identical classifica-

tions but largely agree on the major clinical

forms.

Stress reactions arise from real external situa-

tions and are generally proportionate to them.

Thus they represent a normal response to stress.

Acute stress, e.g. stage fright, is usually short-

term  and  may  be  relatively  benign,  rarely

needing treatment. A key factor is that however

unpleasant it is, there is a definite time when it

will be over. The more chronic forms are known

as  adjustment  reactions.  The  causes  of  this

would be longer-term situations such as bereave-

ment,  which  may  also  cause  depression,  or

chronic  illness  or  unemployment,  where  the

anxiety arises from fear of future inability to

cope. In some cases it might be expected to

resolve along with the external cause, but the

sufferer may require some treatment to help

manage the symptoms temporarily.

Most patients fall into the category of gener-

alized  anxiety  disorder.  There  is  a  chronic,

irrational,  exaggerated  but  all-encompassing

feeling of fear and worry. It may be unfocused

or anticipate future illness, accident or poverty.

Diagnosis depends on presence of symptoms for

more than 6 months. It seems to be related to a

person’s personality, and tends to be chronic or

recurrent.  At  its  mildest  it  takes  the  form  of

chronic worry, but in its severest forms it can be

disabling and merges into panic disorder. It may

account for a third of all psychiatric consultations

in general practice. It is frequently accompanied

by depression.

In panic disorder there are sudden onsets of florid  autonomic  symptoms,  extreme  distress and  often  a  paralysis  of  action  or  decision, usually with no discernible cause or evident

stressful trigger. Patients may feel in imminent danger of death or insanity.

Post-traumatic stress disorder is a failure to

adjust to or overcome a serious, life-threatening

event and is characterized by recurrent sudden,

panic-like symptoms associated with flashbacks.

Phobias or morbid fears are easily understood

to be expressions of underlying anxiety. Exam-

ples are agoraphobia (the fear of going out, of

crowds or of open spaces) and arachnophobia

(fear of spiders). Recall that our definition of

pathological anxiety involves the inability to

cope. Thus simply being upset by spiders does

not count: but if the sight or even just the anti-

cipation of them disrupts everyday life, then

there is a problem. However, the cultural and

social context is also important: a fear of being

eaten by a tiger might be phobic in Britain but a

reasonable precaution in Bengal. Social phobia

may be a form of agoraphobia. The sufferer is

acutely fearful of public failure and humiliation

and so avoids socializing and may cut themselves

off.

Obsessional  and  compulsive  states usually

occur together. Obsessions are inescapable trou-

bling  thoughts:  compulsions  are  irresistible,

uncontrollable  ritualistic  actions.  Like  many

neuroses they are related to the patient’s person-

ality. Obsessions are commonly seen in the rigid,

inflexible person, overly concerned with neat-

ness and order. For example, a patient obsessed

by the idea of cleanliness, and perhaps their own

impurity, may be compelled to wash their hands

20  or  30  times  a  day.  Hypochondriasis  is  a

particular form of obsessional disorder with a

morbid  fear  of  disease  and  inappropriate  and

inaccurate thoughts or ideas about the sufferer’s

own health. Such patients are constantly imag-

ining  themselves  to  be  ill,  serially  consulting

multiple   health   professionals,   and   self-

medicating, when there is almost invariably no

physical  abnormality  whatsoever.  They  may

persuade doctors to carry out multiple invasive

investigations but are never satisfied with the

inevitable   negative   results.   Hypochondriasis

should not be confused with either psychoso-

matic illness (somatization), when there are

genuine physical symptoms, or hysteria, where

there  is  little  insight  and  the  symptoms  are

dramatic (see below).

The  commonly  combined  presentation  of

mild to moderate anxiety mixed with depression

(mild or mixed affective disorder) is of partic-

ular importance to primary carers, including GPs

and community pharmacists. It presents a diag-

nostic difficulty: is the patient depressed in reac-

tion to prolonged anxiety, or perhaps anxious

over the problems caused by their depression?

The distinction in fact is not that important

because as we shall see it does not influence

management.

Insomnia, commonly associated with anxiety

and other psychiatric illnesses, is a symptom

rather than a discrete diagnosis. Nevertheless it

may be a target for therapy, using hypnotics,

which are largely the same range of drugs as used

for anxiety.

Less common forms of chronic anxiety

A number of related neurotic syndromes with characteristic psychiatric symptoms are included in the wider definition of anxiety. They repre-

sent  less  common  behavioural  patterns  with which the chronically anxious personality can present. They are summarized in Table 6.8 and a few will be briefly discussed here.

Psychosomatic illness (somatization) is not a

discrete diagnosis, but frequently accompanies

Anxiety 379

many psychiatric conditions including anxiety.

In it there is a genuine physical dysfunction

(usually autonomic) accounting for the symp-

toms, e.g. dyspepsia caused by gastric hyper-

secretion. The close relationship between the

mind and the peripheral nervous system makes

it easy to see how mental conflicts could alter

physiological function in these cases. Indeed,

psychosomatic symptoms may be the only indi-

cation that a patient is anxious: they may not

complain of any mental stress, and it is the phys-

ical symptoms that are the declared reason for

consultation.  The  clinician  has  to  be  astute

enough to discern the underlying psychiatric

disorder or risk merely treating the secondary

symptoms.

Dissociative states usually start suddenly and

dramatically, either as a neurological or neuro-

muscular deficit (e.g. blindness or paralysis), or

an   altered   state   of   consciousness (e.g.   the

amnesic  patient  found  wandering  aimlessly).

However, no physical cause can be found, e.g. no

problem with the eye, optic nerve or visual cortex

in ‘hysterical’ blindness. Once again its origin

seems  to  lie  in  mental  conflict  or  instability,

which is converted to a physical sign, hence the

alternative term conversion disorder. The older

term hysteria is now avoided to prevent confu-

sion with lay concepts such as ‘mass hysteria’ and

‘hysterical’, which are quite different.

Possible causes of secondary anxiety  must

always be bone in mind when a patient first

presents. Certain diseases (e.g. hyperthyroidism,

phaeochromocytoma)  can  cause  the  physical

symptoms of anxiety, as can certain drugs (e.g.

excessive coffee consumption, alcohol or other

drug  abuse,  adrenergic  stimulants  such  as

bronchodilators).

Clinical features

Anxiety  is  a  neurotic  disorder.  The  patient

retains insight, is not deluded, and does not

have hallucinations. They retain their grasp of

reality and their illness can usually be related

either  to  external  events  or  their  pre-illness

personality. Anxiety is anticipatory (in that a

future threat is perceived), whereas depression is

generally retrospective.

As in all psychiatric illness the disease mani-

fests itself as intangible features such as the

patient’s   demeanour,   body   language,   facial expression or phraseology: precisely measurable physical signs may be absent although there will be variable physical features. Thus great skill is needed in eliciting a psychiatric history. The

presentation   is   highly   variable   but   usually includes many of the following:

Psychiatric features:

•  Feelings of apprehension, tension, fear, panic

            or terror, being ‘on edge’.

•  Hyper-arousal: excitability, labile mood, out-

            bursts of hostility, insomnia.

•  Circling thoughts, inability to concentrate,

            easily distracted, lapses of memory.

Physical (somatic) features:

•  Cardiovascular: palpitations, bradycardia or

            tachycardia; elevated blood pressure; flushing

or pallor.

•  Respiratory: rapid shallow breathing (hyper-

            ventilation), or breathlessness (dyspnoea).

•  Gastrointestinal: diarrhoea, dyspepsia,  dys-

phagia, churning stomach.

•  Musculoskeletal:agitation,restlessness,tremor,

            muscle tension.

•  CNS: initial insomnia, i.e. difficulty in going

            to sleep.

•  Metabolic: elevated blood glucose and gluco-

            corticoids.

•  Miscellaneous:  excessive  sweating,  urge  to

            defaecate or urinate.

This is clearly a picture of sympathetic nervous

system over-activity; even the mental features

resemble   the   central   actions   of   adrenaline

(epinephrine).  However,  psychiatric  presenta-

tions  in  general  may  frequently  be  mixed,

masked or disguised. An anxious patient may

present  as  depressed (or  vice  versa),  or  the

patient  may  complain  only  of  the  physical

symptoms and deny any mental problem.

Course and prognosis

Stress-related   anxiety   has   a   generally   good

outlook as long as the external stressor can be

eliminated or the sufferer learns to cope with it.

Most other forms of primary anxiety disorder

system

tend to follow a chronic relapsing course, with the most serious cases requiring the intermittent refuge of acute psychiatric units. An unfortunate minority   of   patients   suffer   from   chronic disabling anxiety.

Management

Aims and strategy

The aims in managing anxiety are:

•  to discover any immediate cause and deal

            with that if possible;

•  to assess the severity;

•  to relieve the patient’s distress as soon as

            possible;

•  to  institute  long-term,  potentially  curative

            treatment (in chronic anxiety).

Careful  history  taking  and  a  full  physical

examination  are  important  to  eliminate  any

medical or iatrogenic causes. Family and social

history are also vital. Any objective threat to the

patient and its likely duration must be evaluated,

as well as the patient’s personality, the extent of

their  disability  and  their  perceptions  of  the

illness. It will often be necessary to address their

social, economic or domestic situation.

If prompt symptomatic relief is felt essential,

short-term pharmacotherapy may be indicated.

However,  some  psychiatrists  believe  that  this

only helps patients to avoid confronting their

problems  and  may  be  counter-productive  in

the long term. Nevertheless, drugs may enable

the  patient  to  benefit  from  psychotherapy

because   mood   is   stabilized,   concentration

improved,  and  the  patient  becomes  more

receptive.

Drugs obviously cannot alter the reality of

unemployment,  bereavement  or  a  disastrous

marriage. However, it is also vital to understand

that drug therapy does nothing to alter the under-

lying problem in chronic anxiety. Similarly, drugs

cannot  cure  phobias  or  obsessive-compulsive

disorder;   they   merely   suppress   symptoms.

However, the relief of temporary stress-related

anxiety (e.g. stage fright or examination nerves)

with appropriate drugs is simple, harmless and

usually effective.

Non-drug therapy

In most cases the problem lies within the patient,

so  any  hope  of  permanently  overcoming  the

illness usually requires psychotherapy. Chronic

anxiety may benefit from a combination of indi-

vidual  counselling  (even  a  brief  chat  with a

skilled  counsellor  can  be  very  effective)  and

group therapy. Cognitive behavioural therapy

(CBT) is now established as the method of choice

for  achieving  long-term  relief.  It  has  a  good

evidence base and is recommended by NICE. It

aims to help people understand situations that

precipitate their symptoms and to devise strate-

gies to deal with them. Hypnosis and relaxation

therapies also have a place. Behavioural  con-

ditioning  is  sometimes  helpful,  e.g.  desen-

sitization  for  phobias,  aversion  therapy  for

compulsions.  It  has  been  found  that,  as  an

adjunct,  providing  patients  with  explanatory

literature of an appropriate level can help signif-

icantly: this has been dubbed (rather grandly)

bibliotherapy. Psychoanalysis is an option less

commonly  used  in  Europe  than  in  North

America.

Admission to a psychiatric unit may in itself

sometimes be therapeutic, simply by removing

the patient from a stressful domestic situation.

However, some patients are regularly admitted

this way. They recover partially and are returned

after a few weeks or months to a life they cannot

cope with, only to be re-admitted a few months

later, and so on: the so-called revolving door

phenomenon.

Psychiatric  care  requires  a  team  approach,

utilizing   community   nurses,   social   services,

occupational therapists and clinical psycholo-

gists to improve the patient’s social, domestic or

economic circumstances - often a difficult task.

It is also expensive compared to drug therapy

and requires resources that are not always avail-

able. Thus many patients may not get optimal

care.

Drug therapy

Benzodiazepines

The initial enthusiasm for the benzodiazepines

as universal panaceas (based - with some justifi-

Anxiety 381

cation - on their great advantages over the barbi-

turates) has now cooled. Although they cause little generalized CNS depression and are remark-

ably  safe  in  overdose,  both  dependence  and withdrawal symptoms are now known to be

quite common. Their association with accidents and falls is also serious.

Action

The  wide  range  of  benzodiazepines  available

obscures the fact that there are few differences

between them except in their pharmacokinetics.

They all probably act at specific benzodiazepine

receptors by enhancing the inhibitory activity of

GABA at its receptors in many central and spinal

sites, including the reticular activating system.

Thus, they have a wide range of actions and that

makes  them  very  useful  drugs  in  the  right

circumstances.  The  anxiolytic  action  is  quite

selective for the limbic system at low doses, with

only a small, but not insignificant, effect on

cognition and coordination; hence the rather

unsatisfactory description ‘minor tranquillizers’

(the antipsychotic drugs being the ‘major tran-

quillizers’). At higher doses this selectivity is lost

and they act as hypnotics (sleep inducers).

Benzodiazepines   have   other   very   useful

actions. They are muscle-relaxant and anticon-

vulsant: IV lorazepam is the drug of choice for

status   epilepticus.   Parenteral   use   enables

premedication and light general anaesthesia for

minor procedures such as endoscopy, with the

added benefit of short-term amnesia that covers

the  period  of  these  disagreeable  experiences.

They  are  also  useful  adjuncts  in  anti-emetic

regimens to cover cancer chemotherapy and are

used  in  alcohol  withdrawal  therapy  and  for

delirium tremens.

Tolerance   develops   during   use   owing   to receptor down-regulation, although this should have little impact when used in the short courses currently recommended.

Pharmacokinetics

Benzodiazepines are lipophilic drugs that have

good  bioavailability,  are  widely  distributed

throughout the body and are avidly bound to

plasma protein. They are cleared almost exclu-

sively by the liver, the inactive glucuronide being

eliminated  renally.  The  principal  difference

between them is in their metabolism (Table 6.9). The shorter-acting agents (e.g. temazepam) are metabolized directly to an inactive form, and

have  half-lives  of 3-15 h.  The  longer-acting drugs (e.g. diazepam) have one or more active intermediates   and   thus   may   have   effective biological half-lives of up to 100 h.

One result, which is often overlooked, is that

the  longer-acting  benzodiazepines  may  take

several weeks to reach steady-state plasma levels.

If the dose is increased too frequently during this

period there will be accumulation, although the

system

cumulative effect is to some extent offset by the development of tolerance (above). This can also occur  with  normal  dosing  in  the  elderly,  in whom the half-lives given in Table 6.9 may be increased up to threefold. Similarly, the effects of the long-acting agents persist for similar periods after discontinuation.

Advantages

Comparison of the benzodiazepines with their

predecessors,  the  barbiturates,  explains  their

initial appeal to prescribers (Table 6.10). They are

selective, and largely avoid drowsiness at thera-

peutic doses. In addition, they are extremely safe

in overdose because they do not cause depres-

sion  of  the  respiratory  or  other  medullary

centres. Death from benzodiazepine over-dosage

alone is uncommon; the drugs rarely produce

more than a light coma, from which the patient

can be aroused.

Habituation,   dependence   and   withdrawal

phenomena were initially thought to be rare.

Interactions  are  few.  Benzodiazepines  neither

induce  liver  enzymes  nor  have  their  metabo-

lism significantly affected by other drugs that

induce or inhibit liver enzymes. They do not

displace  other  drugs  from  plasma  protein

binding sites.

Disadvantages

The validity of many of these favourable initial impressions has been eroded. Benzodiazepines do cause unwanted CNS depression, especially a subtle impairment of motor coordination that

is  often  unnoticed  by  the  patient.  This  is

thought to be implicated in a growing number of road traffic accidents. In the elderly particu-

larly, the longer-acting agents sometimes cause paradoxical excitement, giddiness, confusion or aggression, owing to disinhibition.

Serious interactions with other CNS depressants can occur, producing potentially fatal respiratory depression. An important example is with opioids or alcohol, or both: this cocktail is commonly used in deliberate overdoses. Respiratory depression can  also  occur  with  benzodiazepines  alone when used in otherwise compromised patients, such as the elderly, the very young, or those

with respiratory impairment.

Flumazenil injection, a specific benzodiazepine competitive antagonist, is available to reverse overdoses rapidly.

Dependence and drug-seeking behaviour and

abuse,  related  to  down-regulation  of  GABA

receptors, are being increasingly recognized as

major problems. Withdrawal symptoms seem to

be more common with the short-acting benzodi-

azepines, especially if they are stopped abruptly,

probably because of the subsequent rapid fall in

plasma level. Dependence may occur after as

little as 4-6 weeks, and up to half of patients

may  become  dependent  after 3  months  of

Anxiety 383

continuous treatment. Currently the BNF recom-

mends courses no longer than 4 weeks except in very exceptional circumstances.

Weaning patients off benzodiazepines is not

easy, if it can be done at all. One recommended

procedure is to convert the patient to diazepam

at an equipotent daily dose (see BNF), and then

reduce the dose by one-eighth of this every

fortnight. Complete withdrawal can take many

months.

Certain  patients  seem  particularly  prone  to

dependence,  notably  those  with  compliant,

passive,  dependent  or  anxious  personalities,

who tend to abuse prescription medication in

low  doses  over  long  periods.  Another  group,

characterised as impulsive, antisocial and more

psychotic  than  neurotic,  may  abuse  illicitly

obtained   benzodiazepines   recreationally,   in

larger doses but with less likelihood of depen-

dence.  Pharmacists  in  the  UK  are  becoming

actively  involved  in  monitoring  the  use  of

benzodiazepines and countering their misuse.

When  used  as  hypnotics,  benzodiazepines

produce abnormal sleep patterns and cause a

rebound sleep debt on withdrawal. There is a

hangover effect from all but the shortest-acting

agents.

Indications and use

The  longer-acting  benzodiazepines  should  be started cautiously with small dose increments. When stopping treatment, the dose should be

reduced slowly, especially with the short-acting drugs (but see above).

A shorter-acting benzodiazepine is preferred

for the elderly, for those with liver disease and

for daytime use generally, even if this means

more doses per day. Longer-acting drugs may

be beneficial if compliance is doubtful, if the

shorter-acting ones do not provide enough sleep,

or if a combination of hypnotic effect and next-

day sedation is desired. Doses in the elderly

should start at one-quarter the usual adult dose.

The choice of agent and dose should be tailored

carefully to the patient because of individual vari-

ations in clearance, response, residual effects and

concurrent medication. If these precautions are

observed, the value of benzodiazepines will be

retained without incurring the potentially serious

problems.

Other agents used in anxiety

Apart from in severe disabling acute anxiety,

benzodiazepines are avoided. There are several alternatives.

Antidepressants, particularly the selective sero-

tonin re-uptake inhibitors (SSRIs), are now the

drugs  of  choice  for  many  forms  of  chronic

anxiety, primarily because of the involvement of

serotonin in anxiety. Their specific antidepres-

sant action is also helpful in mixed anxiety and

depression, but effectiveness does not depend on

the  presence  of  a  diagnosis  of  depression.

Although  not  addictive  like  benzodiazepines,

antidepressants do have common and distressing

adverse effects (see below).

Starting  dosage  is  lower  than  in  primary

depression (pp. 399-400) because of a risk of

briefly exacerbating the anxiety, with gradual

increments to the usual maxima. Owing to the

slow   onset   of   action   of   antidepressants

(3-4 weeks), and an initial increase in anxiety

symptoms, patients with acute severe problems

may need initial cover with a short course of

benzodiazepines.   Treatment   is   assessed   at

12 weeks and if ineffective, changed to a tricyclic

antidepressant. Otherwise treatment may last up

to 6 months, followed by gradual withdrawal

(see p. 400).

Buspirone, a partial 5-HT agonist, is anxiolytic

but not hypnotic. It is probably less liable to

habituation   than   benzodiazepines.   Its   main

drawback is a slow onset of activity. It must be

titrated up to an effective dose and then a further

2-4 weeks are needed for full effect. This makes

it unsuitable for the treatment of acute anxiety.

However, it may be a suitable alternative where

benzodiazepine-like activity is required beyond

1 month.

Beta-blockers are useful for pronounced adren-

ergic physical symptoms in some patients with

stress-related  anxiety  such  as  stage  fright.  In

addition they may indirectly reduce the associ-

ated psychiatric symptoms by sparing patients

further concern about the physical symptoms.

However,   beta-blockers   are   less   useful   in

general  anxiety  or  panic  disorder.  Their  prin-

cipal advantages are that they can be used for

long periods if necessary and have no general

sedative effect.

system

Hypnotics

Alternatives  to  benzodiazepines  as  hypnotic

drugs are the chemically distinct ‘Z-drugs’ (e.g.

zopiclone, zaleplon and zolpidem), which also act

on benzodiazepine receptors. They appear to

offer little advantage, have similar adverse effects

and are equally likely to produce dependence.

Other sedative drugs sometimes used include

first-generation   sedative   antihistamines (e.g.

promethazine,  hydroxyzine),  chloral  derivatives

(e.g. dichloralphenazone) and antipsychotics in

low doses. Barbiturates are now restricted to

anaesthesia   and   continuing   medication   for

epileptic patients long stabilized on phenobar-

bital. NICE recommends the short-term use of

hypnotics, either benzodiazepines or Z-drugs,

only after non-pharmacological methods have

proved ineffective.

Drug selection

Drug  therapy  has  only  a  limited  role  in  the

management  of  anxiety.  It  should  usually  be

restricted  to  short  periods  and  have  specific,

agreed  end  points,  e.g.  to  cover  an  acutely

distressing or disabling temporary exacerbation

of the patient’s circumstances. However, some

chronic  anxiety  disorders  may  require  up  to

6 months of treatment. It must then be moni-

tored closely to ensure that the drug is withdrawn when the specific indications resolve. For most types of anxiety the evidence of effectiveness is in the following sequence:

psychotherapy → [  ? short-term benzodiazepine] → longer-term antidepressant

Psychotherapy  is  always  preferred,  but  the

rapid effectiveness of short-term benzodiazepines

in severe acute situations allows other therapies

time to start working. In particular circumstances

more specific recommendations can be made (see

below).

Stress-related anxiety

Short courses of anxiolytics can be very useful in

acute conditions where the patient definitely

needs support for a brief period before a specific

stressful  event.  There  is  insufficient  time  for

psychotherapy to produce an effect, and the dura- tion  of  drug  therapy  is  usually  too  short  for

dependence  to  develop,  so  that  the  benzodi-

azepines  could  be  used.  If  the  symptoms  are

mainly  somatic (palpitations,  gastrointestinal

upset,  tremor)  beta-blockers  are  preferable,  as

they would be in situations when benzodiazepine

adverse effects would be disadvantageous (e.g.

musicians, actors, examination candidates). For

chronic  adjustment  reactions,  antidepressants

are indicated.

Where the anxiety is secondary to an under-

lying medical condition (e.g. thyrotoxicosis) or drug  use  or  misuse,  obviously  the  aim  is  to diagnose and correct that.

Generalized anxiety disorder (GAD)

This  category  perhaps  presents  the  greatest

dangers for inappropriate drug treatment. The

temptation is strong to suppress the patient’s

unwelcome, distressing symptoms continuously

with drugs - which seem so effective - rather

than uncover the true cause of the patient’s

problems. This is particularly so if the problems

stem from an adverse social environment, such

as poverty or marital distress, when the clinician

may feel frustrated by their inability to tackle the

underlying difficulties. However, this temptation

should be resisted. Short courses of anxiolytic

drugs have a place in stabilizing the patient, but

the only chance of a permanent solution is to

address these underlying problems. The alterna-

tive is the ‘revolving door phenomenon’ and

anxiolytic dependence.

The idea of the ‘pill for every ill’ society is

based mainly on the misguided drug treatment

of this category of illness. The prevailing belief in

the West seems to be that all distress must be

avoided, all stress reactions are pathological, and

life must be perpetually serene. In his Brave New

World, Aldous Huxley prophesied and ruthlessly

satirized this idealistic philosophy. Modern-day

critics also claim that too frequent recourse to

drugs  reduces  people’s  capacity  to  confront

stressful situations and may mask wider socio-

political causes. There are Valium Anonymous

groups in the USA and TRANX in the UK to

counter this attitude and help those who are

benzodiazepine-dependent.

With  growing  recognition  of  the  problems

with the benzodiazepines, attention has turned

Affective disorder: depression   385

to other forms of treatment. In the medium and

long term, SSRIs or venlafaxine are far more effec-

tive than benzodiazepines and are now drugs of

choice. Indeed NICE specifically advises against

benzodiazepines. One notable exception is in

serious organic illness and pain (Chapter 7),

especially  cancer,  where  relief  of  symptoms

becomes the prime objective.

Psychotherapy is essential for the long term management of GAD; even so, it is not always successful.

Panic

For   acute   panic   attacks   a   short   course   of

anxiolytic therapy is often used, but this is not

recommended   by   NICE   for   panic   disorder.

Longer-term   prophylactic   treatment   requires

antidepressants:  SSRIs,  the  tricyclic  clomipra-

mine,   and   MAOIs   are   effective.   However,

psychotherapy (such  as  anxiety  management

or CBT) may be useful, to try to tackle the

underlying problem.

Post-traumatic stress disorder;

obsessive-compulsive disorder (OCD); phobias

Generally all three are treated similarly. SSRIs are the drugs of choice, usually in higher than normal  dosage,  followed  by  tricyclics  then MAOIs. Clomipramine is licensed for the treat-

ment of obsessive-compulsive disorders. MAOIs are effective in phobias.

Mixed anxiety and depression

Antidepressants, preferably those with a sedative action, are especially useful here, as opposed to anxiolytics.

Affective disorder: depression

Most people experience a depressed mood from time to time. Feelings of sadness are an appro-

priate response to a recent loss or disappoint-

ment,   but   they   are   expected   to   remit

spontaneously sooner  or  later,  depending  on the  personality of the sufferer and the cultural norms of their society.

As with all psychiatric conditions this normal

response must be distinguished from an illness.

One criterion is whether it is so severe as to be

socially incapacitating or whether the patient

can  cope.  Another  is  whether  the  sufferer’s

culture regards the apparent cause as meriting a

period of such depressed mood. Grieving for a

close  relative  is  a  good  example:  almost  all

cultures   recognize   a   period   of   mourning,

although the acceptable duration varies greatly.

In  the  UK,  perhaps 6 months  to  a  year  is

regarded  as  normal  for  a  spouse.  Much  less

would seem callous, but to dwell on it for much

longer, remaining pessimistic, tearful and unable

to function, would be regarded as abnormal.

On   the   other   hand,   someone   who   was depressed  but  whose  life  gave  no  apparent evidence of substantial adverse events would be regarded as suffering from a morbid, or ‘clinical’, depression. This distinction must be reflected in the classification of depression. It is also neces-

sary to identify cases where there is an under-

lying primary medical or iatrogenic cause, to

which the depression is secondary.

Note  that  in  psychiatry  the  term  ‘affect’  is

an objective description of a person’s transient

emotional  behaviour,  whereas  mood  describes

their  prevailing  subjective  emotional  state.

Conventionally,   affective   disorder   includes

illnesses with abnormally high or low mood, i.e.

mania and depression. Although it might seem

system

logical to include anxiety, the complexity of

anxiety disorder has always meant that it is

considered  as  a  separate  major  category  of mental illness.

Classification and course

The difficulty in classifying depression is that it occurs in numerous forms, and different systems of classification use different criteria to distin-

guish between them. The range of criteria used by different classifications include:

•  Severity.

•  Presence or absence of physical (somatic or

            biological) features.

•  Presence or absence of psychotic features. •  Course (duration and recurrence).

•  Presence  or  absence  of  intervening  manic

phases.

The  first  three  criteria  apply  to  a  single

depressive or manic episode, while the last two

describe a pattern of recurrence (the course or

natural history). Taken together, these represent

the  basis  of  the  current  ICD  classification

(Figure 6.6).

A depressive episode usually lasts between 3

and 12 months but a fifth of episodes last more

than 2 years. A patient with the recurrent form

suffers on average four episodes over their life-

time. About half of patients only ever experience

a single episode, which can then be differentiated

on the basis of its severity (see below). Others

experience a persistent mild depression for much

of their adult lives, which may be a feature of their

personality. This is dysthymia or chronic affec-

tive disorder, and it may be interspersed with

more  severe  exacerbations,  when  it  is  known

as  double  depression;  one-quarter  of  major

depressive episodes occur against a background of

dysthymia.

More commonly, depressive episodes, although

naturally self-limiting, are prone to recurrence. A

distinction is then made on the basis of whether

the patient also suffers episodes of unnaturally

elevated mood known as mania; this is bipolar

affective disorder (pp. 405-408). A milder form,

cyclothymia, is seen in persons subject to alter-

nating periods of elevated or depressed mood

outside  what  is  considered  as  normal  mood

variation. Such persons are able to cope with the

illness  usually  without  resort  to  psychiatric

support. Both cyclothymia and dysthymia are

sometimes regarded as mild forms of personality

disorder (see Table 6.8).

In unipolar affective disorder there are recur-

rent episodes of depression or (far more rarely)

mania, with periods of remission. With unipolar

recurrent depressive illness, which is probably

the largest single category, each episode is insid-

ious in onset, developing over the course of

months, and lasting for months or even years.

Many patients with apparent unipolar disease

prove on closer examination to have bipolar

illness, which is now regarded as more common

than formerly.

In some cases there are only brief recurrent

depressive episodes, although each is severe. In

seasonal affective disorder (SAD), depression

recurs in the winter months, with either normal

or hypomanic mood (p. 404) in the spring and

summer. A depressive episode, particularly the

milder form, may coexist with anxiety, when the

description mixed anxiety and depression is

appropriate (p. 379).

Thus affective disorder should be regarded as a

continuum  or  spectrum  of  conditions  rather

than a collection of unique ones. At either end

Affective disorder: depression   387

are episodes of either mania or depression; in

between are various combinations in different

degrees of severity and patterns of occurrence.

Secondary depression can be caused by some

antihypertensive drugs, e.g. methyldopa, clonidine,

some  beta-blockers,  steroids (including  oral

contraceptives), and many CNS depressants such

as benzodiazepines, opioids and alcohol. It also

occurs secondary to dementia, stroke, Parkinson’s

disease, hypothyroidism, other psychiatric con-

ditions (schizophrenia,  alcoholism)  and  many

serious   chronic   diseases,   especially   those

involving pain (e.g. cancer, RA).

Epidemiology

Affective disorders comprise about half of all

psychiatric morbidity. Epidemiological statistics

for depression are difficult to obtain because of

the different definitions and the fact that most

depression  is  encountered  in  primary  care  or

is  frequently  unrecognized.  About 20%  of  us

develop a depressive episode at some stage in our

life and in a quarter of cases it becomes chronic.

The prevalence of major depression is approxi-

mately 5%. About a quarter of all depression is

bipolar.

Only 1 in 20 cases is referred to secondary care,

the remainder being handled by GPs, 1 in 10 of

whose consultations probably involve depres-

sion. Half of these patients will only have a few

minor depressive symptoms, which may well be

missed,   while   one-quarter   will   have   severe

depression. Depression is two to three times

more common in women and its prevalence

increases with age.

Depression is frequently associated with other

psychiatric   morbidity   or   substance   abuse,

notably alcohol. Thus it must be regarded as a

chronic  illness,  with  all  the  implications  for

continuing care that this brings. Only IHD has a

greater cost in disability-adjusted life-years. It

increases mortality, principally from suicide.

Aetiology

As might be expected, there are both biochemical and psychological theories.

Biochemical theories

Most   biochemical   or   organic   theories   of

depression  involve  abnormalities  in  central

monoamines  or  their  receptors.  However,  the

situation is still unclear and only a brief outline

is given here.

The  traditional  model  was  deduced  from

several  observations.  First,  depression  is  an

adverse effect of the now obsolete antihyperten-

sive   reserpine,   a   non-specific   central   amine

depleter. Second, the tricyclic antidepressants act

by preventing amine re-uptake (following release

at the nerve terminal) and hence could coun-

teract such depletion. Thirdly, MAOIs, which

enhance amine levels, are effective antidepres-

sants. The amines believed to be most involved

are serotonin (5-HT) and the catechol noradren-

aline (norepinephrine); more recently, dopamine

has been implicated.

Owing to its role in mood control, the limbic system is the most likely location of the abnor-

mality. Both transmitters are found there and the levels of one or other, or both, are affected by most  (but  not  all)  antidepressants.  Moreover, 5-HT is known to be involved in hypothalamic function, which could account for the sleep and appetite abnormalities of depression.

This theory has become considerably modified.

The tricyclics have been found to cause amine

receptor subsensitivity (down-regulation) which

takes about the same time to develop as the anti-

depressant effect, i.e. 2-4 weeks. MAOIs and ECT

have a similar action. This provides a plausible

explanation of the hitherto puzzling delay in the

therapeutic  action  of  antidepressants,  because

amine re-uptake block occurs within hours of

entry of antidepressants into the CNS. On this

newer model, depressed mood is associated with

an initial increase in receptor sensitivity. Amine

depletion could then be explained as a compen-

satory fall in amine release to reduce receptor

stimulation.  Alternatively,   depression   could

represent the failure of a compensatory down-

regulation  mechanism  that  normally  restores

emotional stability.

It is likely that these various transmitters are

interlinked along the polysynaptic neural path-

ways involved, so that action at one type could

have secondary consequences on others. We are

system

still a long way from understanding depression at the molecular level.

Metabolic markers

The diagnostic difficulties of depression have provoked a long search for a biochemical or

metabolic marker so that the degree of depres-

sion could be measured simply and reliably, like blood glucose in diabetes mellitus. The identifi-

cation of such a marker might also provide an insight into aetiology.

Thyroid function was an early candidate in

view of the depression associated with hypothy-

roidism, which must always be a differential

diagnosis in assessing affective disorder. There is

some correlation between depression and TSH

release. There is a better, but still insufficiently

consistent   or   specific,   association   between

depression, raised serum cortisol and a reduced

suppression of corticosteroid secretion by dexa-

methasone (the dexamethasone suppression test).

However, this may be secondary to the feeding

and sleep disturbances of depression. Neither

measure is clinically useful.

At  one  time  it  was  hoped  that  measuring

the  level  of  a  metabolite  of  brain  catechols,

3-methoxy-4-hydroxy-phenylethylene glycol (MHPG)

would yield specific information about central

amine turnover. If the depression were associ-

ated with a catecholamine depletion, then urine

or CSF levels of MHPG would be low: a 5-HT

deficit on the other hand would not affect the

MHPG level. It is also possible, though less prac-

ticable, to measure CSF levels of 5-HT, which are

frequently found to be low in depression. These

measurements should then have an important

bearing on drug treatment because some anti-

depressants   selectively   block   noradrenaline

(norepinephrine) re-uptake (e.g. reboxetine) while

others block 5-HT re-uptake (e.g. fluoxetine). They

might also predict when tryptophan would be a

logical   replacement   therapy.   Unfortunately

this approach too has proved inconsistent, and

more pragmatic considerations still determine

drug selection.

Mood   disorders   are   frequently   seasonal.

Depression  is  usually  more  prevalent  in  the

spring, but in SAD it occurs each winter. Lack of

sunlight is an important aetiological factor in

this.  Abnormalities  in  melatonin  metabolism

may be involved because melatonin secretion is

suppressed by light, and in manic depressive

patients (p. 403) this response is augmented.

However, the relationship between melatonin

and mood is unknown.

Functional theories

An alternative to this reductionist biochemical

approach is the functional or psychoanalytic

one,  where  unconscious  mental  conflicts  are

thought to be the cause of the disease (p. 373).

It  is  likely  that  a  synthesis  of  these  two

approaches will eventually prevail, though at

present   the   biochemical   model   probably

provides a better explanation of major depres-

sion,   while   minor   depression   seems   better

accounted for in terms of mental conflicts and

personality,  or  illness  behaviour.  A  family

history  of  depression  in  this  respect  tells  us

little,   because   the   family   environment   is

as  likely  as  genes  to  influence  personality

development.

Genes versus environment

A family history is common in depression. Envi-

ronmental  factors  such  as  social  class,  unem-

ployment,  domestic  pressures  and  childhood

experiences  are  also  important.  The  contri-

bution  of  each  factor  varies  from  patient  to

patient.

Although  a  serious  loss  or  disappointment

does not always precede major depression, an

adverse or stressful event is frequently found in

the patient’s recent premorbid history. However,

a careful history is necessary, including informa-

tion from friends or relatives, to reduce the

likelihood  of  confusing  cause  with  effect.  A

depressed patient may first present with many

problems in their life, any of which might plau-

sibly have been the immediate cause of their

depression. It is necessary to establish whether

any did in fact predate their illness, or if they are

all the consequences of depression. It may even-

tually   become   evident   that   there   was   no

discernible specific trigger for the initial onset of

symptoms.

Affective disorder: depression   389

Clinical features

When summarizing the various clinical features

of depression it may be more helpful to adopt a

pragmatic approach based primarily on severity.

Psychiatric spectrum

Traditionally, depressive episodes were divided

into two broad groups depending on the pres-

ence  or  absence  of  psychotic  features:  this

roughly  correlates  with  severity.  The  milder

forms resemble an exaggerated response to an

adverse life event that might be expected to

depress  anyone  to  a  certain  extent,  such  as

bereavement,  but  with  which  the  patient  is

unable to cope. As such it is referred to as minor

depression.  However,  the  description ‘minor’

should not be taken to imply that, for the patient,

the depression is somehow more bearable.

This was previously known as neurotic or reac-

tive depression; the term ‘reactive’ being used to

emphasize that the depression is a response to a

life event, or that the patient still retains the

ability to react emotionally to normal everyday

events.

Minor depression is analogous to stress-related anxiety neurosis in that both are associated with real events and insight is preserved. However, depression  is  generally  retrospective,  whereas anxiety is anticipatory.

At the other extreme are patients with severe

or major depression, where there are no obvi-

ously adverse events in the recent past. Such

patients become depressed for reasons they do

not understand and they have poor insight into

their condition. This form is usually associated

with  a  number  of  metabolic  and  physical

(somatic) features and may or may not involve

psychotic  symptoms  such  as  delusions.  It  is

still  sometimes  referred  to  as  biological  or

endogenous  depression,  and  sometimes  the

poetic  medieval  (originally  Greek)  description

melancholia is revived.

Figure 6.7 illustrates in a simplified manner

how this distinction parallels some of the more

important   features   of   depressive   illness.

However,  things  are  rarely  as  simple  as  this

would  imply  and  the  distinctions  are  some-

what  artificial.  Biological  variables  within  a

population tend to occur as finely graded differ-

ences rather than clearly defined categories. The

majority of patients appear somewhere on the

continuum   between   the   two   extremes,   at

different points for different features, at different

times in the course of their illness. Thus, a

patient may have a strong family history and

some delusions of guilt, but have an acute onset

triggered by bereavement.

Severity

For  a  single  depressive  episode  an  alternative approach is to look at common presentations of the  symptomatology  at  different  degrees  of severity (Figure 6.8). The core features of depres-

sion, present even in the mildest form, are:

•           Depressed        mood   (dysphoria)       with

pessimism  (especially  regarding  the  likeli-

hood of recovering) and frequent tearfulness.

•           Loss of interest and capacity for enjoyment

(anhedonia).

•           Lethargy and fatigue.

system

Patients  may  suffer  a  variety  of  additional

neurotic  symptoms  similar  to  those  seen  in

anxiety.  There  are  sleep  and  appetite  distur-

bances and somatization (psychosomatic) prob-

lems such as constipation or dyspepsia. Anorexia

is common, perhaps as an expression of general

loss  of  interest  in  life’s  pleasures,  but  also

possibly resulting from alterations in hypothal-

amic function. Patients tend to feel sorry for

themselves but not guilty, i.e. ‘What have I done

to deserve this?’ rather than ‘I deserve this as

punishment’.

The mood of patients with mild to moderate depression,  although  predominantly  low,  is labile, tending to fluctuate to some extent in

reaction to the minor everyday joys and disap-

pointments of life. They have difficulty getting off to sleep, i.e. initial insomnia or increased

sleep latency, as seen in anxiety. They lose the capacity to initiate any action, and find making even quite trivial decisions difficult.

There is a marked loss of drive, vitality, ambi-

tion, libido and zest for life - a loss of appetite

for more than just food. Patients have difficulty

concentrating  and  suffer  selective  memory

impairment;  past  mistakes  in  particular  are forgotten, or repressed. There is a persistently negative approach to everything.

In severe depression, more physical (somatic)

and metabolic features become evident. There

may also be psychomotor impairment - either

agitation,  or  conversely,  mild  retardation  of

thought, speech and action. Patients wake early,

at 4 or 5 a.m. feeling at their worst, but things

improve slightly by evening. Mood variation

tends to become fixed in this pattern, whatever

the day brings. Rather than feeling sorry for

themselves they somehow blame themselves for

their plight: guilt, self-reproach and feelings of

worthlessness  are  common.  Because  marital

problems can follow the reduced libido, and

impaired performance at work result from the

loss of drive, the patient’s low self-esteem is rein-

forced and justified: it becomes a self-fulfilling

prophecy.

Very severe depression  is accompanied by

intensification  of  signs  and  symptoms,  with

worsening of agitation or retardation. Psychotic

features also become apparent. For example, the

general feeling of worthlessness and inadequacy

becomes a firmly fixed, unrealistic false belief

that no amount of appeals to reason can shift

(i.e. a delusion). They believe themselves to be

guilty, ‘bad people’, responsible for destroying

their family and letting down their colleagues at

work. These delusions may (although rarely), be

reinforced with another classic psychotic sign,

hallucinations (false  perceptions).  They  may

hear  voices  chastising  them  for  the  evil  or

corrupt person they believe themselves to be.

Thus it is possible to get an inkling of how

people may be driven to what is, in biological

terms, the highly anomalous behaviour of inten-

tional self-harm or even self-destruction. The

perpetual misery, the firm conviction that you

are causing serious problems for people around

you, including loved ones, and the belief that

things  will  not  improve,  makes  suicide  an

appealing, almost logical option for them.

Suicide and self-harm

There are some 5000 successful suicides a year in

theUK. About 75% of these involve mental

illness, mostly depression but also alcoholism,

personality        disorder            and    schizophrenia.

Depressed  patients  are  about 25  times  more

likely than others to attempt suicide, and 15%

will eventually succeed. Prevention of suicide

and attempts to change suicidal ideas are major

targets in the management of depression.

In theUK, drug overdose is the usual means

(up to 90% of cases) in both suicide and delib-

erate self-harm. Since barbiturates have become

unavailable,  the  risk  of  serious  consequences

from drug overdose should have been reduced.

However, the barbiturates have been superseded,

and possibly exceeded in danger, by paracetamol

(acetaminophen), which is freely available and

hence widely believed to be safe. Yet as few as

20 paracetamol tablets can have the tragic unin-

tended  outcome  of  fatal  liver  failure.  Overall,

paracetamol and other analgesics are the most

common choice, presumably because of ease of

access (30%); sedatives are used in 15% of cases

and antidepressants in 10%. Alcohol use is also

involved in over half of drug overdoses. The

major advantage of the SSRIs over the older

system

tricyclic antidepressants is their relative safety in

overdose: this ameliorates the paradox that some

of the most toxic medication is given to the

groups of patients most likely to overdose on

medication.

It is important to distinguish between suicide,

which   is   intentional   successful   self-inflicted

death, and deliberate self-harm (DSH), which is

deliberate self-inflicted injury (formerly parasui-

cide); the two are distinct entities presented by

different types of patient. Those who decide to

commit suicide are likely to be thorough: they

will fail only through ignorance or chance, and

when  successful  it  is  known  as  completed

suicide.  They  plan  the  suicide  carefully  for

weeks so that it almost becomes an obsession.

Although ostensibly keeping it secret, they may

well give covert, even unconscious hints of their

intention. More importantly, if asked directly

they will usually admit to suicidal thoughts.

Unfortunately, the topic is usually taboo among

their family so they are driven to veiled hints

that, tragically, are frequently only recognized in

retrospect. Suicide is more often associated with

major  depression  than  minor,  and  patients

frequently utilize means that are more likely to

succeed and less likely to permit reversal (i.e.

large overdoses or jumping in front of a train).

For  patients  who  are  recovering  from  very

severe retarded depression, a high-risk time for

suicide attempts is in the early stage of recovery,

when  their  mood  remains  low  but  their

motivation and drive have started to return.

Patients with less severe depression may be

driven  to  an  ill-conceived,  impulsive  act  of

aggression   directed   at   themselves.   DSH   is

frequently preceded or immediately followed by

a desperate phone call - the literal cry for help to

complement the symbolic one. Self-harm is not

usually intended to be fatal: it rarely employs

irreversible measures. Often it means ingestion

of a few dozen hastily assembled tablets or some

clumsy slashes at the wrist. Yet if someone is so

disturbed that they demonstrate their hopeless-

ness and their need to draw attention to their

plight by self-harm, they surely merit help and

sympathy,   however   manipulative   they   may

appear. And of course if the tablets are para-

cetamol  the  consequences  might  well  be  far

more serious than intended.

The term ‘attempted suicide’ is now little used because it does not convey whether or not there was an intention to die.

Diagnosis

A formal diagnosis of depression is made on

symptomatic grounds. There must be low mood

and at least four other symptoms from among:

•  pessimism;

•  negative thoughts; •  weight change;

•  sleep disorder;

•  psychomotor agitation or retardation; •  fatigue;

•  feelings of inadequacy or guilt; •  difficulty in concentrating;

•  suicidal thoughts or actions.

These must have persisted for at least 2 weeks, be inconsistent with the patient’s prior person-

ality and not be secondary to another medical condition or drug therapy.

The severity of depression may be rated by

applying a standard scale based on a question-

naire, such as the Hamilton Depression Rating

Scale or the Beck Depression Inventory. This can

also be used to assess the efficacy and progress

of  any  treatment.  However,  most  depressive

episodes remit and antidepressant drugs are slow

to act, so any observed remission may be at least

partially natural, although it may be tempting to

ascribe it to the therapy.

In determining treatment, the finding that the depressive episode is associated with an adverse life event, and is therefore in some way under-

standable or ‘reactive’, is no reason to avoid

drugs. The key criterion is severity.

Management

Aims and modes

The aims in managing depression are, in order of priority, to:

•  Prevent   suicide  -   consider   custody   and

compulsory treatment if suicide is a risk.

Affective disorder: depression   393

•  Identify  possible  primary  causes,  such  as

            chronic or iatrogenic illness.

•  Symptomatic therapy to relieve the patient’s

            misery.

•  Investigate any adverse social, domestic or

            financial circumstances.

•  Initiate long-term therapy to prevent relapse

            or recurrence.

The three main modes of treatment are phys-

ical (invasive, with drugs, ECT or custody), social

and psychological therapy. The principal criteria

for  deciding  between  these  for  a  particular

patient are:

•  Urgency of illness or required speed of onset

            of treatment.

•  Effectiveness.

•  Natural history and likelihood of recurrence. •  Presence of psychotic symptoms.

•  Contra-indications and adverse effects. •  Cost.

A multidisciplinary team approach is preferred,

involving clinicians, nurses, social workers and

sometimes  pharmacists.  In  severe  depression,

invasive methods are almost invariably indicated,

if  necessary  including  temporary  custody  and

involuntary   treatment   to   prevent   patients

harming themselves or making unwise decis-

ions. Drugs and ECT are cheap and relatively

rapidly effective, but in the long term, social or

psychological therapies have a better chance of

reducing recurrence. A distinction must be made

between  a  depressive  episode  and  recurrent

depressive illness, for which prophylaxis must be

considered.  If  biological  or  psychotic  features

are prominent, the illness responds better to

physical therapy. In mild depression drugs are

generally ineffective.

Non-drug treatment

Psychotherapy

This is the preferred treatment for mild depres-

sion and frequently all that is needed: there is

little evidence that drug therapy is effective.

Psychotherapy is also an important component,

along with drugs, in the management of major

depression after the acute stage, when patients

have improved insight, although the aim of

some  forms  of  psychotherapy  is  to  further

improve insight. At its simplest, psychotherapy

may involve no more than giving the patient a

sympathetic and concerned ear (see also p. 381).

Preferably the patient should be involved in the

treatment decision process, with discussion of

the possible causes of the illness, an account of

the pros and cons of the treatment options and

a realistic account of possible outcomes.

Patients must be reassured that they are even-

tually going to get better - something that they find difficult to believe. In severe depression it is also important to try to restore the patient’s self-

esteem. Patients should be advised against taking important decisions while depressed.

More specific therapies include group, marital

and family therapy. CBT has been used very

successfully in numerous mental illnesses. In

this approach, unhelpful or negative ways of

thinking are countered: the patient might be

encouraged, for example, to think constructively

about his or her illness or to plan strategies to

overcome specific symptoms. Psychoanalysis has

a place in the management of neurotic depres-

sion. Psychosocial treatment involves improving

the patient’s social situation.

Although  slow,  often  labour-intensive  com-

pared with drug therapy, and not always effective,

psychotherapy does at least hold out hope of a

recovery in that the patient becomes better able

to cope and remains relatively symptom-free.

Electroconvulsive therapy

While it may seem a bizarre and unlikely proce-

dure, and is thought inhumane by some, ECT

is often rapidly effective compared with drug

therapy in very severe and suicidal depression. It

also helps patients who are resistant to other

forms of therapy or intolerant of them, and there

are surprisingly few proven harmful effects. It

involves a brief electrical pulse being passed

through   the   brain   between   two   electrodes

attached  to  the  skull   with   electrode   jelly

(similar  to  that  used  in  EEG  and  ECG  tech-

niques).   The   treatment   induces   electroen-

cephalographic changes characteristic of a major

tonic-clonic seizure (see p. 443). The patient is

anaesthetized with a brief-acting IV anaesthetic

and a muscle relaxant (e.g. suxamethonium) is

used to prevent a physical seizure.

system

A course of ECT consists of treatments given

about three times weekly, for 3-4 weeks. The

procedure often produces a rapid elevation of

mood in the severely depressed, minimizing the

risk of suicide at the most vulnerable phase of

depression  and  enabling  the  patient  to  start

psychotherapy.

Psychiatry has often had recourse to what may

be called trauma therapy. In less enlightened

times, hydrotherapy involved drenching ‘mad’

patients in freezing water. Later, insulin was used

to induce hypoglycaemic shock. ECT was partly

an extension of this into the age of electricity.

Presumably  the  rationale  for  these  so-called

treatments was akin to our modern predilection

for kicking a recalcitrant piece of machinery, and

presumably they met with the same occasional

success.

The original rationale for ECT was the obser-

vation, ill-founded as it transpired, that people

with  epilepsy,  if  they  developed  a  psychotic

illness, had fewer seizures but tended to feel

better after a seizure: it was therefore reasoned

that more fits in psychotic patients might reduce

their  psychosis.  Subsequent  research  showed

that only the cranial events of a seizure are

important, so that the muscular seizure may

safely be suppressed.

Double-blind placebo-controlled trials, using

no actual current in controls, have subsequently

demonstrated  convincingly  that  the  electric

shock is essential. Previously, a variety of other

factors had been suggested, including the anaes-

thetic, the muscle relaxant, the central hypoxia,

the   special   care   and   attention   the   patient

receives (because of the slight but real risks) or

even the punitive element a patient suffering

from  guilt  might  welcome.  ECT  appears  to

produce changes in brain amines and receptor

sensitivity  similar  to  those  produced  by  anti-

depressant drugs.

The   common   adverse   mental   effects   are

headache    and    confusion    on    recovering

consciousness; similar symptoms occur after a

major epileptic seizure. There is some loss of

recent   memory,   which   may   reduce   undue

anxiety about subsequent ECT procedures. These

side-effects are minimized by passing the current

through  only  the  non-dominant  hemisphere

(unilateral ECT). Uncertainty over the possibility

of long-term brain damage means that courses

are kept to a maximum of about 12 treatments,

although   patients   have   had   more   without

evident harm. ECT has none of the adverse

effects or contra-indications of antidepressant

drugs. It can be used safely in pregnant women,

the elderly and those in whom antimuscarinic

drugs  are  contra-indicated.  There  is  no  long-

term toxicity or suicide risk. However, ECT has

no place in the treatment of mild or atypical

depression.

Current NICE recommendations for ECT are that it should only be used short term in severe depression where other treatments have failed, especially where suicide is a serious risk.

Other methods

In SAD, 2 h of exposure each morning to full-

spectrum  artificial  light  equivalent  to  bright

sunlight (phototherapy) seems to be effective

but it must be continued as long as natural

sunlight is unavailable. Transcranial magnetic

stimulation was tried experimentally with some

initial success but has not yet been shown to be

effective in clinical trials. It resembles ECT but

requires no anaesthesia or muscle relaxant.

Drug therapy

Antidepressant drugs

The  antidepressant       (thymoleptic  or  mood-

elevating)  drugs  can  be  divided  into  several groups, on the basis of pharmacological, clinical and adverse actions:

•  Tricyclics, the original group. •  Second-generation cyclics.

•  Selective re-uptake inhibitors, especially for

            serotonin (SSRIs).

•  MAOIs.

The properties of the first two groups (referred to here generically as ‘cyclics’) and SSRIs are

compared in Table 6.11.

Selectivity.   The  nomenclature  of  antidepres-

sants has become somewhat confused owing to

an  informal  classification  by  the  principal

neurotransmitter affected. The original tricyclics

Affective disorder: depression   395

are non-selective. Most block both noradrena-

line (norepinephrine)  and 5-HT  re-uptake  at

central synapses to varying degrees and may also

restore  receptor  sensitivity.  They  also  block

histamine-H1 receptors, peripheral acetylcholine

(muscarinic) and alpha-adrenergic receptors to

a lesser extent. Some also appear to interact

weakly  with  dopaminergic  systems.  There  is

partial selectivity in that some (e.g. nortriptyline)

have a greater effect on noradrenaline (norepi-

nephrine) and others, especially amitriptyline, on

5-HT.

The SSRIs show high selectivity for 5-HT re-

uptake.   Newer   agents   have   noradrenaline (norepinephrine) selectivity (e.g. reboxetine), or

noradrenaline/5-HT   selectivity             (venlafaxine). Mirtazapine and mianserin appear not to affect any of the usual transmitters directly.

Certain    anti-dopaminergic    antipsychotic

drugs also have moderate antidepressant action,

particularly the thioxanthenes, e.g. flupentixol.

However, extrapyramidal and endocrine adverse

effects (pp. 419-421) limit their usefulness in

depression without marked psychotic features.

They are more useful in schizophrenia with asso-

ciated depressive features. Amoxapine also blocks

dopamine.

Despite  this  pharmacological  diversity,  no

particular  pharmacodynamic  properties  have

been  clearly  shown  to  confer  superior  anti-

depressant efficacy. The most important clinical

distinctions are the presence or absence of:

•  Antimuscarinic, anti-adrenergic and antihist-

            aminic  activity  (conferring  adverse  effects).

•  Sedative  action (not  linked  exclusively  to

activity on any single transmitter).

•  Cardiotoxic and convulsant action, especially

            in overdose.

These properties, governing the adverse affect and contra-indication profile, significantly affect

drug   selection.   Antidepressant           ‘selectivity’ means little more than activity predominantly on  one  or  other  of  the  central  transmitters presumed to be involved in depression.

Tricyclics and related drugs

Agents with the traditional three-ring structure,

such  as  amitriptyline,  are  effective  and  cheap,

and their properties well known. However, they

have significant adverse and toxic effects (Table amoxapine,   mianserin,   venlafaxine),   offer   a

6.11). Later and second-generation drugs (e.g.  number of advantages in that they have fewer

lofepramine,  dosulepin (dothiepin)),  some  not adverse effects and are less toxic in overdose

chemically   tricyclic   but          ‘heterocyclic’    (e.g.     (Table 6.12).

Indications.   All the traditional tricyclics and

the  related  newer  agents  are  very  useful  in

moderate to severe depression, but there is no

evidence that they are useful in mild depression.

Amitriptyline and many others also have seda-

tive  as  well  as  antidepressant  action,  making

them  particularly  useful  for  depression  mixed

with anxiety or agitation. A single daily dose

in  the  evening  also  aids  sleep  but  minimizes

daytime  sedation.  The  antidepressant  effect,

being  unrelated   to   plasma   level,   is   more

prolonged.

Less    sedative    agents           (e.g.     imipramine, lofepramine) may be useful where neither anxiety nor retardation are problems. This will minimize daytime sedation, which may otherwise restrict the patient’s activity or occupation.

Pharmacokinetics.   Generally,   tricyclics   are

rapidly and fairly well absorbed but are subject to

considerable first-pass metabolism. The extent of

hepatic clearance (by demethylation and hydrox-

ylation) varies greatly, partly accounting for wide

interpatient variation in response. Among the

older  drugs,  tertiary  amine  derivatives (e.g.

imipramine) are frequently demethylated to active

secondary amines (e.g. desipramine) with a slightly

different  activity  profile,  thus  prolonging  the

action.

The tricyclics are quite highly protein bound.

However, they have a high volume of distribu-

tion with accumulation in extravascular sites so

there  are  no  significant  displacement  interac-

tions. Plasma level monitoring is not a useful

guide to dose titration or efficacy, but can be used

to assess compliance. The overall pharmacoki-

netic effect is usually a long biological half-life,

which may be further prolonged in over-dosage.

Although therapeutic plasma levels are achieved

within 24 h, the clinical (antidepressant) effect is

not seen for several weeks. Metabolism may be

considerably reduced in the elderly, predisposing

them to side-effects.

Side-effects.   Mild to moderate antimuscarinic

adverse effects, such as dry mouth, constipation,

poor visual accommodation, tachycardia, etc.,

are common, as is drowsiness with many agents

(Tables 6.11 and 6.12). Although these effects are

not serious and usually remit on continued use

they may be troublesome at first and contribute

to poor compliance. The visual problems may

result in consultation with an optician, who

should be informed about the drug therapy. If

not   evident   as   drowsiness,   CNS   depressant

effects on cognition and coordination may be

difficult to distinguish from the more retarded

forms of the illness. Owing to this and postural

hypotension (due to peripheral alpha-adrenergic

blockade),  antidepressants  are  believed  to  be

implicated in a large number of falls and other

accidents, especially in the elderly.

Weight gain is quite common. This may be in

part the resolution of depressive anorexia, but a

direct effect on appetite is also seen even in non-

depressives, e.g. when used for migraine prophy-

laxis. In patients who are susceptible to bipolar

mood swings, possible (re)activation of mania is

a significant risk. Indeed they may precipitate

the manic episode of latent bipolar disease in

what had been diagnosed as unipolar recurrent

depression.  A  lowering  of  seizure  threshold

may destabilize patients with epilepsy or cause

problems if a patient overdoses.

Hyponatraemia may occur owing to central

stimulation of ADH secretion, particularly in the

elderly. This causes CNS depression, including

drowsiness and confusion, which may easily be

attributed to other causes. Extrapyramidal symp-

toms  occasionally  occur.  Mianserin  has  been

implicated  in  bone  marrow  depression  with

leucopenia.

Toxic effects and overdose.   The acute toxicity

of antidepressants is important because they are

system

often involved in suicide attempts. For tricyclics there is no specific antidote; all complications must be managed symptomatically as they arise. The main problems are:

•  CNS, with seizures and confusion;

•  cardiovascular, with conduction defects giving

heart  block;  vagal  inhibition  giving  tach-

yarrhythmias;  and  profound  hypotension.

Thus such overdoses are often fatal. A prin-

cipal advantage of the newer agents is a generally reduced   likelihood   of   these   complications, although dosulepin (dothiepin) is the most toxic of all in overdose.

Interactions.   Adrenergic over-stimulation may occur when tricyclics are given with sympath-

omimetic drugs (e.g. as included in some OTC decongestant preparations) and MAOIs causing potentially dangerous hypertension. CNS depres-

sants, especially alcohol, which are frequently taken in association with an overdose, produce excessive sedation or coma. The effect of anticon-

vulsant drugs may be diminished. Arrhythmias may occur with digoxin or quinidine.

Contra-indications   and   cautions.   Patients

with heart disease (e.g. arrhythmias, previous

MI), narrow-angle glaucoma, urinary retention

(e.g.   prostatism)   or   constipation,   in   whom

antimuscarinic   effects   could   be   harmful   or

dangerous,  should  use  other  drugs.  Patients

taking antihypertensive medication should use

mianserin, which does not interfere with amines.

Care is needed in epilepsy. The elderly are prone

to over-sedation because of the long half-lives

and  this  problem  can  increase  until  steady-

state  concentrations  are  reached.  The  blood

count  of  elderly  patients  on  mianserin  should

be monitored.

Dosage and administration.   Drug treatment

should be started as early as possible once major

depression  is  diagnosed;  this  has  beneficial

effects on outcome and relapse. For optimal anti-

depressant effect, it is strongly recommended

that tricyclics must be used in adequate doses,

for example, at least 125 mg daily of amitripty-

line, and treatment failure has been attributed

largely to subtherapeutic dosing. However, one meta-analyis has suggested that lower doses may

be  adequate.  Therapeutic  levels  are  achieved

gradually over 2 weeks by starting at half the

target dose, or one-quarter of it for the elderly or

children. Unacceptable adverse effects must be

warned about and monitored, lest confidence in

drug therapy is impaired. The inconvenience of

the   initial   side-effects   may   be   reduced   by

dividing doses unequally, e.g. two-thirds at night

and one-third in the morning. If suicide is a

likely  or  suspected  risk,  only  small  supplies

should be given at any one time, although deter-

mined patients still accumulate them.

Patients must be warned that the antidepres-

sant effect does not become apparent for several weeks, because most people expect all drugs to work very quickly. During this initial period they may only experience the adverse effects, which may be both disabling and dispiriting.

Antidepressants should not be withdrawn too

quickly; otherwise there may be a discontinua-

tion syndrome with mild mixed gastrointestinal

and central effects. After 8 weeks or more of

therapy, dosage should be tailed off over at least

4 weeks. These effects do not indicate depen-

dence, which does not occur, but are probably

due to receptor sensitivity changes. However,

many people believe antidepressants are addic-

tive, which reduces confidence in them and

damages compliance. Exploring and correcting

this misconception should be part of the initial

discussion with the patient about therapeutic

choice.

Newer cyclic agents

Table  6.12  shows  how  many  of  the  newer

tricyclic-related drugs, which have been devel-

oped to make safer antidepressants, are compa-

rable to the class archetype amitriptyline for the

most common adverse effects. Molecular model-

ling has produced a wide range of drugs with one

or more improvments, greatly increasing the

choice for  specific  circumstances.  They  either

have fewer or less serious side-effects or less life-

threatening toxic effects. However, they are no

better and no faster at relieving depression.

SSRIs

Indications.   These are now the first choice in

most moderate to serious depressive episodes.

Affective disorder: depression   399

The main advantages of 5-HT specificity are the absence of many of the troublesome adverse

effects, toxicity and overdose problems associ-

ated with the older, less selective agents. Early hopes that they might be more effective or faster acting have not been realized, and they are used for  the  same  purposes  as  the  tricyclics,  and generic ones are little more expensive.

Pharmacokinetics.   The  SSRIs  have  broadly similar kinetic properties to the tricyclics, but are less likely to have active metabolites.

Side-effects.   SSRIs   have   a   quite   different

profile from the tricyclics (Table 6.12). Most

important  is  the  almost  complete  absence  of

antimuscarinic, cardiotoxic and convulsant prop-

erties. Also, there may be weight loss as opposed

to weight gain, and psychomotor impairment is

less.  However,  withdrawal  phenomena,  hypo-

natraemia, activation of mania and occasional

extrapyramidal symptoms still occur.

The  most  common  problems  are  relatively

mild   gastrointestinal   and   central   effects.

However, nausea, diarrhoea or constipation may

be intolerable for some. An association with

gastrointestinal   bleeding   has   recently   been

noted,  possibly  related  to  interference  with

serotonergic mechanisms in platelets. Gastro-

protection (p. 777)  has  been  recommended

when SSRIs are to be taken by patients already

taking NSAIDs.

Less common is CNS stimulation, presenting

as anxiety, agitation, restlessness, headache or

confusion. In general (i.e. apart from specific

contra-indications) the SSRIs are tolerated better

than tricyclic antidepressants, as measured by

the dropout rate due to adverse effects; about

one-third more patients cease tricyclic therapy.

Toxic effects and overdose.   Another signifi-

cant advantage of this group is the absence of

serious  cardiotoxic  or  convulsant  actions  in

overdose. Thus they are much safer than the

original tricyclics, although some newer cyclic

antidepressants are also safer than these (Table

6.12). It must also be remembered that SSRIs

offer no greater protection from suicide by other

means.  Both  tricyclics  and  SSRIs  appear  to

increase   the   likelihood   of   suicide   ideas   or

behaviour in the early stages of therapy. Fears

that SSRIs and venlafaxine might be worse in this

respect   are   probably   unfounded,   except   in

patients under 18, where their use is contra-

indicated,   except   for   fluoxetine   in   certain

restricted circumstances. Patients under 30 need

to be monitored closely early in treatment.

Rarely,   SSRIs   may   cause   the        ‘serotonin

syndrome’,  especially  when  interacting  with

other antidepressants and especially the anti-

migraine triptan 5-HT agonists. The liberation of

excessive  5-HT  causes  CNS  stimulation  result-

ing   in   restlessness   and   hyperthermia.   The

syndrome is usually mild, but may prove fatal. It

usually responds to discontinuation of SSRIs if

they are not tolerated, but in severe cases the

5-HT blocker cyproheptadine can be used as an

antidote.

Contra-indications,   cautions,   interactions.

Serotonin selectivity means SSRIs can be used in

many  situations  when  tricyclics  are  contra-

indicated. However, care is still needed in heart

disease and epilepsy. They should be used with

caution in patients susceptible to seizures (or

undergoing  ECT),  cardiac  disease,  mania  or

bleeding. Suitable washout periods, usually of

several weeks, must be observed when changing

to  other  types  of  antidepressant (especially

MAOIs) in order to avoid serious interactions.

SSRIs  inhibit  hepatic  metabolism  but  despite

numerous theoretical interactions, few clinically

significant ones occur in practice. Sertraline and

citalopram may be the safest in this respect.

Dosage and administration.   Most SSRIs can

be given once daily. The dose-response curve

rapidly plateaus above the usual effective (and

tolerable) dose. Therefore full doses can often be

started at once (perhaps originally fostering the

impression of more rapid onset of activity). If

nausea is a problem, starting on half the dose

for a few days can help. However, withdrawal

should still be staged over 2-4 weeks to avoid

discontinuation symptoms.

Monoamine oxidase inhibitors

Indications.   Frequent and hazardous dietary

and drug interactions have always limited the

system

use of MAOIs, especially in general practice.

They may be of value in depression associated

with atypical features characteristic of anxiety-

neurosis (especially phobia or panic disorder),

but they are not usually effective when used

alone in severe depression. Their main role is as

adjuncts or second- or third-line drugs in resis-

tant  disease.  Dietary  compliance  is  easier  to

ensure in hospital.

Side-effects and toxicity.   The range of mild to

moderate adverse effects is very wide, and the

potential toxic effects of MAOIs are serious. Many

central, cardiovascular and gastrointestinal auto-

nomic disturbances occur in normal use, espe-

cially antimuscarinic and anti-adrenergic effects.

Although interactions may cause hypertension,

adverse  effects  frequently  involve  postural

hypotension. Peripheral oedema is also common.

CNS stimulation and activation of mania can

occur.  Overdose  with  MAOIs (except  tranyl-

cypromine) is somewhat easier to manage than

with tricyclics because specific alpha-adrenergic

blockers can be used, e.g. phentolamine.

Interactions.   Normally,  dietary  tyramine  is

metabolized by MAO in the gut wall. Inhibition

of MAO allows high levels of tyramine to be

absorbed, causing widespread release of nora-

drenaline (norepinephrine) from nerve fibres,

with   predictable   hypertensive   cardiovascular

consequences. Interacting prescription medica-

tion (e.g. cyclic antidepressants, CNS sedatives,

opioid analgesics) can usually be avoided by

professional vigilance. Greater potential prob-

lems arise with OTC preparations containing

sympathomimetics         (e.g.   decongestants)   and

tyramine-containing  foods (especially  cheese,

pickled fish and meats or meat extracts, fava

beans), which is why patient counselling and

warning cards are so important. Tranylcypromine,

the  most  stimulant  MAOI,  seems  to  be  the

worst  offender,  especially  if  combined  with

clomipramine.  Phenelzine  is relatively sedating.

Fatalities are still reported and MAOIs are likely

to remain little used.

Because phenelzine causes irreversible enzyme

inhibition,   recovery   of   enzyme   functions

following cessation of therapy does not start for

several days and may take several weeks to return

requires at least a 2-week interval, but the reverse can be done safely after a week, though some SSRIs require longer.

Sometimes a tricyclic/MAOI combination is indicated, in which case a sedative tricyclic with a less stimulant MAOI (e.g. amitriptyline  with phenelzine) is usually safe if managed by experi-

enced psychiatrists. Tranylcypromine is the most dangerous in combination.

Other antidepressants

Moclobemide causes reversible inhibition of MAO

type A (RIMA). This has two principal advan-

tages:  enzyme  function  is  restored  promptly

following drug cessation, allowing quicker and

safer switching to other antidepressants; and

tyramine metabolism is little inhibited, so the

patient is relatively free from dietary restrictions.

It also has few of the autonomic adverse effects

of the non-selective MAOIs and appears to be

equally effective as the cyclics, particularly for

retarded depression. However, it has not yet

found its place in treatment protocols.

Tryptophan,  a  5-HT  precursor,  has  proved

disappointing considering the strong suspicion

of a 5-HT disturbance in some forms of depres-

sion.  It  is  occasionally  used  as  an  adjuvant.

Lithium and its use are discussed on pp. 406-408.

The herbal productSt John’sWort (hypericum

extract) has been shown in some trials to be as

effective  as  other  antidepressants  in  mild  to

moderate   depression   but   is   still   not   yet

adequately  standardized,  nor  have  its  active

ingredients been characterized. Moreover, there is

concern  about  interactions,  especially  as  it  is

available without prescription and induces liver

enzymes.

Lithium and some anticonvulsants (e.g carba-

mazepine, lamotrigine, valproate) are sometimes

used as ‘mood stabilizers’ in bipolar disorder (see

below).

Treatment strategy and drug selection

The strategy for the drug treatment of depression

involves several stages (Figure 6.9). Bazire in the

Psychotropic Drug Directory characterises these

as the 6 Ds:

•  Check the illness is not Drug induced. •  Select the most suitable Drug.

•  Don’t Delay treatment: start acute treatment

            as soon as possible.

•  Titrate to full recommended Dose.

•  Continue   for   the   correct   Duration,   for

6 months   after   symptoms   first   respond, ensuring compliance.

•  Discontinue slowly.

In deciding whether or not to treat an episode

of  primary  depression,  the  criteria  are  the

severity of depression and the patient’s ability to

cope. Whether or not the low mood is consid-

ered an appropriate reaction to an adverse event

is not important. Moreover, prompt treatment

reduces the length of the episode and the likeli-

hood of recurrence. The use of full doses for

4-6 months after symptoms have subsided or

remitted is essential to prevent relapse. Unfortu-

nately, non-compliance is common once the

patient’s   mood   has   improved,   especially   if

adverse effects are still troublesome.

Where  there  is  a  history  of  at  least  one

previous   depressive   episode,   up   to 3 years’

prophylactic therapy has been shown to limit

recurrence. Although some would argue that the

prophylactic dose level should be the same as

that used in the acute phase, lower doses tend to

be used.

Drug selection depends to some extent on a

variety  of  specific  patient  or  disease  factors,

most  of  which  are  related  to  tolerance  or

contra-indications (Table 6.13).  It  is  conven-

tional to distinguish different grades of depres-

sion for the purposes of treatment but there are

not clear thresholds and in the vast majority of

cases SSRIs are first-line when drug therapy is

indicated. Other groups are used when SSRIs fail

or are not tolerated, or in certain exceptional

circumstances.

system

Mild (minor) and moderate depression

Antidepressant medication is usually unneces-

sary for an isolated episode of mild depression

associated with an adverse life event: psycho-

therapy is the first choice. If drugs are needed,

treatment  can  be  stopped  when  the  patient

recovers; continuation and prophylactic therapy

are not indicated. No particular types are any

more effective and SSRIs are first choice. It has

been  claimed  that  MAOIs  are  more  effective

than others if there are atypical features such

as  anxiety,  phobia,  anger,  hypochondria  or

increased  appetite  and  sleep.  However,  the

evidence is not strong and their dietary precau-

tions limit their use. Pharmacotherapy is usually indicated in moderate depression.

For patients with anxiety, initial insomnia or mixed anxiety and depression, short courses of anxiolytics or short-acting hypnotics are some-

times used. However, sedative antidepressants (e.g. amitriptyline) are preferable, given in the evening if needed for sleep problems.

Severe depression

Drugs are almost always necessary in severe or psychotic  depression,  possibly  preceded  by  a course  of  ECT.  SSRIs  are  first-line  and  cyclics or  MAOIs  are  second-line.  Specialist  guide-

lines must be consulted when either switching antidepressants or combining them.

Patients suffering from psychotic symptoms,

especially  hallucinations,  may  benefit  from  a

short  course  of  an  antipsychotic  such  as

haloperidol. Amoxapine and thioxanthenes (e.g.

flupentixol)   have   both   been   recommended

because  they  have  mixed  antidepressant  and

sedative-antipsychotic actions. (In the case of

the former, these effects are due to the metabo-

lite loxapine.) However, there is little evidence

to support these uses.

Where suicide is a suspected risk, the least

toxic agents are essential, especially those with a lower convulsant potential, dispensed in small quantities and preferably under supervision. ECT may also be used.

Contra-indications or intolerance of adverse

effects such as antimuscarinic effects or sedation

can usually be circumvented by judicious drug

selection (see Table 6.13). If there evidence of

a  recurrent  pattern,  lithium  therapy  may  be

considered. With a recurrent unipolar pattern,

lithium may be as effective as tricyclics, but the

need  for  plasma-level  monitoring  is  incon-

venient. However, lithium is the drug of choice in

bipolar illness, i.e. with manic phases (p. 406).

Resistant depression

For the minority of patients who do not improve

after 6-8 weeks on optimal doses of a first-line

antidepressant, measuring the drug plasma level

might first be considered to check for possible

compliance   problems   or   abnormal   drug

handling. Otherwise, changing to another anti-

depressant that acts in a different manner, such

ia and manic-depressive disorder          403

as from a noradrenaline (norepinephrine) to a selective   serotonin   re-uptake   inhibitor,   may help.  A  course  of  ECT  could  be  considered. Thereafter, a wide variety of combinations are employed by psychiatrists, the only logical basis of  which  is  that  components  have  different modes of action (Figure 6.10).

Adding   lithium   to   other   antidepressants (lithium   augmentation)   seems   particularly successful. Switching to venlafaxine has also been successful. Levothyroxine (thyroxine) augmenta-

tion is also used, presumably partly on the basis that hypothyroidism is often associated with

depression.  Tryptophan  augmentation  can  be done on a named patient basis only.

Less common and based on even less evidence

are the addition of olanzapine  plus fluoxetine,

lamotrigine, high-dose tricyclics, or a short course

of  steroids (e.g.  dexamethasone 3 mg/day  for

4 days). Use of these agents and combinations requires close monitoring and careful selection of agent and dose, and is too specialized to

consider here.

Mania and manic-depressive disorder

Mania

Mania is a severe, usually recurrent, psychotic

affective  disorder  that  is  almost  the  precise

opposite of severe depression, although about

one-tenth as common. There is an abnormally

elevated   mood            (i.e.   euphoria   rather   than

dysphoria), unwarranted optimism, exuberance,

over-confidence, inflated self-esteem, hyperac-

tivity of thought and action, excessive libido,

and little sleep. The patient has increased drive

and is outgoing, but often socially tactless. The

over-confidence is similar to that experienced by

otherwise normal individuals during the usually

pleasurable sensations of early alcoholic inebria-

tion.  In  the  same  way,  patients  enjoy  their

episodes of mania, when they also often feel

more creative, thus making compliance with

treatment a problem.

A full-blown manic attack usually lasts no

more than a few days, during which the patient

sleeps  little  but  appears  to  have  boundless

energy. That level of arousal, if sustained for

longer,  is  potentially  life-threatening.  Hypo-

mania is more usual, where symptoms are less

florid.  During  a  manic  phase  sufferers  may

require   compulsory   admission   to   hospital

(‘sectioning’; see p. 421) but others manage to

function adequately in the community. At the

other end of the spectrum an even milder form

is seen as a (potentially advantageous) person-

ality trait in some individuals. Such an indi-

vidual retains insight, and may be thought of as

simply a flamboyant character, a tireless and

exuberant go-getter, a workaholic, even (collo-

quially) as a ‘Don Juan’ or a ‘nymphomaniac’.

Attacks  are  extremely  disruptive  socially,

domestically  and  financially,  because  over-

confidence causes patients to undertake wildly

ambitious   commitments.   Patients   discover

when  they  finally  ‘come  down’  that  they

are  hopelessly  over-committed,  double-booked and in serious debt.

The accelerated mental processes produce a

short attention span, flight of ideas and pressure

of speech as the patient jumps rapidly from

subject   to   subject   on   apparently   irrational

grounds, such as a rhyme or puns (the ‘clang

association’).  The  mood  too  may  suddenly

change from euphoria to irritability and aggres-

siveness, possibly out of frustration. The patient

may  thus  appear  quite  incoherent,  and  the

apparent  thought  disorder  and  delusions  are

not  easily  distinguished  from  acute  schizo-

phrenia. The differential diagnosis depends on

whether mood disturbance or thought disorder

predominates.  An  illness  with  elements  of

both  is  known  as  schizoaffective  disorder.

Clearly  the  lay  term ‘maniac’  is  related,  but manic  patients  are  no  more  criminal  or

violent than the general population.

Psychotic features such as lack of insight and

delusions of grandeur, the counterpart of the

depressive’s   delusions   of   worthlessness,   are

common during attacks, but the patient may

seem quite normal and rational between them.

Attacks rarely last more than a few weeks before

the attention that patients draw to themselves

means they are soon treated, although often

involuntarily.   Otherwise   the   patient   would

become exhausted. Mood returns to normal after

a few months of therapy, or may swing into

depression.

Manic-depressive disorder

Most  patients  experiencing  mania  also  have

episodes  of  severe  depression,  although  the

frequency,  sequence  and  duration  of  mood

swings  vary  enormously  between  patients.

Depressive phases usually last longer than manic

ones and are less obvious, and there are variable

periods of normality between extreme swings.

Mood swings can be quite abrupt, with manic

phases tending to have a more rapid onset. Rapid

cycling is when there are four or more swings per

year. Fluctuations become more frequent, and

more frequently depressive rather than manic, as

the patient ages.

Where the symptoms are serious or psychotic, the terms manic-depressive or bipolar  affec-

tive  disorder  are  used.  If  manic  attacks  are quite frequent or predominate it is known as

type I. Occasional hypomania with predominant depression is type II.

An otherwise normal person with moderately

exaggerated mood swings may be described as

having a cyclothymic personality; this is not a

psychotic illness, and some would argue not an

illness at all.

Aetiology and pathology

Major depression, mania and manic-depressive

disorder   may   have   similar   aetiologies.   The

ia and manic-depressive disorder          405

underlying abnormality could be in the regula-

tion of emotional responses, both low or high,

by the limbic system, and involving monoamine

neurotransmitters,   possibly   dopamine   and

GABA.  This  hypothesis  is  supported  by  the

prophylactic effectiveness of the single agent

lithium in all three. In bipolar disease, however,

cyclic antidepressants, SSRIs and MAOIs are not

just ineffective but can actually be dangerous

because they can trigger an exaggerated upswing

from depression to mania. An electrolyte imbal-

ance has also been proposed to account for

cycling.

As with severe recurrent depression, mania is a genetically linked, chronic but generally non-

progressive disease. External triggers for attacks may sometimes be found in stressful life events but equally, as with some severe episodes of

depression, there may be no apparent cause.

There is usually a family history and the concor-

dance rate in twins is over 70%. The disease

seems evenly spread worldwide.

Management

The aims of management are to:

•  control acute manic or hypomanic attacks;

•  minimize recurrence and intensity of mood

swings with prophylactic therapy.

In acute severe manic attacks, initial sedation

will be required. The drugs of choice are the

newer atypical antipsychotics (see p. 415), partic-

ularly  olanzapine,  risperidone  or  quetiapine  for

psychotic features (e.g. loss of insight, grandiose

delusions), or the mood stabilizer valproate. At the

same time prophylactic lithium can be started, but

it takes 7-10 days to take effect. Patients may

need  compulsory  admission  and  treatment  if

they are a risk to themselves or others. ECT is an

alternative to lithium in the acute phase.

If the patient is already taking lithium  the

plasma level must be checked: often patients will deliberately  stop,  either  because  of  adverse effects or to induce hypomania. (This is an inter-

esting  inversion  of  recreational  drug  misuse, which more usually involves taking something to get high, rather than avoiding  it.) Otherwise,

lithium therapy is started.

If  the  patient  demonstrates  an  alternating

pattern (bipolar disorder) or a recurrent pattern,

which is more common with mania than depres-

sion, they must be assessed for maintenance

therapy. The drug of choice in the prophylaxis of

recurrent mania or bipolar disorder is lithium.

This   is   a   potentially   lifelong   commitment,

including   mandatory   regular   follow-up   to

monitor   both   compliance   and   toxicity.   As

patients age, they will be more likely to start

using medication for comorbidities that poten-

tially will interact with lithium (e.g. diuretics,

NSAIDs), so regular re-assessment is essential.

Lithium  is  the  only  antimanic  that  reduces

suicide   risk.   Theoretically,   long-term   anti-

psychotics could be used, but they are usually

avoided  because  of  adverse  effects.  Once  a

patient   is   stabilized,   social   and   psycho-

therapeutic interventions should be introduced.

Not all patients respond to either of these tradi-

tional  treatments,  and  a  number  of  other

approaches have become established as second-

line alternatives (Table 6.14). The most promising

are anticonvulsant drugs such as carbamazepine,

valproate or lamotrigine, and topiramate (p. 448).

These  drugs  may  be  more  effective  in  rapid

cyclers, or in those who fail to respond to lithium.

Additional   treatment   during   a   depressive

phase of bipolar disease is problematic because

of the propensity of antidepressants to activate mania. The usual antidepressant drugs are used, but with more than usual care.

Lithium

Lithium  is an enigmatic drug. A monovalent

cation seems an unlikely candidate for the stabi-

lization of major affective disturbances. It closely

resembles the sodium ion in charge and ionic

size, and is distributed similarly in body water,

which has lent support to body-water/electrolyte

theories of affective disorder. However, lithium

has also been shown to affect monoamine levels

and post-receptor intracellular signalling.

Indications.   For   single   depressive   episodes

lithium is similar to amitriptyline in potency and

onset. However, its real value lies in the mainte-

nance  prophylaxis  of  recurrent  unipolar  or

bipolar affective disorder. Here it is a safer and

perhaps  more  effective  proposition  than  the

cyclics or SSRIs, and safer than antipsychotics or

benzodiazepines, despite the need for regular

plasma level monitoring.

Lithium  treatment  should  be  considered  if

there is evidence of recurrent episodes, either of

severe depression or mania (more than one in

2-3 years). Lithium  will reduce the frequency,

duration  and  intensity  of  mood  swings  or

abolish them. Some patients complain that it

produces a flattening of normal mood (blunting

of  affect),  and  manic  patients  may  comply

poorly because they value or enjoy some aspects

of their attacks of mania or hypomania. Some

believe themselves to be more creative in that

state.

Dosage  and  side-effects.   Lithium  therapy is

managed by monitoring the serum level. For

acute  attacks,  up  to 1.2-1.5 mmol/L  may  be

needed initially. For prophylaxis, the goal is to

maintain a plasma level within the narrow ther-

apeutic   window (0.6-1.0 mmol/L);   in   many

patients control is obtained below 0.8 mmol/L.

Dosage should aim for a stable plasma level

throughout   the   day   because   acute   adverse

effects, such as nausea, polyuria, polydipsia and

fine tremor, seem to be related to post-dose peak

levels (Table 6.15). Thus, frequent daily doses or

modified-release preparations are preferred, and

the risk of renal damage is reduced by once-daily dosing.   Change   of   formulation   should   be avoided or done gradually. Dose changes should also  be  gradual.  Many  minor  adverse  effects remit on prolonged use.

Plasma  levels  of  lithium  above  1.5 mmol/L

produce warning signs of toxicity including diar-

rhoea and vomiting, coarsening of the tremor

and CNS depression. Sustained toxic levels above

2.5 mmol/L cause hypotension, convulsions and coma, and are potentially fatal; in such circum-

stances dialysis may be needed to bring about sufficiently rapid reversal.

Hypothyroidism is quite a common chronic effect.  The  significance  of  long-term  kidney changes   remains   controversial   and   only   a minority are affected. Renal and thyroid func-

tion tests are recommended before commencing lithium therapy, and regularly thereafter.

Withdrawal of lithium, if it becomes necessary, must be gradual, staged over at least 4-12 weeks, to reduce the likelihood of rebound hypomanic symptoms or relapse.

ia and manic-depressive disorder          407

Monitoring.   Regular measurement of plasma

lithium  (12 h  post-dose)  is  essential,  at  first

weekly until a stable level is obtained. Subse-

quently the interval may be increased (up to 3-4

months) in patients who understand the precau-

tions.  Thyroid  and  renal  functions  tests (see

Chapters 9 and 14) should be done at the same

time.   Extra   measurement   is   indicated   in

suspected drug interaction or poor compliance,

intercurrent  illness,  other  circumstances  that

interfere with lithium level, or unexpected toxi-

city. Table 6.16 shows common circumstances

that can potentiate lithium level and/or toxicity.

Pharmacokinetics and interactions.   Lithium

is distributed throughout body water, generally

following  the  same  pattern  as  sodium.  It  is

cleared entirely by the kidney, which gives rise to

many potential problems. Both fluid retention

(e.g.   from   NSAIDs)   and   circumstances   that

promote increased renal sodium reabsorption

(such  as  pyrexia,  unaccustomed  hot  climate,

electrolyte  depletion,  dehydration  or  diuretic

therapy) tend to reduce lithium clearance and

can   cause   dangerously   high   plasma   levels.

Patients must be warned about this and those

with renal or cardiovascular impairment need

close   attention.   Interactions   are   important

because of the narrow therapeutic index (Table

6.16). While most interactions increase lithium

levels,  antacids  and  theophylline  can  reduce

them.

Clearance is increased in pregnancy. However, lithium may be teratogenic and use in the first trimester will depend on the relative risks of

withdrawal and continuation.

Despite these potential difficulties, lithium is

a valuable drug. It has no antimuscarinic or

extrapyramidal   effects,   causes   no   amnesia,

general CNS depression or psychomotor impair-

ment, and has little effect on normal mood.

Thus it has advantages over the other treatments

available for depression or mania. However, in

overdose it is no less toxic than conventional

antidepressants or antipsychotics. All patients

should be given a Lithium Treatment Card. Dili-

gent monitoring has enabled patients to take

lithium for decades with little ill effect.

system

Schizophrenia

Schizophrenia is a much misunderstood disease.

It  has  nothing  to  do  with  a  so-called ‘split

personality’ or with Jekyll and Hyde. In schizo-

phrenia, the split is between different compo-

nents  of  the  same  personality,  e.g.  between

mood and action, or behaviour and belief. The

normally integrated aspects of a healthy per-

sonality  become  fragmented -  literally  dis-

integrated. A more subtle solecism is the use of

‘schizophrenic’ as a synonym for ambivalent.

There is a very rare, genuine split personality

disorder.   More   properly   called   dissociative

disorder or multiple personality, in this extra-

ordinary condition a person alternates between

two  or  more  usually  completely  integrated,

rational personalities. This has nothing to do

with schizophrenia.

Although schizophrenia is generally what the

lay public understands by madness, the popular

idea (promoted by the mass media) that every

‘mad  axeman’  has  schizophrenia  is  wrong.

Patients with schizophrenia are no more violent

than   the   population   as   a   whole.   Violent criminals may be suffering from a ‘psychopathic personality’,  a  condition  quite  distinct  from schizophrenia.

It must not be assumed that people with schiz-

ophrenia, being psychotic and thus out of touch with reality and lacking insight, do not suffer

from their disease: they do. Moreover, there are lucid periods, possibly due to treatment, during which they have sufficient insight to remember their experiences when acutely ill. Many patients give a strong impression of perplexity and bewil-

derment. They are convinced of the truth of

their own reality and cannot reconcile this with the reaction this causes in others.

Schizophrenia is probably the most difficult

mental illness to understand because, unlike with

depression or anxiety, the experience of sufferers

is very remote from normal. It has been likened to

that twilight world between sleep and waking,

when it is difficult to distinguish between illusion

and reality. Schizophrenic patients also have diffi-

culty distinguishing between themselves and the

outside world. While for most of us dreaming,

thinking, saying and hearing are categorically

distinct phenomena, in schizophrenia the differ-

ences are blurred. Consequently, patients may

feel  irrational  external  influences  on  their

thoughts, or that their thoughts are available for

all the world to read. This loss of the ultimate

privacy - that of our thoughts - is possibly what

makes schizophrenia so miserable and confusing

for the sufferer.

Aetiology and pathology

Schizophrenia is a common condition: about 1%

of the population are likely to have at least one

episode during their life, although not all will

become permanently ill. Nevertheless, schizo-

phrenia accounts for the majority of patients in

continuous psychiatric care. The condition has

a  uniform  global  prevalence.  The  apparent

increased  prevalence  among  lower  socioeco-

nomic groups is attributed to the ‘downward

social  drift’  of  sufferers,  who  have  difficulty

maintaining jobs, eduction or relationships.

The  causes  of  schizophrenia  are  unknown.

Both genetic predisposition (susceptibility), early

Schizophrenia   409

development and various environmental factors

are important, as has been confirmed by studies

on twins born to schizophrenic parents. Several

potential genes or gene linkages have been iden-

tified. Among the biological causes suggested are

autoimmune  or  viral  encephalitis,  abnormally

large brain ventricles (causing reduced functional

brain size), or imbalance between the right and

left hemispheres. Recently, research has focused

on neurodevelopmental abnormalities arising at

a very early age.

Previously, postmortem studies on the brains

of schizophrenic patients failed to distinguish

the effects of chronic disease from those of many

years of antipsychotic therapy. However, newer

non-invasive   imaging   techniques,   such   as

positron emission tomography (PET) and MRI,

which visualize the living brain and monitor

changes in its activity with far greater precision

than the relatively crude electroencephalogram

(EEG), reveal abnormalities in frontal, thalamic

and cerebral areas. These are seen in newly diag-

nosed, drug-naïve patients so cannot be the result

of treatment. But it is still unknown whether

there are structural abnormalities in various brain

centres  or  defective  interconnections between

key centres.

It could be speculated that the classical symp-

toms  of  inappropriate  mood,  delusions  and

hallucinations imply involvement of the limbic

system,  which  is  concerned  with  emotional

responses and beliefs, and the ascending reticular

system, concerned with monitoring and filtering

of perceptions. On the other hand, the negative

symptoms typical of chronic schizophrenia may

involve lesions of frontal lobe cortical areas, and

the impaired thought processes suggest cortical

involvement.

The  predominant  biochemical  abnormality

seems to be functional over-activity of dopamin-

ergic pathways between midbrain and certain

cortical areas (the mesolimbic and mesocortical

tracts).  This  is  compatible  with  the  known

dopamine-blocking action of most antipsychotic

drugs. Whether dopamine excess is a cause or an

incidental  consequence  has  not  been  estab-

lished, and uncertainty remains about the rela-

tive  roles  of  dopamine  DA1,  DA2 and  DA3-6

receptors. GABA and 5-HT pathways are also

probably involved.

Functional theories involve ideas of incom-

plete adjustment to society, especially to the

family. The ‘antipsychiatrists’ suggest, on the

other hand, that schizophrenia is a response to

an irrational, contradictory or hostile world.

A holistic synthesis of these various theories

might propose a genetic predisposition confer-

ring  susceptibility,  maladaption  in  early  life,

followed  by  environmental  triggers  such  as

drug misuse, stress or a hostile family or social

environment,  ultimately  leading  to  disease

onset.

In view of the uncertainty about causes, the

management of schizophrenia remains essen-

tially    symptomatic:    relieve    the    patient’s

suffering, perhaps slow the progression, and help

the patient to cope. As yet prevention is not

possible, nor can anything be done to reverse the

disease process.

Clinical features

Psychotic signs

The classification and symptomatology of schiz-

ophrenia are complex and have been endlessly debated among psychiatrists. Recently, the WHO and the American Psychiatric Association have rationalized and harmonized diagnostic criteria. Table 6.17 describes the common signs in terms of normal brain functions. All are descriptive

psychiatric signs; there are no objective physical, biochemical or metabolic signs.

The defining clinical features of schizophrenia

are thought disorder and delusions. A delusion is

an ill-founded and irrational belief that never-

theless  is  implacably  held,  not  amenable  to

reasoned persuasion and completely at variance

with the patient’s religious, social or ethnic back-

ground. Many patients have very strange ideas

and   make   bizarre   associations,   sometimes

inventing their own language (neologism) or

using common words in an inappropriate way:

“I’m in hospital because of my minarets”. Some

have elaborate paranoid delusions: everybody is

spying  on  them  or  plotting  against  them,

including  relatives,  neighbours,  and  govern-

ments. They may also hallucinate, hearing the

voices of their tormentors talking about them,

system

or they see their own name in newspapers or on television (ideas of reference). No wonder they are so miserable.

Patients   may   believe   their   actions   are

controlled by others - often, among patients

from Western countries, by means of invisible

rays, magnets or electric wires. They may hear on

the  radio  an  echo  of  what  they  have  just

thought, or feel that they have foreign ideas

inserted into their brain. One patient believed he

had a telephone implanted in his head, through

which his every action was dictated to him by a

distant ‘friend’: no X-ray would have convinced

him otherwise. Often their emotional responses

are fatuous or inappropriate, e.g. laughing or

crying at the socially incorrect time; in others

mood variation is generally reduced. The combi-

nation   of   all   these   features   produces   the

markedly unusual behaviour popularly known as

‘madness’.

There is often a tragic consistency about a

patient’s  symptom  complex.  Their  delusions

reflect their hallucinations, or their hallucina-

tions confirm their delusions: both appear to

escape the logical censorship provided by fully

functional  insight.  However  bizarre  it  seems,

their behaviour may be consistent with their

misreading of reality. Someone who believed

that thoughts were constantly put into their

head against their will might well believe there

was a plot against them, especially if they heard

voices apparently confirming this. Their reaction

might well be aggressive behaviour, or shouting

out loud to tell the persons whose voices they

heard to go away. The novels of Franz Kafka give

us some inkling of the terrors of paranoia.

Classification

Two broad syndromes are recognized (Table 6.17).

Patients in the early acute stages have type 1

(classical or florid) schizophrenia with positive

symptoms. Loosely, these resemble disinihibited

exaggerations of normal activity. Type 2 features,

usually  seen  in  chronic  schizophrenia,  are

predominantly  negative,  such  as  flat  mood,

apathy,  social  withdrawal,  and  lack  of  speech

(alogia), pleasure (anhedonia) or initiative (avoli-

tion) are seen. These describe functions found in

normal persons but absent in schizophrenia.

Type 2 features may represent long-term dete-

rioration of chronic disease, burnt-out disease or

possibly  consequences  of  long-term  antipsy-

chotic use. However, careful history taking will

often reveal forerunners of these negative traits

in early life, e.g. a withdrawn lonely child. At

onset they are masked by the florid positive

features but once the latter are under pharmaco-

logical control the former, less affected by drugs,

emerge  as  the  predominant  signs  of  illness

(Figure 6.11).

Another  classification  groups  symptoms  as

either  cognitive (i.e.  thought  disorder),  posi-

tively psychotic (e.g. delusions and hallucina-

tions) or negative (e.g. apathy, social withdrawal,

lack of self-care).

Diagnosis

It is first necessary to eliminate any primary

underlying cause such as iatrogenic psychosis

(e.g. corticosteroids), drug misuse (e.g. ampheta-

mines), brain tumour or infection, head injury, certain rare forms of epilepsy, hyperthyroidism, etc. The symptoms must have been continuously present for at least a month, and usually with

evidence of deteriorating social functioning, at work and with family and friends.

Distinguishing   between   borderline   schizo-

phrenia and other psychoses (mainly mania and

severe depression) can sometimes be difficult.

Traditionally the presence of at least two symp-

toms from the list given in Table 6.17, or just one

‘first-rank’ symptom (positive symptoms, mostly

specific forms of delusion) is needed to confirm

schizophrenia.  Type  I  patients  usually  show

several positive symptoms and the diagnosis is

clear.

In  chronic  schizophrenia,  symptoms  are

mainly  negative  and  are  more  difficult  to

diagnose and to treat. Moreover they must be

distinguished from iatrogenic over-sedation or

extrapyramidal  effects,  clinical  depression  or

‘institutionalization’,  i.e.  the  dependency  and

apathy that can result from long-term hospital

care. The latest editions of both DSM-IV and

ICD-10 include negative  symptoms  as  primary

diagnostic  criteria.

In different cultures and times the symptoms

take different outward forms, but the overall

pattern and prevalence are consistent. Nowadays

the persecutors imagined by paranoid patients

are Martians or secret government spies, who

control  them  with  lasers  or  magnetism.  In

medieval Europe, and present-day pre-industrial

societies, devils or evil spirits using witchcraft or

curses are to blame. This is one disease for which

the stress of modern industrial society cannot be

held responsible.

Course and prognosis

The old medical Latin name for schizophrenia

was  dementia  praecox  (loosely,  the  madness  of

youth) because the most common time of onset is

late adolescence or early adulthood. There may be

a sudden deterioration, a form of ‘nervous break-

down’, which is the generic lay term for the acute

onset of any psychiatric condition. However, this

acute phase will usually have been preceded by a

prodromal  phase  with  gradual  reduction  in

social, academic or work-related functioning, loss

of friends, deterioration in personal hygiene or

other behaviour uncharacteristic of the patient’s

former personality.

About a quarter of patients will suffer just a single episode, then recover and lead normal

lives. Good prognostic signs are:

•  The absence of a family history of schizo-

            phrenia.

•  Stable premorbid personality. •  Acute onset.

•  Preservation of emotional responses, initiative

            and coherent personality.

•  Early recognition and treatment.

Conversely, a poor outlook is indicated by:

•  Positive family history.

•  Disturbed, eccentric, antisocial or withdrawn

            premorbid personality.

•  Difficulty in forming relationships, from an

            early age.

•  Disrupted domestic situation and poor social

            adjustment.

•  Insidious onset.

•  Loss of affect, initiative and drive. •  Delay in treatment.

Although most initially recover from the first

episode, about three-quarters of sufferers will

relapse and eventually enter a chronic phase

suffering gradual decline or repeated relapses

and remissions. This tends to be accompanied by

the development of type 2 features. They have

great difficulty forming relationships, do not

marry, cannot keep jobs and drop out of educa-

tion. Without treatment, follow-up and social or

family support they may drift down the social

scale, becoming progressively more involved in

vagrancy, petty crime, illicit drug use and alco-

holism. Some 10% of all patients need long-term

institutional care. The same proportion commits

suicide.

Between these extremes are those who, with

the   help   of   medication,   adjust   to   their

disability and manage to cope in the commu-

nity, while perhaps seeming just a trifle eccen-

tric.  The  importance  of  social  and  family

support  is  emphasized  by  the  fact  that  the

prognosis for patients in developing countries,

with  their  extended  families  and  perhaps

greater  tolerance  of  eccentricity  and  non-

conformity,  is  better  than  in  the  developed

world. Even in the UK, many people who are

obviously quite ‘mad’, but harmless and able to

look after themselves after a fashion, are free to

roam the streets.

Management

Drugs, psychotherapy and social interventions all

have  a  place  in  the  management  of  schizo-

phrenia. Most patients are managed by a combi-

nation  of  family  or  community  care  with

occasional  hospital  admission,  supported  by

maintenance antipsychotic drug therapy. Nowa-

days, few schizophrenic patients need permanent

institutional care.

Before the discovery of chlorpromazine in the

early 1950s, however, things were very different.

Little  could  be  done  for  people  with  severe

mental illness, if help was offered at all. There

was only heavy sedation with barbiturates or,

before   them,   bromides   and   straitjackets.

Community care was unheard of: the idea was

to keep ‘maniacs’ as far from ‘normal’ people as

possible.

Schizophrenia   413

In the UK in the 19th century, the enlightened

Victorians built asylums. Not then a pejorative

term,  asylum  implied  protection  rather  than

imprisonment and neglect. In these enormous

rambling institutions built at a safe distance

outside the big cities, custodial care may have

been  the  ethos  but  sedatives,  locked  doors,

spiked walls and padded rooms were still the

means (Figure 6.12).

Aims

The aims in managing schizophrenia are to:

•           control acute attacks and prevent self-harm or

harm to others;

•           attend to social and domestic factors;

•           rehabilitate the patient if possible;

•           start  long-term  support  and  maintenance

therapy as appropriate.

psychotherapy

Conventional psychotherapy is of little benefit

in acute schizophrenia, and psychoanalysis even

less   so,   because   patients   have   no   insight.

However,  simple  counselling  can  help  many

patients  to  adjust  to  a  chronic  illness  while

remaining   in   the   community.   Coping-skills

training,  occupational  therapy  and  hostel  or

‘halfway   house’   accommodation   may   be

arranged.  CBT  has  also  been  found  helpful.

Family  education  and  social  therapy,  where

people in contact with the patient can be told

what  to  expect  and  are  encouraged  to  be

supportive, can be very helpful in maintaining

remission. Stormy home relationships and a lack

of  acceptance  of  the  patient  are  the  most

common   reasons   for   relapse,   whereas   a

supportive home environment greatly reduces

the chances of relapse.

For most types of psychotherapy a degree of

insight is required, and drug therapy is usually

needed to establish this. Once patients have been

stabilized, such interventions can be instituted.

It must be emphasized that most patients who

are concordant with their medication and have a

relatively stable domestic and social environ-

ment are to a greater or lesser extent lucid, have

autonomy, can make decisions and should be

encouraged to participate in treatment choices.

Even more than a disruptive social environment,

abruptly stopping medication is the greatest risk

factor for relapse.

Drug therapy

When discovered in the 1950s the antipsychotic

drugs, also less accurately termed neuroleptics,

antischizophrenic agents or major tranquillizers,

revolutionized the care of psychotic illness. For

the first time many patients who would other-

wise have languished in deluded misery inside

custodial institutions were freed from both their

mental and their physical prisons. The prototype

chlorpromazine, perhaps surprisingly, is still used.

However, steady development since the 1950s

has  resulted  in  significant  improvements  in

efficacy, safety and formulation. Antipsychotic

drugs  are  now  divided,  although  not  so

system

distinctly as some manufacturers would like it

believed,  into  two  broad  classes.  The  ‘typical’

antipsychotics include original phenothiazines

(e.g.  chlorpromazine)  and  the  butyrophenones

(e.g.  haloperidol).  The ‘atypical’  antipsychotics

include clozapine and olanzapine. These groups

are distinguished mainly on the basis of their

side-effect profile.

Pharmacodynamic action

The proprietary name Largactil was chosen for

the   original   phenothiazine,   chlorpromazine,

rather prosaically because of its large number

of  actions.  Its  structural  similarity  to  several

natural neurotransmitter amines gives it antag-

onist activity on cholinergic, histaminic, alpha-

adrenergic,  serotoninergic  and  dopaminergic

receptors.  Structural  alterations  during  subse-

quent development have produced a number of

groups of antipsychotics with varying receptor

affinities, giving a diverse range of therapeutic

and adverse profiles (Tables 6.18 and 6.19).

Unfortunately, despite modern non-invasive

imaging of the living brain, there is still much to

learn about the relationship between receptor

blockade,  localization  of  CNS  activity  and

antipsychotic action. One problem is that little is

known of the inter-relationships and interdepen-

dencies between different receptor systems. For

example, although most of the original antipsy-

chotics are potent dopamine blockers, the newer

ones preferentially target 5-HT receptors. This

may be because one type of receptor is upstream

or downstream of the other, with actions at either

having a similar outcome. Again, the adrenergic

alpha2 activity  shown  by  some  agents  could

enhance dopaminergic transmission in the BG

and thus reduce extrapyramidal adverse effects.

Typical antipsychotics

Among the typical antipsychotics, antipsychotic

action  correlates  well  with  blockade  of  the

dopamine D2  receptors assumed to be in the

thalamus, limbic system and cortical projections

of the ascending reticular formation. However,

although the onset of antipsychotic action takes

weeks, receptor blockade occurs within hours of

starting therapy. A similar phenomenon is seen

with antidepressants, and another similarity is

that measurements of changes in the metabolite of the presumed transmitter (homovanillic acid in the case of dopamine) do not correlate with clinical activity.

Dopamine blockade yields other useful actions

(e.g. prochlorperazine used as an anti-emetic) and

many adverse ones, including extrapyramidal

symptoms    and    endocrine     (hypothalamic)

actions.  There  are  at  least  five  subtypes  of

dopamine receptor, and research is attempting to

differentiate these in terms of their anatomical

location and clinical or adverse effect.

Actions  at  other  receptors  also  have  either therapeutic or adverse actions (see Table 6.18). Antimuscarinic   action   may   reduce   extra-

pyramidal  symptoms;  antihistamine  action  is sedative,   exploited   therapeutically   as   with promethazine;  adrenergic  blockade  may  cause postural  hypotension;  and  serotonin  blockade has antipsychotic action.

Following the initial prolific development of

the phenothiazines, other chemical structures

were developed (Table 6.19). These are usually

more specific for dopamine D2  receptors, e.g.

butyrophenones, which appears to accentuate

the antipsychotic clinical and extrapyramidal

adverse actions, but reduces autonomic adverse

Schizophrenia   415

effects. The thioxanthene and benzamide groups are claimed to have, in addition, stimulant and antidepressant activities.

Atypical antipsychotics

The dibenzodiazepines (Table 6.19) are forerun-

ners of the atypical antipsychotics, which have

arguably greater clinical activity combined with

unarguably a significant reduction in extrapyra-

midal adverse effects. They mark a return to

broad-spectrum receptor blockade, but with a

greater relative affinity for 5-HT receptors over

D2  receptors. One of the most recent, aripipra-

zole, appears to be a partial dopamine agonist,

with blocking activity in the presence of high

dopamine levels but mild agonist activity in the

absence of dopamine. Additionally it has activity

at 5-HT receptors.

The advantages of this group lie in superior

activity in treatment-resistant disease, significant

activity against negative symptoms, and freedom

from  EPS,  including  tardive  dyskinesia (see

p. 419). Generally, atypical antipsychotics are as

effective  on  positive  symptoms  as  the  older

agents and share the same autonomic adverse

effects of cholinergic and adrenergic blockade.

Weight gain is greater but hyperprolactinaemia

less. Experience is not yet sufficient to be certain, but some may have a lower incidence of tardive dyskinesia  and  perhaps  neuroleptic  malignant syndrome. There may be a lower risk of suicide with  clozapine.  Nevertheless,  this  is  not  a

completely  homogeneous  group:  there  are

important  differences  in  their  adverse  effects (Table 6.23; compare with Table 6.19).

Whether or not the atypicals really are clini-

cally more effective than the typicals in symptom

control is still debated. Initial evidence suggesting

they were better than the typicals in controlling

negative symptoms and resistant schizophrenia

has been disputed. Re-analysis has indicated that

the use of haloperidol as the comparator drug, and

the doses used, make the results less clear-cut. The

CATIE trial implied that careful selection of the

right agent for a patient is more important than

the class of drug used. The difference, if it does

exist, is unlikely to be great. However, the unique

superiority of clozapine in resistant schizophrenia

is now firmly accepted.

Schizophrenia   417

Experience with clozapine is greatest, and it is

now the benchmark. Unfortunately, because of its

propensity to cause bone marrow suppression, its

use necessitates patient registration on an obliga-

tory, costly and inconvenient blood monitoring

scheme. However, there are now several alterna-

tives.  Mainly  because  of  possible  orthostatic

hypotension, most atypical agents need careful

initiation  and  subsequent  dose  titration.

Psychotropic actions

These drugs have a wide spectrum of psycho-

tropic actions (Table 6.20). Their ability to sedate

without  general  impairment  of  consciousness

(i.e.   tranquillize)   gives   them   an   anxiolytic,

tension-relieving   effect.   The   antipsychotic

action is a remarkable ability to banish halluci-

nations, diminish the power of delusions and

straighten  out  distorted  thought;  this  is  the

origin  of their description as ‘major tranquil-

lizers’.  Psychomotor  inhibition  is  a  specific

depression of overactive thought and physical

activity, again not at the expense of conscious-

ness. Some antipsychotics have a stimulant or

alerting  effect  on  withdrawn  patients,  and

others also a mood elevating or antidepressant

action.   These   properties  are  shown  to  a

different degree by different groups and their

usefulness in treating the common target symp-

toms of various psychiatric illnesses is indicated

in Table 6.20.

Pharmacokinetics and administration

No consistent predictions about antipsychotic

effectiveness in a given patient can be made

either from the dose used or from plasma level

measurements. Most antipsychotics have a half-

life        24 h,  so  single  daily  doses  are  usually

adequate in the maintenance phase; evening

usually is the best time, especially if a sedating

effect is required. There is often a first-pass effect,

so parenteral doses are usually lower than oral

ones. The apparent volume of distribution is

high owing to the lipophilic nature and conse-

quent accumulation in the CNS. Thus, although

plasma binding is usually quite high (e.g. 95% for chlorpromazine), this does not present any potential  interaction  problems.  Clearance  is usually hepatic (an exception is the benzamide group), so the potential exists for hepatic drug interactions.  This  is  also  important  because hepatotoxicity occasionally occurs.

Recommended   maximum   doses   are   only guides: patient response is the principal crite-

rion. Very high doses may be given if adverse effects are absent or tolerable, provided that a clinical effect is achieved. However, persistently high doses must be avoided, especially after the acute  phase  has  been  controlled.  The  Royal College of Psychiatrists has issued precautions about high dose therapy (see BNF).

Side-effects

Many of the adverse effects of the antipsychotics

derive  from  their  various  pharmacodynamic

actions and so are, in principle, predictable. The

widest spectrum of side-effects is seen with the

phenothiazines (Table 6.21), but the prominence

of different adverse effects varies between groups

system

(see Table 6.19). Among the typicals, the greater

the antipsychotic potency, the fewer the auto-

nomic effects and the more likely the EPS. For

the atypicals this trend has been reversed, with

potent  antipsychotics  having  greatly  reduced

EPS.

There are potentially serious non-specific or

idiosyncratic effects. Jaundice and photosensi-

tivity are more common than agranulocytosis.

Clozapine  is particularly liable to depress the

white  cell  count (incidence  approx. 1%  of

patients  per  year)  and  it  is  mandatory  that

patients are regularly monitored. As with many

psychotropic drugs, seizure threshold is lowered,

a problem for epileptics at normal doses and for

all  patients  in  overdose.  ECG  changes  with

prolongation of the QT interval and potential

arrhythmias can occur, especially with thiori-

dazine and, less commonly, the atypicals (usually

in  combination  with  other  drugs).  Atypical

agents should not be used in the elderly with

dementia because of a risk of stroke or death.

Autonomic blockade

The  antimuscarinic  actions  and  consequent precautions are similar to those of the tricyclic antidepressants  (Table  6.11).  Peripheral  alpha-

adrenergic  blockade  can  cause  cardiovascular problems,  particularly  postural  hypotension with reflex tachycardia. Autonomic symptoms tend to remit with chronic use.

Endocrine effects

Blocking   dopamine   in   the   hypothalamus

inhibits  some  endocrine  mechanisms.  Most

important  is  the  rare  but  potentially  fatal

neuroleptic  malignant  syndrome,  which  is

probably hypothalamic in origin. The syndrome

involves hyperthermia, muscle spasm, impaired

consciousness and cardiovascular instability, and

has a 10% mortality. Treatment is with dopamin-

ergic agents (e.g. bromocriptine), muscle relaxants

(e.g. benzodiazepines, dantrolene) and antimus-

carinics (e.g. procyclidine), as well as cooling and

rehydration.

Hyperprolactinaemia is quite common and

has many consequences that are unacceptable to many patients, such as galactorrhoea, amenor-

rhoea, gynaecomastia, and sexual dysfunction (loss of libido and impotence).

Less serious, but equally likely to discourage compliance, is weight gain. The atypicals are

particularly associated with this, notably cloza-

pine and olanzapine, and they can also cause or precipitate diabetes, a condition that is anyway more common even among untreated schizo-

phrenics. Thus regular monitoring of weight and blood glucose is essential.

Extrapyramidal syndromes (EPS)

Movement disorders, although usually harmless,

are  a  major  cause  of  non-compliance  among

schizophrenia  patients.  A  drawback  of  using

dopamine blockers for antipsychotic action in

the limbic and reticular systems is that dopamine

is  also  a  transmitter  crucial  to  the  motor-

controlling functions of the BG. (For a fuller

discussion, see pp. 368-370 and pp. 427-428.)

Both systems seem to involve D2  receptors, for

which most antipsychotics have a high affinity.

Thus disturbance of fine motor control has been

thought  almost  inevitable  with  potent  anti-

psychotics: the two effects seemed inextricable.

There are currently three ways around the

problem:

•  Intrinsic antimuscarinic activity.

•  Selective affinity for dopamine receptors in

            the limbic system.

•  Balance  between  5-HT  and  D2  receptor

blockade

Normally, dopaminergic action in the BG is

counterbalanced by cholinergic activity, whereas

in  the  areas  presumed  to  be  disturbed  in

psychosis, dopamine does not appear to have a

natural antagonist. Thus it is possible to coun-

teract the adverse effects of dopamine blockade

with  a  centrally  acting  antimuscarinic  drug

without significantly diminishing the antipsy-

chotic action: this effect is used to treat some

antipsychotic-induced   EPS.   Note   that,   by

contrast, using the anti-Parkinson drug levodopa,

although it would be effective, would also nullify

the   antipsychotic   activity.   Some   standard

antipsychotics with high antimuscarinic activity,

especially thioridazine, have a lower incidence of

EPS  and  so  reduce  the  need  for  ancillary

antimuscarinic drugs.

Most atypicals have a far lower incidence of

EPS, perhaps because of a selective affinity for

receptors in limbic and cortical areas, but not in

the BG. In addition, the antipsychotic activity of

most  atypicals,  especially  clozapine,  is  more

closely correlated with blockade of 5-HT2 recep-

tors, an action that has little direct effect on

motor coordination but may indirectly prevent

it being disturbed.

EPS can be classified into four groups (Table

6.22). Unfortunately, there are no predictors of

which type will occur to which patient, or when.

Acute dystonia.   Some patients react with an

alarming acute muscle spasm on the first dose or

within   the   first   few   days   of   antipsychotic

therapy, especially if given in high doses or by injection. This usually occurs in the head and

neck region: commonly it presents as an exagger-

ated and uncontrolled rolling upwards of the eyes

(oculogyric  crisis),  a  stiff  jaw  or  a  hyper-

extended  neck.  Occasionally  there  may  be  a

dangerous  choking  laryngospasm.  Fortunately

these reactions are easily treated by parenteral

antimuscarinics (e.g. procyclidine) but they can

severely damage the patient’s confidence in the

therapy.

Pseudo-parkinsonism.   Early in therapy up to 50% of patients develop a motor incoordination syndrome very similar to idiopathic Parkinson’s disease. (Note that Parkinson’s disease involves dopamine  deficit;  pp. 427-428).  Almost  any parkinsonian symptom can occur, and though iatrogenic  parkinsonism  often  remits  sponta-

neously after a few months of treatment, this is of little comfort to the patient, who loses confidence and may become non-compliant.

The  temptation  to  use  oral  antimuscarinic

anti-Parkinson drugs prophylactically is strong.

In  the  past  they  have  been  given  routinely.

However, this is now uncommon because they

themselves may cause various psychotomimetic

effects such as delirium; they have even been

misused  for  such  effects.  In  addition,  their

adverse antimuscarinic effects would be additive

to  those  of  antipsychotics  themselves.  Thus

strong efforts are made to encourage the patient

to tolerate the symptoms without anti-Parkinson

drugs until they eventually subside.

Akathisia.   Marked  restlessness  and  anxiety

seem to be the most disturbing effects for many

patients. Akathisia follows a similar course to

pseudo-parkinsonism  but  responds  poorly  to

conventional   anti-Parkinson   therapy.   Some

patients develop a more persistent form, which

resembles tardive dyskinesia. Sometimes a short

course   of   benzodiazepines   may   help,   and

lipophilic  centrally-acting  beta-blockers  have

also been used.

Tardive dyskinesia (TD).   A form of orofacial

dyskinesia, TD may develop after months or

years of successful therapy, or even after drugs

have  been  withdrawn.  Its  bizarre  symptoms

involve lip-smacking, chewing and grimacing

facial   expressions.   Although   not   directly

distressing for the patient, these provoke unsym-

pathetic   or   hostile   reactions   in   onlookers

because they give the patient the appearance of

the popular idea of ‘craziness’, when paradoxi-

cally the drugs responsible for this bizarre behav-

iour   are   in   fact   controlling   the   psychotic

symptoms.

The pathology of TD is different from that of

other EPS and is poorly understood. It is found

sometimes in schizophrenics who have never

been  treated,  but  certainly  antipsychotics  do

increase its occurrence. The cause may be the

development of dopamine receptor supersensi-

tivity. Prevalence rates of up to 50% and annual

incidences of 5% have been reported, the elderly

being   particularly   susceptible.   It   has   been

suggested  that  TD  is  actually  a  symptom  of

severe schizophrenia rather than iatrogenic, and

these patients are the most likely to be taking

high doses of potent antipsychotics.

TD is unpredictable, but seems to occur more

commonly after long courses and high doses of

antipsychotics. Intermittent therapy, e.g. drug

holidays,  depot  therapy,  and  antimuscarinic

therapy seem particularly to predispose patients

to TD. Perversely, reducing the antipsychotic

dose  temporarily  intensifies  symptoms,  while

increasing the dose may alleviate them.

Aside from a correlation with potency, gener-

ally no one drug is more likely to cause TD than another. However, clozapine definitely causes TD less often and indeed is a drug of choice when patients develop TD on other drugs. Other atyp-

icals would be expected also to be favourable owing to their low EPS potential; however, the evidence for them is not as clear.

Currently the best strategy seems to be to

reserve antipsychotics for serious psychosis and keep doses as low as possible. If TD occurs, it may only be mild and, with the agreement of the

patient and his or her family, may be ignored, especially if the patient would be expected to

relapse without antipsychotics.

Various  drugs  have  been  used,  including

vitamin E, clonazepam, nifedipine, sodium valproate,

reserpine and choline, with little success. Currently

clozapine  and  quetiapine  are  the  best  options.

Antimuscarinic drugs should be stopped. If TD

cannot be controlled, the antipsychotic must be

Schizophrenia   421

gradually reduced until the dyskinesia remits. If psychotic  symptoms  then  recur,  re-starting

antipsychotic  therapy  with  a  different  agent (especially  clozapine)  may  be  possible  without recurrence of TD.

Drug selection

The issue of whether or not to start with an atyp-

ical is still disputed by some, but the current

consensus and the recommendation by NICE is

that   atypicals   are   preferred.   Their   superior

adverse    effects    profile,    with    consequent

improvements in compliance, is now considered

to outweigh by far the extra cost, both in phar-

macoeconomic and patient satisfaction terms.

Depending on which drug a patient is taking

when  treatment  is  being (re)considered,  the

progression is to start with an atypical, usually

olanzapine or risperidone, then move on to cloza-

pine  if control is not achieved (Figure 6.13).

However, if a patient is stabilized on a typical

agent and is content, there is no need to change

to an atypical.

Acute attack

First onset.   It is now clear that prompt appro-

priate antipsychotic treatment at as early a stage

as possible in the course of a developing schizo-

phrenic  illness  is  likely  to  produce  a  better

outcome.  Although  it  is  not  yet  possible  to

prevent the illness progressing, therapy prolongs

remission and reduces relapses and gives the

patient the best chance of maintaining social

functioning. To ensure the new patient is least

disturbed by side-effects and thus encouraged to

continue with maintenance medication, atypi-

cals  are  first  choice.  Wherever  possible,  the

patient  should  be  involved  in  the  decision

about initial drug therapy.

Emergency tranquillization.   An acutely psy-

chotic patient is likely to be deluded, halluci-

nating,  incomprehensible  and  quite  without

insight. The immediate objective is to control

these features and to prevent patients harming

themselves or others while their grasp of reality

is impaired. In the UK, a section of the Mental

Health  Act  allows  compulsory  admission  to

hospital  for  essential  physical  treatments  in circumstances when the patient is likely to be a danger to themselves or others; this process is known colloquially as ‘sectioning’. It is proposed in the UK that compulsory treatment should be extended to community care also.

Prompt  drug  therapy  is  the  only  option.  A

sedating  antipsychotic  from  one  of  the  tradi-

tional or atypical groups may be given, e.g. olan-

zapine or haloperidol. A benzodiazepine may be

added. At first, medication may have to be admin-

istered  parenterally  to  a  reluctant  patient,

although the onset of action is hardly quicker and

there  is  a  much  higher  risk  of  serious  acute

extrapyramidal complications if the dose is too

high.

Drug administration during this time is care-

fully supervised, and the dose gradually titrated

up to the minimum effective level. Tranquilliza-

tion  and  sedation  are  rapidly  and  reliably

achieved, but control of psychotic symptoms will

not be evident for up to a month or so (Figure

6.13),  and  full  stabilization  may  take  several

months. If the patient does not respond in the

first 1-2 months, some clinicians would advo-

cate further increasing the dose; others would

change to a different group for a further month.

If extrapyramidal symptoms or other adverse

effects are intolerable, or if there are specific

contra-indications, there should be a switch from

a typical to an atypical or a change of atypical.

NICE recommends initiation of clozapine when a

patient has failed after 6-8 weeks on each of two

other drugs successively, including one atypical.

Treatment resistance.   This is usually defined

as failure with at least two drugs, typically one typical and one atypical at optimal doses (Figure

6.13). It occurs in about one-third of patients. A true treatment-refractory state must be distin-

guished from poor compliance, drug interaction or possibly enhanced elimination by the patient’s hepatic or renal systems.

Three strategies are currently used to manage

resistance to conventional agents at maximal

recommended doses. High-dose therapy can be

tried,  but  there  is  limited  evidence  and  the

guidelines of the Royal College of Psychiatrists

(see BNF, section 4.2) contra-indicate it for many

at-risk patients, rigorous precautions must be

taken  and  close  monitoring  is  required.  An

adjunct drug can be added, such as lithium (espe-

cially  in  schizoaffective  states),  carbamazepine

(especially in aggression or mood swings) or a

benzodiazepine (especially if extra sedation is

indicated); experience with lithium is greatest.

Probably the treatment of choice now is to use

clozapine. In the rare cases where clozapine  is

inadequate, NICE recommends the trial addition

of olanzapine.

Maintenance and prophylaxis

Patients  who  develop  chronic  disease  need

continuation therapy, either to suppress their

symptoms (maintenance) or to prevent relapses

(prophylaxis). The management of such patients

has undergone a number of changes in recent

decades.

The first was changed social policy, which

encouraged a reduction in the number of long-

stay psychiatric hospital patients, many of them

schizophrenic, and their return to the commu-

nity. The inevitable lack of stimulation in long-

stay hospitals is detrimental to recovery. Patients

become institutionalized and incapable of inde-

pendent existence, even if their disease eventu-

ally remits. Unfortunately, community services

have not always been appropriately equipped or

funded to care for this large increase in their

dependent population. Only now is it realized

that this discharge process has been too thor-

ough  and  has  become  counter-productive,  as

the discharged patients lose contact with the

care services and become homeless, vagrant or

imprisoned. In a very few cases - unfortunately

those that attract most public attention - they

have become dangerously violent.

Schizophrenia   423

Second, the community care of these patients would not have been feasible without the devel-

opment  of  long-acting  depot  forms  of  the

antipsychotic drugs (see below).

Finally, recognition of the extent of TD (p.

420) provoked a further re-assessment of the role

of long-term antipsychotic drugs. Fortunately

this  has  been  considerably  mitigated  by  the

advent of the atypicals, which are less likely to

cause this.

Duration of drug treatment.   Once the acute

phase of schizophrenia has been controlled, drug

dosage  can  be  gradually  reduced.  At  least

12-24 months’ maintenance treatment is needed

after a single acute attack. Some 75-85% of

patients will eventually relapse following a single

attack, and how best to manage these patients

is still debated. A high-risk patient (i.e. who is

violent, aggressive and never fully controlled) is

still likely to need to be maintained on drugs

indefinitely. For those who stay in remission for

lengthy periods, attempts may be made to treat

each acute attack aggressively, but gradually tail

drugs off and stop them completely between

attacks (targeted or intermittent therapy). Even

so,  an  initial  minimum 5-year  maintenance

period is recommended. There is a relapse rate of

about 15%   per  month  on  discontinuation,

which is halved by prophylaxis, so a difficult

judgement has to be made, in association with

the patient and their carer.

Delayed initiation of treatment or continua-

tion therapy for less than 3-5 years predisposes to more frequent, more severe, less easily treated relapses. However, the maintenance dose might be kept low, and depot injections might allow total drug doses to be reduced further.

Depot therapy.   If a patient is stabilized on

antipsychotic   therapy   and   seems   to   need

medium- or long-term maintenance, there are a

number of advantages to transferring to depot

therapy:

•  Lower total dose.

•  Facilitation of community care.

•  Regular contact with the patient by carers.

•  Supervised administration prevents defaulting.

•  No accumulation or abuse of unused tablets.

In most depot formulations the antipsychotic

is esterified and dissolved in an oily vehicle. The

deep IM dose is distributed throughout body fat

during the first few days. Before exerting the

clinical effect, the drug must be partitioned into

the plasma from these lipid depots and then de-

esterified by hydrolysis. Hence a single injection

can maintain effective plasma levels for between

14 and 28 days. Depot therapy, organized either

via  special  outpatient  clinics  or  community

psychiatric  nurses,  has  greatly  facilitated  the

trend to  community  care  for  chronic  schizo-

phrenic patients, enabling many patients who

would otherwise be unable to do so to cope in

the community.

There is at least one depot preparation available

from each main antipsychotic group. Most reach

therapeutic levels within about 1 week and steady

state after two or three doses, and require 2- to 4-

weekly injection. At present only one atypical

preparation is available as a depot: risperidone

absorbed into nylon globules. This has unusual

release  characteristics,  with  a  lag  phase  of

4 weeks,  during  which  oral  therapy  must  be continued.

Ideally, the patient is first titrated for a main-

tenance dose using the oral form of the selected antipsychotic. A test dose of the depot formula-

tion  is  given  to  check  for  sensitivity  to  the vehicle. A formula may then be used to estimate the initial injected dose; e.g. one method would convert a 10-mg daily dose of fluphenazine into a 25-mg fortnightly dose.

A crossover phase follows, with the oral dose

tapered   off   and   the   depot   dose   gradually

increased. Subsequently, the depot dose must be

titrated against effect, which is not as straight-

forward as with oral therapy because of the

delayed   onset,   prolonged   action   and   slow

reversal of effect. Moreover there is no consistent

relationship   between   oral   and   depot   doses

needed to achieve control in a given patient.

Nevertheless, the final dose is usually lower than

the previous total oral dose over the injection

interval, partly because of improved bioavail-

ability but also possibly due to the efficiency of a

constant plasma level.

There  are  disadvantages  to  this  approach.

Injection  site  problems,  including  pain,  are

common. If adverse effects occur that were not

system

identified in the preliminary oral dose-ranging

trial, the depot cannot be cleared quickly from

the body. Although the lower total dose might

be expected to reduce the incidence of adverse

effects, including most EPS, this is not usually

found; TD may be more common. These may

be  incidental  consequences  of  the  imposed

improved compliance, or related to the stable

plasma level as opposed to the constantly varying

levels of oral therapy.

Equally important is the fact that there is often

considerable patient resistance to this form of

therapy, which is seen as controlling, punitive or

detrimental to autonomy. This is likely to be

particularly the case with those patients who are

poorly compliant with oral therapy and there-

fore one of the main target groups. Patient and

family counselling and education are essential to

explain the purpose and potential advantages of

the technique.

Special problems

Adverse effects.   The autonomic effects of the

less  potent  drug  groups (e.g.  simple  pheno-

thiazines)  may  be  contraindicated  in  certain

patients, notably those with CVD, glaucoma,

urinary  retention,  etc.  Excessive  sedation  is

unwanted in a disease associated with depression

or negative features such as apathy and with-

drawal. Many of these problems are found with

the typical drugs but are absent or rarer in the

atypicals, so a switch is indicated if they become

troublesome or intolerable (see Tables 6.19 and

6.23).

The   main   problems   that   occur   more

commonly with the atypicals are weight gain

and diabetes. For the first, either a typical drug is

recommended, or amisulpride; for the second,

either risperidone or amisulpride. In all such cases

regular  monitoring  of  weight  and/or  blood

glucose is essential.

Compliance.   Poor compliance is a perennial

problem  in  psychotic  patients.  The  basis  is

reduced  insight,  combined  with  often  quite

severe   and   frequently   embarrassing   adverse

effects. Paranoid delusional states add to the

difficulty when patients perceive carers as spies

or  tormentors.  Apocryphal  tales  circulate  of  caches of medication found under floorboards or

behind radiators when psychiatric hospitals are

refurbished.   In   the   community,   once-daily

dosing or depot therapy ameliorate the problem.

In hospital, or in other situations allowing super-

vision of drug administration, nurses and carers

become adept at ensuring that patients are not

hiding tablets in their cheek pouch rather than

swallowing them.

Concentrated oral liquid forms help because

they cannot be hidden from a search of the oral

cavity, but syrup-based preparations are inconve-

nient, messy and, in the long term, cause tooth

decay and obesity. Moreover, doses are difficult

to measure accurately. Newer orally dispersible

formulations of risperidone  and olanzapine  are

preferred nowadays.

In all cases counselling or other psychotherapy

should also be used to encourage concordance.

Target symptoms.   There is little evidence that

particular  antipsychotics  have  a  preferential

effect  on  specific  target  symptoms.  The  less potent agents are indicated where tranquilliza-

tion or psychomotor inhibition is needed, e.g. in hypomania  or  acute  panic  attacks.  However, antipsychotics should not be used for simple

anxiety or agitation. Medium-potency agents are usually sufficient for the short-term treatment of the less intense psychotic features occasionally found in severe depression.

If there is an affective, especially depressive,

component to schizophrenia then a thioxan-

thene may be indicated. However, fixed-dose

combination   preparations   of   tricyclic   anti-

depressants and antipsychotics are almost never

used by psychiatrists. Zuclopenthixol is reputed

to   be   effective   in   diminishing   aggressive

symptoms.

Combinations.   There are almost no situations

where antipsychotic drugs should be combined,

for no advantage is to be gained and adverse

effects are likely to be inceased. Only in severely

treatment-resistant cases, under specialist care, might this be attempted.

Cost.   Pharmacoeconomic  analysis,  in  com-

paring the conventional agents with atypicals,

weighs the far greater drug costs of the latter

against the resulting savings in reduced hospital

Neurological disorder

Parkinson’s disease and the

extrapyramidal syndromes

The basal ganglia (BG) and the extrapyramidal

pathways play an important role in modulating

and smoothing voluntary muscular movement

(pp. 368 and 370; see also Figure 6.3). Disorders

of  movement  and  tone,  collectively  termed

extrapyramidal syndromes (EPS), can result from

an imbalance between excitatory and inhibitory

influences   in   these   structures.   Idiopathic

Parkinson’s disease is the most common form;

other  similar  conditions  are  often  described

system

admissions for relapse or inability to achieve

control. Less easily quantifiable factors, such as

improved quality of life and reduced drain on

social  services  costs,  are  taken  into  account.

Looked at this way, the balance between risk and

benefit definitely favours using an atypical agent

as soon as possible.

generically as parkinsonism. Iatrogenic EPS are

well-known side-effects of antipsychotic treat-

ment (p. 419), and they can also be caused by

numerous   closely   related   neurodegenerative

diseases or lesions (Table 6.24). Other causes can

be circulatory, toxic or traumatic. This section

will refer mainly to the idiopathic form.

Some important structures in the BG, and

interconnections with other brain centres, are

shown in Figure 6.14. Of the numerous stimu-

lant and inhibitory transmitters involved, the

best characterized are dopamine (mainly at D2

receptors), acetylcholine, GABA and glutamate.

These complex pathways and circuits, which are           

still being traced, allow the brain to exercise a

high  degree  of  fine  adjustment  of  voluntary

movement;  they  are  also  responsible  for  the

resting muscular tone required, for example, for

posture. One important input comes from the

substantia nigra (strictly speaking, a midbrain

structure) and output from this system descends

via the corticospinal tract and the extrapyramidal

pathways.

Aetiology and pathology

Parkinson’s disease arises from lesions of the

inhibitory  dopaminergic  nigrostriatal  path-

way  (Figure 6.14). The resulting reduction in

dopaminergic inhibition allows a preponderance

of unopposed cholinergic action in the nigro-

striatum and increased GABA-ergic inhibitory

tone  downstream.  These  imbalances  have  a  major destabilizing impact on the whole motor

system.

Parkinson’s disease is a chronic progressive

neurodegenerative disease. It involves destruc-

tion of the melanin-pigmented dopaminergic

cells of the substantia nigra and their axonal

connections to the striatum. This results in a fall

in  the  nigrostriatal  dopamine  output  and  a

consequent increase in the activity of inhibitory

GABA-ergic neurones. It is not simply a case of

cellular   dopamine   deficiency   or   reduced

turnover:   there   is   a   reduced   number   of

dopamine-secreting  cells.  Other  brain  centres

and transmitter systems may also be involved

but most characteristic clinical signs, i.e. the

motor defects, derive mainly from this lesion.

The underlying cause of idiopathic Parkinson’s

disease is unknown, so preventative measures

cannot be taken. Neither auto-immunity nor

infection is implicated. Many people developed

Parkinson’s disease as part of the encephalitis

lethargica   syndrome   following   an   influenza epidemic in the 1920s, but although these are

now dying out the prevalence of Parkinson’s

disease is unchanged.

There seem to be no strong genetic links, but

there are suggestions of inherited susceptibility,

and some studies have implicated single genes

in a few patients, especially those with early-

onset disease, which has been associated with

mutations of the parkin gene. Within the broad

classification  of  idiopathic  Parkinson’s  disease

there  are  possibly  several  subtly  different

subtypes.

Pathogenesis

The  process  by  which  nigrostriatal  cells  are

destroyed is still not clear. A histological charac-

teristic of affected cells is a build-up of ‘Lewy

bodies’ (dark-staining  hyaline  cell  inclusions).

However, the specificity and direct pathological

significance  of  such  inclusions  is  unknown,

because they also occur in some of the dementias.

Theories on the pathology include accelerated

ageing with increased apoptosis (programmed

cell death; see Chapter 10), excessive build-up of

highly destructive oxidative free radical interme-

diates,  which  are  normally  neutralized,  and

impaired energy handling.

A chance finding among drug misusers opened

a fruitful line of research. A contaminant in

illicitly   synthesized   pethidine   (meperidine)

produced symptomatic and histological features

very similar to idiopathic Parkinson’s disease.

This   contaminant,   MPTP       (methyl   phenyl

tetrahydropyridine), is oxidized by monoamine-

oxidase-B to MPP . This toxic free radical seri-

ously damages mitochondrial energy pathways.

This  discovery  provided  a  primate  model

for studying the pathology and treatment of

Parkinson’s disease. However, its relevance to the

naturally-occurring  disease  in  humans  it  still

uncertain. MPTP is not common in the environ-

ment, although many herbicides and pesticides

are pyridine-based and might be metabolized in

a similar way. One possibility is a genetic defect

in the ability to detoxify oxidative intermedi-

ates,  leading  to  mitochondrial  damage.  It  is

unclear why only nigrostriatal cells should be

affected by any of these effects.

system

Epidemiology

The overall prevalence of Parkinson’s disease is

1-2 per 1000. Although it can affect people as

young as 40, it is predominatly a disease of the

elderly. The prevalence increases sharply with

age and is about 3% among people aged over 65.

Men and women are affected equally and there

seems to be little racial or social variation. This is

not what we would expect if the cause were

wholly environmental or toxic. The prevalence

in Europe and North America is twice that of

China and Japan, which implies genetic factors,

but  confirmation  could  only  come  from  a

large-scale follow-up of migrating sufferers.

Thus the current thinking is that people may

have a genetically conferred susceptibility, but

whether or not they develop the disease depends

on   probably   several   environmental   factors,

which they may or may not meet during their

life. In this, Parkinson’s disease resembles many

cancers.

Course

Parkinson’s disease has an insidious onset, with

slowly  progressive  non-specific  signs  such  as

vague muscle pain, stiffness, mild depression

and general slowing down. A late onset may

protect some patients from the worst ravages of

advanced  disease.  Over 80%  of  nigrostriatal

dopamine must be lost before symptoms become

apparent,  possibly  because  neurological  and

behavioural   compensation   mask   symptoms

before this stage. This prodromal phase may last

for up to 5 years. Parkinson’s disease can be

considered to proceed through five phases:

1. Prodromal phase - asymptomatic.

2. Early symptomatic phase - little disability;

drugs may not be needed.

3. Main treatable phase (5-7 years) - levodopa

            effective.

4. Late phase - declining levodopa effectiveness.

5. Terminal phase - disease extremely difficult to

control.

The early symptoms can easily be mistaken for

simple ageing, although when the frank clinical

features emerge the condition is unmistakable.

Intellect  is  initially  unimpaired,  which  may      stroke,  hypothyroidism,  dementia,  etc.            (Table

exacerbate the patient’s distress. However, drug-

resistant  dyskinesias,  cognitive  degeneration,

dementia  and  various  psychiatric  and  other

non-motor features can all occur as the disease

progresses.  These  may  be  due  partly  to  the

involvement of structures other than the nigro-

striatum, within or outside the BG, becoming

affected by the same or related degenerative

processes. It is known that brainstem areas such

as the olfactory and autonomic centres and later

the limbic areas can become involved.

The rate of progression of Parkinson’s disease

is very variable and it is difficult to determine

whether treatment slows the condition in any

one individual. However, progression is relent-

less, with no sustained periods of remission.

Untreated, the median survival is about 10 years

from the onset of symptoms, and patients have

three times the mortality of a matched normal

population. Modern management with levodopa

has improved the prognosis, partly by reducing

the mortality and morbidity from the secondary

complications  of  immobility.  Optimal  drug

therapy  has  at  least  doubled  the  average

survival   time,   and   life   expectancy   now

approaches  normality.  Death  is  usually  from

pneumonia, resulting from immobility or from

aspiration  due  to  impaired  swallowing  and

respiratory musculature.

Clinical features

The classic signs of parkinsonism are tremor,

rigidity, slowness and abnormal posture. Most

may  be  traced  to  disorders  of  muscle  tone

(dystonias) or muscle movement (dyskinesias),

which are usually asymmetrical. There are also

signs   of   excess   cholinergic   parasympathetic

activity. The clinical features are classified and

described in Table 6.25.

The clinical presentation is relatively uniform

and diagnosis may be straightforward, but differ-

ential diagnosis between Parkinson’s disease and

other  neurodegenerative  diseases (Table 6.24)

can  be  problematic.  It  is  also  necessary  to

consider possible primary causes, e.g. antipsy-

chotic drug therapy, and to distinguish other

conditions   presenting   similar   features,   e.g.

6.24). Parkinson’s disease should be excluded

when investigating falls, ankle oedema (possibly

the result of immobility) and reduced mobility

attributed to ‘ageing’. In doubtful cases imaging

techniques   may   be   helpful,   including   MR

(magnetic  resonance)  and  CT (computerised

tomography)  scanning  and  nuclear  medicine

techniques   such   as   PET (positron   emission

tomography).

Clinical diagnosis is based on the following gross features:

•  Flexed posture and shuffling gait.

•  Expressionless face and reduced blinking.

•  Distal tremor that is abolished by purposive

movement.

Non-motor   symptoms.   A   variety   of   non-

motor  problems  can  complicate  Parkinson’s

disease at all stages in the disease (Table 6.25).

These  include  psychological  and  psychiatric

disease, autonomic dysfunction, sleep disorders

and falls. They may be overlooked or mistaken

for other diseases in the early stages. In the later

stages they are more significant because they are

generally unresponsive to dopaminergic therapy

and  can  seriously  impair  quality  of  life  and

reduce   life   expectancy.   Possibly   they   have

become more prominent as the prognosis has

improved following the introduction of levodopa

therapy.

Their pathogenesis is still unclear. Some may

be iatrogenic, originating from dopaminergic or

antimuscarinic therapy; others may result from

involvement of other brain centres (see above).

Management

Aims

Ideally,   the   aims   in   the   management   of Parkinson’s disease would be to:

•  reduce symptoms;

•  provide general support;

•  prevent further degeneration;

•  induce reversal or regeneration.

At present, most success has been obtained

with the relief from troublesome symptoms and maintenance of the patient’s independence and general  health.  Attempts  at  prevention  and reversal are hampered by the lack of a clear under-

standing of the aetiology and pathogenesis, and so are still largely experimental.

Support

Exercise,  physiotherapy,  speech  therapy  and

occupational therapy are essential to help the

patient to cope with their progressive disability

and maintain their independence as long as possible. It is important to maintain muscle and         and levodopa therapy, particularly in the elderly.

tendon strength in the face of reduced mobility. Psychiatric help and medication may be needed for depression, which is common.

Symptomatic treatment: drug therapy

The  primary  objective  of  drug  therapy  is  to enhance   dopaminergic   activity   within   the damaged areas of the BG, and this is achieved in various  ways  (Table  6.26).  Parkinson’s  disease has provided an exceptionally fertile field for

rational drug design and formulation to specific clinical requirements.

Residual dopaminergic activity can be mildly

enhanced by inhibiting neuronal dopamine re-

uptake (amantadine) and the excessive cholin-

ergic tone that dopamine deficiency causes can

be countered with antimuscarinics. Neither are

first-line drugs but are sometimes used as adju-

vants. However, antimuscarinics exacerbate the

psychiatric complications of Parkinson’s disease

They are helpful in tremor, but are of little use

for bradykinesia and have only limited effective-

ness. Thus some form of dopamine augmentation

or replacement invariably becomes necessary.

Levodopa is currently the standard treatment. It

is discussed in detail below. As levodopa inevitably

becomes less effective over time, its availability

can  be  enhanced  by  reducing  its  metabolism

using a catechol-O-methyl transferase (COMT)

inhibitor. Alternatively, dopamine’s half-life in

the brain can be increased with a monoamine

oxidase B (MAO-B) inhibitor or a COMT inhibitor

(Figure 6.15 and Table 6.26). Where levodopa is

ineffective or cannot be tolerated at all (primary

failure), or has become ineffective or intolerable

(secondary  failure),  direct-acting  dopaminergic

agonists,  e.g.  ropinirole,  can  be  substituted  or

added. These do not require central activation by

dopa-decarboxylase and have a longer duration

of  action  than  levodopa.  However,  they  have

worse peripheral dopaminergic adverse effects, and although centrally they cause less dyskinesia, they cause more psychotic reactions.

Although drugs are the mainstay of treatment, they do not relieve all symptoms in all patients, nor do they work indefinitely. In particular, the efficacy  of  levodopa  tends  to  decline  as  the disease progresses.

Retard or prevent progression

With the discovery of the possible role of free

radical damage in the pathogenesis of Parkinson’s

disease there was hope that neuroprotection with

antioxidants (e.g. vitamin E, co-enzyme Q10) or

inhibitors of dopamine metabolism (i.e. MAO-B

inhibitors, such as selegiline in the DATATOP

trial) might prevent or retard the disease. Unfor-

tunately, these hopes have proved unfounded.

Early observations on selegiline probably mis-

attributed simple symptom control to disease

retardation. On the other hand, the suggestion that selegiline might actually increase mortality has also been refuted. A more recent theory that selegiline might after all retard progression, by inhibiting apoptosis, remains unconfirmed. The final judgement on selegiline is still awaited, but MAO-B inhibitors currently retain a role in early disease and as adjuvants later.

Although the progression of the disease seems inexorable   at   present,   modern   therapy   has undoubtedly improved both quality of life and survival if started promptly.

Reversal or regeneration

Rarely, highly selective surgery or deep brain

stimulation to the subthalamic nucleus or globus pallidus  for  treatment-resistant  symptoms  has the CNS. The free dopamine produced causes

been used with some benefit. This is designed to

reduce or reverse the abnormal BG output. Trials

of  implantation  of  dopamine-secreting  tissue

into the brain have so far proved disappointing.

Initially  the  patient’s  own  adrenal  tissue  was

used; more recently the transplantation of nigral

tissue from aborted fetuses has been investigated.

Levodopa therapy

Rationale and development

Clearly,  the  ideal  treatment  for  Parkinson’s

disease   would   be   to   replace   the   depleted

dopamine in the BG. However, there is an impor-

tant drug delivery problem because dopamine,

being polar, is poorly absorbed orally and does

not readily cross the blood-brain barrier. Further,

dopamine has potent peripheral adverse effects.

Thus direct delivery of dopamine to the CNS

is  impractical  and  its  natural  amino  acid

precursor levodopa (L-dopa, L-dihydroxy pheny-

lalanine) is used.

Levodopa is extremely effective for all symp-

toms of Parkinson’s disease, and especially for

bradykinesia; it is up to five times more effective

than antimuscarinics. Unfortunately levodopa is

poorly tolerated, especially if given orally, when

it produces severe gastrointestinal side-effects.

About  90%  of  Parkinson’s  disease  patients have  a  good  to  excellent  initial  response  to levodopa and failure to respond should prompt a  re-evaluation  of  the  diagnosis.  However, some  patients  cannot  tolerate  levodopa  at  all, while  others  may  have  involvement  of  other neurotransmitter systems.

Pharmacokinetics

Levodopa  is  well  absorbed  from  the  GIT  and,

because it is the brain’s natural source of neuronal

dopamine, a proportion of the orally administered

dose is transported into the brain from the plasma

across the blood-brain barrier by an active uptake

pump (Figure 6.15). However, owing to peripheral

decarboxylation, mainly by dopa-decarboxylase

(DD) in the gut wall during absorption, only about

1-3% of the administered dose actually reaches

undesirable peripheral dopaminergic effects on the muscle of the gut, heart and blood vessels.

Once in the brain, levodopa is decarboxylated intraneuronally to dopamine. Decarboxylation is  probably  not  confined  to  neurones  of  the nigrostriatum but also occurs in dopaminergic neurones elsewhere.

Levodopa augmentation

Dopa-decarboxlyase inhibitors

The problems of poor central levodopa delivery

and excessive peripheral dopaminergic adverse

effects are neatly ameliorated by using a periph-

erally  acting  inhibitor  of  dopa-decarboxylase

(DDI), e.g. carbidopa or benserazide. These have

been designed to be insufficiently lipophilic to

cross  the  blood-brain  barrier  and  so  cannot

interfere with the central activation of levodopa

to dopamine. When a DDI is combined with

levodopa  (as  co-careldopa  or  co-beneldopa)  the

proportion of the levodopa delivered to the CNS

rises to 10%, the levodopa dose may be cut by

75% and peripheral side-effects are substantially

reduced.  The  main  problem  is  a  predictable

increase in dopaminergic adverse effects in the

CNS. The DDIs themselves have few side-effects.

COMT inhibitors

Another route for the peripheral metabolism of

levodopa is methylation by COMT (Figure 6.15).

While this does not produce a metabolite with

the adverse effects of dopamine, it does reduce

the central availability of levodopa. Moreover,

brain COMT is partly responsible for the clear-

ance of dopamine. Thus further increases in

levodopa bioavailability can be obtained by using

blockers of peripheral COMT, e.g. entacapone,

tolcapone. In addition, because tolcapone crosses

the   blood-brain   barrier   it   also   enhances

dopamine activity in the CNS by reducing its

intraneuronal catabolism. However, the use of

tolcapone is limited by liver toxicity so it must be

carefully monitored under specialist supervision.

MAO-B inhibitors

The intraneuronal catabolism of dopamine can

be   further   reduced   by   inhibiting   central

MAO-B. MAO-B inhibitors provide a moderate

improvement in cases where levodopa effective-

ness is waning (Figure 6.15). Because they do

not  block  MAO-A,  the  form  of  the  enzyme

affected by conventional antidepressant MAOIs

such as phenelzine, they are not subject to the

usual MAOI restrictions and dietary precautions.

Unfortunately, even these augmented combi-

nations do not work indefinitely, so problems

remain. First, all drug treatment is only really

effective  against  dyskinetic  symptoms.  Other

motor symptoms remain and indeed tend to

deteriorate, especially dystonias causing gait and

postural problems (Table 6.25), which are pos-

sibly mediated by defects in non-dopaminergic

neuronal systems. These may be the patient’s

most  disabling  complaints  despite  otherwise

good control. Secondly, cholinergic features may

still be troublesome. Finally, there is an inex-

orable, imperfectly understood decline in the

effectiveness of levodopa after a number of years

of satisfactory control.

Dopamine agonists

The final option where levodopa is ineffective or

intolerable is to use a dopamine agonist, either

as an adjuvant or a substitute. There are several

advantages to using a direct-acting dopamine

receptor agonist. They do not require activation

by DD, which is useful because concentrations of

this enzyme may be reduced in late disease. Also,

there may be some receptor subtype selectivity,

e.g. pergolide (D1 and D2 receptors), ropinirole (D2

only), although the clinical significance of this is

unclear. The main disadvantage is an increase in

peripheral  dopaminergic  side-effects  because,

unlike levodopa, they are active peripherally as

soon as assimilated.

There are two main subgroups (Table 6.26).

The original ones were ergot derivatives, e.g.

bromocriptine, pergolide, cabergoline  and lisuride,

and  these  were  widely  used  in  Parkinson’s

disease. Later drugs are not derived from ergot,

e.g. ropinirole, pramipexole, rotigotine. Unrelated to

either  group  is  apomorphine.  The  effects  and

adverse reactions of all are broadly similar to

levodopa,  with  one  important  exception.  All

ergot   derivatives         (not   just   those   used   in

Parkinson’s disease) can cause potentially serious

fibrotic reactions in the lungs, heart and peri-

toneum, and so drugs in this subgroup are now

system

rarely used. If they are needed, patients must

first be screened for CXR, renal function and

ESR.

Dopamine  agonists  may  be  used  as  initial therapy, to delay the introduction of levodopa, or in primary or secondary levodopa failure.

Administration

Levodopa is routinely given with a decarboxylase

inhibitor. Low doses are used to establish toler-

ance and then gradually increased every 2-3 days

until symptoms are controlled, the limit of toler-

ance is reached, or a compromise between these is

achieved. The precise regimen, including the best

ratio of levodopa to inhibitor, must be carefully

individualized to balance tolerance, benefit and

toxicity,  and  must  remain  continually  under

review. A variety of different combinations of

strengths  and  relative  proportions  are  avail-

able. Considerable ingenuity needs to be exer-

cised  in  tailoring  levodopa  drug  regimens  to

extract maximum benefit as efficacy declines.

Divided  doses  are  necessary  to  minimize

gastric intolerance, caused in part by dopamine

generated in the gut wall, and to prevent swings

in plasma levels, which would be reflected in

uneven clinical action. Levodopa is always taken

with  food;  however,  high-protein  meals  can

reduce CNS penetration owing to competition

for amino acid transport mechanisms at the

blood-brain barrier.

Side-effects

The side-effects of levodopa, both peripherally and  centrally,  are  caused  exclusively  by  the dopamine   produced   after   decarboxylation, which  acts  on  dopaminergic  and  adrenergic receptors (Table 6.27). Elderly patients are more prone to all side-effects.

Gastrointestinal tract

Locally   formed   dopamine   reduces   gastro-

intestinal motility, slowing the absorption of

further levodopa. Severe dyspepsia is minimized

by taking small frequent doses with food. Nausea

and vomiting are less common with the lower

levodopa  doses  that  decarboxylase  inhibitors

allow. However, if such doses remain troublesome a peripherally acting dopamine blocking anti-

nauseant such as domperidone may be used. Meto-

clopramide  and  phenothiazines  are  unsuitable because they cross the blood-brain barrier and so would antagonize the intended central thera-

peutic action of dopamine, as well as causing

their own extrapyramidal effects.

Cardiovascular system

Dopamine is active at both beta1-adrenergic and

dopaminergic receptors (it is used therapeutically

as an inotrope and vasodilator; see Chapter 4).

Most parkinsonian patients are in an age group

that is prone to heart disease, and so care is

needed in this respect. Graduated compression

hosiery is helpful to minimize postural hypoten-

sion, but serious arrhythmias such as tachycardia

and  premature  ventricular  beats  may  require

an  anti-arrhythmic  drug,  or  the  cessation  of

dopamine treatment. Fortunately, the cardiovas-

cular effects seem to remit on continued therapy.

CNS

Dopamine is a transmitter in other areas of the

BG besides the nigrostriatal pathway, and also in

centres outside the BG, and actions in these areas

may be intensified by decarboxylase inhibitors.

Dopamine may, ironically, produce other move-

ment   disorders,   chiefly   writhing        (choreoa-

thetosis) or restless legs. These tend to occur in

the later stages of treatment and are difficult to

distinguish  from  late  manifestations  of  the

disease or declining disease control. The precise picture of these complex interactions is far from clear,   and   strategies   for   countering   them, discussed below, are largely empirical.

Similarly, a variety of psychiatric effects can

result from an excess of dopamine in, presumably,

mesolimbic  and  mesocortical  centres. (Recall

that one theory of the pathogenesis of schizo-

phrenia attributes it to excess dopamine here; p.

409.) Psychotic features such as hallucinations

and  paranoia  may  occur,  as  may  delirium,

depression or mania. These may be exacerbated

by the use of antimuscarinic drugs, e.g. anti-

parkinson agents and antidepressants. They may

be confused with symptoms of advanced disease

itself (e.g. dementia), or represent the unmasking

of latent psychiatric illness, which is a relative

contra-indication to levodopa use.

For    psychosis,    traditional    antipsychotic dopamine blockers such as the phenothiazines clearly cannot be used. The atypical antipsy-

chotics  are  preferred,  especially  quetiapine  or clozapine, partly on the basis that the psychiatric symptoms may involve serotoninergic receptors, and partly because these drugs themselves cause few extrapyramidal problems. The 5-HT3 blocker ondansetron has also been tried.

Care  is  needed  when  treating  depression. Conventional   tricyclic   antidepressants   with antimuscarinic effects must be avoided. There is  a  theoretical  possibility  of  SSRIs  worsening the  parkinsonism (see  above),  and  they  also

interact  with  selegiline  causing  hypertension.

Conventional   non-selective   MAOIs   interact

with levodopa, causing hypertensive crisis, and

with selegiline, causing hypotension. A sensible

choice  would  appear  to  be  an  antidepressant

with  little  antimuscarinic  activity,  such  as

lofepramine.  However,  in  practice  SSRIs  are

quite frequently used.

Endocrine

There  is  a  theoretical  possibility  of  levodopa

mimicking dopamine’s inhibition of hypothal-

amic-releasing hormones. However, the poten-

tial  results (e.g.  hypoprolactinaemia)  are  not

seen and are likely to be less significant in the

parkinsonian age group. (This effect is exploited

therapeutically in the use of bromocriptine  for

hyperprolactinaemia.)

Long-term and late complications

Fortunately, no serious long-term haematolog-

ical,  renal  or  hepatic  toxicity  has  yet  been

observed. However, the long-term dyskinetic and

psychiatric side-effects of levodopa are so varied

and so difficult to distinguish from late complica-

tions of the underlying disease that Parkinson’s

disease  management  has  become  almost  a

subspecialty  in  itself,  with  its  own  confusing

taxonomy of complications. Fluctuating thera-

peutic  responses,  dyskinesias  and  psychiatric

symptoms frequently become seriously disabling,

with about half of patients experiencing prob-

lems after 5 years on levodopa and three-quarters

after 15 years.

Numerous  ingenious  pharmacological,  bio-

pharmaceutical and formulation strategies have

been devised to optimize delivery of dopamine

to its intended site of action and to minimize

adverse   systemic   effects.   This   represents   a

tremendous clinical pharmacological challenge.

Early experience with levodopa had suggested

that there was a limited window for effective

levodopa treatment in the course of the illness

(phase 3, p. 428), after which these problems

would  arise.  Thus,  it  was  felt  that  levodopa

therapy should be delayed as long as possible,

conserving it for the later, more severe phases.

However, many of the effects ascribed to long-

term  levodopa  therapy  are  now  considered

system

likely to be related to disease progression, and treatment is now introduced earlier.

Difficulties in interpreting inconsistent levodopa activity arise because of:

•  Lack of correlation between plasma and brain

            levels,   because   CNS   uptake   of   levodopa

depends on an active pump, the activity of

which may change.

•  Lack  of  correlation  between  CNS  levodopa

            levels and synaptic dopamine levels, because

of reliance on neuronal uptake and neuronal DD action.

•  Declining ability of a reducing BG neuron

            population  to  decarboxylate  levodopa  and

store dopamine, either within or outside the striatum.

•  Changes in post-synaptic receptor sensitivity.

•  Involvement of dopaminergic systems outside

the nigrostriatal system.

•  Involvement of non-dopaminergic systems.

•  Normally dopamine activity in the BG varies

smoothly, whereas in treatment it is pulsatile.

Management of levodopa complications

The motor problems of Parkinson’s disease and

levodopa therapy are classified in Table 6.28. It is

unhelpful to attempt to specify a particular solu-

tion for each one, especially because many treat-

ments are still experimental. Moreover, when a

particular  problem  is  not  responding,  other

methods are tried empirically. In general, three

basic approaches are employed, depending on

whether problems are related to reduced levodopa

activity,  excessive  levodopa  activity,  or  mixed

intractable fluctuations.

For reduced effectiveness (e.g. ‘end of dose’

and  other ‘off’  phenomena),  the  aim  is  to

increase delivery of levodopa, modify its time

course or smooth out variations in its plasma

level.   Increased   frequency   of   lower   doses

(without allowing plasma levels to fall below the

threshold  of  activity),  liquid  formulations  or

enteral systems may help. The Duodopa intra-

jejunal system provides continuous co-careldopa

dosing, avoiding the stomach, and the dosage

rate  from  the  portable  external  pump  can  be

varied. Rotigotine is available as a transdermal

patch.  Continuous  SC  infusion  of  dopamine

analogues such as apomorphine is cumbersome

but effective. The rectal and intranasal routes

are also being considered. Modified-release oral

preparations  are  often  used,  but  these  have

reduced bioavailability and so require dosage

adjustment.

Dopaminergic  analogues  generally  have  a

longer half-life and may be added to levodopa. If

a  regimen  can  be  developed  with  the  right

balance between levodopa precursor and direct-

acting  agonist,  it  may  enable  an  acceptable

compromise  between  prolonged  activity  and

increased  peripheral  dopaminergic  problems.

The addition of an MAO-B or a COMT inhibitor

increases    levodopa   bioavailability.    Delayed

levodopa onset may be countered by avoiding

simultaneous high-protein meals or by enteral

administration.

Increased toxicity (i.e. ‘on’ phenomena with

troublesome dyskinesias) usually necessitates a

reduction in dose, although spacing out doses or

using modified-release preparations may help.

Reducing the dose is likely to be at the expense

of increased disease symptoms and decreased

mobility. Again, an acceptable balance must be

agreed with the patient. Gastric toxicity can also

be circumvented by the non-oral formulations.

A wide range of drugs have been tried for

intractable motor problems, including atypical

antipsychotics,   adrenergic   beta-blockers   and

SSRIs, but evidence is lacking. Painful dystonias

may benefit from the antispastic agent baclofen.

‘Drug holidays’, once recommended, are now thought unhelpful. Because no treatment defi-

nitely retards progression of the condition, it is to be expected that even a successfully managed patient will eventually go through increasing

periods of instability.

Combination products (e.g. levodopa   DDI COMT inhibitor) may be helpful for patients

taking multiple drug therapy at this stage.

Drug selection

Drug treatment is started when the degree of

functional   disability   caused   by   the   disease

outweighs likely adverse effects. The NICE guide-

lines do not indicate a specific sequence but

suggest a range of drugs from which selection

must be made on the grounds of patient accep-

tance and tolerability, contra-indications, and

concurrent morbidity or drug therapy (Figure

6.16).

The trend has been to start reliable evidence-

based drugs as soon as patients find their lives appreciably affected. The principal controversy concerns the stage at which levodopa  therapy should be introduced. The arguments in favour of early initial therapy with levodopa are:

•  It is the most effective drug.

•  It is now widely agreed that levodopa does not

            have a strictly limited window of activity - it

continues to benefit most patients to some

degree.

•  It is no longer believed that levodopa has long-

            term  neurodegenerative  effects  that  could

accelerate disease progression.

•  The   elderly   are   very   sensitive   to   the

            adverse  effects of the possible alternatives

(antimuscarinics and dopamine agonists).

•  Evidence   does   not   support   the   role   of

selegiline in disease retardation.

On the other hand, the inevitable long-term

adverse effects of levodopa, which usually start after 5-10 years of therapy, do mean that these are experienced earlier with earlier initiation of therapy, as would be the case with younger

patients. The evidence is not conclusive yet on the optimal policy in all cases.

The first choice in the early stages is between

levodopa, in as low a dose as relieves symptoms, a

dopamine agonist or an MAO-B inhibitor. Less

potent and less effective drugs with a poorer

evidence base could be added or substituted if

control is not achieved. Amantadine may help,

antimuscarinics can be used in young patients

with severe tremor, and beta-blockers may help

where there is postural tremor. Older patients

should usually be started on levodopa as initial

therapy and should avoid antimuscarinics.

Whatever treatment patients start with, all will

eventually need to take levodopa  in gradually

increasing dosage and frequency (assuming they

can tolerate it). However, at some stage, when

long-term complications supervene or resistance

develops (i.e. secondary levodopa failure), other

strategies will be required. These fall into two

groups. First, modify the dose form and route

of administration, as described above. Second,

use one or more of the other groups of drugs,

with dopamine agonists being the first choice

adjuvants (Figure 6.16).

For a few patients it seems that, eventually, no drug combination will be satisfactory and surgical options may have to be considered.

The word ‘epilepsy’ is derived from the Greek,

meaning ‘to take hold of, seize’. For centuries,

there has been the ancient fear that the sufferer

of epilepsy has been possessed, literally taken

hold of, by some malign external force. The

following  quotation  describes  many  of  the

features of a major seizure in graphic terms.

He begun to groan then like some terbel thing

wer taking him and got inside him. He startit to

fall and I easit him down I knowit he wer having

a fit I seen that kynd of thing befor. I stuck the

clof . . . be twean his teef so he wunt bite his

tung. I wer on my knees in the mud and holding

him wylst he twissit and groant . . . I cud feal

how strong he wer tho he wernt putting out no

strenth agenst me he wer sturgling with what

ever  wer  inside  him.  I  wunnert  what  wud

happen it got pas him and out. It dint tho. It

roalt him roun and shook him up it bent him

like a bow but finely it pult back to where ever it

come out of. When it gone he wunt do nothing

only sleap nor I coulnt get him to walk 1 step.

Russell Hoban, Ridley Walker

(Reproduced with kind permission

of the publishers)

For  this  reason  people  with  epilepsy  have

always encountered as much prejudice as those

suffering from psychiatric disorders. The lack of

self-control evident during a seizure was feared,

and the sufferer was therefore spurned. Yet only

rarely does epilepsy directly cause psychiatric

symptoms: it is predominantly a neurological

condition involving disorders of movement or

consciousness; moreover, patients are asympto-

matic for most of the time. There may, however,

be  secondary  psychiatric  morbidity  in  many

long-term sufferers.

Epilepsy  is  a  chronic,  paroxysmal,  non-

progressive  disorder  of  intermittent  disorga-

nized  electrical  activity  in  the  brain,  which

causes  the  seizure.  Seizures  are  characterized

most commonly by impairment of motor activ-

ity (convulsions),  consciousness,  perception  or

behaviour.

The term ‘fit’ is now avoided, as is ‘epileptic’ in reference to a patient, for whom the term ‘person with epilepsy’ (PWE) is gaining currency.

Epilepsy           439

Aetiology and classification

An isolated seizure can be precipitated in anyone

if their brain is suitably provoked.  Transient

reversible triggers include drugs (e.g. tricyclic

antidepressants), cerebral infection or inflamma-

tion,  intracranial  hypertension,  head  injury,

stroboscopic lights and metabolic disturbances

such as glycaemic, osmotic or pH imbalance.

Fever can trigger seizures, especially in children.

However, an isolated attack is not considered

epilepsy, which is better defined as a reduced

seizure threshold with a continuing tendency to

experience  seizures.  Because  it  may  have  a

variety   of   causes   and   triggers,   epilepsy   is

regarded as a syndrome rather than a discrete

disease.

The   problem   may   be   symptomatic   of   a

structural abnormality in the brain. In about

one-third of epilepsy patients, a congenital or

neurodevelopmental abnormality is found in a

specific   part   of   the   brain,   especially   the

hippocampus. Other identifiable causes include

ischaemia (arteriosclerosis, stroke or perinatal

hypoxia), head trauma (post-infective, perinatal,

post-operative  or  other  injury),  tumour (5%

overall;   up   to 40%   of   adult-onset   partial

epilepsy) or alcoholic brain disease. Other cases

are referred to as either idiopathic (unknown

cause) or cryptogenic (hidden cause).

Interestingly, even in cases where there is a

clear cause (such as head injury), a family history

may be found. This suggests that it is a tendency

to lowered seizure thresholds that is inherited.

There are strong genetic links in the generalized

epilepsies,  but  no  consistent  environmental

factors  have  been  identified.  The  likelihood

is that for many forms of epilepsy there may

be   a   genetic   predisposition   that   requires

environmental triggers to cause active disease.

The  classification  of  epilepsy  has  recently

become more complex, following the system

proposed by the International League Against

Epilepsy (ILAE). Traditionally a patient’s disease

was classified simply to reflect the predominant

seizure   type.   However,   following   improved

investigation    and    diagnostic    techniques,

evidence has shown that outcomes are better if

management takes more account of aetiology

and other clinical features, and this is reflected

in the new classification. Thus it is necessary first

to review how individual seizures are classified,

then see how this can be adapted to a more

refined classification of epilepsy syndromes.

Seizure type

Partial or generalized

Many seizures clearly originate in one particular

area of one side of the brain, the epileptogenic

focus. The symptoms a patient displays in a

partial seizure (focal or location-related, Table

6.29)  are  usually  readily  identified  as  over-

activity in this area, e.g. a particular sensory

experience  or  an  abnormal  muscular  action,

implicating an area in the sensory or motor

cortex respectively. Usually, a specific anatomical

lesion will be found in the area predicted from

the  symptom.  In  other  words  most  partial

seizures are secondary, and even when no lesion

can be traced, one is assumed to exist. There is

little evidence of genetic links, and a family

history is unusual.

A primarily generalized seizure involves the

whole of the brain, on both sides, from the

outset,   with   symptoms   involving   impaired

consciousness, major muscle groups, or both.

This category includes the most familiar forms, tonic-clonic (‘grand mal’) and absence (‘petit mal’) seizures.

Although   some   partial   seizures   may   be

restricted to their area of origin, in other cases

they spread rapidly to many other areas on both

sides of the brain. This is known as secondary

generalization. In such cases patients may expe-

rience, before the spread, a specific sensory or

other warning symptom, the aura. The aura is

characteristic of the epileptogenic focus, and in

the restricted partial form would represent the

entire seizure. In some cases the partial onset is

masked or unrecognized because of very rapid

generalization, but an attempt should be made to

identify them because it affects prognosis and

treatment.

Simple or complex

This  distinction  among  the  partial  epilepsies is based on whether or not consciousness is

impaired during the seizure. Generalized seizures are  almost  invariably  complex.  The  relation-

ship between these different forms of seizure is illustrated schematically in Figure 6.17.

Epilepsy syndrome

The most recent classification by the ILAE at first

sight seems unduly complex. The rationale is to

group epilepsies taking into account their clin-

ical features, age of onset and presumed primary

causes (Table 6.30). It is then possible, on the

basis of empirical trial evidence, to treat different

patients in a far more targeted, if still empirical,

manner than when basing selection simply on

seizure type alone.

Nevertheless it still starts with the primary

distinction between general and partial seizures as its chief characteristic. Next it classifies the syndrome as idiopathic, symptomatic or crypto-

genic. Subsequently the syndrome is described by its various clinical factors.

Pathology

The ultimate defect in epilepsy may be a reduced

threshold for neuronal membrane depolariza-

tion, for example, a reduced resting potential.

Little is known of how the neuronal instability

occurs or why it should be set off by particular

triggers. One possibility could be that there is an

increased tendency to allow random discharges,

which are normally suppressed, to spread. A

neurotransmitter imbalance may be involved,

such as a reduced level of an inhibitory trans-

mitter, e.g. GABA or glutamate. Such inhibitors

act   physiologically   by   promoting   chloride

uptake  into  the  neuron  using  the  chloride

ionophore  membrane  pump,  which  increases

membrane potential and so stabilizes it. Recent

research has focused on defective ion channels

and a number of monogenic defects in these

channels have been identified. However, this

accounts for only a very small proportion of

patients.

An amine imbalance theory is consistent with

the   apparent   action   of   antiepileptic   drugs

(AEDs), many of which facilitate the stabilizing

action of inhibitory amines. However, although

several specific mechanisms have been identified

for the various AEDs, it is not certain that this is

how they exert their antiepileptic effect.

Epidemiology and course

About 1% of the population in the West and Asia

have epilepsy, which is half the rate in Africa.

There are about 250 000 patients taking AEDs

in the UK. Onset is usually below the age of

30 years, with another peak in the elderly owing to cerebrovascular disease.

In general, epilepsy does not deteriorate, i.e. it

is not progressive, and children especially may

grow out of it. Even adult epilepsy can remit

spontaneously but this can be extremely difficult

to predict, and quite long seizure-free periods

may, if medication is stopped, be followed by a

seizure. The chances of remission are best for

generalized  tonic-clonic  and  absence  seizures

and  poorest  where  these  is  a  structural  but

inoperable  lesion.  Overall,  the  median  period

from diagnosis to being drug- and seizure-free

system

for  at  least  5 years  is  20 years.  However,  30% of  patients  continue  to  have  seizures  despite optimal therapy.

Partial and secondarily generalized epilepsies account for up to two-thirds of cases, tonic-clonic about  one-third,  and  absence  seizures,  which only occur in children, about 5%. The disease usually  causes  no  intellectual  impairment,

although  long-term  AED  therapy  may  do  so. However, repeated uncontrolled convulsions can produce brain damage owing to cerebral hypoxia, and in some cases the seizure disorder is in fact the consequence of brain damage.

Clinical features

The features of different seizures are summarized in Table 6.31. Seizures are usually very short-

lived, and although usually unpredictable are

sometimes triggered in a characteristic way, e.g. by flashing lights, altered mood, stress or relax-

ation.  In  all  but  simple  partial  seizures  the patient is unaware of the seizure and may be

unable to recall it afterwards.

If a patient presents with a fall, blackout or

syncope (faint), or a transient absence, jerking,

odd  behavioural  phenomenon  or  psychiatric

symptom, this must be thoroughly investigated for a non-epileptic primary cause before epilepsy is diagnosed (see below).

Partial seizures

The effects of these highly localized seizures are

mostly self-explanatory once the focus is known:

or, more precisely, the features point to the

focus.  Temporal  lobe  epilepsy  is  the  most

common form, representing about half of all

cases. This condition can be manifested in a very

wide variety of neurological, and occasionally

psychiatric, symptoms, including aphasia (the

inability to find words), mood disorder, halluci-

nations and fainting. This can make differential

diagnosis very difficult. There are many different

types of partial seizures (see Table 6.31 and the

References and further reading section), each of

which usually lasts for a matter of minutes.

Tonic-clonic seizures

The classic seizure, as described in the quotation on p. 439, is the type most widely associated with epilepsy, and goes through up to five phases. In the prodromal phase, which is not experienced by all patients, the advent of a seizure is sensed subjectively, e.g. by a mood change.

Some patients then experience a more specific

symptom,  the  aura  immediately  before  the

attack; this may be a sensory phenomenon, e.g a

smell, a tingling feeling or an epigastric sensa-

tion. Auras are always the same for a given

patient and suggest a primary partial seizure that

subsequently becomes generalized. (There is an

interesting   parallel   with   classical   migraine,

which  also  is  preceded  by  such  auras;  see

Chapter 7.)

The actual convulsion then follows. In the

tonic (contractile) phase there is a generalized

contraction   of   many   muscle   groups,   both

somatic and visceral. Consequently, the patient

loses balance and falls. The respiratory muscle

spasm causes an initial brief involuntary cry, like

being  winded  by  a  blow  to  the  abdomen,

followed by cyanosis.

After 30 s or so the tonic-clonic phase starts.

This series of alternating contractions and relax-

ations causes the jerking that is so alarming for

the onlooker, although by this time the patient

Epilepsy           443

is usually unconscious. There may be frothing at

the mouth, incontinence and tongue biting, and

this is the most dangerous phase because of the

risk of self-harm. There may also be inconti-

nence of urine or faeces. The tonic-clonic phase

lasts a couple of minutes. The only first aid prac-

ticable is, if possible, to get the patient into the

semi-prone (recovery) position to prevent aspira-

tion (inhalation) of vomitus or profuse saliva.

The  popular  idea  of  putting  a  cloth  in  the

patient’s mouth is now strongly discouraged;

any potential damage would usually have been

done by the time this could be arranged, and

the patient might choke on it. There is also a

significant chance of damage to the helper’s

fingers.

In the post-ictal (after seizure) phase there is relaxation, with flaccid paralysis and continued stupor, gradually merging into sleep. After a few hours  the  patient  wakes  with  a  headache,

confused, and often bruised, but with no recol-

lection of the events. A similar state follows ECT as used to treat depression.

Absence seizures (‘Petit mal’)

In a typical absence seizure patients will seem

briefly to lose concentration. There may be an

obvious stare and fluttering of the eyelids. After a

few seconds they continue with what they were

doing, unaware of the hiatus. In more severe

forms there may be a loss of consciousness for up

to 30 s, when the patient may fall; there may be

muscular jerks (myoclonic seizure) but there will

be  no  tonic-clonic  convulsion.  There  are  no

prodromal signs and no post-ictal phenomena.

Such seizures usually occur in children and

can  easily  be  mistaken  for ‘daydreaming’  or learning difficulty, or else they may be over-

looked. Patients may have many attacks a day, sometimes hundreds, but the condition tends to remit as children grow up.

Status epilepticus

If a seizure does not terminate spontaneously,

becoming a series of successive short fits or one

long  one,  it  is  defined  as  status.  This  may

happen   with   any   seizure   type,   but   most

commonly with the tonic-clonic form, and it is

a medical emergency. Patients may cause them-

selves  some  physical  injury,  but  the  main problem is cerebral hypoxia owing to compro-

mised respiration.

Complications

In addition to the seizures, patients face con-

siderable psychosocial difficulties. There is the general ignorance already referred to, as well as the associated stigma, causing difficulties with education, work, leisure and social relationships. The disease itself also imposes restrictions on such   activities   as   driving,   swimming   and bathing. There may be secondary risks in infant care with an epileptic mother.

Temporal lobe epilepsy may be associated with

psychiatric morbidity, including a schizophrenia-

like   psychosis,   and   depression   is   common.

Unfortunately, most antidepressant and antipsy-

chotic drugs lower seizure threshold and thus

make  treatment  of  these  symptoms  difficult.

Examples of drugs with the lowest proconvul-

sant effects are SSRIs and haloperidol. The possi-

bility  of  drug  interactions  then  needs  to  be

monitored.

Morbidity and mortality

There is an approximate doubling of the stan-

dardized mortality ratio, mostly related to refrac-

tory cases or from poor management leading to

complications from seizures. There may be falls,

home accidents, etc., where the link to a seizure

is  evident.  In  sudden  unexpected  death  in

epilepsy (SUDEP), which affects about 0.5% of

refractory cases per year, there is no obvious

cause  of  death,  which  is  unobserved,  but  is

assumed to be seizure-related. There is a fivefold

increased risk of suicide.

Investigation and diagnosis

The first aim when a patient presents with an

unaccountable fall, blackout, etc. is to ascertain

if there is any identifiable underlying cause, e.g.

medical or toxic (Figure 6.18). A description of

the seizure, from both the patient and especially

a witness, is important. In certain circumstances

a  seizure  may  be  induced  artificially  under

controlled conditions with EEG recording, or

system

there   may   be   continuous   ambulatory   EEG recording. An ECG may also be needed to elimi-

nate a cardiac cause.

The EEG lacks specificity and sensitivity. Some

15% of otherwise normal people will give an

abnormal trace, while only 50% of patients with

epilepsy  will  show  abnormalities  on  random

testing. An EEG is most useful in defining the

seizure type, because it is possible to distinguish

generalized from partial or secondarily general-

ized seizures. MRI has replaced radiography as

the procedure for the accurate identification of

cranial lesions, which it can identify in over half

of chronic cases, of which 75% are partial epilep-

sies, 25% generalized forms. Surgery can occa-

sionally rectify such problems, but usually it is

reserved for cases where the disease is refractory

to drug treatment, which is almost always tried

first.

Decision to treat

Formerly it was suspected that, following only a

single seizure, delaying treatment and allowing

further attacks would worsen prognosis. It now

seems  clear  that  this  is  unlikely  and  further,

that reduced quality of life results from prema-

ture initiation of drug therapy. This is particu-

larly the case in children. In the UK at present

the general rule is that a single attack does not

automatically warrant the initiation of regular

medication.

A diagnosis of epilepsy has serious legal and

social implications (e.g. for driving, pregnancy,

education, employment and leisure activities),

and   treatment   involves   the   likelihood   of

adverse drug effects. Epidemiological evidence is

ambiguous as to the probability of subsequent

fits  following  a  single  episode,  especially  in

childhood. Estimates of such probability range

from 25% to 75%.

The  circumstances  of  the  first  seizure  are

crucial to the decision. How likely is there to be

a recurrence? Factors to consider are the cause or

provoking event, if known, and whether this is

persistent, e.g. was it a manifestation of a tran-

siently reduced threshold or a unique event,

such as high fever. More than two-thirds of

children who have a febrile convulsion have no

further  problems.  Is  there  objective  evidence (EEG, MRI, etc.) or neurological abnormality? Certain syndromes are more likely to recur (e.g. absences, juvenile myoclonic epilepsy).

Another factor is how serious would be the

consequences of a further attack. Obviously if a

patient  drives  a  public  service  vehicle,  for

example, the risk cannot be taken. Even if treat-

ment  is  delayed,  avoiding  driving  for 1 year

following a single seizure is advisable. Similar

prudence  might  be  exercised  with  a  mother

nursing  an  infant,  who  may  be  dropped  or

drowned if the mother has a seizure.

In all circumstances, decisions about starting

drug treatment must involve informed discus-

sion with the patient. Some patients who have

fits  no  more  than  once  every  few  years  may

prefer  to  risk  these  rather  than  to  undergo

long-term  drug  therapy  and  its  associated

problems.

Management

The aims of the management of epilepsy are:

•  Investigate possible primary causes.

•  Minimize  social  and  psychological  conse-

quences and complications.

•  Decide if active prophylaxis is necessary.

•  Use drug therapy to prevent seizures or to

keep seizures to minimum compatible with acceptable adverse reactions.

The over-riding consideration is to maximize the patient’s quality of life by careful balancing of risks and benefits.

General measures

Careful, sensitive and thorough counselling is

important, especially with children and their

parents. Patients must be prevented as far as

possible from being or feeling stigmatized. It

must   be   emphasized   that   between   attacks

patients  are  perfectly  normal,  and  that  the

disease causes no impairment of intelligence and

no psychiatric disorder. It is helpful if not only

the family but also a teacher, school friend or

work colleague can be taught how to deal with a

seizure.

Patients are encouraged to lead as normal a life

as possible. There are certain legal constraints,

notably on driving. In the UK the current require-

ment  for  epilepsy  patients  to  hold  a  driving

licence is that the patient must have had no fits at

all  for  a  year,  or  no  daytime  fits  for 3 years.

Patients who are well controlled and seizure-free

may have normal schooling, and even swim. The

number of constraints is much reduced if the

patient has recognizable prodromal signs or a

consistent aura.

Principles of pharmacotherapy

Drug therapy does not alter the basic lesion,

whatever  it  is,  but  can  be  very  effective  in

preventing seizures. The limiting factor is the

adverse effects, especially as treatment is likely to

be prolonged. Sometimes the doses needed for

complete seizure suppression may be intolerable

and it is preferable simply to aim for a reduced

seizure  frequency.  The  risks  from  continued

seizures in the absence of treatment must be

weighed against the risks of adverse effects and

the   potential   benefits   of   reduced   seizure

frequency with treatment.

A decision  tree  for  managing  drug  therapy is given in Figure 6.19. The following general principles should be observed.

Monotherapy

Monotherapy is now the accepted ideal. There

are considerable benefits to avoiding polyphar-

macy in epilepsy in view of the many possible

drug interactions. If the first drug proves inade-

quate or is not tolerated after optimal dose titra-

tion, another should be substituted, rather than

adding another drug. Two first-line agents appro-

priate for the patient and their seizure type and

syndrome should be tried singly before a combi-

nation is tried. Over two-thirds of patients are

system

satisfactorily controlled on a single drug, and

those who are not are likely to prove the most difficult to control.

Formerly, insufficient care was taken to achieve

the  optimal  plasma  level  with  the  first  drug

chosen, and instead of trying a second single

drug, a second or even third adjunctive drug was

commonly added quite early. Some older patients

may  still  be  on  polypharmacy (and  should

remain so), but this approach probably produces

more problems than it solves and is now avoided.

Dual therapy

If two first-line agents used alone fail, a further

15% of patients may be controlled on a combi-

nation of two drugs. Selection of drugs to be

used in combination involves consideration of

minimal overlap in adverse effects, least chance

of  interaction  and  complementary  modes  of

action. There are few evidence-based data on the

relative effectiveness of different combinations.

The  remaining  patients  can  be  difficult  to

control, but triple therapy should be avoided if

possible. Failure to achieve control with a variety

of drugs as monotherapy or dual therapy at

optimal  plasma  levels  suggests  the  need  for

further investigation and perhaps a revised diag-

nosis.  In  all  such  cases  specialist  referral  is

mandatory.

Gradual initiation

Low drug doses are used initially and gradually

increased until control is achieved or unaccept-

able toxic effects occur, carefully adjusting doses

according to individual patient response. This

requires patience: a half-life of 36 h or more (e.g.

phenytoin) means that a new steady state will

not be reached until about 7-14 days after any

change in dose.

Compliance

The importance of compliance must be stressed

to the patient, and this must include a clear

explanation about the range and likelihood of

potential  adverse  effects.  Seizure-free  patients

can easily decide that they are cured and stop

taking medication: this is a common cause of

‘failure’ of therapy. This is especially true of

younger patients. Abrupt withdrawal of drugs

may precipitate a seizure.

Switching

Particular care is needed when switching AEDs. There must be a gradual process of tapering the dose  of  the  original  while  titrating  up  the

replacement.   Close   monitoring   is   essential throughout to ensure continuity of control with minimal side-effects. Similar precautions apply when adding in an adjunctive agent.

Plasma level monitoring

It is not necessary to monitor patients’ blood

levels regularly. However, the non-toxic thera-

peutic  ranges  of  most  AEDs  are  known  and

therapeutic  drug  monitoring  may  be  helpful in  a  number  of  circumstances,  to  check  for possible abnormalities in absorption, clearance, interaction and compliance:

•  During initiation, if expected response is not

            achieved.

•  When previously stable control deteriorates. •  In pregnancy.

•  In suspected toxicity.

•  When doses are changed or other drugs are

            initiated.

•  When there is renal or hepatic impairment.

Consistent formulation

Efforts should be made to ensure that patients continue taking the same manufacturer’s form of AED. Unpredictable changes in bioavailability often follow changes in dose form or formula-

tion, with loss of control or increased toxicity. Special  vigilance  is  necessary  when  patients move between primary and secondary care, or between different prescribers or pharmacies, and for patients taking phenytoin.

Medication record

The   patient   should   carry   and   maintain   a

complete  medication  record,  including  OTC

drugs.

Antiepileptic drugs

It is possible to make some generalizations about AEDs where there are broad similarities, but the reader is referred to an official formulary for

detailed information on each drug. Some of the more important features are given in Table 6.32 and discussed below.

Mode of action

Although most AEDs can be assigned a specific

neuropharmacological  mode  of  action,  it  is

uncertain to what extent this correlates with

their clinical action. Nevertheless, knowledge of

the class to which a drug belongs facilitates

rational combinations when these are necessary.

There  are  three  main  classes.  Most  AEDs

interfere with voltage-dependent high-frequency

sodium  channels  to  limit  the  spread  of  the

neuronal instability by inhibiting unnaturally

rapid firing. This group includes the common

agents used in generalized and partial seizures,

such as carbamazepine, lamotrigine, phenytoin and

perhaps valproate. Others facilitate the inhibitory

transmitter  GABA,  which  stabilizes  neuronal

membranes; this includes especially the CNS

sedative  agents  phenobarbital  and  the  benzo-

diazepines.   Vigabatrin   irreversibly   inhibits

GABA   catabolism.   Although   valproate   also

inhibits GABA-catabolic enzymes, other actions

contribute to its wide spectrum of antiepileptic

activity. Some drugs active in absence seizures,

system

especially ethosuximide, block voltage-dependent calcium T-receptors.

The modes of action of some of the newer AEDs

have not yet been fully elucidated, e.g. topiramate,

gabapentin. Approaches currently being explored

include inhibition of excitatory transmitters such

as glutamate.

Pharmacokinetics

The handling of AEDs is varied and complex.

There may be great variation in plasma levels for

a given dose, both between patients and even in

the same patient at different times. For some

drugs this may make plasma level monitoring

advisable at certain times, but routinely it is

usually   sufficient   to   monitor   clinically,   by

freedom from seizures and adverse effects.

Absorption

This  may  be  highly  formulation-dependent,

especially  with  phenytoin,  which  makes  it

unwise to change brands or dosage forms. This

does  not  preclude  using  generic  drugs,  as  is

sometimes erroneously believed, provided that

one particular proprietary formulation is used

consistently.  For  drugs  like  phenytoin,  with  a

narrow therapeutic index, variation in bioavail-

ability may permit seizure breakthrough on the

one  hand  or  excessive  toxicity  on  the  other.

Similar considerations apply to potential inter-

actions.  Absorption  rate  and  bioavailability

may  be  reduced  by  food  and  antacids.  IM

phenytoin  is  very  poorly  absorbed,  so  the  IV

route is essential if parenteral therapy is neces-

sary. The prodrug fosphenytoin causes less local

irritation.

Distribution

Clearly,  all  AEDs  are  sufficiently  lipophilic  to

enter the brain but they do so at different rates.

Benzodiazepines enter quickly to act most rapidly

and  so  are  the  standard  treatment  for  status

epilepticus. They have the highest volumes of

distribution due to central accumulation. Most

AEDs are highly plasma protein bound. Conse-

quently,  free  drug  levels  may  be  sensitive  to

displacement owing to competition for binding

by other drugs, including other AEDs. Whether or

not this has clinical consequences is not easy to

predict, partly because hepatic clearance will be

increased by higher plasma levels, and a formu-

lary should always be consulted for the latest

guidance.

Clearance

Most AEDs undergo extensive hepatic metabo-

lism  but  vigabatrin,  gabapentin  and  topiramate are cleared renally. Valproate, carbamazepine and phenobarbital have active metabolites. Rates and extents  of  metabolism  vary  according  to  age, other disease, genetic factors and enzyme induc-

tion or inhibition by other drugs, usually in a

predictable manner (see Chapter 1). A number of factors further complicate this.

Many  AEDs  are  hepatic  enzyme  inducers,

especially carbamazepine, phenytoin  and pheno-

barbital.   Some   also   cause   auto-induction,

producing   subsequent   difficulties   in   dosage

adjustment; with carbamazepine the half-life can

reduce from 50 h at the start of therapy to less

than one-third of this later on. The very similar

oxcarbazepine  is  far  less  troublesome  in  this

regard. There is commonly a mutual interaction

between AEDs, each enhancing the clearance of

the other. AEDs will also enhance the clearance

of other drugs by the same mechanism; pheny-

toin and others reduce the efficacy of oral con-

traceptives  in  this  way,  which  is  especially

important in view of the potential teratogenic

action of many AEDs. Valproate uniquely is an

enzyme inhibitor, so can enhance the action of

other, hepatically cleared AEDs.

The enzyme systems involved are also poten-

tially saturable, especially with phenytoin. This may permit plasma levels to enter the toxic

range inadvertently while treatment is initialized or the dose changed, because a dose increase

similar to the previous one may produce an

unexpectedly large effect (Figure 6.20). Phenytoin dose  changes  in  particular  must  be  gradual, allowing the usual five half-lives to attain steady state before alteration, and in small increments (25 mg) above about 200 mg/day.

Duration of action

Most AEDs have a sufficiently long half-life as

to  require  only  once-daily  dosing.  Notable

exceptions  are  valproate,  carbamazepine  and

gabapentin.

Side-effects

Consideration of the adverse effects of the AEDs

is important because they may have to be given

for long periods and compliance is sometimes a

problem (see Table 6.32). This occurs in partic-

ular with adolescents on phenytoin because of the

disagreeable, though not dangerous, cosmetic

side-effects. In addition, as noted above, pheny-

toin causes dose-related toxic effects to occur at

plasma levels that are only a little above the ther-

apeutic range, and these are important markers

of over-dosage (Figure 6.21). Other AEDs do not

have such a narrow therapeutic index; for this

reason  phenytoin  has  fallen  from  favour  as  a

first-line AED.

CNS.   Most   AEDs   have   some   non-specific

depressant  or  sedative  action,  but  especially

phenytoin, phenobarbital and its prodrug, primi-

done. In the long term there may be cognitive

and behavioural impairment with these older

drugs, especially phenobarbital; nystagmus and

ataxia may also occur. Vigabatrin and tiagabine

can   cause   psychiatric   disturbances.   Carba-

mazepine and valproate, and many of the newer

agents,  are  significantly  less  troublesome  in

these respects, which is one reason why they

have replaced the former three as first choices.

Gastrointestinal   tract.   Many   AEDs   cause gastrointestinal distress (especially valproate), but care is needed with the use of antacids as they may impair absorption of some AEDs. Enteric-

coated preparations are preferred where available. Rarely,  valproate  causes  hepatic  or  pancreatic injury (monitoring essential).

Skin.   There  are  serious  cutaneous  reactions

with phenytoin, including acne, hirsutism (exces-

sive hair growth), coarsened looks and gingival

hyperplasia (gum  overgrowth).  Valproate  can

cause hair loss. Lamotrigine  and carbamazepine

can cause rash that is severe enough to force

discontinuation.

Eyes.   Both  vigabatrin  and  topiramate  require ophthalmological   monitoring.   The   former causes visual field defects and the latter raises intraocular pressure.

Metabolic.   A number of AEDs, notably pheny-

toin  and carbamazepine, lower vitamin D and folate levels, partly because enzyme induction accelerates  the  catabolism  of  these  vitamins. Osteomalacia (rickets) or megaloblastic anaemia may follow. Folate supplementation is recom-

mended before and during pregnancy to prevent neural tube defects (see Chapter 11).

Systemic.   Haemopoietic,   hepatic   or   renal disturbances   are   rare.   Nevertheless,   blood and   liver   monitoring   should   be   performed regularly.   Carbamazepine  occasionally   causes hyponatraemia.

Teratogenic  effects.   Many  AEDs  present  a

serious dilemma in the therapy of young women.

They  can  enhance  the  metabolism  of  oral

contraceptives, possibly promoting contraceptive

failure. Because most teratogenic effects occur

very early in pregnancy, possibly before it is real-

ized, any damage would already have been done.

AED  therapy  carries  a  small  but  definite  risk

(about two to three times normal) of fetal abnor-

mality;  particular  offenders  are  valproate  and

phenytoin. Yet both fetus and mother may suffer,

the former irreversibly from anoxia, if a pregnant

woman has frequent major seizures. A further

difficulty  in  evaluating  the  risk-benefit  is  the

fact that fetal abnormalities are slightly more

common among female epilepsy sufferers, even if

untreated  with  AEDs.  Other  potential  toxic

effects  on  the  fetus  include  sedation,  enzyme

induction,   and   even   neonatal   withdrawal

seizures. Neonatal bleeding may occur due to

impaired vitamin K transplacental transport, and

vitamin K is used immediately before delivery. In

addition, the clearance of many AEDs increases during pregnancy, so plasma levels have to be

monitored carefully. AEDs are secreted in small

quantities in breast milk, but this does not seem

to be a serious clinical problem. Whether or not

AED therapy should be continued depends on the

risks of withdrawal. The current view is that treat-

ment in women should be continued, but with

careful monitoring to maintain control at the

lowest  possible  dose,  preferably  with  a  single

AED.  Carbamazepine  appears  to  be  the  safest

option.

Drug selection

Despite determined efforts to produce a precise classification of epilepsy and frequently quoted recommendations for different types of seizure, a systematic rationale has not emerged for linking drug to seizure type. Evidence is slim; few direct comparisons have been made and recommen-

dations are still basically empirical. The classifi-

cation by epilepsy syndrome has enabled some improved  targeting  and  the  current  NICE

guidelines (2004) give best choices.

Table 6.33 provides the currently agreed best

choices, based on the NICE guidelines and the

BNF. There seems to be no systematic distinction

between  choices  for  generalized  tonic-clonic

Epilepsy           453

seizures and those for partial seizures, but there

is  between  these  two  seizures  types  and  for

absences.  Patient  factors  such  as  tolerability,

adverse effects and interactions are more impor-

tant  criteria  informing  the  decision  on  drug

therapy.

Epilepsy is the sole remaining indication for

phenobarbital,  which  escapes  the  strict  legal

control of barbiturates in the UK, but its use is

declining.

Interest continues in the use of high-fat keto-

genic diets as an adjunct in resistant childhood epilepsy. This is based loosely on the apparent antiepileptic effect of starvation, but it is not an established or widely used approach.

Benzodiazepines  are  the  first-line  drugs  of

choice in most forms of status epilepsy. The

current recommendations for convulsive status

are:

•  Threatened  status  (i.e.  premonitory  stage;

seizures  lasting  longer  than  5 min,  or  of

greatly inreasing frequency): rectal diazepam or buccal midazolam.

•  Established status: IV lorazepam, adding IV

            phenytoin if control not achieved.

•  Refractory status: referral to an intensive care

            unit for anaesthesia.

Withdrawal of drug therapy

After a suitable seizure-free period the possibility of gradual withdrawal should be discussed with the patient. The likelihood of a lengthy remis-

sion  is  increased  in  patients  with  childhood onset,  epilepsy  of  short  duration,  previous

good control, a normal EEG, and no evidence of  a  primary  cause,  e.g.  head  injury,  mental retardation or an adverse MRI scan.

At  present  it  is  generally  agreed  that  the

patient must have at least two seizure-free years

before stopping treatment can be considered,

but the more cautious would wait for up to

5 years. It also depends on patient and clinician

preference (e.g. does the patient need to drive?).

Withdrawal  should  be  phased  over  at  least

6 months, one drug at a time if on multiple

therapy. Even if withdrawal is successful, the drug therapy itself cannot be considered to have brought about a cure: it is more likely that the disease has remitted spontaneously.