72ournal of Neurology, Neurosurgery, and Psychiatry 1993;56:742-751

NEUROLOGICAL EMERGENCIES

Delirium

David Taylor, Sh6n Lewis

The concept of delirium has been with us forover two thousand years. Greek physicians,such as Hippocrates, described its essentialfeatures' and the Roman writer Celsus distin-guished it from mania and depression.2 Galendifferentiated between primary (idiopathic)and secondary (symptomatic) forms of thedisorder.'

Delirium is common in the hospital set-ting.4 It often goes undiagnosed in its earlystages5 and may therefore present as a neuro-logical emergency. If untreated, it is associa-ted with a high mortality.6 The followingarticle will therefore pay close attention to thediagnosis of the condition and its propermanagement. In addition, we will review itsepidemiology, aetiology and pathogenesis.

Department ofPsychiatry, CharingCross Hospital,London, UKD TaylorDepartment ofPsychiatry, CharingCross andWestminster MedicalSchool, St Dunstan'sRoad, LondonW6 8RP, UKS Lewis

Correspondence to:Dr Lewis

TerminologyA general problem in psychiatry has been thecomplex etymology of clinical terms. Thisproblem has been largely rectified as a resultof the move towards operational definitions inboth the International Classification ofDiseases, 10th edition (ICD 10) (table 1)7and the American Diagnostic and StatisticalManual of Mental Disorders, 3rd edition,revised (DSMIIIR) (table 2).8 "Delirium" isnow the accepted term for acute, transient,global, organic disorders of higher nervous

system function involving impaired con-

sciousness and attention. It is synonymouswith the "acute confusional state" of ICD-9,9which term it replaced, and is also referred toas "acute organic reaction" and "acute brainsyndrome".The problem remains of a lack of consen-

sus as to what constitutes the definition ofsome particular symptoms or signs in psychi-atry. Neither ICD1 07 nor DSMIIIR8 containsglossary definitions for individual clinical fea-tures in delirium. In this paper, symptomsand signs will be defined with regard to theirclinical utility in diagnosing delirium. Itshould be borne in mind, however, that manyterms such as "consciousness"'0 and "mem-ory"" are used differently by different groups,even within medicine. The terms "confu-sion". "clouding" and "sensorium" in partic-ular should be avoided on account of theirlack of standard definitions.

EpidemiologySophisticated epidemiological data are nowavailable for most psychiatric disorders.

However, accurate data on incidence, preva-lence and mortality in delirium are difficult tocome by. The comparison of estimates acrossstudies is hampered by methodological differ-ences. Methods of case finding and diagnosiswill influence rates obtained, as will thepatient population and the setting (commun-ity, general medical, surgical or geriatric inpa-tient) in which the disorder is diagnosed.'2Most incidence and prevalence studies ofdelirium have been conducted in inpatientsettings. One notable exception is the EasternBaltimore Mental Survey, a large communitybased survey forming part of theEpidemiologic Catchment Area (ECA) pro-gramme."3 This study aimed to look at theprevalence of delirium in the general adultpopulation aged between 18 and 64 years. Atotal of 810 individuals were subject to psy-chiatric evaluation, and a point prevalencerate of 04% was calculated, rising to 1-1%for those aged 55 years and over. Severalother risk factors could be identified. Whencompared with dementia sufferers and toindividuals of the same age who did not sufferfrom a psychiatric disorder, delirium suffererswere found to have more medical conditions,take more medication and have higher levelsof physical disability.Whereas prevalence rates in the commun-

ity are low, in hospital settings they are high.It has been estimated that 10% of all medicaland surgical inpatients meet criteria for delir-ium at some point during their stay.4Delirium may therefore represent the mentaldisorder with the single highest incidence.'4As with community subjects, advanced age

is a potent risk factor for delirium in inpatientsurveys. Amongst elderly general medicalinpatients, rates of 30%'5 or even 50%16 havebeen reported, although a more conservativeestimate of 17%"7 is probably more realistic.Rates of delirium in elderly surgical patientsare broadly similar, although following opera-tion for hip fracture the rate may be as highas 5O%.'8A pre-existing dementia seems also to be a

risk factor, independently of age. A largeFinnish study'9 found that in 2000 patientsaged 55 and over admitted to medical wards,delirium was present in over 40% of thosepatients with evidence for an establisheddementia, compared with 15% on admissionfor the group as a whole.The diagnosis of delirium carries a signifi-

cant mortality rate,6 particularly in theelderly.20 Development of delirium doubles

742 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Delirium

Table I ICD 10 Diagnostic Criteria for Delirium

For a definite diagnosis, symptoms, mild or severe, should be present in each one of thefollowing areas:

A) impairment of consciousness and attention (on a continuum from clouding to coma;reduced ability to direct, focus, sustain and shift attention);

B) global disturbance of cognition (perceptual distortions, illusions and hallucinations-mostoften visual; impairment of abstract thinking and comprehension, with or without transientdelusions, but typically with some degree of incoherence; impairment of immediate recall andof recent memory but with relatively intact remote memory; disorientation for time as well as,in more severe cases, for place and person);

C) psychomotor disturbances (hypo- or hyperactivity and unpredictable shifts from one tothe other; increased reaction time; increased or decreased flow of speech; enhanced startlereaction);D) disturbance of the sleep-wake cycle (insomnia or, in more severe cases, total sleep loss or

reversal of the sleep-wake cycle; daytime drowsiness; nocturnal worsening of symptoms;disturbing dreams or nightmares, which may continue as hallucinations after awakening);

E) emotional disturbances, for example, depression, anxiety or fear, irritability, euphoria,apathy or wondering perplexity.The onset is usually rapid, the course diurnally fluctuating, and the total duration of the

condition less than six months. The above clinical picture is so characteristic that a fairlyconfident diagnosis of delirium can be made even if the underlying cause is not firmlyestablished. In addition to a history of an underlying physical or brain disease, evidence ofcerebral dysfunction (such as, an abnormal electroencephalogram, usually but not invariablyshowing a slowing of the background activity) may be required if the diagnosis is in doubt.

the risk of the patient dying"l over a period ofhours to weeks, depending on the underlyingcondition. Of equal importance, however, isthe fact that successful treatment of the con-dition eliminates much of this excess mortal-ity. Even so, death occurs in 25% of elderlyhospitalised patients with delirium." Thisunderlines the importance of early detectionand urgent treatment.

AetiologyDelirium is a consequence either of a primarybrain lesion or of cerebral involvement sec-

ondary to systemic illness, including thosecases caused by exogenous substances such asdrugs and poisons. A wide variety of causesact by way of a common pathway of electricalor neurochemical disturbance to produce theclinical syndrome."Some of the more common causes of delir-

ium are given in table 3. Almost any suffi-ciently severe acute medical or surgicalcondition may, under the right circum-stances, cause the syndrome. Whilst mostcases of the clinical syndrome present littledifficulty in determining a cause, a substantialminority of patients (as high as 5-20% of theelderly delirious)20 never receive an aetiologi-cal diagnosis. In this situation, it may be nec-

Table 2 DSMIIIR Diagnostic Criteria for Delirium

A) Reduced ability to maintain attention to external stimuli (for example, questions must berepeated because attention wanders) and to appropriately shift attention to new external stimuli(for example, perserverates answer to a previous question).

B) Disorganised thinking, as indicated by rambling, irrelevant or incoherent speech.C) At least two of the following:

(1) reduced level of consciousness, for example, difficulty keeping awake duringexamination

(2) perceptual disturbances: misinterpretations, illusions or hallucinations(3) disturbance of sleep-wake cycle with insomnia or daytime sleepiness(4) increased or decreased psychomotor activity(5) disorientation for time, place, or person(6) memory impairment, for example, inability to learn new material, such as the names of

several unrelated objects after five minutes, or to remember past events, such as history ofcurrent episode of illness.D) Clinical features develop over a short period of time (usually hours to days) and tend to

fluctuate over the course of a day.E) Either (1) or (2):(1) evidence from the history, physical examination or laboratory tests of a specific organic

factor (or factors) judged to be aetiologically related to the disturbance(2) in the absence of such evidence, an aetiologic organic factor can be presumed if the

disturbance cannot be accounted for by any nonorganic mental disorder, for example, manicepisode accounting for agitationand sleep disturbance.

essary to reconsider the diagnosis, but casesremain where the clinical diagnosis is not indoubt, and the aetiological agent simplyremains unidentified. It must also be remem-bered that more than one aetiological factormay be contributing to the patient's delirium,particularly in the elderly.2'The commonest causes of delirium include

alcohol abuse and withdrawal, stroke, dia-betes, ischaemic heart disease, pneumoniaand urinary tract infections.'4 Almost anymedically prescribed drug can cause deliriumif taken in sufficient overdose, but anticholin-ergic and hypnotic agents in particular areknown for their propensity to provoke the dis-order, especially in the elderly.'5

Pathology and pathogenesisThe neuropathology associated with deliriumis obviously determined by the cause of theunderlying disorder, but it is important tonote that brains of patients dying with delir-ium often reveal no obvious macroscopic ormicroscopic pathological changes,26 reflectingthe fact that many cases -are related to sys-temic disturbance resulting in global highernervous system dysfunction at a cellular ormolecular level. This consideration limitsinquiry into which sites are critical in mediat-ing the psychological disturbance in delirium.The frontal lobes are known to subserve psy-chological functions considered to be of pri-mary importance in delirium, such asattention.'7 However, localised lesions else-where in the brain may also cause the condi-tion.'8 This observation has three possibleexplanations: a widely distributed substratefor consciousness and attention (whichappears to be the case, including the pre-frontal, cingulate and parietal cortices, thereticular activating system, and thalamicprojections)," functional diaschisis,'9 orincreased intracranial pressure.30 Any or all ofthese three explanations may be relevant inan individual patient. Likewise, whilst it hasbeen argued that lesions of the right cerebralhemisphere are more prone to result in delir-ium,2' as a result of attentional processes pos-sibly being dominant to the right side," leftsided lesions may also give rise to the condi-tion.32

Although it is often assumed that deliriumis associated with impaired cerebral oxidativemetabolism, there is little direct evidenceeither to support or refute this claim. Normalcerebral function relies almost entirely on theoxidative metabolism of glucose." In the nor-mal subject there is usually a close correlationbetween neural activity, regional cerebralmetabolism and cerebral blood flow.34 Studiesof cerebral metabolic rate in delirium are rareand lack consistency in their results. In onestudy, a correlation was demonstratedbetween cerebral metabolic rate and level ofconsciousness:33 cerebral oxygen consump-tion decreased as the subjects passed fromnormal consciousness through "confusion" tocoma. However, delirium can occur in theabsence of, or precede, failure of cerebral

743 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Taylor, Lewis

Table 3 Causes of delirium

A) Primary central nervous system causes.Head injuryCerebrovascular: stroke, subdural haematoma, transient ischaemic attack.Raised intracranial pressure.Intracranial infection: Encephalitis, meningitis.Epilepsy: Ictal, post-ictal.

B) Secondary to systemic illness.Infections: chest, urinary tract, septicaemia, malaria, HIV.Cardiovascular: Infarction, cardiac failure, arrhythmia.Metabolic: hypo/hyperglycaemia, uraemia, hepatic failure, electrolyte disturbances.Endocrine: Addisonian crisis, disturbance of thyroid, parathyroid.Alcohol: Wernicke's encephalopathy, delirium tremens.Prescribed drugs: psychotropic drugs, steroids, digoxin, cimetidine,anticonvulsants, anticholinergics, overdosage.

Illicit drugs.C) Rare causes.

Systemic lupus erythematosus, porphyria, vitamin B12 or folate deficiency, pellagra,heavy metal poisoning, hypothermia, Wilson's disease, remote effect of carcinomahypertensive encephalopathy.

metabolism35 and additional mechanismsmust therefore play a part.

Studies on delirious patients measuringregional cerebral blood flow are equallysparse and contradictory. In the consciousstate, the frontal lobes are activated relative topost-frontal regions of the brain. A case ofdelirium has been reported showing an

absence of this normal physiological "hyper-frontality" on measurement of cerebral bloodflow.36 Another study reported no correlationbetween cerebral blood flow and psychologi-cal impairment in delirium tremens.37

In the delirious patient, the deviation ofelectrical activity on the EEG from that of thenormal waking state correlates well with theclinical condition38: in most cases, the greaterthe slowing of the EEG trace, the more clini-cally impaired the patient.38 Delirium associ-ated with withdrawal states such as deliriumtremens do not follow this general rule andtend to be associated with fast wave activitysuperimposed on generalised slowing of thetrace, suggesting additional pathogeneticmechanisms.39 The rate of change of fre-quency is important: single EEG recordingsmay occasionally give a false impression in

delirious patients with unusually high or lowpremorbid baseline frequencies,40 implyingthe desirability of serial EEG recordings inthe diagnosis of delirium. These changes are

reversible and mirrored in the patient's clini-cal recovery.39 41EEG recordings made during the day in

delirious patients reveal disorder of the circa-dian rhythm, with traces characteristic ofwakefulness, somnolence and sleep emergingin a disorganised manner. Night traces like-wise display breakdown of the normal rhythmof sleep stage progression as well as reductionin the overall sleep time.42The exact nature of the neurochemical cor-

relates of delirium are poorly understood.Claims have been made for the primacy ofacetylcholine in this context on both clinical43and experimental44 grounds. In the past, simi-lar claims were made in Alzheimer's dementiaresearch which proved to be unfounded, andcaution is therefore essential in ascribing a

primary role for acetylcholine in the patho-genesis of delirium. Nonetheless, hypoxiaoccurring with decreased cerebral oxidative

metabolism results in a generalised decreasein neurotransmitter synthesis, preferentiallyaffecting acetylcholine.45 Blockade of centralcholinergic projections certainly can result indelirium44 and it is of some interest that anti-cholinergic agents reduce cerebral perfusionin the frontal cortex.46 However, cerebralstructures and pathways implicated in highercentral nervous system processes such as con-sciousness and attention are legion47 48 and theprimary neurochemical abnormality in delir-ium has yet to be clarified.

Clinical featuresAs noted by Lishman,49 one of the mostintriguing aspects of delirium is the constancyof the clinical picture despite the wide varietyof different causes.

Traditionally, impaired consciousness is thecardinal sign of delirium. The main problemhere is that consciousness is a complex, ill-defined concept. A number of differentaspects of consciousness have been describedincluding awareness (of the extemal environ-ment), alertness and wakefulness.22Awareness is always impaired in delirium.Alertness to environmental stimuli may beabnormally elevated or lowered. If hyperalert,the patient responds to stimuli without dis-crimination, with the clinical result that he isdistracted by irrelevant events at the expenseof the more relevant. It is characteristic ofdelirium tremens and may be accompaniedby overactivity. If hypoalert, there is an over-all reduction in response to environmentalstimuli; for example, the patient may remainmute as the physician attempts to take a his-tory. This is characteristically seen in meta-bolic encephalopathies and may be associatedwith underactivity. Any one individual deliri-ous patient, however, may alternate betweenhyper- and hypoalertness, as well as exhibit-ing relatively lucid periods.The sleep-wake cycle is invariably disrup-

ted in delirium: the patient may exhibitexcessive drowsiness by day and/or insomniaat night. In severe cases there may be totalinsomnia or reversal of the cycle. Sleep maybe accompanied by nightmares which maythen merge into frank hallucinations on wak-ening.The difficulty in clinically detecting mild

degrees of impaired consciousness is reflectedin DSMIIIR,8 where this feature need nolonger be elicited to make a diagnosis of delir-ium. Instead, the related, more robust notionof impaired attention is seen as the centralfeature. Impaired ability to direct, focus, sus-tain or shift attention is seen in delirium.8Thus questions may need to be repeated andthe patient will have difficulty with such basicbedside tasks as Serial Sevens (subtraction ofseven from one hundred, then seven fromninety three and so on), and Digit Span(immediate verbal recall of a string of six orseven digits). It is, however, essential toappreciate that impairment on any basic testof attention is not an absolute indication ofdysfunction: possible confounding factors

744 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Delirium

include the patient's age, level of educationalattainment and coexisting anxiety. It has beenclaimed that serial sevens and digit span donot even allow differentiation of organic from"functional" disorders, let alone deliriumfrom dementia.505' If doubt remains aboutthe presence of attentional deficit, one meansof helping to decide the matter is to giveincreasingly simple tasks, such as requestingthe months of the year or days of the week inreverse, serial threes or serial ones (asking thepatient to subtract one from twenty, then onefrom nineteen and so on).

These impairments of attention are essen-tially impairments of immediate memory. Anumber of forms of memory impairment maybe seen in delirium. Immediate ("working")memory relates to the ability to hold informa-tion for very short periods of only a few sec-onds. In contrast, short-term memory refersto information recalled after a delay of min-utes. Long-term memory refers to the recallof events that occurred days, weeks, monthsor years previously.52

In delirium, disordered immediate memoryis primary. Secondary, anterograde short-term deficits result, and long-term memorydeficits may also be present,39 although gener-ally there is relative preservation of long-termmemory except in the most severe cases.Thus the anmestic deficit, whilst primarilyanterograde, may in severe cases be retro-grade.

Short-term memory deficit may be elicitedby the patient's inability to recall completelyafter five minutes a seven unit name andaddress or even three nouns (for example,dog, wall, sun) after correct registration.Registration (the immediate repetition of theunits) is obviously contingent on adequateattentional processes and it therefore followsthat if those processes are impaired to such anextent that the patient cannot correctly regis-ter the units, then formal testing of short-term memory is not possible.

Disorientation for time, place and personreflects abnormalities in immediate and short-term memory.53 It does not refer to the pari-etal lobe phenomenon of "topographicaldisorientation". Disorientation for the exacttime of day is considered a sensitive indicatorof delirium.54 As the disorder progresses andbecomes more established the patient charac-teristically loses the ability to correctly iden-tify the day of the week, the month of theyear or even the year itself. In relatively mildcases of the disorder the patient may be fullyorientated for place and person but disorien-tated for time.

Disorientation for place refers to inabilityto identify correctly where he or she is physi-cally located at the time of interview.However, some consideration must be givento the context in which the interview occurswhen assessing this. For example, a patient athome may be able to identify this correctly,yet become totally disorientated for placewhen taken from his familiar surroundingsand admitted to hospital.5556 Presumably insuch cases a relatively unimpaired access to

the long term memory store compensates forthe underlying functional deficit.

Misidentification of persons often occursand usually involves mistaking unfamiliar per-sons for familiar.57 Here again context isimportant. For example, a patient's inabilityto identify his recently acquired hospital doc-tor would not give as much cause for concernas his being unable to recognise his wife.Global misidentification of persons is com-mon in delirium and must be differentiatedfrom prosopagnosia which is a selectiveinability to recognise an individual by theirfacial characteristics; this occurs as a result ofdamage to the non-dominant occipitotempo-ral gyrus.

Thought content in delirium has oftenbeen noted to have an oneroid (dream-like)quality.58 59 Delusions, when they occur, aretypically transitory and fragmentary, and usu-ally of a persecutory nature. Different studiesput their prevalence at between 40 and100%.6062

Patients often experience abnormal percep-tions. These have been shown to be related toimpairments of perceptual processing.6' Thepercept may be distorted, and happens mostoften in the visual modality. Examplesinclude metamorphopsia (distortions in theperceived size of objects including macropsiaand micropsia), dysmorphopsia (alterations inshape) and polyopsia (a single object per-ceived as multiples). In addition, the patientmay have difficulty in distinguishing betweeninternal mental events and external percepts.Perceptual misidentifications (illusions) arealso common, as in delirium tremens where,for example, an innocuous wallpaper patternmay be perceived by the patient as a host ofunpleasant scurrying animals. Less commonlythe abnormal percepts have absolutely nobasis in external reality (hallucinations). Theyoccur in between 40 and 75% of all cases ofdelirium60-62 64 and are more common inyounger patients. Again, they are most oftenvisual.60 They vary in complexity from simplegeometric shapes and colours to the highlydetailed, such as people, animals, or the phys-ical environment. They are most frequentlyassociated with hyperalert states of deliriumsuch as occur commonly with withdrawalstates associated with alcohol, hypnotics andanxiolytics, and intoxication states associatedfor example with anticholinergic drugs.Mood disturbance is frequent, in particular

irritability, agitation, aggression, anxiety,depression, apathy, perplexity or suspicious-ness.606' Euphoria may also occur, but is lesscommon.22 The mood and its behaviouralconsequences shift frequently and rapidlyfrom one state to another, giving a character-istic lability of mood. Anxiety or fear is oftenseen in hyperalert states and will be accompa-nied by overactivity of the autonomic nervoussystem. Mood disturbances can be markedand present an urgent management problem:undertreated agitation may result in injury tothe patient or others.The abnormalities in mentation present in

delirium result in directly observable abnor-

745 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Taylor, Lewis

malities of behaviour, including speech. Thepatient may be either underactive or overac-tive or may irregularly alternate between thetwo. The level of activity tends to correlatewith the level of conscious alertness exhibitedby the patient at the time. In general, hyper-alertness is associated with general motoroveractivity which is at best semi-purposeful,such as repeatedly getting out of bed in anattempt to leave the hospital. On other occa-sions the behaviour is essentially purposeless,for example, picking at the bedclothes andpurposeless tossing and turning in the bed.Speech may be increased in quantity, rateand volume; in some cases it may be franklyincoherent. There may be associated defectssuch as alexia65 and agraphia.66 In contrast,hypoactivity is associated with general under-activity, prolongation of the reaction time andspeech which is reduced in quantity, slow,monotonous and low in volume. In severecases the patient may be stuporous and mute.It is important to emphasise that these twodimensions of the clinical syndrome are notdistinct subtypes of delirium, as both mayoccur in irregular alternation during any oneillness. It is equally important to bear in mindthat the popular medical conception of delir-ium corresponds to hyperactivity; it followsthat in hypoactive patients a diagnosis ofdelirium may be missed.'9

Clinical information on both short andlong term memory can sometimes be inferredfrom the patient's spontaneous behaviour andspeech. For example, he may insist that he isseeing his wife for the first time that day eventhough she had been talking with him at hisbedside just a few minutes earlier (impairedshort-term memory). He may believe, despitebeing in hospital, that he is living at a locationhe left many years earlier and may talk ofgoing out to work at what was his occupationthere at that time (impairment of short-termmemory with inappropriate accessing of long-term memory). If severely ill, he may claimnever to have met his wife and childrenbefore (impairment of long-term memory).Confabulation is the term for the giving offalse past personal information in the contextof a short-term memory deficit, and is pre-sumed to be an attempt by the patient toobscure the memory deficit.The clinical course of an episode of delirium

is influenced by individual patient variablessuch as age and general physical condition,the cause of the disturbance and the speedand efficacy of management interventions.However, certain statements apply to delir-ium in general. Onset is typically sudden,developing over hours or days, and oftenbegins at night. In those cases with a rela-tively gradual onset, there may be a prodro-mal period in which the patient complains ofvarious non-specific symptoms such asmalaise, irritability, sleep disruption, night-mares and difficulty in concentrating. Onceestablished, the clinical picture tends to fluc-tuate in intensity. Lucid intervals occur mostcommonly during the day, and the usual pat-tern is nocturnal worsening, when levels of

awareness and attention are lowered, withabnormally raised or lowered levels of alert-ness. If the underlying disease is severe andprogressive, the patient may lapse into anincreasingly pronounced stupor, followed bycoma and death. This downward spiral issometimes marked by a period of agitatedoveractivity. The two most robust predictorsof a fatal outcome are, unsurprisingly,advanced age and the presence of multiplephysical illnesses.6 However, full recovery isthe most frequent outcome, with a return tothe patient's premorbid level of consciousnessand attention, and often occurs in less thanone week. Subsequent episodes may occur inindividual patients at high risk such as elderlypeople prone to recurrent chest or urinarytract infections. In a minority of cases, resolu-tion of the delirium leaves a chronic organicbrain syndrome such as dementia followinghead injury, or Korsakoffs psychosis follow-ing Wernicke's encephalopathy.67 Unipolardepressive disorder68 and post-traumaticstress disorder69 have been described follow-ing delirium. In all cases, following recoverythere is a dense or patchy retrograde amnesiafor the period of delirium which can be usefulfor clinching the diagnosis in hindsight.

InvestigationThe list of investigations shown in table 4aims at identifying the physical agent respon-sible for delirium in an individual patient.The choice of investigations ordered willobviously be dictated by the findings on his-tory taking and examination. If, after initialinvestigation, no cause can be found for thedelirium, it may then be appropriate to run afull test screen, given the serious nature of theuntreated condition. In so doing one maydetect uncommon presentations of relativelycommon diseases, such as systemic lupus ery-thematosus. Even after thorough investiga-tion, some patients remain without anidentified cause for their disorder. In suchcases, it may be necessary to reconsider thediagnosis.

Table 4 Physical investigations of delirium

First lineFull blood pictureESRElectrolytes, ureaLiver functionThyroid functionBlood glucoseUrine analysis and cultureEEGECGChest x-raySecond lineSerum folate, B12Syphilis serologyLumbar punctureUrinary illicit drug screenUrinary porphyrinsHIV antibodiesCardiac enzymesBlood gasesAutoantibody screenBlood culturesCalcium, phosphateCranial CT scan

746 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Delirium

Whilst the process of physical investigationis successful in most cases in identifying thecause of the disorder once delirium has beendiagnosed, it is of strictly limited value inconfirming the diagnosis itself. Only the EEGmay be helpful in this respect.70 In deliriumnot associated with psychoactive substancemisuse, the EEG usually shows evidence ofgeneralised slowing of background activity. Indelirium associated with psychoactive sub-stance misuse, for example alcohol or hyp-notic/anxiolytic withdrawal, the EEG usuallyshows an excess of low amplitude fast waveactivity. The limited utility of the EEG as adiagnostic aid in delirium is related to its lackof specificity: although claims have beenmade to the contrary.71 The EEG often can-not enable delirium to be differentiated fromdementia, as both Alzheimer's disease andmulti-infarct dementia (which togetheraccount for about 90% of all dementias) areassociated with a pattern of generalised slow-ing of the background activity similar to thatseen in delirium not associated with psy-choactive substance misuse.7273

Differential diagnosisIn uncomplicated dementia, consciousnessand attention remain essentially intactwhereas in delirium impairment of thesepsychological functions is a cardinal featureof the disorder. Often the first feature ofdementia is a short-term memory defect inthe absence of attentional deficit, reflectingthe early involvement of the medial temporallobe in the commonest form of dementia,Alzheimer's disease.74 Involvement of long-term memory for distant personal and imper-sonal events implies widespread corticalinvolvement by the disease process. Indementia the patient may exhibit short-termmemory impairment for many years untilsuch long-term stores are impaired. In delir-ium by contrast, short term memory defectdoes not occur in the absence of attentionaldeficit, and there is no clear-cut temporalprogression from short-term to long-termimpairment: the two may occur simultane-ously within hours of the disorder developingif the illness is severe.

Dementia is usually associated with aninsidious onset over many months. In delir-ium onset is usually sudden.20 Mode of onsetis an important differentiating featurebetween delirium and dementia, and thisunderlines the importance of obtaining a reli-able corroborative history whenever possible,as the patient's account of the onset of his ill-ness may well be incorrect. Determiningwhether the course of the illness is relativelyfluctuating or stable over hours or days (sug-gesting delirium and dementia respectively) isbest facilitated by careful observation in theclinical setting.The diagnostic situation is complicated by

the fact that delirium and dementia maycoexist in certain patients, particularly theelderly. Delirium may be diagnosed in aknown demented patient if the diagnostic cri-

teria are fulfilled. However, if a patient isdelirious, an additional diagnosis of dementiacannot be made either until the delirium hasresolved or until a diagnostic corroborativehistory is obtained, as from past medicalrecords. A known demented patient withreduced level of consciousness should notautomatically be assumed to have developeddelirium, as the disturbance of consciousnessmay be due to other factors, such as sedativemedication the patient is receiving. Thedifferential diagnosis in such cases can bedifficult, and requires expert psychiatricassessment.The rare dementia of cortical Lewy body

disease can resemble delirium. The finding ofa Parkinsonian syndrome in a patient appar-ently suffering from delirium should alwaysarouse suspicion of cortical Lewy bodydementia, as its course is much more fluctu-ating than that of other dementias.75The other organic mental syndromes do

not usually present serious difficulties in dif-ferentiation from delirium. In delirium, thereis global psychological dysfunction in thepresence of impaired consciousness andattention. By contrast, organic mental syn-dromes involving amnesia, hallucinations,delusions, depression, mania or anxietyexhibit relatively de-limited mental or behav-ioural abnormalities, with essentially intactconsciousness and attention.

Certain of these syndromes may, however,be associated with delirium in any individ-ual patient. For example, Wernicke'sencephalopathy (delirium) may be followedby Korsakoffs psychosis (amnestic disorder);alternatively, a depressive organic mood syn-drome associated with hypothyroidism maybe followed by delirium ("myxoedematousmadness").

Incoherent speech, gross incongruity ofaffect, delusions and hallucinations are clini-cal features associated with schizophrenia,but all may occur in delirium. In general, allsuch features are more fragmentary and fluc-tuating in delirium. To avoid misdiagnosis,however, detailed examination of mentationand behaviour is necessary.76 Careful atten-tion should be paid to consciousness, atten-tion, and memory, all of which are relativelyunimpaired in schizophrenia, but which maynot be amenable to examination in an acutelydisturbed patient. In such situations, theimportance of physical examination andinvestigation cannot be overstressed. In somecases, this is only possible after the patienthas been sedated. Care is necessary to avoidoversedation of a possibly delirious patient,with the risk of worsening of the condition.

In schizophrenia, delusions are often rela-tively stable over time and form part of anelaborate system of abnormal beliefs. In delir-ium, hallucinations are usually visual; inschizophrenia, auditory hallucinations aremuch more common. So-called first-ranksymptoms of schizophrenia77 are rare in delir-ium, even in severe cases.

Certain variants of schizophrenia are ofsudden onset, where the patient may appear

747 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Taylor, Lewis

"confused" and exhibit signs of altered con-sciousness78 or disorientation.79 These disor-ders are referred to as acute and transientpsychotic disorders (ICD-1i0),7 tend to beprecipitated by situational stress and have abetter prognosis than schizophrenia. They aremore common in developing countries butalso occur in the developed world. Such"functional" cases of "pseudo-delirium"20may rarely present particular diagnostic diffi-culties. The correct diagnosis may only bereached following a period of careful observa-tion accompanied by exclusion of a physicalcause by physical examination and investiga-tion. A minority of patients have physical dis-orders which may be causally related to theirschizophrenia.808'

In mood disorders such as hypomania ormania, the patient may be overactive, behav-iourally disturbed, and unamenable to psychi-atric examination. "Confusion"82 andtransient global cognitive dysfunction83 84 havebeen reported in mania. Visual hallucinationsmay be present in a substantial proportion ofmanic patients.85 The mood state in mania,particularly in the acute stages, may be pre-dominantly irritable as opposed to elevated orexpansive, further complicating the picture.

In unipolar depressive disorder or bipolarmood disorder in the depressive phase, thereis usually little difficulty in differentiating thedisorder from delirium. When apparent cog-nitive abnormalities are present, they are usu-ally more suggestive of dementia ("depressivepseudodementia").86 However, an acutely agi-tated depressed patient may present similardifficulties to the assessor as an acutely dis-turbed schizophrenic or manic patient. Again,detailed examination, investigation andobservation may be needed to help decide theissue.

Dissociative disorders are uncommon butmay be associated with impaired consciousawareness (fugue), or memory (amnesia).8Mixed pictures may occur. The characteristiccourse of these disorders is of sudden onset,short course and sudden termination. Theymay therefore be mistaken at first for deli-rium. In dissociative states, however, the dis-turbance of consciousness is often associatedwith a loss of sense of personal identity,which is rare in delirium. Likewise, amnesiatends to be relatively circumscribed to episo-dic (past personal) memory, with semantic(past impersonal) memory remaining rela-tively intact.87 Such a picture is not seen indelirium. In many cases of dissociative dis-order a psychosocial stressor may be identi-fied, which may have meaningful connectionswith an event in the patient's distant past.There are no abnormalities of the EEG indissociative disorder, and abreaction canresult in sudden remission,88 whereas in delir-ium it will usually cause worsening of theclinical condition.Thus the diagnostic process in delirium is

identical to that for any other medical condi-tion. Careful history taking, examination andinvestigation should result in the clinicianidentifying a syndrome.The 10th edition of the International

Classification of Diseases (ICD 10)7 (table 1)was adopted in 1993 and henceforth the diag-nosis of delirium should be made with refer-ence to the guidelines provided in the sectionon Mental and Behavioural Disorders.Delirium is divided into those cases "notinduced by alcohol or other psychoactive sub-stances" ("Organic including symptomatic,mental disorders, FOO-F09"), and thosewhich are ("Mental and behavioural disordersdue to psychoactive substance use, F10-F19"). The American Diagnostic andStatistical Manual of Mental Disorders, thirdedition, revised (DSMIIIR)8 (table 2) hasbeen an important influence on ICD 10, notleast in the designation of the disorder as"delirium". The reader will note that the defi-nition of delirium in ICD 10 is of a globaldisorder of mentation and behaviour andincludes information regarding aetiology,symptoms and signs, mode of onset, course,duration and physical investigations. Suchdiagnostic systems continue to undergorefinement, however, and it is likely, forexample, that DSMIV Delirium will containradical changes as a result of extensive empir-ical field testing.89

In delirium, the diagnostic process is intwo stages, namely identification of the clini-cal syndrome followed by identification of thecausative aetiological factor(s). The first stageis facilitated by history taking, mental stateexamination and careful observation of thebehaviour of the patient. The second stagemay be suggested by the history and physicalexamination, and is usually confirmed byphysical investigation. In determining theclinical syndrome, emphasis should be placedon taking a detailed history not just from thepatient (whose account may be unreliable),but also from those informants who have use-ful information to impart about the patient'srecent and past behaviour. A history of sud-den onset of impaired consciousness andattention with a fluctuating course is stronglysuggestive of delirium.90Once a diagnosis of delirium has been

made, rating scales can be used to help quan-tify the severity of the disorder. An initialrating made for example at the time of admis-sion to hospital acts as a useful baseline fromwhich to assess progression of the disorder.9'The most commonly used rating scale is theMini-Mental State Examination (MMSE),9'but its validity in delirium is compromised byits lack of specificity in differentiatingbetween delirium and dementia, and its poorsensitivity in detecting mild global organicconditions and focal brain disorders.9394In addition, increasing age and level ofeducation attainment have been shown toconfound its interpretation.95 An alternativerating scale is the NeurobehaviouralCognitive Status Examination (NCSE),which is claimed to have greater sensitivityand specificity than the MMSE.9394

ManagementThe management of delirium consists of twogeneral principles: general supportive and

748 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Delirium

symptomatic measures, and treatment of thespecific underlying condition or conditionsgiving rise to the delirium.90 It is important toemphasise that general supportive and symp-tomatic measures should be initiated as soonas the clinical syndrome has been diagnosed,without waiting for identification of thecausative organic factors. Delirium can becaused by almost any acute medical or surgi-cal condition and a complete account of itsmanagement would include an account of thespecific treatment of all such illnesses. Wewill emphasise general supportive and symp-tomatic measures in the management ofdelirium.

Delirium requires expert liaison betweendisciplines. Where the diagnosis is in doubt,the responsible medical or surgical teamshould have access to a psychiatrist withexpertise in diagnosis of the condition.5Referral should be made at an early stagerather than later. The psychiatrist in turnrequires access to all members of the wardstaff, including doctors, nurses and paramed-ical staff, as well as to close relatives andfriends, who may be able to contributeimportant information on the behaviour ofthe patient. If a clinical diagnosis of deliriumis made, the liaison psychiatrist should adviseon investigation and management, includinglegal questions of consent.

Whilst the level of conscious awareness isalways reduced in a delirious patient, the levelof alertness may be decreased or increased. Ifincreased, the patient may be overactive andacutely disturbed in his behaviour. This mayresult in accidental or non-accidental self-injury by the patient, worsening of thepatient's condition through exhaustion, orattacks on staff members. In such situations,pharmacological sedation may be necessary.Extreme care is necessary in sedating a deliri-ous patient in whom the cause of the deliriumis unknown, as the choice of sedative shouldideally be informed by the physical cause ofthe patient's condition, thus avoiding seda-tives whose side effects may complicate thatcondition. For example, benzodiazepinesshould be avoided in patients suffering fromrespiratory failure. Attention must also bepaid to the dose of the drug prescribed.Elderly patients are sensitive to the side-effects of drugs, particularly anticholinergicagents,25 and care is therefore necessary toprevent iatrogenic delirium or worsening of apre-existing delirium. In rare cases, such asfulminant hepatic encephalopathy, all formsof pharmacological sedation must be usedwith extreme care.

Only three classes of drugs can be cur-rently recommended in the treatment of theagitated patient with delirium: butryophe-nones, group three phenothiazines and ben-zodiazepines. Group one and twophenothiazines include chlorpromazine andthioridazine respectively. They should not beused in the treatment of delirium on accountof their side-effect profiles, especially anti-cholinergic effects. Anti-adrenergic effectsmay lead to cardio-vascular complications

such as postural hypotension, tachycardia andfatal cardiac arrhythmia. Antihistaminergiceffects lead to excessive sedation. Groupthree phenothiazines, such as trifluoperazine,and butyrophenones, such as haloperidol,lead to more severe extrapyramidal side-effects such as a Parkinsonian syndrome.However, they are less likely to cause theother side-effects associated with group oneand group two phenothiazines,96 and are thusto be preferred in the treatment of delirium.The drug of choice for most delirious

patients is held to be haloperidol,97 but shouldbe avoided in cases of benzodiazepine andalcohol withdrawal, anticholinergic toxicityand hepatic failure. It is highly efficacious inthe management of agitation and has rela-tively low toxicity.98 The dose regime mustalways be tailored to the individual circum-stances of the case in hand, and here onlygeneral guidelines will be given.

Oral haloperidol is detectable in the blood-stream after 60 to 90 minutes, and peakplasma concentrations are obtained after fourto six hours.98 It is therefore not appropriatefor the immediate control of severe agitation.In less severe cases however, it should be pre-scribed as 1-5 to 20 mg per day in divideddoses, gradually increasing to 60 mg per dayin severely agitated patients.99 Elderly ordebilitated patients should initially receivehalf the adult dose. Five to 10 mg orally twiceper day have been suggested for mild to mod-erate agitation in young or middle aged deliri-ous patients.22

Parenteral administration of haloperidol isassociated with almost 100% bioavailabilityas opposed to approximately 66% by the oralroute. Thus parenterally administeredhaloperidol should be given in a lower dosethan the oral form to achieve equivalentplasma concentrations.'00 It has the additionalbenefit of causing less extrapyramidal sideeffects than oral administration.'0' This is ofmore than academic importance, as severeextrapyramidal side-effects require treatmentwith anti-cholinergic agents such as procycli-dine, which may further exacerbate the delir-ium. Intramuscular haloperidol reaches peakserum levels in 20 to 40 minutes98 and thusmay be used in the treatment of severe agita-tion. It is prescribed in a dose of 2 to 10 mg(increasing to 30 mg for emergency control)then up to 5 mg every hour if necessary(intervals of four to eight hours may be satis-factory).99 Intravenous haloperidol is notlicensed for use in the United Kingdom orUnited States. In a case of extreme emer-gency, where even delay in intramuscularabsorption is unacceptable, intravenousdroperidol may be prescribed as 5 to 15 mg,repeated every four to six hours if necessary.99The elderly should receive half the initialadult dose.

Benzodiazepines are the drug treatment ofchoice in delirium associated with benzodi-azepine withdrawal,'02 alcohol withdrawal'03and hepatic failure.'04 Delirium tremensbegins about 48 hours after acute withdrawalfrom severe alcohol abuse. It is associated

749 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Taylor, Lewis

with fever, sweating, prominent visual hallu-cinations, tachycardia and a marked tremor.Urgent management is necessary.Benzodiazepines should be prescribed, up to30 mg of diazepam per day in divided doses,gradually reducing to cessation over a sevento ten day period. Convulsions may occur,requiring anticonvulsant therapy. Thiaminereplacement is necessary and should be givenparenterally. Acute dehydration is commonand intravenous fluids are usually neededonce the patient is sedated.

Careful attention must be paid by medicaland nursing staff to the patient's state ofnutrition, fluid and electrolyte balance, andbowel and bladder function, as neglect ofthese will worsen the patient's condition. Inpatients confined to bed for long periods,expert nursing care and physiotherapy is nec-essary to prevent the development of pressuresores and contractures.The impairment in consciousness and

attention in delirium renders the patientliable to misinterpret incoming sensory stim-uli. This is particularly liable to occur if thestimuli are new. It is not uncommon for apatient who is only mildly delirious at hometo deteriorate once admitted to hospital. Thismay be related to exposure to a new, foreignenvironment. A careful balance is needed toavoid overstimulation or understimulation,both of which can have deleterious effects onthe patient's mental state. The patient is bestplaced in a side-room which should be quietand well lit by day and night. The patient'sorientation may be improved by surroundinghim with familiar objects from home.Likewise, close relatives should be encour-aged to visit and should be allowed to remainoutside normal visiting times when indicated.Expert nursing care is necessary for the opti-mal management of the delirious patient.This includes accurate observation andreporting.

Delirium is underdiagnosed, underinvesti-gated and undertreated: correct detection andtreatment will improve outcome and reducecomplications.

1 The medical works of Hippocrates. Trans, Chadwick J,Mann WN. Oxford: Blackwell, 1950.

2 Celsus. De Medicina. Trans, Spencer WG. London:Heinemann, 1938.

3 Jackson SW. Galen-on mental disorders. J Hist BehavSci 1969;5:365-84.

4 Lipowski ZJ. Delirium: acute brain failure in man.Springfield, III: Charles C Thomas, 1980.

5 Perez EL, Silverman M. Delirium: the often overlookeddiagnosis. IntJ PsychiatryM 1984;14: 181-9.

6 Trzepacz PT, Teague GB, Lipowski ZJ. Delirium andother organic mental disorders in a general hospital.Gen Hosp Psychiatry 1985;7:101-6.

7 World Health Organisation. The ICD-1OClassification ofMental and Behavioural Disorders. Geneva: WorldHealth Organisation, 1992.

8 American Psychiatric Association. DSM-III R: Diagnosticand Statistical Manual of Mental Disorders, 3rd ed, rev.Washington DC: American Psychiatric Association,1987.

9 World Health Organisation. Mental disorders: Glossaryand guideto their classification in accordance with theNinth revision of the International Classification ofDiseases. Geneva: World Health Organisation, 1978.

10 Oakley DA. Animal awareness, consciousness and self-image. In: Oakley DA, ed. Brain and mind. London:Methuin, 1985:132-51.

11 Baddeley AD. Memory Theory and Memory Therapy.In: Moffat M, ed. Clinical management of memory prob-lems. London: Chapman Hall, 1992:4-31.

12 Levkoff S, Cleary P, Liptzin B, et al. Epidemiology ofdelirium: an overview of research issues and findings.Int Psychogeriat 1991;3: 149-69.

13 Folstein MF, Bassett SS, Romanoski AJ, et al. The epi-demiology of delirium in the community: the EasternBaltimore Mental Health Survey. Int Psychogeriat1991;3:169-76.

14 Horvath TB, Siever LJ, Mohs RC, et al. Organic MentalSyndromes and Disorders. In: Kaplan HI, et al, eds.Comprehensive textbook of psychiatry. Baltimore:Williams and Wilkins, 1989:599-641.

15 Gillick MR, Serrell NA, Gillick LS. Adverse conse-quences of hospitalisation of the elderly. Soc Sci Med1982;16: 1033-8.

16 Warshaw GA, Moore JT, Friedman SW, et al.Functional disability in the hospitalised elderly. JAMA1982;248:847-50.

17 Johnson J, Sullivan E, Gottlieb E, et al. Delirium inelderly patients on internal medical services. J AmGeriatr Soc 1987;35:972.

18 Williams MA, Campbell EB, Raynor WJ, et al.Predictors of acute confusional states in hospitalisedelderly patients. Res Nurs Health 1985;8:31-40.

19 Erkinjuntti T, Wikstrom J, Palo J. Dementia amongmedical inpatients. Arch InternMed 1986;146: 1923-6.

20 Lipowski ZJ. Transient cognitive disorders (delirium,acute confusional states) in the elderly. Am JPsychiatry 1983;140: 1426-1526.

21 Hodkinson HM. Mental impairment in the elderly. J RColl Physicians Lond 1973;7:305-17.

22 Lipowski ZJ. Delirium: acute confusional states. OxfordUniversity Press, 1990.

23 Lipowski ZJ. Acute confusional states (delirium) in theelderly. In: Albert ML, ed. Clinical Neurology ofAging.Oxford: Oxford University Press, 1984:277-97.

24 Sirosis F. Delirium: 100 cases. Can J Psychiatry 1988;33:375-8.

25 Blazer DG, Federspiel CF, Ray WA, et al. The risk ofanticholinergic toxicity in the elderly: a study of pre-scribing practices in two populations. J Gerontol1983;38:31-5.

26 Fordyce G. Five dissertations on fever. Boston: Bedlingtonand Ewes, 1823.

27 Ingvar DH. "Memory of the future": An essay on thetemporal organisation of conscious awareness. HumanNeurobiol 1985;4: 127-36.

28 Schmidlry NW, Messing RO. Agitated confusionalstates in patients with right hemisphere infarctions.Stroke 1984;15:883-5.

29 Jamieson D, Alavi A, Jolles P, et al. Positron emissiontomography in the investigation of central nervous sys-tem disorders. Radiol Clin North Am 1988;26:1075-88.

30 Assal G, Zander E, Hadjiantonion J. Les troubles men-taux au cours des tumeurs de la fossa posterieure. ArchSwisses Neurol Neurochir Psychiatr 1975; 116: 17-27.

31 Mori E, Yamadori A. Acute confusional state and acuteagitated delirium. Arch Neurol 1987;44: 1139-43.

32 Devinsky 0, Bear D, Volpe BT. Confusional states fol-lowing posterior cerebral artery infarction. Arch Neurol1988;45: 160-3.

33 Sokoloff L. Neurophysiology and neurochemistry ofcoma. Exp Biol Med 1971;4:15-33.

34 Hawkins R. Cerebral energy metabolism. In:McCandless DW, ed. Cerebral energy metabolism andmetabolic encephalopathy. New York: Plenum Press,1985:3-23.

35 Siesjo BK, Carlsson C, Hagerdal M, et al.Brain metabo-lism in the critically ill. Crit Care Med 1976;4:283-94.

36 Berglund M, Nielsen S, Risberg J. Regional cerebralblood flow in a case of bromide psychosis. ArchPsychiatr Nervenkr 1977;223: 197-201.

37 Hemmingsen R, Vorstrup S, Clemmesen L, et al.Cerebral blood flow during delirium tremens andrelated clinical states studied with xenon-133 inhala-tion tomography. Am J Psychiatry 1988;145:1384-90.

38 Markand ON. Electroencephalography in diffuseencephalopathies. J Clin Neurophysiol 1984;1:357-407.

39 Engel GL, Romano J. Delirium: a syndrome of cerebralinsufficiency. J Chronic Dis 1959;9:260-77.

40 Romano J, Engel GL. Physiologic and psychologic con-siderations of delirium. Med Clin North Am 1944;28:629-38.

41 Adams RD, Victor M. Principles of neurology. New York:McGraw-Hill, 1981:281.

42 Ey H, Lairy GC, de Barros-Ferreira M, et al.Psychophysiologie du Sommeil et Psychiatrie. Paris:Masson, 1975.

43 Smiler BG, Bartholomew EG, Sivak BJ, et al.Physostigmine reversal of scopolamine delirium inobstetric patients. Am J Obstet Gynecol 1973;116:326-49.

44 Ital T, Fink M. Anticholinergic drug-induced delirium:Experimental modification, quantitative EEG andbehavioural correlations. J Nerv Ment Dis 1966;143:492-507.

45 Gibson GE, Peterson C, Sansone J. Decreases in aminoacid and acetylcholine metabolism during hypoxia. 7Neurochem 1981;37:192-201.

46 Honer WG, Prohovnik I, Smith G, et al. Scopolaminereduces frontal cortex perfusion. JfCereb Blood FlowMetab 1988;8:635-41.

47 Robbins TW, Everitt BJ.Psychopharmacological studiesof arousal and attention. In: Stahl SM, et al, eds.Cognitive neurochemistry. Oxford: Oxford University

750 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from

Delirium

Press, 1987:135-70.48 Hobson JA, Lydic R, Baghoddoyan HA. Evolving con-

cepts of sleep cycle generation: From brain centers toneuronal populations. Behav Brain Sci 1986;9:371-448.

49 Lishman WA. Organic psychiatry. Oxford: BlackwellScientific, 1987.

50 Rosen AM, Fox HA. Tests of cognition and their rela-tionship to psychiatric diagnosis and demographicvariables. J Clin Psychiatry 1986;47:495-8.

51 Hinton J, Withers E. The usefulness of the clinical testsof the sensorium. BrJ Psychiatry 197 1;1 19:9-18.

52 Squire LR. Memory and brain. Oxford: OxfordUniversity Press, 1987.

53 Lezak MD. Neuropsychological assessment. Oxford:Oxford University Press, 1983.

54 Levin M. Thinking disturbances in delirium. A M AArch Neurol Psychiatry 1956;75:62-6.

55 Litin EM. Mental reaction to trauma and hospitalisationin the aged. NEnglJMed 1956;162:1522-4.

56 Levin M. Toxic delirium precipitated by admission tohospital. J Nerv Ment Dis 1952;1 16:210-4.

57 Levin M. Varieties of disorientation. Jf Ment Sci 1956;102:619-23.

58 Loftus EL, Schooler JW. Information processing con-ceptualizations of human cognition: past, present andfuture. In: Ruben BD, ed. Information and Behaviour.New Brunswick, NJ: Transaction Books, 1985:225-50.

59 Mettler CC. History of medicine: a correlative text,arranged according to subjects. Philadelphia: Blakiston,1947.

60 Wolff HG, Curran D. Nature of delirium and alliedstates. A M A Arch Neurol Psychiatry 1935;33:1175-215.

61 Farber U. Acute brain syndrome. Dis Nerv System1959;20:296-9.

62 Simon A, Cahan RB. The acute brain syndrome in geri-atric patients. Psychiatr Res Rep 1963;16:8-2 1.

63 Morris GO, Singer MT. Sleep deprivation: the contentof consciousness. J Nerv Ment Dis 1966;143:291-304.

64 Bleuler M, Willi J. Buhler HR. Akute Psychische Begleiterscheinungen Korperlicher Krankheiten. Stuttgart:Thieme, 1966.

65 Kitselman K. Language impairment in aphasia, delir-ium, dementia, and schizophrenia. In: Darby JK, ed.Speech evaluation in medicine. New York: Grune andStratton, 1981:199-214.

66 Chedru F, Geschwind N. Writing disturbances in acuteconfusional states. Neuropsychologia 1972;10:343-53.

67 Victor M, Adams RD, Collins GH. The Wernicke-Korsakoffsyndrome. Philadelphia: F A Davis, 1971.

68 Blank K, Perry S. Relationship of psychologicalprocesses during delirium to outcome. Am J7 Psychiatry1984;141:843-7.

69 MacKenzie TB, Popkin MK. Stress response syndromeoccurring after delirium. Am Jf Psychiatry 1980;137:1433-5.

70 Brenner RP. The electroencephalogram in altered statesof consciousness. Neurol Clin 1985;2:615-31.

71 Rabins PV, Folstein MF. Delirium and dementia: diag-nostic criteria and fatality rates. Br J Psychiatry1982;140: 149-53.

72 Rae-Grant A, Blume W, Lau C, et al. The electroen-cephalogram in Alzheimer-type dementia. Arch Neurol1987;44:50-4.

73 Leuchter FA, Spar JE, Walter DO, et al.Electroencephalographic spectra and coherence in thediagnosis of Alzheimer-type and multi-infarct demen-tia. Arch Gen Psych 1987;44:993-8.

74 Tomlinson BE. The structural and quantitative aspectsof the dementias. In: Roberts PJ, ed. Biochemistry ofdementia. Chichester: Wiley, 1980.

75 McKeith I, Fairbaim A, Perry R, et al. Neuroleptic sen-sitivity in patients with senile dementia of Lewy Bodytype. Brj Psychiatry 1992;305:673-7.

76 Dubin WR, Weiss KJ, Zeccardi JA. Organic brain syn-drome. The psychiatric imposter. JAMA 1983;249:60-2.

77 Schneider K. Clinical psychopathology. New York: Gruneand Stratton, 1959.

78 Van Praag GE. About the impossible concept of schizo-phrenia. ComprPsychiatry 1976;17:481-97.

79 Newmark CS, Raft D, Toomet T, et al. Diagnosis ofschizophrenia: Pathognomonic signs or symptom clus-ters. Compr Psychiatry 1975;16: 155-63.

80 Davison K, Bagley CR. Schizophrenia-like psychosesassociated with organic disorders of the central ner-vous system: a review of the literature. In: HerringtonRN, ed. British Journal of Psychiatry special publication,no. 4. Ashford, Kent: Headley Bros, 1969:113-84.

81 Johnstone EC, Macmillan JF, Crow TJ. The occurrenceof organic disease of possible or probable aetiologicalsignifican§ce in a population of 268 cases of firstepisode schizophrenia. PsycholMed 1987;17:371-9.

82 Carlson GA, Goodwin FK. The stages of mania: A lon-gitudinal analysis of the manic episode. Arch GenPsychiatry 1973;28:221-8.

83 Van Sweden B. Disturbed vigilance in mania. BiolPsychiatry 1986;21:311-13.

84 Mentzos S, Lyrakos A, Tsiolis A. AkuteVirwirrtheitszustande. Nervenarzt 1971;42: 10-17.

85 Taylor MA, Abrams R. The phenomenology of mania.Arch Gen Psychiatry 1973;29:520-2.

86 La Rue A, Spar J, Hill CD. Cognitive impairment inlate-life depression: Clinical correlates and treatmentimplications.3 AffectDis 1986;11:179-84.

87 Kopelman MD. Amnesia: Organic and psychogenic. BrJYPsychiatry 1987;150:428-42.

88 Perry JC, Jacobs D. Clinical applications of the amytalinterview in psychiatric emergency settings. Am YPsychiatry 1982;139:552-9.

89 Moskowitz J, Davidson M, Harvey PD. Effect of con-current distraction on communication failures inschizophrenic patients. II. Medication status correla-tions. SchizophrRes 1991;5:153-9.

90 Lipowski ZJ. Delirium (acute confusional states). JAA1A1987;258: 1789-92.

91 Fields SD, MacKensie CR, Charlson ME, et al.Reversibility of cognitive impairment in medical inpa-tients. Arch Intern Med 1986;146:1593-6.

92 Folstein MF, Folstein SE, McHugh PR. "Mini-Mentalstate". Y Psychiatr Res 1975;12:189-98.

93 Nelson A, Fogel BS, Faust D. Bedside cognitive screen-ing instruments: A critical assessment. YNervMent Dis1986;174:73-83.

94 Schwamm LH, Van Dyke C, Kiernan RJ, et al. Theneurobehavioural cognitive status examination: Com-parison with the cognitive capacity screening examina-tion and the mini-mental state examination in aneurosurgical population. Ann Int Med 1987;107:486-91.

95 Anthony JC, Le Resche L, Niaz U, et al. Limits of the"Mini-Mental State" as a screening test for dementiaand delirium among hospital patients. Psychol Med1982;12:397-408.

96 Salzman C. Treatment of the elderly agitated patient.J Clin Psychiatry 1987;48(5, Suppl): 19-22.

97 Moore DP. Rapid treatment in critically ill patients. AmJ Psychiatry 1977;134:1431-2.

98 Settle EC, Ayd FJ. Haloperidol: a quarter century ofexperience. YClin Psychiatry 1983;44:440-8.

99 British National Formulary: London: British MedicalAssociation and the Royal Pharmaceutical Society,1993.

100 Forsman A, Ohman R. Applied pharmacokinetics ofhaloperidol in man. Curr Ther Res 1977;21:396-41 1.

101 Menza MA, Murray GB, Holmes VF, et al. Decreasedextrapyramidal symptoms with intravenous haloperi-dol. J Clin Psychiatry 1987;48:278-80.

102 Foy A, Drinkwater V, March S, et al. Confusion afteradmission to hospital in elderly patients using benzo-diazepines. BMJ 1986;293:1072.

103 Dubin WR, Weiss KJ, Dorn JM. Pharmacotherapy ofpsychiatric emergencies. Y Clin Psychopharnacol 1986;6:210-22.

104 Misra P. Hepatic encephalopathy. Med Clin North Am1981;65:209-26.

751 on N

ovember 17, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.7.742 on 1 July 1993. D

ownloaded from