ACUTE CONFUSIONAL STATE (DELIRIUM)

UMEÅ UNIVERSITY MEDICAL DISSERTATIONSNew series No 309 - ISSN 0346-6612'

From the Departm ents of Geriatric Medicine and Internal Medicine, University of Umeå, and the Departm ent of Geriatric Medicine, Karolinska Institu te, Stockholm, Sweden.

ACUTE CONFUSIONAL STATE (DELIRIUM)Clinical stud ies in hip-fracture and stroke patien ts

iy

Yngve Gustafson

Umeå 1991

UMEÅ UNIVERSITY MEDICAL DISSERTATIONSNew series No 309 - ISSN 0346-6612

From the Departments of Geriatric Medicine and Internal Medicine, University of Umeå, and the Department of Geriatric Medicine, Karolinska Institute, Stockholm, Sweden.

ACUTE CONFUSIONAL STATE (DELIRIUM)Clinical studies in hip-fracture and stroke patients

AKADEMISK AVHANDLING som med vederbörligt tillstånd av Rektorsämbetet vid

Umeå universitet för avläggande av doktorsexamen i medicinsk vetenskap kommer att offentligen försvaras i

Tandläkarhögskolan 9 trp sal B, i Umeå lördagen den 8 juni 1991 kl 10.00.

Yngve Gustafson

Umeå 1991

ISBN 91-7174-594-7

ABSTRACT

ACUTE CONFUSIONAL STATE (DELIRIUM).Clinical studies in hip-fracture and stroke patients.

Yngve Gustafson, Departments of Geriatric Medicine and Internal Medicine, Umeå University, Umeå, Sweden. Department of Geriatric Medicine, Karolinska Institute, Stockholm, Sweden.

Acute confusional state (ACS) or delirium according to DSM-III-R holds a central position in the medicine of old age. ACS is a common and sometimes the only symptom of diseases and medical complications in the elderly patient.The aim of this study was to elucidate ACS in patients with femoral neck fractures and patien ts with acute stroke with regard to frequency, predictors, possible pathogenetic mechanisms, associated complications, assessm ent and documentary routines and the clinical outcome for the patients. An intervention program to prevent postoperative ACS based on our results was developed and evaluated.The m ain findings of the study were high frequencies of ACS in elderly patients with femoral neck fractures (61 %) and in patients with acute stroke (48 %). The main risk factors for ACS in patients with femoral neck fractures were old age, diseases and drug trea tm ent interfering with cerebral cholinergic metabolism. There was no link between anaesthetic technique and ACS bu t the connection between peroperative hypotension, early postoperative hypoxia and ACS was close.In stroke patients the degree of extremity paresis and old age were independent ACS risk factors. ACS was commonly associated with post stroke complications such as myocardial infarction, pneumonia, urinary infection and urinary retention. In stroke patients there was a close connection between high hypothalam ic-pituitary-adrenal axis (HPA-axis) activity and ACS. High HPA-axis activity and disturbances in the cerebral cholinergic system may be two im portant ACS mechanisms.A correct diagnosis is a prerequisite for proper treatm ent of ACS and its underlying causes. In the orthopaedic wards both physicians and nurses diagnosed and documented ACS poorly and therefore associated complications were insufficiently treated.The intervention program for postoperative ACS, aimed mainly at protecting the cerebral oxidative metabolism and thereby the cerebral cholinergic metabolism which is especially sensitive to hypoxia. Postoperative complications associated with ACS were also treated. The intervention resulted in reduced frequency, duration and severity of postoperative ACS and in shorter orthopedic ward stay for patients with femoral neck fractures.

Key words: Acute confusional state, delirium, elderly, stroke, femoral neck fractures, acetylcholine, cortisol.

UMEÅ UNIVERSITY MEDICAL DISSERTATIONSNew series No 309 - ISSN 0346-6612

From the Departments of Geriatric Medicine and Internal Medicine, University of Umeå, and the Department of Geriatric Medicine, Karolinska Institute, Stockholm, Sweden.

ACUTE CONFUSIONAL STATE (DELIRIUM)Clinical studies in hip-fracture and stroke patients

Yngve Gustafson

A l v

Umeå 1991

Copyright © 1991 by Yngve Gustafson

ISBN 91-7174-594-7

Printed in Sweden by

LARSSON & CO:S TRYCKERI AB

Umeå 1991

‘To my ÿamity

3

CONTENTSAbbreviations 6

Abstract 7

Original papers 8

Introduction

History 9

Terminology 9

Diagnostic criteria 11

Clinical Features 13

Differential diagnosis 14

ACS frequency 16

Postoperative ACS 17

ACS in stroke patients 18

Predisposing ACS factors 18

Anesthetic technique and hip-surgery 20

ACS and clinical outcome 21

Etiology and pathogenesis 22

Hip-fracture and stroke 26

Aims of the study 2 7

Patients 28

Methods 29

Results

ACS frequency 32


Anesthetic technique and ACS 37


ACS diagnosis and documentation 39

Intervention 41

4

Discussion

ACS frequency 43


Anesthesia and ACS 45


ACS diagnosis and documentation 47

Intervention 49

Etiology and pathogenesis 51

Prevention and treatm ent 58

General summary and conclusions 5 9

Acknowledgements 61

References 63

Original papers

I 77

II 85

III 95

IV 117

V 135

VI 151

5

ABBREVIATIO NSACS = Acute confusional state

ADL = Activities of daily living

C.I. = Confidence interval

CNS = Central nervous system

CT = Computed tomography

DEX = Dexamethasone

DSM-III = Diagnostic and statistical m anual of mental disorders (third edition)

DSM-III-R = Diagnostic and statistical m anual of mental disorders (third edition-revised)

DST = Dexamethasone Suppression Test

GA = General anesthesia

HPA axis = Hypothalamic-pituitary-adrenal axis

IL = Interleukin

MMSE = Mini mental state examination

PAI = Plasminogen activator inhibitor

RA = Regional anesthesia

RAS = Reticular activating system

TLA = Transitory ischemic attack

6

ABSTRACTACUTE CONFUSIONAL STATE (DELIRIUM).Clinical studies in hip-fracture and stroke patients.

Yngve Gustafson, Departments of Geriatric Medicine and Internal Medicine, Umeå University, Umeå, Sweden. Department of Geriatric Medicine, Karolinska Institute, Stockholm, Sweden.

Acute confusional state (ACS) or delirium according to DSM-III-R holds a central position in the medicine of old age. ACS is a common and sometimes the only symptom of diseases and medical complications in the elderly patient.The aim of this study was to elucidate ACS in patients with femoral neck fractures and patien ts with acute stroke with regard to frequency, predictors, possible pathogenetic mechanisms, associated complications, assessm ent and documentary routines and the clinical outcome for the patients. An intervention program to prevent postoperative ACS based on our results was developed and evaluated.The main findings of the study were high frequencies of ACS in elderly patients with femoral neck fractures (61%) and in patients with acute stroke (48%). The main risk factors for ACS in patients with femoral neck fractures were old age, diseases and drug trea tment interfering with cerebral cholinergic metabolism. There was no link between anaesthetic technique and ACS b u t the connection between peroperative hypotension, early postoperative hypoxia and ACS was close.In stroke patients the degree of extremity paresis and old age were independent ACS risk factors. ACS was commonly associated with post stroke complications such as myocardial infarction, pneumonia, urinary infection and urinary retention. In stroke patients there was a close connection between high hypothalam ic-pituitary-adrenal axis (HPA-axis) activity and ACS. High HPA-axis activity and disturbances in the cerebral cholinergic system may be two im portant ACS mechanisms.A correct diagnosis is a prerequisite for proper treatm ent of ACS and its underlying causes. In the orthopaedic wards both physicians and nurses diagnosed and documented ACS poorly and therefore associated complications were insufficiently treated.The intervention program for postoperative ACS, aimed mainly at protecting the cerebral oxidative m etabolism and thereby the cerebral cholinergic m etabolism which is especially sensitive to hypoxia. Postoperative complications associated with ACS were also treated. The intervention resulted in reduced frequency, duration and severity of postoperative ACS and in shorter orthopedic ward stay for patients with femoral neck fractures.

Key words: Acute confusional state, delirium, elderly, stroke, femoral neck fractures, acetylcholine, cortisol.

7

ORIGINAL PAPERS

I. G ustafson Y, Berggren D, Brännström B, Bucht G, Norberg A, Hansson L-I, Winblad B. Acute confusional states in elderly patients treated for femoral neck fracture. J Am Geriatr Soc 36:525-530, 1988.

II. Berggren D, Gustafson Y, Eriksson B, Bucht G, Hansson L-I, Reiz S, Winblad B. Postoperative confusion in elderly patients with femoral neck fractures. Anesth Analg, 394:497-504, 1987.

III. Gustafson Y, Olsson T, Eriksson S, Asplund K, Bucht G. Acute confusional states in stroke patients. Cerebrovascular Dis, accepted for publication.

IV. Gustafson Y, Olsson T, Asplund K, Hägg E. Acute confusional states (Delirium) early after stroke are associated with hypercortisolism. Submitted.

V. G ustafson Y, Brännström B, Norberg A, Bucht G, Winblad B. Underdiagnoses and poor documentation of acute confusional states in elderly hip-fracture patients. Submitted.

VI. Gustafson Y, Brännström B, Berggren D, Ragnarsson J-I, Sigaard J , Bucht G, Reiz S, Norberg A, Winblad B. A geriatric-anesthesiologic program aimed a t reducing acute confusional s ta tes in elderly pa tien ts treated for femoral neck fractures. J Am G eriatr Soc, accepted for publication.

8

INTRODUCTION

HISTORYAcute confusional state (ACS) or delirium was one of the first mental d isorders described in the lite ra tu re . About 2500 years ago, Hippocrates described ACS as a mental disorder of physiologic origin in a different terminology though. Several of his observations are still relevant, for example "When a delirium or raving is appeased by sleep, it is a good sign" and "Difficulty of breathing and delirium in continual fevers are mortal". Greek and Roman writers used the term phrenitis which refers to both diaphragm and mind. The soul or seat of life was considered to be in the diaphragm and the disruption of the union between mind and senses was suggested as the cause of phrenitis. This view on ACS as a disruption of the integrative system of m ental functioning is in line with modern perspectives on the pathophysiology of ACS. For a review of the history of ACS see the monograph by Lipowski (Lipowski 1990).The literature from the 19th century contains excellent clinical descriptions of ACS. One of them made by Savage in 1887 proposes a m ultifactorial approach to ACS because "there are several predisposing causes which may have been in operation for a long time, as well as one or more exciting causes which may have been in action for m uch shorter periods" (Savage 1887).

TERMINOLOGYACS is one of the m ost common and im portan t form s of psychopathology in the elderly (Lipowski 1989). ACS is perhaps the most frequent presenting symptom of disease in the medicine of old age (Hodkinson 1976, Lipowski 1989). The study of ACS has been plagued by terminological confusion obstructing research, nursing and medical care, as well as communication and education in the field. In Table 1 some of the synonyms of ACS used in the literature are presented. The lack of uniform terminology would have been a minor problem if consistent diagnostic criteria had been used. The use of DSM-III and DSM-III-R criteria for delirium has made it possible to compare research resu lts (APA 1980, APA 1987). For reasons discussed under diagnostic criteria we have chosen to use the term Acute Confusional State (ACS) in this study.

9

TABLE 1. »SYNONYMS1 FOR ACUTE CONFUSIONAL STATE (ACS) USED IN THE ENGLISH LITERATURE.

Acute brain dysfunctionAcute brain failureAcute brain syndromeAcute cerebral insufficiencyAcute cognitive disorderAcute confusionAcute confusional insanityAcute organic brain syndromeAcute organic psychosisAcute organic reactionAcute organic syndromeAcute psychoorganic syndromeAgitated confusionAgitated deliriumBrain dysfunctionCerebral insufficiency syndromeClouded stateConfusionConfusional stateDelayed psychosisDeliria of feverDelirious stateDeliriumDelirium nervosum Dysergastic reaction Emergency delirium Encephalopathy Exogenous psychosis Infective exhaustive psychosis Intensive care syndrome Mental confusion

Metabolic encephalopathy Organic confusion Pharmacotoxic psychosis Phrenitis PhrensyPostanesthetic delirium Postoperative confusion Postoperative delirium Postoperative insanity Postoperative psychosis Pseudodementia PseudosenilityReversible cognitive dysfunction Reversible dementia Reversible m adness Reversible toxic psychosis Senile delirium States of excitement Subacute befuddlement Symptomatic psychoses Toxic confusion Toxic confusional state Toxic delirious reaction Toxic delirium Toxic encephalopathy Toxic-infectious psychoses Toxic psychosisTransient behavioural syndrome Transient cognitive disorder Vascular psychotic organic brain syndrome

10

DIAGNOSTIC CRITERIADSM-III criteria for delirium (APA 1980) are:A. Clouding of consciousness (reduced clarity of awareness of the environment), with reduced capacity to shift, focus and susta in attention as to environmental stimuli.B. At least two of the following:

1. perceptual disturbance, misinterpretations, illusions, or hallucinations2. speech at times incoherent3. disturbance in the sleep-wakefulness cycle, with insomnia or day-time drowsiness4. increased or decreased psychomotor activity

C. Disorientation and impaired memory (if possible to test).D. Clinical features developing over a short period of time (usually hours or days) and tending to fluctuate during the course of a day.E. Evidence, from the history, physical examination or laboratory tests of a specific organic factor judged to be etiologically related to the disturbance.

In 1987 the DSM-III-R was published as a revision of the DSM-III. DSM-III-R criteria for delirium (APA 1987) are:A. Reduced ability to m aintain attention to external stimuli (e.g., questions m ust be repeated because a ttention wanders) and to appropria te ly sh ift a tten tio n to new external stim uli (e.g., perseverates answer to a previous question).B. Disorganized thinking, as indicated by rambling, irrelevant, or incoherent speech.C. At least two of the following:

1. reduced level of consciousness, e.g., difficulty in keeping awake during examination2. perceptual disturbances: misinterpretations, illusions, or hallucinations3. disturbance in sleep-wake cycle with insomnia or daytime sleepiness4. increased or decreased psychomotor activity5. disorientation to time, place, or person6. impaired memory, e.g., inability to learn new material, such as the names of several unrelated objects after five

11

minutes, or to remember past events, such as the history of the current episode of illness

D. Clinical features developing over a short period of time (usually hours or days) and tending to fluctuate during 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) tha t can be etiologically related to the disturbance2. in the absence of such evidence, an etiologic organic factor can be presumed if the disturbance cannot be accounted for by a nonorganic mental disorder, e.g., manic episode accounting for agitation and sleep disturbance.

The DSM-III and DSM-III-R criteria for delirium are essentially the sam e. The difference is an altered hierarchy which m akes no difference in substance. When we have applied DSM-III-R criteria on patients previously assessed as acutely confused according to DSM- III, no diagnosis has been changed.U nfortunately DSM-III and DSM-III-R criteria (E) do not accept psycho-social factors as etiological for delirium. However, in most cases of delirium, there are several contributing risk factors and triggering factors for delirium and therefore it is difficult to d istinguish between organic and psychological m echanism s for delirium. A m ultifactorial approach to delirium is necessary in assessm ent as well as in treatm ent. Jolley sum m arizes the basic mechanisms producing the ACS as one: "in the elderly it is usual for m any factors to contribute a little, ra th e r th an one factor to contribute the whole" (Jolley 1981). In most cases there are probably psychosocial factors acting as risk factors or contributing triggering factors. This is particularly true of patients with dementia. Regarding ACS as a threshold phenomenon, it is obvious th a t the demented patient who has reduced cerebral spare capacity can develop ACS from minor strain. We have no instrum ent to differentiate between the biochemical disturbance in the brain caused by psycho-social factors and those caused by organic/metabolic factors. As the etiology of delirium in most cases is multifactorial we have chosen to use the term Acute Confusional State (ACS) in this study. However, the DSM- III and DSM-III-R criteria for delirium are fulfilled in all aspects in

12

patients classified as being in an acute confusional state (ACS) in all the papers.Another problem with the DSM-III and DSM-III-R criteria is tha t the diagnosis is sometimes technically difficult in a single given situation due to two criteria, namely rapid onset and fluctuating symptoms. The fulfilm ent of these criteria dem ands an observation of a fluctuating course or tha t relatives or caregivers can report a rapid onset and a fluctuating course. In clinical practice, when in doubt, it is necessary to make a provisional ACS diagnosis and immediately assess the patient for potential causes.

CLINICAL FEATURESIn ACS the highest integrative functions of the brain, such as perception, processing and retrieval of information are disorganized (Geschwind 1982). This makes the acutely confused patient more or less incapable of thinking and acting in a rational and goal-directed m anner (Lipowski 1990).The essential features of ACS are the reduced ability to m aintain the attention paid to external stimuli and to appropriately shift attention to new external stimuli; the disorganized thinking resulting in reduced clarity of speech th a t appears ram bling, fragm entary, disjointed, irrelevant or incoherent. The syndrome also includes a reduced level of consciousness, sensory m isin te rp re ta tio n s , d istu rbances in the sleep-wake cycle and increased or reduced psychomotor activity. The patient is usually disorientated to time, place, situation a n d /o r person. The onset is often rapid and the course fluctuates in a typical m anner and the duration is short, at least if the etiologic factor/factors is /a re treated (Lipowski 1989, DSM-III, DSM-III-R).ACS may appear in three clinical variants: 1. a hyperactive variant, characterized by psychomotor over-activity; 2. a hypoactive variant, characterized by reduced psychomotor activity and apathy; 3. a mixed variant which shifts rapidly between hyperactive and hypoactive behaviour (Lipowski 1990). Many studies, old studies in particular, have focused mainly on ACS with an agitated, restless and disturbing symptomatology. The hypoactive, sometimes stuporous type may be misdiagnosed by the clinician and the necessary assessm ent of its cause is thus not performed.

13

DIFFERENTIAL DIAGNOSISACS may simulate many types of mental disorders in the elderly. In most cases the differential diagnosis concerns dem entia (Lipowski 1982). The two syndromes often coincide which sometimes makes a distinction between them difficult. ACS in a patient not assessed previously m akes it impossible to diagnose dem entia as the ACS symptoms interfere with the proper assessm ent of dementia. Both diagnoses are given only when there is a definite history of preexisting dementia. In most cases the typical rapid onset, fluctuating course, disturbed attention and clouding of consciousness make the ACS diagnosis simple. When there is uncertainty a provisional ACS diagnosis m ust be made and a consequent assessm ent of underlying d iseases m ust be performed. In time the proper diagnosis will appear.In some cases depression with cognitive impairment and psychotic sym ptom s may also cause diagnostic problems. Schizophrenia, schizophreniform disorders and other acute psychoses may appear with hallucinations, delusions and disordered thinking (Daniel 1985). In an elderly patient with cognitive impairment due to pre-existing diseases these diagnoses can cause diagnostic problems. Table 2 presents the typical symptoms of the syndromes or diseases tha t can cause diagnostic problems, constructed and modified after the DSM- III-R (APA 1987) and the monograph made by Lipowski (Lipowski 1990).

14

TABLE 2. A SIMPLIFIED GUIDE TO THE TYPICAL CLINICAL FEATURES OF ACS, DEMENTIA, DEPRESSION AND ACUTE FUNCTIONAL PSYCHOSIS.FEATURE ACS DEMENTIA DEPRES

SIONACUTE FUNCTIONAI PSYCHOSIS

Age at initial >60 >60 All ages <40onsetOnset Minutes- Months Weeks Days-

hours WeeksCourse Fluctuating Stable Fluctuating Stable

Duration Hours- Months- Weeks- Weeks-weeks years months m onths

Consciousness Reduced Clear Clear Clear

Emotions Fear, agony Indifferent Discomfort Various dependingupon type

Awareness Fluctuating Normal Normal May bedisordered

Alertness Fluctuating Normal Usually Usuallynormal normal

Attention Globally Usually Usually Usuallydisordered normal normal normal

Cognition Globally Globally Normal or Usuallydisordered impaired fluctuating normal

Orientation Impaired Impaired Usually Usuallyfluctuates normal normal

Memory Impaired Impaired May be May beimpaired impaired

Thinking Disorga Impaired Normal Oftennized abstraction disordered

Perception Distorted Often Normal Sometimesnormal disturbed

Hallucinations Visual and/ Absent Absent Predominantlyor auditory auditory

Delusions Fleeting un Absent Absent Sustained andsystematized systematized

Psychomotor Increased Normal Normal or Varies dependingactivity decreased slightly on type of

shifting decreased psychosisSpeech Incoherent Dysphasia Normal Normal,

slow or rapidSleep-Wake Disturbed Normal or Sometimes Sometimescycle often fragmentary disturbed disturbed

reversedInvoluntary Asterixis or Often Absent Usuallymovements tremor absent absent

commonPhysical illness Present Usually Usually Absent

absent absentDrug toxicity Often Absent Absent Absent

presentAwareness of Unaware Unaware or Highlights Usuallysymptoms conceals disabilities unawareAnswers Wrong "Near miss" I don't Usually

and conceals know correctEEG Abnormal Normal in Normal Usually

early phase normal

15

ACS FREQUENCYDifferent diagnostic ACS criteria and various patient populations in clinical studies have resulted in incidence and prevalence rates from0.7 - 80% (Liston 1982. MacDonald 1989, Lipowski 1990). In older studies of cognitive impairment in the elderly, few authors have made any distinction between dem entia and ACS. Even fewer au thors describe ACS in patients with dementia. The results of recent studies on the frequency of ACS in different patient groups are summarized in Table 3.

TABLE 3. FREQUENCY OF ACS IN RECENT STUDIES ON ELDERLY PATIENTS.

POPULATIONSetting Age Pat.

(N)ACS(%)

Author Year

GeneralhospitalGeneral

>60 99 56 Chisholm 1982

m edicineGeneral

>70 173 30 Gillick 1982

hospitalGeneral

>65 282 21 Erkinjuntti 1987

m edicine

Medical in

Allages

133 15 Thomas 1988

tensive care Nonintensive

>60 71 38* Foreman 1989

medical care General

>65 80 30 Rockwood 1989

hospital >75 146 40 Bucht 1990

Nursing home >75 203 65 Bucht 1990

Home forthe aged >75 196 25 Bucht 1990

Home care >75 172 25 Bucht 1990

Generalm edicine >70 235 20* Johnson 1990

Generalm edicine >70 229 22* Francis 1990

* Patients with dementia excluded** ACS diagnosed according to DSM-III or DSM-III-R criteria

1 6

POSTOPERATIVE ACSPostoperative or postanesthetic confusion has been given a lot of interest in the literature. As early as in 1887 George Savage carefully described 'postoperative insanity' based on his clinical experience and proposed possible etiologies (Savage 1887). A farsighted review of post-operative psychoses, risk factors and organic features such as 'endocrine upsets, changes in chemistry and nutritional disturbances' was written in 1938 by Milton Abeles (Abeles 1938).Postoperative confusion has been reported to occur after different types of surgery such as cataract surgery (Summers 1979, Burrows 1985), prostatectom y (Ghoneim 1988), hip-fracture surgery (Morse 1771, W illiams 1979), pelvic floor repair (Chung 1987) and cardiovascular surgery (Morse 1969, Morse 1971). For review on postoperative confusion see Whitaker 1989. In Table 4 the frequency of postoperative ACS in recent studies are summarized.Postanesthetic confusion has been associated with different types of premedication an d /o r anesthetic agents such as alcohol, belladonna, fentanyl and halothane (Savage 1887, Simpson 1976).

TABLE 4. FREQUENCY OF POSTOPERATIVE ACS IN RECENT STUDIES

Patientsample

Age Pat.(N)

ACS(%)

Author Year

Elective hipsurgery Allages

60 12* Hole 1980

General surgery >65 100 14 Millar 1981

Hip-fracture >60 170 52* W illiams 1985a

Hip-fracture(Intervention)

>60 57 44* Williams 1985b

General surgery Allages

92 42* Dieckelmann 1989

Elective orthopaedic surgery

>60 46 26* Rogers 1989*

Hip-fracture >60 35 43 Brännström 1989*

Hip-fracture (6 m onths survivors)

>65 536 23 Magaziner 1990

* Patients with dementia excluded** ACS diagnosed according to DSM-III or DSM-III-R criteria

1 7

ACS IN STROKE PATIENTSACS in stroke patients has never been studied prospectively in a rep resen ta tive sam ple of stroke p a tien ts . There are a few retrospective studies (de Reuck 1982, Dunne 1986), several studies on selected patients with different types and localisation of their strokes (Mullaly 1982, Schmidley 1984, Mori 1987, Garcia-Albea1989) and several case reports (Horenstein 1967, Medina 1974, Mesulam 1976, Medina 1977, Levine 1982, Graff-Radford 1984, Santam aria 1984, Price 1985, Balter 1986, Bogousslavsky 1988, Devinsky 1988). It is difficult to apply DSM-III or DSM-III-R criteria on retrospective study samples and a retrospective approach is likely to lead to an underestim ation of ACS. In one retrospective study, using DSM-III criteria for ACS, 150 of the 450 (33%) patients with cerebra l in farc tions were d iso rien ta ted or ”confused" upon presentation, and 112 of the 211 (53%) with spontaneous brain hem orrhage were disorientated or "confused” (Dunne 1986). In a study of patients with right middle cerebral artery infarction, 25 out of 41 (61%) were judged to be acutely confused (Mori 1987). ACS has been reported to be more common in patien ts with right hem isphere lesions than in patients with left hem isphere brain lesions (Dunne 1986). In different case reports, ACS has been associated with cerebrovascular lesions affecting a large variety of specific locations such as right middle cerebral artery infarctions, left posterior infarctions, medial temporo-occipital infarction, after infarctions of the hippocampal region, the fusiform and lingual gyri or in the thalam us (Horenstein 1967, Medina 1974, Medina 1977, M esulam 1976, Mullally 1982, de Reuck 1982, Levine 1982, Schmidley 1984, Graff-Radford 1984, Santam aria 1984, Price 1985, Dunne 1986, Mori 1987, Bogousslavsky 1988, Devinsky 1988, Garcia-Albea 1989). ACS has also been described in patients with m ultiple small cerebral infarctions (Balter 1986). However, the definition of ACS differs between various studies, and only one retrospective study used the DSM-III criteria (Dunne 1986).

PREDISPOSING ACS FACTORSA m ultifactorial approach considering the p a tie n t 's previous diagnoses, medication and psycho-social capacity is necessary in the assessm en t of ACS (Arie 1981). There are often one or more

1 8

preventable or treatable risk factors or triggering factors in elderly patients (Jolley 1981, Lipowski 1990).The literature contains an overwhelming num ber of case reports but few system atic analyses of risk factors and potential triggering mechanisms for ACS.The following is a summary of factors reported to increase the risk for ACS:Age: Im portant ACS risk factor according to most studies. (Morse 1969a, Morse 1969b, Morse 1971, Hodkinson 1973, Varsamis 1978, Liston 1982, Williams 1985b, Dickelmann 1989, Lipowski 1990). Gender: Most studies report a similar ACS frequency in men and women (Judge 1977, Lipowski 1990). However, in three studies a higher ACS frequency has been reported among men (Simon 1963, Kay 1972, Seymour 1983).Dementia: Progressive degenerative brain diseases lower the ACS th resho ld by reducing the neu ro tran sm itte r m etabolism , e.g. acetylcholine (Gottfries 1983, Gottfries 1987). It is easy for a demented patien t to become acutely confused due to biological, psychological or environmental factors (Morse 1969a, Morse 1969b, Hodkinson 1973, Varsam is 1978, Beresin 1988, Hege-Scheuing 1989, Lipowski 1990, Francis 1990, Johnson 1990b, Thienhaus1990).Previous stroke: Stroke has also been reported to be a predisposing and triggering factor for ACS (Flint 1956, Lipowski 1990).Cardiac disease: Several cardiac diseases are reported to predispose to ACS (Wolff 1935, Flint 1956, Morse 1969a, Morse 1969b, Varsamis 1978).Depression increases the patien t's vulnerability (Gold 1988a, Gold 1988b) and increases the risk for ACS (Wolff 1935, Morse 1969a, Morse 1969b, Varsam is 1978). O ther preoperative psychological s ta tes , e.g. anxiety, h as also been reported to p red ict the postoperative psychological course (Abraham 1961, Varsamis 1978). Impaired hearing is an important ACS risk factor according to several studies (Hodkinson 1973, Judge 1977).Impaired vision has also been reported to be an ACS risk factor (Hodkinson 1973, Judge 1977, Lipowski 1990).Drugs: Treatm ent with many different groups of drugs, drugs with anticholinergic effects in particular, has been reported to increase

19

the ACS risk or to trigger ACS in the elderly (Itil 1966, Tune 1981, Liston 1982, Judge 1977, Judge 1987, Gordon 1988, Miller 1988, Foreman 1989a, Hege-Scheuing 1989, Francis 1990) and especially in the elderly with dementia diseases (Thienhaus 1990).

ANESTHETIC TECHNIQUE AND HIP-SURGERYThe incidence of hip-fracture is rapidly increasing in all the western world (Jensen 1980b, Nickens 1983). Old age is an im portant risk factor for hip fractures but explains only a minor part of the increase (Zetterberg 1982). A nesthesists would like to select an anesthetic technique resulting in minimal postoperative morbidity for the old and vulnerable patient with a fracture of the femoral neck (Covert 1989). General anesthesia is reported, in older studies, to cause more complications in elderly patients than regional anesthesia (McLaren 1978). A general opinion in the literature before 1980 was th a t general anesthesia was complicated with more adverse cerebral effects than regional anesthesia (Gauthier 1963). G authier on the other hand reported a similar morbidity and m ortality rate after spinal anesthesia and general anesthesia. In Table 5 some randomized studies on hip surgery are summarized. Neither general nor regional anesthesia are convincingly supported by the resu lts regarding mortality or postoperative cognitive functioning.

2 0

TABLE 5. GENERAL ANESTHESIA (GA) VERSUS REGIONAL ANESTHESIA (RA).

Patientsample

N GA or RA better for postop cognition

Difference Author in mortality

Year

Hip-fracture 55 - GA>RA McLaren 1978

Hip-replacem. 60 RA better than GA No Hole 1980

Hip-fracture 100 - No McKenzie 1980

Hip-fracture 60 - No White 1980

Hip-fracture 132 - No Davis 1981

Hip-fracture 169 - No Wikström 1982

Hip-replacem. 30 GA=RA No Riis 1983

Hip-fracture 150 - GA>RA McKenzie 1984

Hip-fracture 40 GA=RA No Bigler 1985

Hip-fracture 578 - No Valentin 1986

Hip-fracture 538 - No Davis 1987

Hip-replacem.Prostatectom yHysterectomy

105 GA=RA No Ghoneim 1988

Hip or knee replacem. 146 GA=RA No Jones 1990

- = not assesed

ACS AND CLINICAL OUTCOMEThe mortality rate for patients developing ACS has been reported to be higher compared with th a t of patients who do not develop ACS (Bedford 1959, Roth 1959, Simon 1963, Rabins 1982, Liston 1982, Weddington 1982, Trzepacz 1985, Rockwood 1990, Francis 1990). ACS is probably underestimated as a risk factor contributing to death in many groups of patients (Weddington 1982).In several studies prolonged ward-stay has been reported for patients who develop ACS (Glass 1977, Sheppeard 1980, Lamont 1983, Thom as 1988, Levkoff 1988, B rännström 1989, Rockwood 1990, Brännström 1991). The prolonged ward-stay is often associated with

21

many complications which eure caused at least partly by the ACS. ACS has also been reported to be associated with poorer rehabilitation result (Magaziner 1990). The acutely confused patient is reported to be a t the risk of being hospitalized as staff do not expect him to be able to manage on his own. This could contribute to prolonged dependency upon caregivers also after the ACS h as been reversed (Brännström 1991). It has been shown tha t elderly patients' mental im pairm ent influences the nurse-patient interaction negatively. In confused patients the psychosocial interaction was particularly poor and the physical care was given priority (Armstrong-Esther 1986). ACS patients have difficulties in cooperating with staff as they do not understand nor remember instructions. They often exhibit behaviour disturbances and /o r act perilously demanding continuous supervision. The ACS-associated complications need to be attended to and often an acutely confused patient is probably incapable of experiencing th irst or hunger and even more incapable of satisfying these needs (Brännström 1989).

ETIOLOGY AND PATHOGENESISThere are two m ajor hypotheses about the pathogenesis and pathophysiology of ACS (Lipowski 1987). The first hypothesis suggests th a t a reduction in the cerebral m etabolism and the consequent reduction in neurotransm itters, especially acetylcholine, contributes to the development of ACS (Blass 1979, Blass 1983). This hypothesis is supported by studies showing th a t the cerebral acetylcholine synthesis is especially sensitive to hypoxia and hypoglycemia (Gibson 1981, Hirsch 1984). A close link between ACS and anticholinergic activity has been reported by several authors (ltd 1966, Tune 1981, Mondimore 1983, Miller 1988, Thienhaus 1990). ACS caused by anticholinergic medication can be reversed by means of physostigm ine, a cho lineesterase inh ib ito r (Green 1971, Aquilonius 1978). The second hypothesis suggests th a t ACS is a reaction to stress mediated by elevated plasma cortisol and its effects on the brain (Kraal 1962, Kraal 1975, Carpenter 1982, McEwen 1987, McEwen 1988). High cortisol levels have been suggested to affect the neuron function of the central nervous system (Sapolsky 1985, de Kloet 1987) and the cognitive function as a result of this (Micco 1980, Reus 1987, Issa 1990, Wolkowitz 1990). This may be

2 2

m ost pronounced in the hippocam pus, where the num ber of glucocorticoid receptors is particularly high and thus compromising the cognitive function (Micco 1980, Gilad 1987, Joels 1989, A rm anini 1990, Sapolsky 1990a). G lucocorticoids endanger hippocampal neurons probably by impairing their energy metabolism (Sapolsky 1986a). This is why glucocorticoids probably increase the neuronal vulnerability to hypoxia, ischem ia and hypoglycemia. Supplem enting with 'brain fuels' in anim al models reduces the toxicity of glucocorticoids in the hippocam pus (Sapolsky 1986a). It has also been reported tha t cortisol modulates cholinergic receptors in the subcortical limbic forebrain (von Euler 1990).In several studies, ACS has been found to occur after hippocampal stroke (Medina 1974, Graff-Radford 1984, S an tam aria 1984, Bogousslavsky 1988).Acetylcholine is an important direct and indirect HPA-axis regulator (Gilad 1987, Calogero 1988). A close connection between stress, h ip p o cam p al cho linerg ic system , cognitive fu n c tio n and glucocorticoids has been dem onstrated in anim al models (Gilad 1987, Lai 1990). High cortisol levels have been reported to be associated with postoperative ACS 2-4 days after elective surgery (McIntosh 1985).In most cases of ACS one or both of these mechanisms are involved in the development of the cognitive disturbances. O ther mechanisms interfering with cerebral metabolism and transm itter activity may also be of importance such as the factors interfering with glutamate metabolism (Sapolsky 1990a). Protection of the cerebral oxidative metabolism and a reduction in stress mediated by high cortisol levels thus seem to be the most im portant possibilities in ACS prevention and treatment.In Table 6, some suggested an d /o r documented etiological factors are presented, structured according to the two main hypotheses of ACS pathophysiology. Psycho-social factors, which have been suggested to contribute to or cause ACS, could act through the stress-cortisol system. Some psycho-social factors, which have been suggested to be risk factors an d /o r ACS triggering factors are presented in Table 6 under II B.

2 3

TABLE 6. SUGGESTED AND/OR DOCUMENTED ACS CAUSES IN THE LITERATURE. THE ACS CAUSES ARE STRUCTURED ACCORDING TO THE TWO MAIN HYPOTHESES OF ACS PATHOPHYSIOLOGY.

I. ACS MEDIATED MAINLY BY DISTURBANCES IN THE CHOLINERGIC FUNCTION OF THE BRAIN. (The acetylcholine synthesis is especially sensitive to disturbances in the brain energy metabolism, e.g. hypoxia, hypoglycemia. Disturbances in the cholinergic system is closely associated with cognitive disturbances).

A. DISTURBED CEREBRAL OXYGENATION1. Pulmonary diseases, e.g. pneumonia, pulmonary embolism2. Cardiac diseases, e.g. congestive heart diseases, pulmonary oedema, cardiac arrhythmia, myocardial infarction3. Anaemia4. Hypoperfusion, e.g. peroperative hypotension, hypovolemia

(bleeding, dehydration), ortostatism, aortic stenosis, ischemic brain disease, vasculitis, hyperventilation syndrome, disseminated intravascular coagulation, increased blood viscosity (e.g. polycyternia)

5. Carbon monoxide poisoning6. Methemoglobinemia

B. DEPRIVATION OF ENERGY OR NUTRITIVE SUBSTANCES1. Hypoglycemia, e.g. insulin-coma, spontaneous (i.e.insulinoma, liver disease, starvation, cortisol deficiency), drug-induced (i.e.oral antidiabetics, haloperidol)2. Cofactor deficiency, e.g. thiamine, niacine, pyridoxine, vitamin B12, vitamin E, vitamin C, folate3. Hypoproteinemi

C. TOXIC DISTURBANCES IN THE CHOLINERGIC SYSTEM1. Drugs, e.g. neuroleptics, tricyclic antidepressants, cortisone, antihistamines, other drugs with anticholinergic effects

II. ACS AS A REACTION TO STRESS PROBABLY MEDIATED MAINLY BY HYPERCORTISOLISM. (Cholinergic neurons in hippocampus might be especially sensitive to hypercortisolism and glucocorticoids endanger hippocampal neurons by impairing their energy metabolism. Glucocorticoids probably thereby increase the damage to cholinergic neurons in the hippocampus induced by hypoxia or ischemia for instance).

A. ORGANIC FACTORS1. Trauma, e.g. fractures, bums, contusions2. Acute medical diseases, e.g. myocardial infarctions, congestive heart failure, acute stroke, deep vein thrombosis, pulmonary embolism3. Acute surgical diseases, e.g. pancreatitis, cholecystitis, gastric ulcer4. Diseases with hypercortisolism, e.g. Cushing's syndrome5. Urinary retention6. Allergic reactions7. Fecal impaction

B. TREATMENT WITH CORTICOSTEROIDSC. PSYCHOSOCIAL FACTORS

1. Severe emotional stress, e.g. fatigue, pain, grief, anxiety, relocation2. Secondary to psychiatric disorders, e.g. depression, mania, cycloid psychoses3. Immobilization4. Sensory deprivation, e.g. blindness, deafness5. Sensory overload e.g. noise6. Sleep deprivation

2 4

III. OTHER METABOLIC OR TOXIC ACS-MECHANISMS DISTURBING THE BRAIN METABOLISM IN A MORE GENERAL WAY (Causes probably partly acting through previous mechanisms, hypoxia, hypoperfusion and/or otherwise disturbed transmitter metabolism partly caused by stressmediated hypercortisolism).

A. FEBRILE STATES1. Urinary infections2. Pneumonia3. Septical infections

B. HYPOTHERMIAC. ENDOCRINE DISORDERS

1. Pituitary disorders2. Hyper/hypothyreoidism3. Hyper/hypoparathyreoidism4. Addison's disease5. Pheochromocytoma

D. WATER AND ELECTROLYTE DISTURBANCESDifferent osmolality and electrolyte disturbances can cause ACS, e.g. hypo-osmolality (water intoxication), hyper-osmolality (nonketotic diabetic coma), hypo- and hypercalcemia, hypo- and hyperkalemia hypo- and hypernatremia, hypo- and hypermagnesemia, hypo- and hyperchloremia, hypo- and hyperphosphatemia

E. ACID-BASE DISTURBANCES1. Alkalosis/Acidosis

F. KIDNEY DISEASES1. Uraemia

G. LIVER DISEASES1. Liver precoma2. Porphyria

H. DRUGSMost drugs in toxic or therapeutic doses can cause ACS, e.g. penicillin, anticonvulsants, cardiac glycosides, sedative drugs, salicylates, analgetics, 1-dopa, amantadine, selegiline, bromokriptine, Cimetidine, timolol, anti tumour agents, barbiturates

I. ALCOHOL1. Intoxication2. Withdrawal

J. POISONSDifferent poisons can cause ACS, e.g. paraldehyd, methyl alchohol, ethylene glycol, heavy metals, cyanide, bromid, insecticides

IV. OTHER CENTRAL NERVOUS DISORDERS (Causes probably partly acting through previous mechanisms, hypoxia, hypoperfusion and/or otherwise disturbed transmitter metabolism partly caused by stressmediated hypercortisolism).

A. HEAD TRAUMA1. Concussion2. Subdural hematoma

B. EPILEPSY1. Post-ictal state2. Seizure disorders

C. INFECTIONSMost cerebral infections can cause ACS, e.g. meningitis, encephalitis, neurosyphilis, borreliosis, cerebral abscess, toxoplasmosis, malaria

D. BRAIN TUMOURSE. MULTIPLE SCLEROSIS

25

HIP-FRACTURE AND STROKE

Hip fractures and stroke cause a great deal of suffering for many elderly people and are quantitatively im portant diseases in the elderly in the w estern world (Ceder 1979, Jensen 1980a Jensen 1980b, Jensen 1980c, Luthje 1982, Goldstein 1983, Nickens 1983, Wallace 1983, W hishnant 1984, Eriksson 1987).

Femoral neck fractures predom inantly strike the old woman. As m entioned previously the incidence of hip fractures is rapidly increasing in all the western world (Zetterberg 1982, Jensen 1980b, Jensen 1980c, Nickens 1983, Falch 1985). In four of the papers ACS was studied in patients operated on for femoral neck fractures which is one of the m ost common operations among old people. The operation is usually rapid, and standardized and peroperative bleeding is rare. When studying the postoperative consequences of anesthetic technique it is im portant th a t the surgical procedure remains relatively constant.

In the United States alone there are 1.7 million stroke survivors at any given time (Grotta 1988). Patients with stroke are increasing in num ber in Sweden and represent the patient group consuming the highest percentage of hospital care (MFR 1986). Stroke patients like hip fracture patients have a high mean age but stroke is somewhat more common among men. The organisation of stroke care into nonintensive stroke un its has improved the care and thereby the prognosis for stroke patients (Strand 1985, Eriksson 1987). The stroke unit in Medical Department 1 in Umeå was opened in 1978 and has been proved to be a good basis for research and for the development of stroke care (Strand 1985, Eriksson 1987).

2 6

AIMS OF THE STUDY

The aims of this study were:- the elucidation of ACS in patients with femoral neck fractures and patien ts w ith acute stroke with regard to frequency, predictors, possible pathogenetic m echanism s, associated com plications, assessm ents and documentary routines and the clinical outcome for the patients.- the development and evaluation of an intervention program to prevent and trea t ACS in patients operated on for femoral neck fractures.

2 7

PATIENTSPaper I: Onehundred and eleven consecutive patients 65 years old or older, with femoral neck fractures adm itted to the Orthopaedic departm ent of Umeå University Hospital between March 1983 and June 1984.Paper II: Fifty-seven patients, lucid a t admission, 65 years old or older, with femoral neck fractures, th a t could be randomized to receive either general or regional anesthesia and adm itted to the Orthopaedic departm ent of Umeå University Hospital between March 1983 and November 1984. Forty-five of the fifty-seven patients were also included in Paper I.Paper III: Onehundred and forty-five consecutive stroke patients of all ages adm itted to the stroke un it of the departm ent of Internal Medicine of Umeå University Hospital between April 1983 and December 1984.Paper IV: Eighty-three selected stroke patien ts of all ages with supratentorial ischemic stroke adm itted to the stroke un it of the departm ent of In ternal Medicine of Umeå University Hospital between June 1983 and March 1986.Paper V: All patients in Papers I (N = lll) and II (N=57) and two retrospective patient samples comparable with tha t of Paper I. The first retrospective control sample included: all patients 65 years old or older, irrespective of prefracture m ental state, adm itted to the Orthopaedic departm ent of Umeå University Hospital during 1980 (N=66). The second retrospective control sample, with the same inclusion criteria was admitted to the departm ent of General Surgery of Piteå County Hospital during 1980 and 1981 (N=68).Paper VI: Onehundred and three consecutive patients, 65 years old or older, with femoral neck fractures admitted to the Orthopaedic departm ent of Umeå University Hospital between December 1986 and January 1988.

2 8

METHODS

ACS was diagnosed according to the DSM-III or DSM-III-R criteria for delirium (APA 1980, APA 1987), after clinical assessm ents including: 1. P re-frac tu re /p re-stroke history checked th rough medical records and by m eans of interviews with the pa tien ts ' families or care-givers. 2. Behaviour on the ward assessed by direct patient observations and interviews with the staff. 3. Orientation, su sp ic iousness, em otions, depression, speech, delusions and hallucinations, recognition, motor function, sociability and changes in the patients' mental function assessed and registered by m eans of the Organic Brain Syndrome Scale (Gustafson 1985, Hallberg 1989). 4. A Mini-Mental State Examination (MMSE) was made on admission and then once a week during the hospital stay for all patients included in the studies on stroke patients (Folstein 1975). MMSE was performed when considered necessaiy for the ACS diagnosis in the patients with femoral neck fractures.All patients in this study were observed several times per day and assessm ents including tests and interviews were carried out on the first day of admission and then a t several times during the patients' hospital stay. All tests, interviews with patients, relatives and staff were carried out by the same physician (YG) in the studies on stroke patients and by three different raters in the studies on patients with femoral neck fractures. Before the s ta rt of the studies the three raters assessed ten patients to test the in terrater reliability of the registration of the items included in the OBS-scale. The tests of these ratings were analyzed, and the agreement between the ra ters was above 90% in all ratings. While these studies were performed two of the ra te rs collaborated in two other studies (Brännström 1989, B rännström 1991) using the sam e assessm ent routines. In both studies the agreement on the ACS diagnosis according to DSM-III, was above 95% between them.In Paper I, all pa tien ts 65 years old or older, irrespective of prefracture mental state, operated on for femoral neck fractures with various anesthetic techniques, were studied regarding ACS frequency. In Paper II, patients lucid on admission were randomized to receive either general or regional anesthesia.

2 9

In Paper III the same methods, as in Paper I were used for stroke patients.In Paper IV a selected sam ple of pa tien ts w ith supraten torial ischemic stroke was assessed with the dexamethasone suppression test (DST) and compared with a healthy control population.In Paper V data from the review of medical records were compared with clinical studies from the care occasions, presented in Papers I and II. To find out w hether the on-going clinical studies had influenced the physicians' and nu rses ' ACS documentation in their case-notes, two other retrospective case-note sam ples on the corresponding patient groups were studied.The medical records were studied in order to find the noted ACS frequency and treatment, notes on its consequences and the nursing activities associated with these notes. In the study of the medical records, ACS criteria were explicit statem ents on ACS a n d /o r documented symptoms or behaviour indicating ACS. The analysis of the records and the classification of p a tien ts were m ade independently by two of the authors of Paper V (BB, YG). There was 89% exact agreement between the authors regarding the diagnoses made from the case-note analyses. Every case of disagreement was subject to diagnostic discussions ending up in full agreement on the ACS and the dementia diagnoses.The intervention program (Paper VI), was based on the results of our previous studies (Papers I and II). The intervention aimed a t preventing postoperative ACS by protecting the patients cerebral oxidative m etabolism which was achieved by the prevention of hypoxia and hypotension/hypo-perfusion. Patients who developed postoperative ACS were assessed and trea ted for associated complications. The intervention study could not be performed as a randomized study since the resu lts of our previous studies had initiated changes in the treatm ent routines for these patients. It was also regarded as unethical, considering our previous results, not to prevent severe hypoxemia and peroperative hypotension. The results of the intervention were therefore compared with the outcome of the patients in Paper I.

3 0

STATISTICSThe Systat™ statistical package was used for calculations (Wilkinson 1987, Wilkinson 1990). The chi-square test, Yates' corrected chi- square te s t. S tu d en t's t-tes t, M ann W hitney U -test, Pearson correlation coefficients, the odds ratios and 95% confidence intervals for odds ratios were used when relevant as indicated in the text (Feinstein 1985, Sandercock 1989). The Bonferroni correction was used to adjust for probabilities. To find independent clinical ACS predictors, multiple linear regression analyses were used in Papers I, II and III (Draper 1966). In the multiple linear regression models the F-ratio and the P-value were used to dem onstrate the statistical significance of the model. The squared multiple R was used to illustrate the explanatory degree of the model. The predictors of the geriatric stroke rehabilitation in Paper III were calculated by the use of orthogonal scores from a factor analysis representing clinical variables. In Paper IV a logistic regression model (Dobson 1982) was used in the SAS program package. Also in Paper VI logistic regression models were used to find independent ACS predictors in the two patient samples but the calculations were made in the SPSS program package. In this thesis logistic regression models are also presented for the prediction models presented in Paper I and III. A logistic regression model is preferable as a multiple linear regression model may have some unwanted properties such as the risk of the predictions of new observations ending up outside the range [0,1]. An iterative maximum likelihood procedure was used in the SAS and SPSS program packages where the results obtained are interpreted from the exact distribution (binomial) of the response variable (ACS). The choice of different program packages for the logistic regression analyses was made for technical and financial reasons. A P-value of less than 0.05 was regarded as statistically significant.

3 1

RESULTS

ACS FREQUENCYA summary of the ACS frequency in different papers are presented in Tables 7 and 8.

TABLE 7. ACS FREQUENCY IN THE STUDIED SAMPLES OF PATIENTS WITH FEMORAL NECK FRACTURES.

Paper N Mean Age (range)

Male/Female

Dementia(%)

ACS(%)

ACS>days

I. All patients n . Patients

111 79(65-96)

28 /83 15 61 40

lucid atadmissionV.Retrospect.

57 78(65-95)

11/46 0 44 28

of paper I

Retrospect.

111 79(65-96)

28 /83 43

of paper II

Retrospect.

57 78(65-95)

11/46 32

control 1

Retrospect.

66 79(65-96)

22 /44 44

control 2

VI. In ter

68 78(66-95)

24 /44 47

vention all patients Patients

103 80(65-102)

28 /75 22 48 29

lucid at admission

66 78(65-94)

14/52 0 27 9

- = Not assessed

3 2

TABLE 8. ACS FREQUENCY IN THE STUDIED SAMPLES OF STROKE PATIENTS.

Paper N Mean Age Male/ Dementia ACS ACS>7(range) Female (%) (%) days(%)

m . Ailstrokepatients

145 73 90/55(40-101)

6 48 31

IV. Selectedstrokepatients

83 75(44-89)

52/31 42

ACS was common both in the representative prospective sample of patients with femoral neck fractures (61%) and in th a t of stroke patients (48%). The majority of patients developing ACS in these two samples were acutely confused for more than one week. The ACS frequency of the intervention study (VI) was lower than th a t of the control study (I) (61% vs 48%, p<0.05). When comparing patients lucid a t admission in Paper VI with the corresponding patient sample in Paper II, there were fewer patients in the intervention study who were acutely confused for more than seven days (9% in the intervention study compared with 28% in the control study, pcO.Ol).

3 3

PREDISPOSING ACS FACTORSACS predictors in the non-intervention hip-fracture patient sample in Paper I are shown in Table 9.

TABLE 9. ACS PREDICTORS IN PATIENTS WITH FEMORAL NECK FRACTURES. Results of logistic regression model in Paper I. Dependent variable: postoperative acute confusional state (ACS) in logistic transform ation (N = lll).

Independentvariables:

WaldChi-square

P

Age 8.13 0.004Dementia 3.05 0.08Drugs with anticholinergic effects* 2.09 0.15Depression 2.06 0.15Cerebrovascular diseases** 1.92 0.17Cardiac diseases*** 0.86 0.35Sex 0.001 0.98

Chi-Square for the model: P-value<0.0001. Connection between predicted probabilities and responses observed : Concordant: 74%* Drugs with anticholinergic effects include neuroleptics, antidepressants and other drugs with anticholinergic effects.** Cerebrovascular diseases include cerebral infarction, cerebral hemorrhage and TIA*** Cardiac diseases include heart failure, previous myocardial infarction, atrial fibrillation and angina pectoris.

In the logistic regression model (Table 9) of Paper I, old age was the only significant ACS predictor. It should however be observed that 16/17 (94%) patients with dementia, 15/17 (88%) with depression, 15/18 (83%) with previous stroke and 33/39 (85%) on regular treatm ent with drugs with anticholinergic effects developed ACS even though these variables did not reach statistical significance in the logistic regression model.

3 4

In Paper II, where postoperative ACS in a selected sample of patients was studied, regular treatm ent with drugs with anticholinergic effects and depression were found to be predictors of postoperative ACS.ACS predictors in consecutive stroke patients were studied in Paper III and are shown in Table 10.

TABLE 10. ACS PREDICTORS IN STROKE PATIENTS.Results of logistic regression model in Paper III. Dependent variable: Acute confusional state (ACS) in logistic transformation (N=145).

Independentvariables:

WaldChi-square

P

Degree of paresis 21.49 0.0001Age 5.23 0.02Left-sided brain lesion 3.44 0.06Previous ACS 3.19 0.07Drugs with anticholinergic effects* 3.14 0.08Cardiac diseases** 2.06 0.15Sex 0.23 0.64Dementia 0.14 0.71

Chi-Square for the model: P-value<0.0001. Connection between predicted probabilities and responses observed: Concordant: 85%* Drugs with anticholinergic effects include neuroleptics, antidepressants and other drugs with anticholinergic effects.** Cardiac diseases include heart failure, previous myocardial infarction, atrial fibrillation and angina pectoris.

In Paper III, extensive motor impairment and old age were independent ACS predictors in stroke patients. Left-sided brain lesions, previous episodes of ACS and regular treatm ent with drugs with anticholinergic effects were on the verge of being statistically significant ACS predictors.ACS predictors for selected stroke patien ts with supraten torial ischemic stroke, in whom the hypothalam ic-pituitary-adrenal axis

3 5

was studied by the dexamethasone suppression test (Paper IV), are shown in Table 11.

TABLE 11. ACS PREDICTORS IN PATIENTS WITH SUPRATENTORIAL CEREBRAL INFARCTION. Results of a logistic regression model used to estimate the predicting value of included variables, after dexamethasone suppression test (Paper IV). Dependent variable: Acute confusional state (ACS) in logistic transformation (N=83).

Variable WaldChi-Square

P

Plasma cortisol* 6.45 0.01Degree of paresis 6.05 0.01Left-sided brain lesion 4.02 0.05Age 3.58 0.06Anticholinergic medication 0.86 0.36Male sex 0.22 0.64

C hi-square for the model: P-value=0.0002. Connection between predicted probabilities and responses observed : Concordant=80%. *Plasma cortisol m easured at 7 am after giving the patient 1 mg dexamethasone at 11 pm the day before.

Elevated plasm a cortisol, which was common early after stroke, the degree of paresis on day four after admission and left-sided brain lesions turned out to be significant ACS predictors in the logistic regression model of Paper IV. Old age was on the verge of being a statistically significant ACS-predictor.In Table 12 the logistic regression model used to analyze the ACS predictors in a consecutive sample of elderly patients treated for femoral neck fractures is shown (Paper VI). These patients were treated in accordance with the intervention program.

3 6

TABLE 12. ACS PREDICTORS IN PATIENTS WITH FEMORAL NECK FRACTURES. Results of logistic regression model of the intervention study (Paper VI). Dependent variable: Postoperative acute confusional state (ACS) in logistic transformation (N=103).

Independent Wald Pvariables: Chi-square

Dementia 10.62 0.001Sex* 8.85 0.003Depression 2.34 0.13Cardiac diseases 1.88 0.17Age 1,17 0.28Cerebrovascular diseases 0.25 0.62Drugs with anticholinergic effects 0.02 0.90

Chi-Square for the model: P-value<0.0001. Connection betweenpredicted probabilities and responses observed: Concordant: 76% *Male sex was associated with an increased ACS risk in the intervention study.

Dementia and male sex were the independent ACS-predictors of the intervention study (VI).

ANAESTHETIC TECHNIQUE AND ACSIn paper II, 57 patients lucid on adm ission were randomized to receive either epidural or halothane anesthesia. One aim was to see if the anesthetic technique influenced the frequency of postoperative confusion. We found th a t 44% of the patien ts developed ACS correlating closely to a history of mental depression (PcO.Ol) and to the use of drugs with anticholinergic effects (P<0.005). There was no difference in the ACS frequency between the two anesthetic groups. In patients given halothane, however, early postoperative hypoxemia was associated with ACS (P<0.05). Patients who developed ACS had significantly more postoperative complications and almost four times longer total hospitalization time (Table 13). It was concluded th a t anticholinergic medication and a history of mental depression were

3 7

predom inant risk factors for the development of postoperative ACS and more important than the anesthetic technique in this respect. In paper I there was a close connection between peroperative blood- pressure falls and postoperative ACS. Also in the intervention study (Paper VI) despite oxygen therapy, active prevention and treatm ent of blood-pressure falls, some patien ts exhibited hypoxia an d /o r peroperative hypotension, and alm ost all of them developed postoperative ACS.

ACS AND CLINICAL OUTCOMEWard stay for patients with and without ACS in the different studies are shown in Table 13.

TABLE 13. WARD STAY IN THE ACUTE CARE WARDS FOR PATIENTS WITH AND WITHOUT ACS.

Paper Ward stay (days)Type of Lucid Patients P-valueward stay patients with ACS

I Orthopaedic dept 13 20 <0.05II Orthopaedic dept

Total ward stay of 1 year for previously non

13 25 <0.01

hospitalized patients 22 77 <0.0001III Stroke unit 13 19 <0.001IV Stroke unit 16 23 <0.005VI Orthopaedic dept

(Intervention)11 13 =0.17

ACS in Papers I-IV was associated with prolonged ward stay. In the intervention study, where ACS frequency was lower and ACS duration shorter, the difference in ward stay between patien ts with and without ACS was not significant.Mortality for patients with and without ACS in the different papers are shown in Table 14.

3 8

TABLE 14. MORTALITY IN PATIENTS WITH AND WITHOUT ACS.

Paper Total mortality in Total 6 monththe acute care ward. mortalityLucid ACS Lucid ACS

I(111 pat.) 1/43 2 /6 8 4 /4 3 14/68

II(57 pat.) 0 /3 2 1/25 0 /32 2 /2 5

I ll(145 pat.) 2 /7 6 11/69* 6 /7 6 16/69**

IV(83 pat.) 0 /4 8 2 /3 5 4 /4 8 6 /3 5

VI(103 pat.) 4 /5 4 2 /4 9 6 /5 4 7 /4 9

* p<0.05 (Yates' corrected chi-square test)

** p<0.05 (chi-square test)

The other comparisons showed no statistical significant differences.

ACS in stroke patients (Paper III), was associated with a significantly higher mortality rate both during the acute care ward stay and at six m onths. In the studies on patients with femoral neck fractures the mortality rate was low. The differences in mortality between patients w ith and w ithout ACS in Paper I, did no t reach s ta tis tica l significance. In the intervention study there was the sam e low mortality rate both in patients with and in patients without ACS.

ACS DIAGNOSIS AND DOCUMENTATIONIn Paper V the assessm ent and treatm ent of ACS documented by the physicians and nurses who treated the patients in Papers I and II were studied. The ACS documentation in the records is summarized in Table 15.

3 9

TABLE 15. THE PHYSICIANS' AND NURSES' ACS DIAGNOSES AND INDICATIVE NOTES OP ACS IN PAPERS I AND H AND IN THE RETROSPECTIVE CONTROLS.Comparison II includes only preoperatively lucid non-demented patients. The other three samples are comparable and include all patients irrespective of preoperative mental state.

ACS FREQUENCY C l inical s t u d y Ca s e n o t e s t u d y

n % n % p

Comparison I (N=l 11)

68 61 48 43* <0.01

Comparison II (N=57)Umeå retrospective

25 44 18 32 ns

control (N=66) - - 29 44*Piteå retrospective control (N=68)

32 47*

* There was no statistically significant difference between the three comparable samples.

The study compared the results of two clinical studies with the results of case note studies of the same populations of patients and on the same care occasions. All patients were 65 or older and operated on for femoral neck fractures. All comparisons showed that both physicians and nurses diagnosed ACS unsatisfactorily and documented the patients’ mental states poorly. This was especially true of patien ts who had prefracture dem entia or were acutely confused even on admission. This was the probable explanation of the difference between com parisons I and II. The analysis of two retrospective case note control samples gave the same results as the retrospective analysis of the two clinical studies. Neither the physicians nor the nurses used any kind of diagnostic instrum ent to detect cognitive disorder in the patients.Nurses documented ACS better in the records than the physicians b u t the docum ented nursing actions associated with ACS were

4 0

insufficient with a few exceptions. The most common nursing action associated with ACS was the administering of extra medication, often drugs with anticholinergic effects, on request.

INTERVENTIONThe main results of the intervention can be seen in Tables 7, 13, 14, 16 and 17. The ACS frequency was lower, 48%, in the intervention study compared with 61% (p<0.05) in the control study (Table 16). Furthermore, the ACS occurring in the intervention study was less severe and of shorter duration than the ACS of the control study. The frequency of postoperative decubital ulcers, severe falls and urinary re ten tion was also lower (Table 17). The m ean dura tion of orthopaedic ward stay was 17 days in the control study and 12 days in the intervention study (pcO.OOl) (Table 13). The intervention program reduced the frequency, severity and duration of ACS which resulted in shorter orthopaedic ward stay (Tables 7 and 16).The ACS predictors of the intervention study were different from those of the control study (Compare Tables 9 and 12). Old age was the independent ACS-predictor in Paper I and old age and male sex in Paper VI.

TABLE 16. ACS FREQUENCY IN THE CONTROL STUDY AND IN THE INTERVENTION STUDY.

Control Intervention(N= 111) (N=103)N % N % P

ACS on admission 32 28.8 30 29.1 N.S.Preoperative ACS 37 33.3 30 29.1 N.S.

Postoperative ACS 68 61.3 49 47.6 <0.05ACS >7 days 44 39.6 30 29.1 N.S.Severe ACS* 33 29.7 7 6.8 <0.0001

*ACS was defined as severe if it caused documented severe caring problems.

41

The ACS frequency was lower in the intervention study. The number of patients with severe ACS was also lower in the intervention study. Among patients who were lucid on admission, fewer patients in the intervention study were acutely confused for more than one week (Table 7) (16/57, 28% in Paper II compared with 6 /66 , 9% in the intervention study pcO.Ol, chi-square analysis).

TABLE 17. POSTOPERATIVE COMPLICATIONS IN THE CONTROL STUDY AND IN THE INTERVENTION STUDY.

Control Intervention(N= 111) (N= 103)N % N % P

Urinary incontinence 26 23.4 15 14.6 N.S.Urinary infections 26 23.4 33 32.0 N.S.Urinary retention 21 18.9 9 8.7 <0.05Heart failure 8 7.2 9 8.7 N.S.Stroke 5 4.5 2 1.9 N.S.*Pneumonia 4 3.6 8 7.8 N.S.*Decubital ulcers 14 12.6 4 3.9 <0.05Feeding problems 8 7.2 5 4.9 N.S.*Severe falls 6 5.4 0 0 <0.05

*Yates’ corrected chi-square test.No differences remained significant after the Bonferroni adjustm ent for probabilities.

Fewer patients had urinary retention, decubital ulcers and severe falls in the intervention study. Five of the severe falls with new fractures and other complications in the control study occured among patients with postoperative ACS who also had the longest orthopaedic ward stays.

4 2

D ISC U SSIO N

ACS FREQUENCYThe ACS frequency in the paper including a prospective representative sample of elderly patients with femoral neck fractures was 61% (Paper I). In the corresponding sample of stroke patients of all ages the ACS frequency was 48% (Paper III). The patients were observed during a period, long enough to detect changes in their mental state. Their prefracture and prestroke mental state was also assessed by m eans of interviews with relatives or caregivers. The in terra ter reliability was controlled as described under methods. These procedures support the reliability of the ACS diagnosis, and may also explain the high percentage of ACS found in this study.The criteria for entry to the stroke unit (Paper III and IV) exclude patients with stroke causing only cognitive disturbances without focal neurological deficits. Therefore, our ACS frequency data early after the incident of stroke probably indicate the minimum frequency rate. The DSM-III-R ACS (delirium) criteria may occasionally be difficult to apply to patients with acute stroke because of a fluctuating ACS course and various neuropsychological symptoms from the stroke. In the p resen t studies, careful m onitoring with repeated cognitive testing and frequent patient observations have probably reduced the ACS underdiagnoses to a minimum. The risk for overdiagnoses were probably reduced by following the tim e-course of the p a tien t's cognitive function b u t diagnostic problem s rem ained in a few patients with aphasia, dementia and psychosis.Despite the selection of patients without signs of mental confusion or dementia on admission and the use of short operative procedures in Paper II, 35% (20/57) developed confusion during the first postoperative week. In ano ther five p a tien ts (9%), confusion appeared before and persisted after surgery.In the retrospective studies of Paper V, a lower ACS frequency was found than in the prospective studies on the corresponding groups of patients. The results of Paper V showed that ACS was unsatisfactorily diagnosed and poorly documented in the case note m aterial when compared with the results of the clinical studies. A retrospective approach leads to an underestimation of the ACS frequency as shown

4 3

in the comparisons between the clinical studies and the case note studies. However, when comparing the retrospective case note ACS frequency with the case note analysis of the clinical study samples, a similar ACS frequency can be found. This supports the fact tha t the high ACS frequency is a general problem in the care of elderly hip- fracture patients. The ACS frequency in the various papers is not essentially different from tha t reported in the literature, summarized in Tables 3, 4 and on page 16 regarding stroke patients.The most im portant results of the intervention, Paper VI, were the lower ACS frequency (47.6% compared with 61.3%, p<0.05), the shorter duration and the decreased severity of postoperative ACS. The intervention aimed mainly a t protecting the cerebral oxidative m etabolism , which is one of the m ain hypotheses about the pathophysiology of ACS. The second hypothesis, that ACS is a reaction to stress mediated by hypercortisolism, was not included in the prevention of postoperative ACS, bu t the treatm ent of complications associated with postoperative ACS has probably reduced the stress for the patien ts and thereby reduced the duration and severity of postoperative ACS.

PREDISPOSING ACS FACTORSThe independent significant ACS predictor in Paper I was old age. Dem entia, regular use of drugs with anticholinergic effects, depression and previous stroke were on the verge of being significant. In Paper II, the conclusion was th a t anticholinergic m edication and a history of depression were predom inant risk factors for the development of postoperative confusion and more important than the anesthetic technique in this respect.In the stroke patients of Paper III the degree of extremity paresis and old age were independent ACS predictors. Left-sided brain lesions, previous ACS and treatm ent with drugs with anticholinergic effects came close to being significant ACS predictors. In Paper IV, ACS in pa tien ts with supraten torial stroke was independently associated w ith high cortisol levels after a dexam ethasone suppression test, extensive motor impairm ent and left-sided brain lesion. Old age was on the verge of being a significant independent ACS predictor.In Paper VI dementia and male sex were significant ACS predictors.

4 4

A high proportion of the patients with femoral neck fractures in Paper I (35.1%) were regu larly trea ted w ith d rugs having anticholinergic effects, which have also been shown to be an ACS risk factor in Papers II and III and in the literature (Itil 1966, Tune 1981, Mondimore 1983, Miller 1988, T hienhaus 1990). Also in the retrospective studies of Paper IV trea tm en t with drugs having anticholinergic effects was closely associated with ACS (data not shown). This is perhaps the most important ACS predictor tha t could be prevented by a more restrictive use of drugs with anticholinergic effects in the elderly. Unfortunately many old vulnerable patients are treated with this type of drugs; 35% of the patients with femoral neck fractures in Paper I and according to the literature up to 60% of nursing home residents (Blazer 1983).However, the prediction models of the different Papers explain only a minor part of the ACS frequency. All patients with femoral neck fractures are exposed to traum a, pain, sudden relocation, anesthesia and surgery. Stroke patients are exposed to sudden disablement, sudden relocation and several medical complications. All these events, often unavoidable, bring about stress, especially for the old and vulnerable patient and are probably the main reasons for the high ACS frequency (Coccaro 1984, Arnetz 1985). The m ost vulnerable pa tien ts will probably develop ACS irrespective of treatm ent. In these cases it is im portant to treat and prevent the consequences of ACS. In many cases ACS in itself seems to place the patient in a vicious circle tha t may prolong his ACS. This may partly be due to the d isturbed sleep-wake cycle and the increased psychomotor activity which often results in total exhaustion.

ANESTHESIA AND ACSThere was no difference in the frequency of postoperative mental confusion or mortality between patients given epidural analgesia and patients given general anesthesia. Our results are in agreement with m ost of the studies summarized in Table 5. Only few authors have studied postoperative ACS though. However, in a study by Chung (Chung 1987), the retention of the m ental function after spinal anesthesia in patients operated on for transurethral resection of the prostate or undergoing pelvic floor repair was better than th a t of patients given general anesthesia. In Paper II, confusion following

4 5

general anesthesia correlated with a significant reduction in arterial oxygen tension values from the preoperative to the early postoperative period. In contrast to Hole (Hole 1980), we found a high frequency of postoperative confusion in patients who had been operated on under epidural analgesia. Patients with confusion in the epidural group had no postoperative reduction in arterial oxygen tension. Thus, hypoxemia was not the only cause of postoperative ACS.Spinal an esth esia was associated w ith a significantly higher intraoperative decrease in arterial blood pressure than halothane or narcotic anesthesia . This is consisten t w ith previous studies comparing regional and general anesthesia, in comparable patient groups (McKenzie 1980, Davis 1981). In Paper II we found no correlation between the degree or duration of hypotension and the frequency of postoperative confusion. However in Paper I and in Paper VI there was a close connection between peroperative hypotension and postoperative ACS. In Paper I 18/21 (86%) and in Paper VI 9 /1 0 (90%) of the patients with peroperative hypotension to 80 mmHg or lower developed postoperative ACS.

ACS AND CLINICAL OUTCOMEIn Paper I the consequences of ACS were prolonged ward-stay at the orthopedic departm ent, a greater need for long-term care on discharge, and poor walking ability on discharge and six m onths after surgery. In addition the confused patients had more complications, such as urinary problems, feeding problems and decubital ulcers, as compared with the nonconfused patients.In Paper II pa tien ts developing postoperative confusion had significantly more postoperative com plications and the to tal hospitalization time during the first year after the fracture was four times longer for patients who developed postoperative ACS.In Paper III few stroke patien ts with prolonged ACS could be discharged to their homes. Furthermore, as in other groups of elderly patients, ACS in stroke patients has a profound influence on the length of hospital stay and mortality (Table 13 and 14). In Paper IV, as shown in these tables, the ward-stay for patients with ACS was also prolonged.

4 6

In the intervention study (Paper VI), the shorter ward stay was considered to be the result of the lower frequency, shorter duration and decreased severity of postoperative ACS and its complications. In the intervention study, patients with and without ACS had almost the same length of orthopaedic ward stay and the same mortality rate both during the orthopaedic ward stay and at 6 months.There is a full support in the literature for the opinion th a t the patien t's mental state is perhaps the most im portant predictor for the clinical outcome of the patient. This opinion has taken into consideration th a t ACS causes nursing problems, results in medical complications and prolonged hospital stay and is associated with an increased mortality rate (Carroll 1969, Glass 1977, Baker 1978, Williams 1979, Rabins 1982, Keene 1982, W eddington 1982, Nickens 1983, Williams 1985b, Arm strong-Esther 1986, Bergman 1986, Fields 1986, Maguire 1986, Cummings 1988, Thomas 1988, Dickelmann 1989, Rogers 1989, B rännström 1989, Binder 1990, Francis 1990, Rockwood 1990, Brännström 1991).

ACS DIAGNOSIS AND DOCUMENTATIONThe underdiagnoses and poor documentation of ACS, found in Paper V, and the poor documentation of other cognitive disturbances in the elderly, is consistent with several other studies (Cavanaugh 1983, M cCartney 1983, Perez 1984, Waxman 1984, M cCartney 1985, Palmateer 1985, Trzepacz 1985, Rubin 1987, Barclay 1988, Lyness 1990).ACS has been reported to be a common, im portant symptom, and sometimes the only symptom of an organic disease in the elderly (Hodkinson 1976, Dixon 1984, Cobden 1984, Bayer 1986, Wroblevski 1986, Black 1987, Fox 1988). The cost and burden of undiagnosed and untreated organically induced mental disturbances is probably very high (Horvath 1986). Therefore, assessm ent of cognitive ability in elderly patients is necessary; if the patient is cognitively d isturbed , the cause of the d istu rbance m ust be established. It is essential to distinguish between ACS and dementia, and, when in doubt, the patien t should be regarded as acutely confused and assessed for the etiology (Lipowski 1989). If ACS is neglected or misdiagnosed as dementia, the following assessm ent and treatm ent may be inadequate. The differences between the case

4 7

notes and the clinical studies of the two com parisons (Paper V; Tables II and III) indicated tha t the risk for poor assessm ent and docum entation is m uch greater for patien ts who are cognitively d isturbed already on admission. The big difference between the clinical and case note frequency of ACS in comparison I, including demented patients, is considered to be due to the fact th a t neither the physicians nor the nurses, in m any cases, made any clear distinction between ACS and dementia. Furthermore, non-demented patien ts, who were acutely confused on adm ission, were often considered to be dem ented, and there were no docum ented attem pts to assess or treat these patien ts ' cognitive disturbances. This indicates th a t the physicians and nurses were less aware of m ental changes in patients who they regarded as demented. The retrospective Umeå case note study supports the contention th a t the on-going clinical studies on ACS did not influence the assessm ent and docum entary routines. Less than half of the preoperatively acutely confused patients were demented and for many of them, both dem ented and non-dem ented, the causes of their ACS could be prevented or treated (Paper I). Medical assessm ent and care are often limited in patients with cognitive im pairm ent and diagnosed dem entia (Palmer 1983, Volicer 1986a, Volicer 1986b, Fleishm an 1987). It is therefore extremely im portant th a t patien ts are not diagnosed as dem ented w ithout proper assessm ent. It is also im portant to diagnose ACS in demented patients, as it is often a symptom of a complicating condition.The Piteå retrospective control indicated th a t the poor assessm ent and documentation is probably a widespread problem in the care of elderly hip-fracture patients. It has also been reported th a t other groups of physicians are unaware of the patients' mental function and its implication for the assessm ent and treatm ent (Rubin 1987). These results stress the fact tha t assessm ent and documentation of ACS has to be improved in the care of elderly patients (Brady 1987) The poor assessm ent and documentation of ACS is a threat to the patients, as a correct ACS diagnosis is a prerequisite for further assessm ent of its underlying causes and the consequent necessary medical and nursing care.

4 8

INTERVENTIONAs early as in 1955, Bedford suggested th a t cerebral circulatory failure, including anoxia, anaemia, deficiency in nutrients and stroke were im portant etiological m echanisms for postoperative confusion (Bedford 1955). In Paper VI, despite oxygen treatm ent, severe hypoxemia occurred in some patients and contributed to their ACS. In Paper I there was a close relationship between peroperative hypotension and postoperative confusion. The num ber of patients with severe peroperative hypotension was lower in the intervention study, b u t hypotension rem ained an im portan t risk factor for postoperative confusion in the intervention study. These results are thought to support the hypothesis tha t a reduction in the cerebral oxidative metabolism is an important ACS cause (Blass 1979, Gibson 1981, Blass 1983, Hirsh 1984, Lipowski 1987).In the control study (Paper I), the only significant ACS predicting variable was old age. It should however be observed that 16/17 (94%) patients with dementia, 15/17 (88%) with depression, 15/18 (83%) with previous stroke and 33 /3 9 (85%) on regular treatm ent with drugs with anticholinergic effects developed ACS even though these variables did not reach statistical significance in the logistic regression model. What these variables have in common is a reduced metabolic capacity with decreased transm itter levels in the brain (Blazer 1982, Gottfries 1983, Hardy 1985, Gottfries 1987).In the intervention study (Paper VI), the ACS predicting variables were dementia and male sex. Patients with dementia have a low ACS threshold, so postoperative ACS is probably difficult to prevent. It was also im possible to adm inister oxygen therapy to some of the dem ented patien ts. Nevertheless, it is im portant to trea t these patients carefully and to realise tha t ACS in demented patients might be a symptom of a complication or condition which can be treated. There is a certain num ber of clinically relevant differences between men and women that might explain the higher ACS frequency among the men of the intervention study.Despite th a t depression did not reach significance in the logistic regression models; the great majority of patients with depression in both studies developed ACS ( 15/17 (88%) in the control study and 8 /11 (73%) in the intervention study). The serotoninergic and the noradrenergic transm itter systems are the most affected in the case

49

of depression (Blazer 1982). These systems are reported to be less sensitive to hypoxia than the cholinergic system (Gibson 1981, Hirsch 1984). The intervention program does not seem to protect p a tien ts w ith depression from ACS which m ight indicate an alternative ACS mechanism. Several studies (Kraal 1962, Kraal 1975, Murphy 1981, Carpenter 1982, Kosics 1985, Wolkowitz 1990) have shown th a t depressed patients have elevated cortisol levels and might thus explain why they are more likely to develop ACS. This ACS mechanism is supported in Paper IV. Hypercortisolism in depression is also reported to be connected with agitation, delusions, disturbed sleep, psychomotor alterations, cognitive impairment and left-sided brain lesions (Lipsey 1985, Kocsis 1985, Miller 1987, Brown 1988, Bolla-Wilson 1989).The changes in the ACS prediction model (Paper VI) are considered to be a result of the efforts made to improve the brain oxygenation. Old age, stroke, and treatm ent with drugs with anticholinergic effects have in common a decreased cholinergic activity in the brain. It has been suggested th a t decreased cholinergic activity might be the most im portant transm itter disturbance in ACS (ltd 1966, Blass 1979, Tune 1981, Mondimore 1983, Miller 1988, Thienhaus 1990). The cholinergic system has also been reported to be particularly sensitive to hypoxia (Blass 1979, Gibson 1981, Hirsch 1984). The intervention program might have protected some patients with such risk factors from ACS.The assessm en t of pa tien ts with postoperative ACS, and the treatm ent of associated complications have also reduced the duration and severity of their ACS, thereby preventing or inhibiting cerebral m etabolic d is tu rb an ces m ediated a t least partly by s tress- hypercortisolism.From a methodological point of view, a randomized experimental study would have been preferable to the use of a historic control. However, this was not possible due to changes in the anesthesiologic treatm ent routines initiated by the earlier studies. Ethical aspects supported the decision not to deny certain patients oxygen treatm ent and a more active prevention and treatm ent of severe hypotension. Because of the method used, the results of the intervention should be interpreted with some caution. However, during both studies the routines from admission to discharge were carefully observed and we

5 0

have found no other differences between the study periods than those described in the intervention program . It should also be observed th a t the patients of the two studies are comparable as for m ost of the relevant background variables. The only prefracture variable, with an ACS predicting value, tha t differed between the two study sam ples was cerebrovascular disorders, which were in fact more common in the intervention sample. With the Bonferroni correction of probabilities there were no differences in background variables between the two studies. There were no differences in the psycho-pharmacological treatm ent of patients with ACS or in the organization of the nursing care between the groups. During both studies, the patients were observed and tested by the same methods and by the same physicians a t the same intervals. The caregivers were also interviewed about the patients a t the same intervals and to the same extent in both studies.The lower frequency and severity and the shorter duration of ACS in the intervention study are considered to be the resu lts of the geria tric-anesthesio logic program . However, the frequency of postoperative ACS is still very high. All patients with femoral neck fractures are exposed to traum a, pain, sudden relocation, anesthesia and surgery. All these unavoidable incidents bring about stress, especially for the old and vulnerable hip-fracture patients (Arnetz 1985) and are probably the main reasons for the high ACS frequency. As in Paper IV, where hypercortisolism was an independent ACS- predictor in stroke patients, several studies have reported prolonged hypercortisolism in hip-fracture patients (Frayn 1983, Barton 1987, Roberts 1990). Unfortunately no attention has been paid to the patien ts ' cognitive function in relation to the hypercortisolism in these studies. However, it is probably not a daring assum ption tha t also in hip-fracture patients ACS is associated with hypercortisolism bu t it remains to be established.

ETIOLOGY AND PATHOGENESISThere are two m ajor hypotheses about the pathogenesis and pathophysiology of ACS (Lipowski 1987). The first hypothesis suggests tha t a reduction in cerebral oxidative metabolism and the consequent reduction in neurotransm itters cause ACS (Blass 1979, Blass 1983). This hypothesis is supported by studies showing tha t

51

the cerebral acetylcholine synthesis is especially sensitive to hypoxia (Gibson 1981, Hirsch 1984). A close link between ACS and impaired cholinergic activity has been reported by several authors (Itil 1966, Tune 1981, Mondimore 1983, Miller 1988, Thienhaus 1990). The second hypothesis suggests that ACS is a reaction to various types of stress mediated by elevated plasm a cortisol levels influencing the neuron function (Kraal 1962, Kraal 1975, Carpenter 1982, McEwen1987, McEwen 1988). High cortisol levels are closely related to postoperative ACS (McIntosh 1985). Hypercortisolism probably disturbs cerebral metabolism, especially for the cholinergic neurons in hippocam pus (Paper IV). Both of these m echanism s may be involved in the ACS pathogenesis in stroke patients and in patients with femoral neck fractures.A simplified model of the pathophysiology of ACS is presented in Figure 1.ACS has been associated with cerebrovascular lesions affecting a large variety of specific locations, such as thalam us, midbrain, diencephalic and posterior-inferior brain areas, the right frontostriatal region, the temporo-occipital cortex, the left posterior cerebral artery territory and the hippocam pus (Horenstein 1967, Medina 1974, Medina 1977, Mesulam 1976, Mullally 1982, de Reuck 1982, Levine 1982, Schmidley 1984, Graff-Radford 1984, Santam aria 1984, Price 1985, Balter 1986, Dunne 1986, Mori 1987, Bogousslavsky 1988, Devinsky1988, Garcia-Albea 1989). A common denominator for many of these lesion sites is th a t the neocortex hemisphere is disconnected from the limbic structures. It seems likely th a t acute confusion after stroke is not the result of one or several specific brain lesions bu t ra ther a consequence of a disruption of an integrative system. This disruption may be structural or functional in its nature and may involve neuroendocrine systems. The fact tha t confusion often clears after the first few days after stroke onset (Paper III) points to functional ra ther than structu ral damage as the major cause of confusion.Stroke causes ischem ia and disturbed oxygenation in the brain. Hypoxia is an im portant mechanism for brain damage (Siesjö 1981, Gibson 1988a). Hippocampus has been reported to be especially sensitive to hypoxia (Gibson 1988b). Hypoxia is found in a great proportion of patients with hip fractures (Bedford 1955, Katz 1972,

52

M artin 1977, Phillips 1977, Paper II and VI). A disturbance in cerebral oxidative metabolism is an im portant ACS m echanism as suggested as early as in 1955 by Bedford and by Engel in 1959. In animal models hippocampal ischemia produced necrosis in the CAI pyramid cells and resulted in memory deficits suggested to be relevant to hum an memory deficits caused by mild ischemic brain damage in clinical situations (Auer 1989).Hypercortisolism is common early after stroke (Feibel 1977, Olsson 1989, Olsson 1990, Korsic 1990, Paper IV) and prolonged and persisten t hypercortisolism has been reported for femoral neck fracture patients by several au thors (Frayn 1983, Barton 1987, Roberts 1990). It is clear from animal models tha t hypercortisolism d istu rbs neuron function, causes neuronal death and d istu rbs cognitive function (Micco 1980, Sapolsky 1986b, de Kloet 1987, Armanini 1990, Woolley 1990). It has also been shown in animal models th a t the toxic effect of cortisol on hippocam pal neurons increases with aging (Kerr 1989). Sapolsky suggested th a t cortisol endangers h ippocam pal neu ro n s by im pairing th e ir energy m etabolism (Sapolsky 1986a, Horner 1988) and thereby also potentiating ischemic injury to neurons (Sapolsky 1985). He also showed tha t the ischemic brain damage was worsened when cortisol levels were m anipulated after the ischemic insult (Sapolsky 1985). Glucocorticoids has been reported to cause hippocampal damage in m onkeys and Sapolsky therefore suggests th a t "susta ined hypercortisolism (whether due to stress, C ushing 's syndrome or exogenous administration) might damage the hum an hippocampus" (Sapolsky 1990b). A close connection between hypercortisolism and disturbances in cognitive function has also been described in man; both in patients with Cushing's syndrome (Whelan 1980, Starkm an 1981, Starkm an 1986), in healthy volunteers (Wolkowitz 1990) and in medical trea tm en t with corticosteroids (Reckart 1990). In patients with chronic cognitive disturbances, such as Alzheimer's d isease , a decreased su p p ressab ility of the HPA-axis by dexamethasone has been demonstrated (Balldin 1983, Charles 1986, Abou-Saleh 1987, Shrim ankar 1989, Martignoni 1990). The nonsuppression in Alzheimer's disease has been reported to be linked to hippocampal atrophy (de Leon 1988).

5 3

The ACS can probably place the patient in a vicious circle, with prolonged ACS. The ACS is in itself stressing for the patien t (MacKenzie 1980) which m ight m ain tain the hypercortisolism (Gaillard 1987, Caggiula 1989). Animals exposed to repeated stress has been reported to develop sustained hypercortisolism (Vernikos 1982). ACS is often associated with complications, e.g. urinary re te n tio n an d m yocardial in fa rc tio n w hich m ay prolong hypercortisolism. Other complications, e.g. pulmonary embolism and pneumonia may also cause or aggravate hypoxia and thereby prolong ACS. ACS causes caring problems (Brännström 1989) which often resu lts in treatm ent with drugs with anticholinergic effects (Paper V). Anticholinergic medication is an im portant risk factor for ACS according to the literature (Itil 1966, Tune 1981, Preskorn 1982, Mondimore 1983, Miller 1988, Thienhaus 1990). In experimental studies, on healthy volunteers, anticholinergic medication caused cognitive impairment and changes in the EEG’s of the same type as in the case of ACS (Roth 1959, Ostfeld 1960, Itil 1966, Pro 1977, Zisook 1986, Koponen 1989).In Papers I, II and III, treatm ent with drugs with anticholinergic effects was an independent ACS risk factor. According to the records, in Paper V, the patients' medication was never suggested to be the cause of ACS. In many cases the n u rses ' ACS documentation was followed by notes on extra medication, in many cases drugs with anticholinergic effects. No diagnostic reasoning or indications for the drug treatm ent were documented.Patients with ACS often exhibited postoperative nutritional problems (Brännströn 1990, Paper I) which might also place the patient in a vicious circle (Goodwin 1983). Nutrition deficiency of various kinds has been reported to cause and prolong ACS (Leeder 1981, Day 1988). M alnutrition is common in hip-fracture patients and also associated with prolonged w ard-stay and higher m ortality (Older 1980, Bastow 1983, Cooper 1987).The risk of developing decubital ulcers among patients with cognitive dysfunction, e.g. ACS, has been reported to increase (Paper I, Ek 1982, Nicholson 1988). During ACS the patient does probably not experience the development of ulcer an d /o r can not protect himself. During the intervention, the lower frequency of decubital ulcers was probably the result of the lower frequency and duration of ACS.

5 4

There has been some discussion as to when confusional state should be regarded as dementia. Many of the ACS causes might of course lead to permanent brain damage if they are not diagnosed and rapidly treated (Byrne 1987). The prolonged ACS could also, through disturbed cerebral oxidative metabolism and hypercortisolism, cause progressive irreversible brain damage and dem entia. In anim al models it has been shown th a t glucocorticoid hypersecretion, hippocam pal neuron death, and cognitive im pairm ents form a com plex degenerative cascade of aging (Meaney 88). The mechanisms of the development of dementia after stroke have to be elucidated in research in the future (Blank 1984, Lipsey 1985, Agarwaal 1987, Tatemichi 1990).The results of this study support the two main hypotheses of the ACS pathophysiology. There is substantial evidence, especially in animal models, tha t the two ACS-mechanisms interact closely bu t tha t their importance probably varies in different individuals and situations. These possible pathways are of course not the whole scenario of the ACS pathophysiology. Other mechanisms and their interaction with the cortisol-axis, the acetylcholine synthesis and neuron function m ust be further elucidated.The Reticular Activating System (RAS) and its noradrenergic system are probably involved to some extent. In Korsakoff s psychosis a close connection between depletion of the noradrenergic system and cognitive activation has been dem onstrated (Mair 1983). The d istu rbed sleep-wake-cycle m ight indicate involvement of the serotoninergic system. In animal models the serotoninergic system seems to activate the HPA-axis (Calogero 1990).ACS in patients treated for Parkinson's disease indicates th a t the dopaminergic system interacts with the cognitive function. Several neurotransm itters have been shown to be involved in the regulation of the HPA-axis (Tuomisto 1985).Interleukin (IL-1) is an im portant m ediator of generalized host responses to tissue injury or infection (Sipe 1990). Interleukin IL-1 has been reported to have a stimulatory effect on cortisol secretion (Whinter 1990, S undar 1990). Cortisol stim ulates PAI-1 which in h ib its th e fib rino lysis th e reb y in c reas in g the r isk of throm boem bolic com plications for the pa tien ts (Durum 1990). Several patients, both with femoral neck fractures and with stroke,

5 5

had thromboembolic complications associated with prolonged ACS. If these thromboembolic complications are mediated by interleukin- cortisol-PAI and if ACS can be mediated direct by interleukin or by interleukin-hypercortisolism rem ains to be confirmed. There is evidence th a t interleukin has a direct central nervous effect, shown for instance by the ability to induce slow-wave sleep in anim als (Chedid 1984).Excitatory am ino-acids, especially glutam ate perhaps (Sapolsky 1990a), which, like acetylcholine, is reported to be particularly sensitive to hypoxia are probably also involved in the ACS pathophysiology.The m olecular m echanism s, altering neuron function, such as electrolytes and free radicals are also likely to be involved either prim arily or secondarily. Even if we know very little of the complicated neuronal ACS mechanisms the present model appears to be a good basis for further basic research and intervention studies.

5 6

FIGURE 1. A SIMPLIFIED MODEL FOR THE PATHOPHYSIOLOGY OF ACS IN ELDERLY PATIENTS DISCUSSED IN THE PRESENT STUDY.

COMPLICATIONS(e.g. Urinaiy retention, Pulmonary embolism, Pneumonia, Myocardial infarction)

HYPER-CORTISOLISM

DISTU R BEDOXYGENATION

ANTICHOLINERGICMEDICATION

ENERGY/NUTRIENTD E P R IVATION

H IP-FRACTURE STROKESom atic and psychosocial stress

CARING PROBLEMS(e.g. Disturbing behaviour, Nutrition problem,Decubital ulcers, Sleep disturbances)

CEREBRAL METABOLISM (neurotransm itter m etabolism e.g. acetylcholine)

PROLONGED HYPERCORTISOLISM AND/OR REPEATED OR PROLONGED HYPOXIA CAUSES PERMANENT BRAIN DAMAGES AND DEMENTIA?

5 7

PREVENTION AND TREATMENTThe value of the two main hypotheses is tha t they can be applied to m ost of the suggested ACS causes (Table 7). The two performed interventions in ACS, Paper VI mainly with the protection of the cerebral oxidative metabolism, and the study by Williams (Williams 1985b) which could be said to lessen the psychosocial stress of the patient, reduced ACS.A prevention of factors compromising cerebral m etabolism or prevention of the neuronal toxic effect of high plasm a cortisol induced by stress might be two approaches that could prove effective in the prevention and treatm ent of ACS. High cortisol levels could prolong ACS by the direct or indirect neurotoxic effect of cortisol with the consequent disturbance of cognitive function. Also ACS in itse lf is a s tre ss factor for the pa tien t which m ight cause hypercortisolism, placing the patient in a vicious circle. The most common m istake in the m anagem ent of patients with ACS is to neglect environm ental and psychological in tervention (Beresin 1988). The prevention, tre a tm en t and care m u s t also be individualized (Wolanin 1981, Blank 1984, Foreman 1984, Bergman 1986, Foreman 1989b). Calm and careful nursing of the patient often reduces the s tress and in some cases improves the p a tien t's cognitive function. Etiologic ACS factors of organic and psychological origins may act through either of these mechanisms or through both.

5 8

GENERAL SUMMARY AND CONCLUSIONS

ACS was common in stroke patients and in hip fracture patients and was often associated with risk factors and complications that could be prevented or treated.

The main ACS risk factors in the patients with femoral neck fractures were old age, dementia, previous stroke, depression and regular treatm ent with drugs with anticholinergic effects.

In patients with femoral neck fractures, ACS -was not associated with the type of anesthesia but with peroperative blood pressure falls and early postoperative hypoxia and the conclusion is th a t it is the performance rather than the type of anesthesia th a t is im portant to the development of postoperative ACS.

In stroke patients hypercortisolism, extensive paresis, old age, leftsided brain lesion, previous ACS and regular treatm ent with drugs with anticholinergic effects were the m ost im portant ACS risk factors.

Most of the risk factors found in this study are consistent with the following two major ACS hypotheses:1. Reduced capacity or disturbance of the cholinergic system in the brain due to factors compromising the cerebral oxidative metabolism or caused direct by anticholinergically acting drugs.2. Elevated plasma cortisol level as a reaction to stress or as a direct result of factors disturbing the hypothalamic-pituitary-adrenal axis. Hypercortisolism might endanger the neuron metabolism thereby causing ACS.

ACS in elderly patien ts relates to various com plications, high mortality, prolonged ward stay and a great need for geriatric care.

As a correct ACS diagnosis is a prerequisite for the assessm ent of its causes and consequent trea tm en t, underd iagnoses and poor documentation of ACS is a threat to the patients.

5 9

The intervention in postoperative ACS confirmed the conclusions drawn in Papers I-V, since the result of the intervention was reduced frequency, duration and severity of postoperative ACS. Intervention reduced the orthopaedic ward stay for the patients, and thus, a great deal of suffering for elderly patients and high costs for society could be prevented if intervention was applied.

6 0

ACKNOWLEDGEMENTSThis study was carried out a t the Departments of Geriatric Medicine, In te rn a l Medicine, O rthopaedic Surgery and Anesthesiology, University of Umeå, and at the Departm ent of Geriatric Medicine, Karolinska Institute, Stockholm, Sweden.

I would like to thank everyone who has helped me throughout this study, and in particular:

Professor Gösta Bucht and Professor Bengt Winblad, my supervisors, for providing me with excellent facilities and generous support all along this study.

Mr Benny Brännström RNT, my friend and collaborator for fruitful cooperation and discussions for many years.

Dr Tommy Olsson, my friend and colleague, for his support and interest in this study.

Dr Diana Berggren for her enthusiastic and diligent cooperation in this study.

Professor Astrid Norberg for her constructive criticism and excellent improvement of my papers.

Associate Professor Kjell Asplund for his scientific excellence and for his support during many years.

Professor Sebastian Reiz for his constructive criticism and interest.

Professor Per-Olov Wester for his enthusiastic support and never ending interest in stroke patients.

The late Professor Lars-Ingvar Hansson for his great interest in hip- fracture patients and invaluable support in our studies on hip-fracture patients in the Orthopaedic department.

Dr Erik Hägg for guiding me in my clinical development and for cooperating in this study.

Dr Jon-Ingvar Ragnarsson, Dr Bengt Eriksson and Dr Jarl Sigaard for their fruitful cooperation.

Dr Sture Eriksson and Associate Professor Anders Baudin for their statistical advice.

Mrs Åsa Sundh and Mr Robert Elston for their skilful linguistic revision.

My colleagues a t the Geriatric departm ent and the Internal Medicine departm ent for granting me time for my research.

61

Mrs Karin Stenm ark and Mrs Karin Gladh for their excellent typing of m anuscripts, especially when data-viruses attacked my hard-disc.

Mrs K atharina Nilsson-Hallén for helping me with many practical things.

The staff of the geriatric, orthopaedic, anesthesiologic and internal medicine departm ents for their helpfulness and for their concern for elderly patient.

To all patients and relatives who made this study possible.

My wife Evy, and my children Emma, Martin, Ingvar and Karl and my parents G unnar and Freja Gustafson for their love, patience and support.

These studies have been supported by g ran ts from the Jo in t Committee of the Northern Health Region of Sweden, the 1987 Year's Foundation for Stroke Research, the Foundation of "Gamia Tjänarinnor", Loo and Hans O sterm an's foundation, Gun and Bertil Stohne’s foundation. King Gustaf V's 80th Anniversary Fund, Tore Nilsson's foundation, Trygg Hansa Insurance Company, Thuring's foundation, King Gustaf V's and Queen Victoria's Foundation, Petrus and Augusta Hedlund's foundation, the Foundations of the Medical Faculty, University of Umeå and the Swedish Medical Research Council (No 27X-07192, No 12X-5664).

62

REFERENCESAbeles M. Post-operative psychosis. Am J Psychiatry 1938;94:1187-1203.

Abou-Saleh MT, Spalding EM, Kellett Jm, Coppen A. Dexamethasone suppression test in dementia. Int J Geriatr Psychiatry 1987;2:59-65.

Abraham H, Gill B. Prediction of postoperative psychiatric complications. N Engl J Med 1961;265:1123-1128.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders, Third Edition (DSM-III), Washington, DC, APA, 1980.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders, Third Edition, Revised (DSM-III-R), Washington DC, APA, 1987;97-107.

Aquilonius SM, Hedstrand LI. The use of physiostigmine as an antidote in tricyclic antidepressant intoxication. Acta Anaesthesiol Scand 1978;22:40-45.

Arie T. Management of confused old people in general hospital wards. 1981. New Zealand Medical Journal, 1981;94:143-145.

Armanini MP, Hutchins C, Stein BA, Sapolsky RM. Glucocorticoid endangerment of hippocampal neurons is NMDA-receptor dependent. Brain Res 1990;532:7-12.

Armstrong-Esther CA, Browne KD. The influence of elderly patients' mental impairment on nurse-patient interaction. J Adv Nurs 1986; 11:379-387.

Ametz BB. Endocrine reactions during standardized surgical stress: The effect of age and methods of anaesthesia. Age Ageing 1985; 14: 96-101.

Auer R, Jensen N, Whishaw. Neurobehavioural deficit due to ischemic brain damage limited to half of the CAI sector of hippocampus. J Neurosci 1989;9:1641-1647.

Baker BR, Duckworth T, Wilkes E. Mental state and other prognostic factors in femoral fractures of the elderly. J Coll Gen Pract 1978;28:557-559.

Balldin J, Gottfries C-G, Karlsson I, Lindstedt G, Långström G, Wålinder J. Dexamethasone suppression test and serum prolactin in dementia disorders. Brit J Psychiat 1983;143:277-281.

Balter R, Fricchione G, Sterman AB. Clinical presentation of multi-infarct delirium. Psychosomatics 1986;27:461-462.

Barclay LL, Weiss EM, Mattis S, Bond O, Blass J. Unrecognized cognitive impairment in cardiac rehabilitation patients. J Am Geriatr Soc 1988;36:22-28.

Barton RN, Stoner HB. Why does a high cortisol concentration persist after fracture of the femoral neck in the elderly? Circulatory Shock 1987;21:292

Bastow MD, Rawlings J, Allison SP. Undemutrition, hypothermia, and injury in elderly women with fractured femur: an injury response to altered metabolism? Lancet 1983;1:143-145.

Bayer AJ, Chada JS, Farag RR, Pathy MSJ. Changing presentation of myocardial infarction with increasing old age. J Am Geriatr Soc 1986;34:263-266.

Bedford PD. Adverse cerebral effects of anaesthesia on old people. Lancet 1955;II:259- 263.

6 3

Bedford PD. General medical aspects of confusional states in elderly people. Br Med J 1959;2:185-188.

Beresin E. Delirium in the elderly. J Geriatr Psychiatr Neurol 1988;1:127-143.

Bergman R Nursing the aged with brain failure. J Advanced Nursing 1986; 11:361-367.

Bigler D, Adelhöj B, Petring OU, Pederson NO, Busch P, Kalhke P. Mental function and morbidity after acute hip surgery during spinal and general anaesthesia. Anaesthesia 1985;40:672-676.

Binder E, Robins L. Cognitive impairment and length of hospital stay in older persons. J Am Geriatr Soc 1990;38:759-766.

Black DA. Mental state and presentation of myocardial infarction in the elderly. Age Ageing 1987;16:125-127.

Blank K, Perry S. Relationship of psychological processes during delirium to outcome. Am J Psychiatry 1984;141:843-847.

Blass JP, Plum F. Metabolic encephalopaties in older adults, in Katzman R, Terry RD, (eds): Neurology of Aging. Philadelphia, FA Davis Co Publishers, p 189, 1983.

Blass J P, Gibson G E. Carbohydrates and acetylcholine synthesis: Implications for cognitive disorders. In: Davis K L and Berger P A, (eds): Brain Acetylcholine and Neuropsychiatrie Disease. Plenum Publishing Corporation, 1979, p. 215-236.

Blazer D, Federspiel C, Wayne R . The risk of anticholinergic toxicity in the elderly: A study of prescribing practices in two populations. J Gerontol 1983;38:31-35.

Blazer DG. Depression in Late Life. St. Louis, CV Mosby Co, 1982.

Bogousslavsky J, Ferrazini M, Regli F, Assai G, Tanabe H, Delaloye-Bischof A. Manic delirium and frontal-like syndrome with paramedian infarction of the right thalamus. J Neurol Neurosurg Psychiatry 1988;51:116-119.

Bolla-Wilson K, Robinson R, Starkstein S, Boston J, Price T. Lateralization of dementia of depression in stroke patients. Am J Psychiatry 1989;146:627-634.

Brady PF. Labeling of confusion in the elderly. J Gerontological Nursing 1987;6:29- 32.

Brown R, Stoll P, Stokes P, Frances A, Sweeney J, Kocsis J, Mann J. Adrenocortical hyperactivity in depression: Effects of agitation, delusions, melancholia, and other illness variables. Psychiatry Research 1988;23:167-178.

Brännström B, Gustafson Y, Norberg A, Winblad B. Problems of basic nursing care in acutely confused and non-confused hip-fracture patients. Scand J Caring Sci 1989;3: 27-34.

Brännström B, Gustafson Y, Norberg A, Winblad B. ADL performance and dependency on nursing care in patients with hip-fracture and acute confusion in a task allocation care system. Scandinavian Journal of Caring Sciences, 1991;5:3-12.

Bucht G, Brännström B, Gustafson Y. Acute confusional state, its prevalence and documentation in different care settings, in K Hasegawa and A Homma, (eds): Psychogeriatrics. Biomedical and Social Advances. Excerpta Medica, Current Clinical Practice Series, 1990;59: 269-274.

6 4

Burrows J, Briggs RS, Elkington AR. Cataract extraction and confusion in the elderly. J Clin Expr Gerontol 1985;7:51-70.

Byrne J. Reversible dementia. Int J Geriatr Psych1987;2:73-81.

Caggiula AR, Antelman SM, Aul E, Knopf S, Edwards DJ. Prior stress attenuates the analgesic response but sensitizes the corticosterone and cortical dopamine responses to stress 10 days later. Psychopharmacology 1989;99:233-237.

Calogero A, Gallucci W, Bernardini R, Saoutis C, Gold P, Chrousos G. Effect of cholinergic agonists and antagonists on rat hypothalamic corticotropin releasing hormone secretion in vitro. Neuroendocrinology 1988;47:303-308:

Calogero A, Bagdy G, Szemeredi K, Tartaglia M, Gold P, Chrousos P. Mechanisms of serotonin receptor agonist-induced activation of the hypothalamic-pit uitary- adrenal axis in the rat. Endocrinology 1990;126:1888-1894.

Carpenter WT, Gruen PH. Cortisol's effects on human mental functioning. J Clin Psychopharm 1982; 2:91-101.

Carroll DG. Prediction of performance of patients with hip fracture. Md State Med J 1969;18:69-75.

Ceder L, Ekelund L, Inerot S, Lindberg L, Odberg E, Sjölin C. Rehabilitation after hip fracture in the elderly. Acta Orthop Scand 1979;50:681-688.

Charles GA. Lefevre A. Mirei J. Rush AJ. Suppression of cortisol following dexamethasone in demented patients. Psychiatry Res 1986; 17: 173-182.

Chedid L, Bahr GM, Riveau G, Krueger JM. Specific absorbtion with monoclonal antibodies to muramyl dipeptide of the pyrogenic and somnogenic activities of rabbit monokine. Proc Natl Acad Sci USA 1984;81:5888-5891.

Chisholm SE, Deniston OL, Igrasan RM, Barbus AJ. Prevalence of Confusion in elderly hospitalized patients. Journal of gerontological Nursing, 1982;8:87-96.

Chung F, Meier R, Lautenschlager E, Carmichael FJ, Chung A. General or spinal anesthesia: Which is better in the elderly? Anesthesiology 1987;67:422-427.

Cobden I, Lendrum R, Venables CW, James OFW. Gallstones presenting as mental and physical debility in the elderly. Lancet 1984;1:1062-1064.

Coccaro E, Prudic J, Rothpearl A, Nürnberg G. Effect of hospital admission on DST results. Am J Psychiatry 1984; 141:982-985.

Cooper J. Does nutrition affect surgical outcome? J Am Geriatr Soc 1987;35:229-232.

Covert C. Anaesthesia for hip surgery in the elderly. Can J Anaesth 1989;36:311-319.

Daniel D, Rabin P. Disguises of delirium. Southern Medical Journal 1985;78:666-672.

Davis FM, Laurenson VG. Spinal anaesthesia or general anaesthesia for emergency hip surgery in elderly patients. Anaesth Intens Care 1981;9:352-358.

Davis FM, Woolner DF, Frampton C, Wilkinson A, Grant A, Harrison RT, Roberts MTS, Thadaka R. Prospective, muti-centre trial of mortality following general or spinal anaesthesia for fracture surgery in the elderly. Br J Anaesth 1987:59:1080- 1088.

6 5

Day J, Bayer A, McMahon M, Pathy M, Spragg B, Rowlands D. Thiamine status, vitamin supplements and postoperative confusion. Age Ageing 1988;17:29-34.

Devinsky O, Bear D, Volpe BT. Confusional states following posterior cerebral artery infarction. Arch Neurol 1988;45:160-163.

Dickelmann A, Haupts M, Kaliwoda A, Rembs E, Haan J, Zumtobel V. Akute postoperative Psychosyndrome. Eine prospektive Studie und multivariate Analyse von Risikofaktoren. Der Chirurg 1989;60:470-474.

Dixon PE, Hoey C, Cayley ACD. Tuberculous meningitis in the elderly. Postgrad Med J 1984;60:586-588.

Dobson AJ. An introduction to statistical modelling. London: Chapman and Hall, 1982.

Draper N, Smith H. Applied Regression Analysis. John Wiley and Sons, Inc., New York, 1966.

Dunne JW, Leedman PJ, Edis RH. Inobvious stroke: A cause of delirium and dementia. Aust N Z J Med 1986;16:771-778.

Durum SK, Oppenheim JJ, Neta R. Immunophysiologic role of interleukin 1. In Oppenheim JJ, Shevach EM, (eds): Immunphysiology. The role of cells and cytokines in immunity and inflammation. Oxford University Press1990 pp 210-225.

Ek A-C, Bohman G. A descriptive study on pressure sores: The prevalence of pressure sores and the characteristics of the patients. J Adv Nurs 1982;7:51-57.

Engel G, Romano J. Delirium, a syndrome of cerebral insufficiency. J Chronic Dis 1959;9:260-277.

Eriksson S, Asplund K, Hägg E, Lithner F, Strand T, Wester PO. Clinical profiles of cerebrovascular disorders in a population-based patient sample. J Chron Dis 1987;40:1025-1032.

Erkinjuntti T, Sulkava R, Wikström J, Autio L. Short portable mental status questionnaire as a screening test for dementia and delirium among the elderly. J Am Geriatr Soc 1987;35:412-416.

von Euler G, Fuxe K, Finnman U-B, Agnati LF. Acute and subchronic corticosterone treatment differentially modulates subcortical limbic nicotinic cholinergic receptors. Brain Research 1990;526:122-126.

Falch J, Ilebekk A, Slungaard V. Epidemiology of hip fractures in Norway. Acta Orthop Scand 1985;56:12-16.

Feibel JH, Hardy PM, Campbell RG, Goldstein MN, Joynt RJ. Prognostic value of stress response following stroke. JAMA 1977; 238:1374-1376.

Feinstein AR. Clinical epidemiology. The architecture of clinical research. W.B. Saunders Company, Philadelphia, 1985.

Fields S, MacKenzie R, Charlson M, Sax F. Cognitive impairment: Can it predict the course of hospitalized patients? J Am Geriatr Soc 1986;34:579-585.

Fleishman R, Rosin A, Tomer A, Schwartz R. Cognitive impairment and the quality of care in long-term care institutions. Compr Gerontol B 1987;1:18-23.

6 6

Flint FJ, Richards SM. Organic basis of confusional states in the elderly. Br Med J 1956;2:1537-1539.

Folstein MF, Folstein SE, McHugh PR. 'Minimental state': A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189- 198.

Foreman MD. Acute confusional states in the elderly: An algorithm. Dimensions of Critical Care Nursing 1984; 3:207-215.

Foreman MD. Confusion in the hospitalized elderly: incidence, onset, and associated factors. Res Nurs Health 1989a; 12:21-29.

Foreman MD, Gillies DA, Wagner D. Impaired cognition in the critically ill elderly patient: Clinical implications. Crit care Nurs Q, 1989b; 12:161-173.

Fox RA. Atypical presentation of geriatric infections. Geriatrics 1988;43:58-68.

Francis J, Martin D, Kapoor W. A prospective study of delirium in hospitalized elderly. JAMA 1990;263: 1097-1101.

Frayn KN, Stoner HB, Barton RN, Heath DF, Galasko CSB. Persistence of high plasma glucose, insulin and cortisol concentrations in elderly patients with proximal femoral fractures. Age Ageing 1983; 12:70-76.

Gaillard R-C, Al-Damluij S. Stress and the pituitary-adrenal axis. Baillieres Clin Endocrinol Metab 1987; 1:319-354.

Garcia-Albea E. Confusional state and cerebral infarcts. Postgrad Med J 1989:65:286- 290.

Gauthier JL, Hamelberg W. Hip fractures; Influence of anesthesia on the patients hospital course. Anesth Analg 1963;42:609-615.

Geschwind N. Disorders of attention: a frontier in neuropsychology. Phil Trans R Soc Lond 1982;298:173-185.

Ghoneim MM, Hinrichs J, O’Hara M, Mehta M, Pathak D, Kumar V, Clark C. Comparison of psychologic and cognitive functions after general or regional anesthesia. Anesthesiology 1988;69:507-515.

Gibson G, Mykytyn V. An in vitro model of anoxic-induced damage in mouse brain. NeurochemRes 1988a; 13:9-20.

Gibson G, Freeman G, Mykytyn V. Selective damage in striatum and hippocampus with in vitro anoxia. Neurochem Res 1988b; 13:329-335.

Gibson GE, Peterson C, Sansone J. Neurotransmitter and carbohydrate metabolism during aging and mild hypoxia. Neurobiol Aging 1981;2:165-172.

Gilad GM. The stress-induced response of the septo-hippocampal cholinergic system. A vectorial outcome of psychoneuroendocrinological interactions. Psychoneuroendocrinology 1987; 12:167-184.

Gillick MR, Serrel NA, Gillick LS. Adverse consequences of hospitalization in the elderly. Soc Sci Med 1982;16:1033-1038.

Glass RI, Mulvihill MN, Smith H, Peto R, Bucheister D, Stoll BJ. The 4 Score: An index for predicting a patient's non-medical hospital days. AJPH 1977;67:751-755.

6 7

Gold P, Goodwin F, Chrousos G. Clinical and biochemical manifestations of depression. Relation to the neurobiology of stress. Part I. N Engl J Med 1988a;319:348- 353.

Gold P, Goodwin F, Chrousos G. Clinical and biochemical manifestations of depression. Relation to the neurobiology of stress. Part II. N Engl J Med 1988b;319:413-420.

Goldstein M. Cerebrovascular epidemiology - Economic factors. J Neuroradiology 1983;10:160-164.

Goodwin JS, Goodwin JM, Garry P. Association between nutritional status and cognitive functioning in a healthy elderly population. JAMA 1983;249:2917-2921.

Gordon M, Preiksaitis H. Drugs and the aging brain. Geriatrics 1988;43:69-78.

Gottfries CG. Pharmacology of mental aging and dementia disorders. Clin Neuropharmacol 1987;10:313-329.

Gottfries CG, Adolfsson R, Aquilonius SM, Carlsson A, Eckemäs SÅ, Nordberg A, Oreland L, Svennerholm L, Wiberg Å, Winblad B. Biochemical changes in dementia disorders of Alzheimer type (AD/SDAT). Neurobiol Aging 1983;4:261-271.

Graff-Radford NR, Eslinger PJ, Damasio AR, Yamada T. Nonhemorrhagic infarction of the thalamus: behavioural, anatomic and physiologic correlates. Neurology 1984;34:14-23.

Green LT. Physiostigmine treatment of anticholinergic delirium. Anesth Analg 1971;50:222-226.

Gustafson L, Lindgren M, Westling B. Organic Brain Syndrome Scale (OBS-scale). A new rating scale for evaluation of confusional states and other organic brain syndromes. Abstract No. 128, Und International Congress on Psychogeriatric Medicine, Umeå, Sweden, 1985.

Hallberg IR, Norberg A, Eriksson S. Functional impairment and behaviour disturbances in psychogeriatric wards compared with controls. Int J Geriatr Psychiatry 1990;5:53-61.

Hardy J, Adolfsson R, Alafuzoff I, Bucht G, Marcusson J, Nyberg P, Pehrdahl E, Wester P, Winblad B. Transmitter deficits in Alzheimer's disease. Neurochem Int 1985;7:545- 563.

Hege-Scheuing G. Postoperatives Durchgangssyndrom und Delir. Anaesthesist 1989;38:443-451.

Hirsch JA, Gibson GE. Selective alteration of neurotransmitter release by low oxygen in vitro. Neurochem Res 1984;9:1039-1049.

Hodkinson HM. Mental impairment in the elderly. Journal of the Royal College of Physicians of London 1973; 7:305-317.

Hodkinson HM. Common symptoms of disease in the elderly. Oxford, Blackwell, 1976.

Hole A, Teijesen T, Breivik H. Epidural versus general anaesthesia for total hip arthroplasty in elderly patients. Acta Anaesthesiol Scand 1980;24:279-287.

Horenstein S, Chamberlain W, Conomy J. Infarction of the fusiform and calcarine regions with agitated delirium and hemianopsia. Trans Am Neurol Assoc 1967;92:85- 89.

6 8

Horner H, Sapolsky R. Glucocorticoids decrease glucose transport in cultured hippocampal neurons. Neuroscience 1988:14:921.

Horvath TB. The psychological presentation of somatic disorders. In Berger PA, Brodie KH, (eds): American handbook of psychiatry. Basic Books, Inc, Publishers, New York 1986;8:900-943.

Issa AM, Rowe W, Gauthier S, Meany MJ. Hypothalamic-pituitaiy-adrenal activity in aged, cognitively impaired and unimpaired rats. J Neurosci 1990;10:3247-3254.

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

Jensen SJ, Töndevold E, Sörensen PH. Costs of treatment of hip fractures. Acta Orthop Scand 1980a;51:289-296.

Jensen SJ, Töndevold E. A prognostic evaluation of the hospital resources required for the treatment of hip fractures. Acta Ortop Scand 1980b;51:515-522.

Jensen SJ. Incidence of hip fractures. Acta Orthop Scand 1980c;51:511-513.

Joels M, de Kloet ER. Effects of glucocorticoids and norepinephrine on the excitability in the hippocampus. Science 1989; 245:1502-1505.

Johnson J. Delirium in the elderly. Emergency Medicine Clinics of North America 1990;8:255-265.

Johnson J, Gottlieb G, Sullivan E, Wanich C, Kinosian B, Forciea MA, Sims R, Hogue C. Using DSM-III criteria to diagnose delirium in elderly general medical patients. J Gerontology 1990;45:113-119.

Jolley D. Acute confusional states in the elderly. In D Cloakley, (ed): Acute Geriatric Medicine, London, CroomHelm 1981 pp 175-189.

Jones MJT, Piggott SE, Vaughan RS, Bayer AJ, Newcombe RG, Twining TC, Pathy J, Rosen M. Cognitive and functional competence after anaesthesia in patients aged over 60: controlled trial of general or regional anaesthesia for elective hip or knee replacement. Br Med J 1990;300:1683-1687.

Judge F, Levy RJ, Segali JD. Mental confusion in the elderly. Emergency Medicine Clinics of North America 1987;5:797-805.

Judge TG. Drug treatment of brain failure. Age Ageing 1977;6:70-72.

Katz NM, Agle DP, DePalma RG, DeCross JJ. Delirium in surgical patients under intensive care. Arch Surg 1972;104:310-313.

Kay DWK. Epidemiological aspects of organic brain disease in the aged. In Gaits CM, (ed): Aging and the brain. New Tork, Plenum Press. 1972 pp 15-27.

Keene JS, Anderson CA. Hip fractures in the elderly. Discharge predictions with a functional rating scale. JAMA 1982;248:564-567.

Kerr S, Campbell L, Hao S-Y, Landfield P. Corticosteroid modulation of hippocampal potentials: increased effect with aging. Science 1989;245:1505-1509.

de Kloet ER, Reul JMHM. Feedback action and tonic influence of corticosteroids on brain function: a concept arising from the heterogeneity of brain receptor systems. Psychoneuroendocrinology 1987; 12:83-105.

6 9

Koponen H, Partanen J, Pääkkönen A, Mattila E, Riekkinen P. EEG spectral analysis in delirium. J Neurol Neurosurg Psychiatr 1989;52:980-985.

Korsic M, Brinar V, Plavsic V, Mihajlovic Dt Giljevic Z. Some aspects of adrenocortical stress response following stroke. Acta Med lug 1990;44:137-145.

Kosics J , Davis J, Katz M, Koslow S, Stokes P, Casper R, Redmond E. Depressive behaviour and hyperactive adrenocortical function. Am J Psychiatry 1985:142:1291- 1298.

Kraal VA. Stress and mental disorders of the senium. Medical Services Journal, Canada, 1962; 18:363-370.

Kraal VA. Confusional states: Description and management. In Howell JG, (ed): Modem perspectives in the psychiatry of old age. New York, Brunner/Mazel Inc. 1975 pp 356-362.

Lai H, Monsarrat AC. Effects of noise on high-afllnity choline uptake in the frontal cortex and hippocampus of the rat are blocked by intracerebroventricular injection of corticotropin-releasing factor antagonist. Brain Research 1990; 527:354-358.

Lamont C. Sampson S, Matthias R, Kane R. The outcome of hospitalization for acute illness in the elderly. J Am Geriatr Soc 1983;31:282-288.

Leeder SR, Gunasekera D, Gibson RM. Preventing acute confusional states in elderly patient. Australian Family Physician 1981;10:129-130.

De Leon MJ, McRae T, Tsai JR, George AE, Marcus DL, Freedman M, Wolf AP, McEwen B. Abnormal cortisol response in Alzheimer's disease linked to hippocampal atrophy. Lancet 1988;11:391-392.

Levine DN, Finklestein S. Delayed psychosis after right temporoparietal stroke or trauma: Relation to epilepsy. Neurology 1982;32:267-273.

Levkoff SE, Safran C, Cleary PD, Gallop J, Phillips RS. Identification of factors associated with the diagnosis of delirium in elderly hospitalized patients. J Am Geriatr Soc 1988;36:1099-1104

Lipowski ZJ. Differentiating delirium from dementia in the elderly. Clin Gerontologist 1982;1:3-10.

Lipowski ZJ. Delirium (Acute Confusional States). JAMA 1987;258:1789-1792.

Lipowski ZJ. Delirium in the elderly patient. N Engl J Med 1989;320:578-582.

Lipowski ZJ. Delirium: Acute confusional States. Oxford University Press, Inc, New York, Oxford 1990.

Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR. The dexamethasone suppression test and mood following stroke. Am J Psychiatry 1985; 142:318-323.

Liston EH. Delirium in the aged. Psychiatr Clin North Am 1982;5:49-66.

Luthje P. Cost analysis and a forecast of the fractures of the proximal femur in the province of the middle Finland. Ann Chir Gynaecol 1982;71:312-6.

Lyness J. Delirium: Masquerades and misdiagnosis in elderly inpatients. J Am Geriatr Soc 1990;38:1235-1238.

70

MacDonald AJD, Simpson A, Jenkins D. Delirium in the elderly: A review and a suggestion for a research programme. Int J Geriatr Psychiatry 1989;4: 311-319.

MacKenzie TB, Popkin MK. Stress response syndrome occuring after delirium. Am J Psychiatry 1980;137:1433-1435.

Magaziner J, Simonsick E, Kashner M, Hebel R, Kenzora J. Predictors of functional recovery one year following hospital discharge for hip fracture: A prospective study. J Gerontol 1990;45: 101-107.

Maguire P, Taylor I, Stout R. Elderly patients in acute medical wards: Factors predicting length of stay in hospital. Br Med J 1986;292:1251-1253.

Mair R, McEntee W: Korsakoffs Psychosis. Noradrenergic systems and cognitive impairment. Behavioural Brain Research 1983;9:1-32.

Martignoni E, Petraglia Ff Costa A, Bono G, Genazzani AR, Nappi G. Dementia of the Alzheimer type and the hypothalamus-pituitary-adrenocortical axis: changes in cerebrospinal fluid corticotropin releasing factor and plasma cortisol levels. Acta Neurol Scand 1990;81:452-456.

Martin VC. Hypoxaemia in elderly patients suffering from fractured neck of femur. Anaesthesia 1977;32:852-867.

McCartney JRt Palmateer L. Dementia and delirium: Detection in the general hospital. RI Med J 1983;66:361-363.

McCartney JR, Palmateer LM. Assessment of cognitive deficits in geriatric patients. A study of physicians behavior. J Am Geriatr Soc 1985;33:467-471.

McEwen B, Brinton R Glucocorticoid receptors and behaviour: Implication for the stress response. Adv Exp Med Biol 1988;245:35-45.

McEwen B. Glucocorticoid-biogenic amine interactions in relation to mood and behaviour. Biochemical Pharmacology 1987;36:1755-1763.

McIntosh TK, Bush HL, Yeston NS, Grasberger R, Palter M, Aun F, Egdahl RH. Beta- endorphin, cortisol and postoperative delirium: A preliminary report. Psychoneuroendocrinology 1985; 10:303-313.

McKenzie PJ, Wishart HY, Dewar KM, Gray I, Smith G. Comparison of the effects of spinal anaesthesia and general anaesthesia on postoperative oxygenation and perioperative mortality. Br J Anaesth 1980;52:49-53.

McKenzie PJ, Wishart HY, Smith G. Long-term outcome after repair of fractured neck of femur. B rJ Anaesth 1984;56:581-584.

McLaren AD, Stockwell MC, Reid VT. Anaesthetic techniques for surgical correction of fractured neck of femur. Anaesthesia 1978;33:10-14.

Meaney M, Aitken D, van Berkel C, Bhatnagar S, Sapolsky R. Effects of neonatal handling on age-related impairments associated with the hippocampus. Science 1988;239:766-768.

MFR, Spri. Hjärnblödning och hjärninfarkt (stroke) - diagnostik och behandling. Spri, Stockholm 1986; 103.

Medina JL, Rubino FA, Ross E. Agitated delirium caused by infarctions of the hippocampal formation and fusiform and lingual gyri. Neurology 1974;24:1181-1183.

71

Medina JL, Chokroverty S, Rubino FA. Syndrome of agitated delirium and visual impairment: a manifestation of medial temporo-occipital infarction. J Neurol, Neurosurg Psychiat 1977;40:861-864.

Mesulam M-M, Waxman SG, Geschwind N, Sabin TD. Acute confusional states with right middle cerebral infarction. J Neurol Neurosurg Psychiatry 1976;39:84-89.

Micco DJ, McEwen B. Glucocorticoids, the hippocampus, and behavior: Interactive relation between task activation and steroid hormone binding specificity. J Comp Physiol Psych 1980;94:624-633.

Millar HR. Psychiatric morbidity in elderly surgical patients. Br J Psychiat 1981;138:17-20.

Miller K, Nelson C. Does the dexamethasone suppression test relate to subtypes, factors, symptoms, or severity? Arch Gen Psychiatry 1987;44:769 -774.

Miller PS, Richardson JS, Jyu CA, Lemay JS, Hiscock M, Keegan DL. Association of low serum anticholinergic levels and cognitive impairment in elderly presurgical patients. Am J Psychiatry 1988;145:342-345.

Mondimore F, Damlouij N, Folstein M, Tune L. Post-ECT confusional states associated with elevated serum anticholinergic levels. Am J Psychiatry 1983;140:930-931.

Mori E, Yamadori A. Acute confusional state and acute agitated delirium. Arch Neurol 1987;44:1139-1143.

Morse R. Postoperative delirium: A syndrome of multiple causation. Psychosomatics 1969a; 11:164-168.

Morse R, Litin E. Postoperative delirium: A study of etiologic factors. Am J Psychiatry 1969b; 126:388-395.

Morse R, Litin E. The anatomy of a delirium. Am J Psychiatry 1971; 128:111-116.

Mullally W, Huff J, Ronthal M, Geschwind N. Frequency of acute confusional states with lesions of the right hemisphere. Ann Neurol 1982; 12:113.

Murphy MF, Davies KL. Biological perspectives in chronic pain, depression, and organic mental disorders. Clin North Am 1981;4:223-237.

Nicholson PW, Leeman AL, O'Neill CJA, Dobbs SM, Deshmukh AA, Denham MJ, Royston JP, Dobbs RJ. Pressure sores: Effect of Parkinson's disease and cognitive function on spontaneous movement in bed. Age Ageing 1988; 17:111-115.

Nickens HN. A review of factors affecting the occurrence and outcome of hip fractures, with special reference to psychosocial issues. J Am Ger Soc 1983;31:166-170.

Older MWJ, Edwards D, Dickerson JWT. A nutrient survey in elderly women with femoral neck fractures. Br J Surg 1980;76:884-886.

O lsson T, Åström M, Eriksson S, Forsell Å. Hypercortisolism revealed by dexamethasone suppression test in patients with acute ischemic stroke. Stroke 1989; 20:1685-1690.

Olsson T. Urinary free cortisol excretion shortly after ischemic stroke. J Int Med 1990;228:177-181.

72

Ostfeld AM, Machne X, Unna KR. The effects of atropine on the electroencephalogram and behaviour in man. J Pharmacol Exper Therap 1960; 128: 265-272.

Palmateer LM, McCartney JR. Do nurses know when patients have cognitive deficits? Journal of Gerontological Nursing, 1985; 11: 6-16.

Palmer MH. Alzheimer's disease and critical care. J Gerontol Nurs 1983;9:86-116.

Pérez EL, Silverman M: Delirium. The often overlooked diagnosis. Int'l J Psychiatry in Medicine 1984;14: 181-189.

Phillips G, Tomlin PJ. Arterial oxygen tensions in elderly and in injured elderly patients. Br JAnaesth 1977;49: 514.

Preskom S, Simpson S. Tricyclic-antidepressant-induced delirium and plasma drug concentration. Am J Psychiatry 1982;139:822-823.

Price BH, Mesulam M-M. Psychiatric manifestations of right hemisphere infarctions. J Nerv Ment Dis 1985;173:610-614.

Rabins P, Folstein M. Delirium and dementia: Diagnostic criteria and fatality rates. Brit JPsychiat 1982;140:149-153.

Reckart MD, Eisendrath SJ. Exogenous corticosteroid effects on mood and cognition: Case presentations. Int J Psychosomatics 1990;37:57-61.

de Reuck J, Sieben G, de Coster W, vander Eecken H. Dementia and confusional state in patients with cerebral infarcts. Eur Neurol 1982;21:94-97.

Reus V. Behavioural implications of hypothalamic-pituitary-adrenal dysfunction. In Halbreich U, (eds): Hormones and depression, Raven Press, New York. 1987 pp 385- 401.

Riis J, Lomholt B, Haxholdt O, Kehlet H, Valentin N, Danielsen U, Dyrberg V. Immediate and long-term mental recovery from general versus epidural anaesthesia in elderly patients. Acta Anaesth Scand 1983;27:44-49.

Roberts NA, Barton RN, Horan MA, White A. Adrenal function after upper femoral fractures in elderly people: Persistence of stim ulation and the roles of adrenocorticotrophic hormone and immobility. Age Ageing 1990,19:304-310.

Rockwood K. Acute confusion in elderly medical patients. J Am Geriatr Soc 1989;37:150-154.

Rockwood K. Delays in discharge of elderly patients. J Clin Epidemiol 1990:43:971- 975.

Rogers M, Liang M, Daltroy L, Eaton Holley, Peteet J, Wright E, Albert M. Delirium after elective orthopedic surgery: Risk Factors and Natural history. Int'l J Psychiatry in Medicine 1989;19:109-121.

Roth M. Some diagnostic and aetiological aspects of confusional states in the elderly. Geront Clin 1959;1:83-95.

Rubin SM, Glasser ML, Werckle MA. The examination of physicians’ awareness of dementing disorders. J Am Geriatr Soc 1987;35:1051-1058.

Sandercock P. The odds ratio: A useful tool in neurosciences. J Neurol Neurosurg Psychiatry 1989;52:817-820.

73

Santamaria J, Blesa R, Tolosa ES. Confusional syndrome in thalamic stroke. Neurology 1984;34:1618.

Sapolsky RM, Pusinelli WA. Glucocorticoids potentiate ischemic injury to neurons: Therapeutic implications. Science 1985; 229:1397-1400.

Sapolsky RM. Glucocorticoid toxicity in the hippocampus: Reversal by supplementation with brain fuels. J Neurosci 1986a;6:2240-2244.

Sapolsky RM, Krey LC, McEwen BS. The neuroendocrinology of stress and aging: the glucocorticoid cascade hypothesis. EndocrRev 1986b;7:284-301.

Sapolsky RM. Glucocorticoids, hippocampal damage and the glutaminergic synapse. In Coleman P, Higgins G, Phelbs C, (eds): Progress in Brain Research, Elsevier Science Publishers BV. 1990a;86:13-23.

Sapolsky RM, Uno H, Rebert C, Finch C. Hippocampal damage associated with prolonged glucocorticoid exposure in primates. J Neurosci 1990b; 10:2897-2902.

Savage G. Insanity following the use of anaesthetics in operations. Br Med J 1887; 2:1199-1200.

Schmidley JW, Messing RO. Agitated confusional states in patients with right hemisphere infarctions. Stroke 1984;15:883-885.

Seymour DG, Pringle R. Post-operative complications in the elderly surgical patient. Gerontology 1983;29:262-270.

Sheppeard H, Gleak DK, Ward DJ, O'Connor BT. A review of early mortality and morbidity in elderly patients following Chamley total hip replacement. Arch Ortop Traumat Suig 1980;97:243-248.

Shrimankar J, Soni SD, McMurray J. Dexamethasone suppression test in dementia and depression. Clinical and biological correlates. Br J Psychiatry 1989;154:372-377.

Siesjö B. Cell damage in the brain: A speculative synthesis. J Cerebral Blood Flow Metabolism. 1981;1:155-185.

Simon A, Cahan RB. The acute brain syndrome in geriatric patients. Psychiatr Res Rep 1963;16:8-21.

Simpson JEP, Glynn CJ, Cox AG, Folkard S. Comparative study of short-term recovery of mental efficiency after anaestesia. Br Med J 1976; 1:1560-1562.

Sipe JD. The acute-phase response. In Oppenheim JJ, Shevach EM, (eds): Immunphysiology. The role of cells and cytokines in immunity and inflammation. Oxford University Press 1990 pp 259-273.

Starkman MN, Schteingart DE. Neuropsychiatrie manifestations of patients with Cushing's syndrome: Relationship to cortisol and adrenocorticotropic hormone levels. Arch Intern Med 1981;141:215-219.

Starkman MN, Schteingart DE, Schork A. Cushing's syndrome after treatment: Changes in cortisol and ACTH levels, and amelioration of depressive syndrome. Psychiatry Research 1986;19:177-188.

Strand T, Asplund K, Eriksson S, Hägg E, Lithner F, Wester PO. A non-intensive stroke unit reduces functional disability and the need for long-term hospitalisation. Stroke 1985;16:29-34.

74

Summers W, Reich T. Delirium after cataract surgeiy: Review and two cases. Am J Psychiatry 1979;136:386-391.

Sundar SK, Cierpial MA, Kilts C, Ritchie JC, Weiss JM. Brain IL-l-induced immunosuppression occurs through activation of both pituitary-adrenal axis and sympathetic nervous system by corticotropin-releasing factor. Neuroscience 1990;10:3701-3706.

Tatemichi T. How acute brain failure becomes chronic: A view of the mechanisms of dementia related to stroke. Neurology 1990;40:1652-1659.

Thienhaus OJ, Allen A, Bennett JA, Chopra YM, Zemlan FP. Anticholinergic serum levels and cognitive performance. Eur Arch Psychiatry Clin Neurosci 1990;240:28-33.

Thomas RI, Cameron DJ, Fahs MC. A prospective study of delirium and prolonged hospital stay. Arch Gen Psychiatry 1988;45:937-940.

Trzepacz PT, Teague GB Lipowski ZJ. Delirium and other organic mental disorders in a general hospital. General Hospital Psychiatry 1985;7:101-106.

Tune LE, Holland A, Folstein MF, Namir FD, Gardner TJ, Coyle JT. Association of postoperative delirium with raised serum levels of anticholinergic drugs. Lancet 1981;11:651-653.

Tuomisto J, Männistö P. Neurotransmitter regulation of anterior pituitary hormones. Pharmacological Reviews 1985; 37:249-332.

Valentin N, Lomholt B, Jensen JS, Hejgaard N, Kreiner S. Spinal or general anaesthesia for surgery of the fractured hip? A prospective study of mortality in 578 patients. Br JAnaesth 1986;58:284-291.

Wallace WA. The increasing incidence of fractures of the proximal femur: An orthopaedic epidemic. Lancet 1983;11:1413-1414.

Varsamis J. Clinical management of delirium. Psychiatric Clinics of North America 1978;1:71-80.

Waxman HM, Camer EA. Physicians' recognition, diagnosis and treatment of mental disorders in elderly medical patients. Gerontologist 1984;24:593-597.

Weddington WW. The mortality of delirium: An underappreciated problem. Psychosomatics 1982; 23:1232-1235.

Wells CE. Dementia, pseudodementia, and dementia praecox. In FannWE et al., (eds): Phenomenology and treatment of schizofrenia. New York: Spectrum Publications, 1978.

Vemikos J, Dallman MF, Bonner C, Katzen A, Shinsako J. Pituitaiy-adrenal function in rats chronically exposed to cold. Endocrinology 1982; 110:413-420.

Whelan TB, Schteingart DE, Starkman MN. Neuropsychological deficits in Cushing's syndrome. J Nerv Ment Dis 1980; 168:753-757.

Whitaker JJ. Postoperative confusion in the elderly. Int J Geriatr Psych 1989;4:321- 326.

Whinter JSD, Gow KW, Perry YS, Greenberg AH. A stimulatory effect of interleukin-1 on adrenocortical cortisol secretion mediated by prostaglandins. Endocrinology 1990;127:1904-1909.

75

White IWC, Chappell WA. Anaesthesia for surgical correction of fractured femoral neck. A comparison of three techniques. Anaesthesia 1980;35:1107-1110.

Wikström I, Holmberg I, Stefansson T. Survival of female geriatric patients after hip fracture surgery. A comparison of five anaesthetic methods. Acta Anaesth Scand 1982;26:607-614.

Wilkinson L. SYSTAT. The system for statistics. Evanston, IL, SYSTAT Inc. 1987.

Wilkinson L. SYSTAT. The system for statistics. Evanston, IL: SYSTAT Inc, 1990.

Williams MA, Campbell EB, Raynor WB Jr, Musholt MA, Mlynarczyk SM, Crane LF. Predictors of acute confusional states in hospitalized elderly patients. Research in Nursing and Health 1985a;8:31-40.

Williams MA, Campbell EB, Raynor WB Jr, Mlynarczyk SM, Ward SE. Reducing acute confusional states in elderly patients with hip fractures. Research in Nursing and Health 1985b;8:329-337.

Williams MA, Holloway JR, Winn MC, Wolanin, MO, Lawler ML, Westwick CR, Chin MH. Nursing activities and acute confusional states in elderly hip-fractured patients. Nursing Res 1979;26:25-35.

Wolanin MO, Phillips LRF. Confusion, prevention and care. The CV Mosby Company, St. Louis, 1981.

Wolff HG, Curran D. Nature of delirium and allied states. The dysergastic reaction. Archives of neurology and psychiatry 1935;33:1175-1215.

Volicer L. Need for hospice approach to treatment of patients with advanced progressive dementia. J Am Geriatr Soc 1986a;34:655-658.

Volicer L, Rheaume Y, Brown J, Fabiszewski K, Brady R. Hospice approach to the treatment of patients with advanced dementia of the Alzheimer type. JAMA 1986b;256:2210-2213.

Wolkowitz OM, Reus VI, Weingartner H, Thompson K, Breier A, Doran A, Rubinow D, Pickar D. Cognitive effects of corticosteroids. Am J Psychiatry 1990; 147:1297-1303.

Woolley CS, Gould E, McEwen BS. Exposure to excess glucocorticoids alters dendritic morphology of adult hippocampal pyramidal neurons. Brain Research 1990:531:225- 231.

Wroblewski M, Mikulowski P, Steen B. Symptoms of myocardial infarction in old age. Age Ageing 1986;15:99-104.

Zetterberg C, Andersson GBJ. Fractures of the proximal end of the femur in Gothenburg, Sweden, 1940-1979. Acta Orthop Scand 1982;53:419-26.

Zisook S, Braff DL. Delirium: Recognition and management in the older patient. Geriatrics 1986;41:67-78.

7 6

Date post:	02-Jan-2017
Category:	Documents
Upload:	dangdan
View:	222 times
Download:	1 times

ACUTE CONFUSIONAL STATE (DELIRIUM)

Documents