UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Aspects of acute hospital admission in the elderly

de Rooij, S.E.J.A.

Publication date2006

Link to publication

Citation for published version (APA):de Rooij, S. E. J. A. (2006). Aspects of acute hospital admission in the elderly.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an opencontent license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, pleaselet the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the materialinaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letterto: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. Youwill be contacted as soon as possible.

Download date:14 Jul 2021

SUMMERY Y

SUMMARY Y

AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

OOverr 13% of the population of the Netherlands is older than 65 years. Acute hos-

pitalizationn is common among elderly and very elderly persons. Annually 16%

off these age groups is acutely admitted to hospital. Overall they represent 33% of all

hospitall admissions and account for 48% of the total number of admission days. Data

collectedd as part of the Dutch National Intensive Care Evaluation (NICE) registry (over

200 participating hospitals) show that 50 % of the patients are 65 years or older, and

10%% of the patients are 80 years or older.

Thiss graying of society not only counts for the Netherlands. Population worldwide

iss aging. In the United States the number of persons aged 85 years or older is likely to

groww from about 4 million in 2000 to 19 million by 2050. Consequently the number of

elderlyy patients admitted to hospital, including intensive care units (ICUs) is increasing as

well.. An increasing part of the elderly patients is admitted to hospital often at the end

off their lives. At present, the mean age for a Dutch man is 78 years, and for a Dutch female

811 years. After reaching this mean age, this high age is not really important anymore for

theirr hospital outcome, although health care providers often use high age as a factor in

thee decision making process. There are of course differences between elderly patients,

butt you wil l have to find them on other areas.

Thiss thesis describes the population of two different hospital departments often dealing with

acutee admissions in elderly patients and reports the short-term and long-term outcomes

off the hospital admission.

Chapterr 1 shows an overview of the aims of this thesis. The first part of the thesis including

Chapterss 2,354,5,and 6, reports the study on the characteristics of elderly patients acutely

admittedd to the department of Internal Medicine. The outcomes of the hospital admission and

thee short-term outcome after discharge are described. Attention has been paid especially

too the presence of delirium. Delirium is an acute neuropsychiatric syndrome, characterized

223 3 AspectsAspects of acute hospital admission in the elderly

S U M M E R Y Y

byy deranged consciousness, cognitive and attentional disturbances, and with a typical

fluctuatingfluctuating course. Delirium is common in elderly hospitalized patients and is associated

withh many serious short - and long-term consequences including increased length of stay,

increasedd morbidity and mortality, and higher hospital costs.

Thee risk factors of delirium, the subtypes of delirium, prognosis and measurement

instrumentss wil l be discussed in the next chapters.

Thee second part of this thesis includes Chapters 7,8,9 & 10. In these chapters the

characteristicss of very elderly ICU patients are described. Special attention was paid to

thee physical and cognitive outcomes of the ICU survivors.

Thee aim of the study described in Chapter 2 is to determine factors associated with a

prevalentt delirium among acutely admitted elderly patients to an internal medicine

ward.. In this prospective cohort study, nearly 30% of acutely admitted elderly patients

hadd delirium within 48 hrs after admission. An increased risk for delirium at admission

wass associated with cognitive and physical impairment, and elevated urea nitrogen level.

Noo increased risk for developing delirium was noticed for comorbidity or type or number of

medication.. These observations might contribute to an earlier identification and treatment

off delirium in acutely admitted elderly patients.

Chapterr 3 describes a prospective cohort study among consecutive acutely admitted

elderlyy patients aged 65 years or older at the Department of Internal Medicine to

investigatee risk factors of hospital mortality together with risk factors for mortality

withinn three months after hospital discharge.

Deliriumm is a common condition in this population of acutely hospitalized elderly

patientss (29.1%), see also Chapter 2. Our results indicate that the presence of delirium

inn elderly medical patients is even after adjustment, predictive for both hospital and

224 4 AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

shortt term mortality, increasing this risk by about two fold. Interestingly, concerning

hospitall mortality, this is in agreement with previous studies, however, we demonstrated

alsoo a higher risk in short-term mortality.

Inn addition, we found not only that delirious patients were at risk for short term

mortality,, but also patients that were functionally impaired at admission experienced

withh every declining point on the Activity of Daily Living scale an additional risk of

10%% for short term mortality after being discharged alive from hospital. We furthermore

developedd a simple predictive model based on all factors from the multivariate analysis

thatt can be used to predict which short term mortality risk can be calculated for elderly

patientss at hospital discharge. We opted to use as few variables as possible and to use

variabless and measurement instruments that were used in many previous studies. The

greaterr spread in mortality risks created with diagnosis, comorbidity, functional abilities and

thee presence or absence of delirium enhances the use of this tool as a clinical prediction

instrumentt by allowing the identification of groups at lower and higher post discharge

riskk for mortality. However, the findings of this study suggest the need for further research

onn the improvement of quality of care (after discharge) of acute admitted elderly patients,

andd its effect on post discharge mortality. Despite their important impact on outcome,

delirium,, functional dependence, psychosocial concerns, and other geriatric conditions

aree often underdiagnosed, poorly documented, and inadequately addressed during and

afterr acute hospital admission.

Thee early and systematic identification of delirium and functional decline, and thus

recognizingg high-risk elderly patients, is the first step in targeting appropriate interventions

too promote positive patient outcomes. In addition future studies wil l have to reveal the

mechanismss by which delirium and functional impairment are affecting mortality, and

whetherr modification of risk factors or strategies can decrease adverse outcomes during

andd after acute hospital admission.

225 5 AspectsAspects of acute hospital admission in the elderly

S U M M E R Y Y

Inn Chapter 4 we report the investigation and validation of the DRS-R-98 in a consecutive

seriess of hospitalized elderly patients in Internal medicine and surgical wards of a general

hospital.. For this purpose, the DRS-R-98 was translated into the Dutch language. A

secondd aim of the study was to assess whether the DRS-R-98 is capable of identifying

deliriumm subtypes within a cohort of elderly patients.

Thiss study shows that the Dutch version of the DRS-R-98 was able to differentiate

patientss with delirium from demented as well as from non-psychiatric patients. The

DRS-R-988 also proved to be a valid and reliable severity measure for delirium as

demonstratedd by a high inter-rater reliability and internal consistency. Interestingly, the

partt of this study concerning the delirium subtypes may also offer some practical treatment

consequences.. A positive score on the DRS-R-98 items 'affect liability ' and/or 'motor

agitation'' excludes a hypoactive delirium, and therefore specific and symptomatic,

medicall treatment can be started. The possibility that patients with a hypoactive delirium

aree misdiagnosed is 11%. In case of excluding a hypoactive delirium a quick and safe

decisionn concerning the start of symptomatically medical treatment interventions can

thereforee be provided. This is in order to protect a patient properly from adverse events

orr hospital complications directly or indirectly caused by his delirious symptoms.

Inn Chapter 5 we reviewed the current knowledge about the occurrence of the different

subtypess of delirium in various patient populations and whether a relationship can be

foundd between different subtypes of delirium and their etiology, pathophysiology, outcome

andd optimal treatment strategies. Clinical observations suggesting the existence of delirium

subtypess have been confirmed by several studies, using several methods. The systematic

investigationn of subtypes of delirium based on their clinical symptoms may provide us

withh information concerning the etiology and pathophysiology of delirium, the prognosis

off a delirium, and also may have therapeutic consequences. The studies described in this

226 6 AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

revieww show different results, partly due to methodological differences and possibly by

lackk of a standard classification for delirium subtypes. According to the present literature

aa useful and reproducible method to classify (patterns of) symptoms in delirium subtypes

seemss to be the general rating of a division in to psychomotor subtypes. The MDA S and

thee DAS appear to be reliable methods together with the new version of the DRS, the

DRS-R-98.. However, to compare the results of future studies concerning the relationship

betweenn the pathophysiology of delirium and its clinical symptoms and outcomes, one

methodd of subtype classification in delirium should be available, preferably based on

internationall consensus statements.

Althoughh a variety of factors is associated with delirium, such as psychiatric illness,

olderr age, and cerebral vascular disease, the pathophysiology of delirium remains poorly

understood.. Interestingly, delirium has been recognized as a frequent manifestation of

infectionss in the elderly. Delirium usually disappears as the underlying illness causing

deliriumm has been resolved and is a fully reversible phenomenon similar to cytokine-

inducedd sickness behaviour. Moreover, animal studies have demonstrated that cytokines

cann cause a reduction in the acetylcholinergic pathways which are supposed to be impaired

inn delirium. Based on this information delirium may be considered as a distinct part of

sicknesss behaviour that can be seen as the outward expression of a potentially reversible

episodee of brain inflammation and is triggered by peripheral immune stimulation.

Thesee and other studies resulted in several hypotheses suggesting that cytokines may be

involvedd in the pathogenesis of delirium. There are, however, no data on the association

betweenn peripheral cytokine levels and delirium.

Chapterr 6 reports a study amongst consecutive elderly patients acutely admitted to the

hospitall to compare the expression patterns of pro- and anti-inflammatory cytokines in

227 7 AspectsAspects of acute hospital admission in the elderly

S U M M E R Y Y

patientss with and without delirium. In this sample of 185 patients 34.6% met the criteria

forr delirium. Patients with delirium had significantly more often IL-6 and IL-8 levels

abovee the detection limit . These differences remained after stratifying for infectious

disease.. This is the first study that shows a relationship between peripherally measured

cytokinee levels and delirium as a symptom/exponent of sickness behaviour in acutely

admittedd elderly patients. It can be concluded that more research is necessary to study

thee possibility that inflammatory mechanisms are involved in pathogenetic pathways of

delirium.. The present study suggested a role for proinflammatory cytokines in delirium,

independentt of infectious diseases.

Chapterss 7, 8 , 9 and 10 report studies in very elderly critically il l patients.

Inn Chapter 7 we focus on the most important factors that may influence outcomes in

veryy elderly critically il l patients, on models that predict short-term and long-term outcome,

andd on the available data on patients' preferences regarding life-sustaining treatment

andd how these preferences are influenced by the likelihood of a beneficial outcome.

Thee ICU population is ageing, and it may be concluded that very elderly patients

admittedd to ICUs represent a distinct and important subgroup of patients. In general,

veryy elderly patients have poorer outcomes than do younger patients, but prognosis is

moree dependent on severity of illness and functional status before admission than on

highh age itself. A number of prognostic models have been developed that predict survival

inn critically il l patients, but these models are not calibrated for use in very old patients.

Furthermore,, they do not take into account some known risk factors, such as comorbid

conditions,, and functional and cognitive status before ICU admission. Finally, the

modelss do not provide a prognosis regarding (long-term) functional status after hospital

discharge.. We suggest that a model should be developed for predicting outcome of ICU

treatmentt in very old patients, taking into account all discussed prognostic factors. Such

J 1 28 8 AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

aa model could more precisely predict the (long-term) discharge outcome of these patients

andd support informed decision making, in accordance with the preferences of the

patientss and their relatives.

Thee aim of the study reported in Chapter 8 was to examine short-term and long-term

mortalityy in very elderly medical and surgical patients and to determine factors that are

associatedd with poor outcome. In accordance with earlier publications, we considered

patientss aged 80 years or older to be "very elderly". Because important differences in

outcomee have been reported between (very) elderly patients admitted to the ICU after

plannedd surgery vs. patients admitted after urgent surgery or for medical reasons, we

reportt data for these three patient groups separately.

Ourr results indicate that outcome after intensive care in patients aged 80 years or

olderr is highly dependent on the reason for ICU admission. Unadjusted ICU mortality

wass significantly higher in patients who were admitted because of unplanned surgical

orr medical causes than in patients after planned surgical admission. Similarly, survival

untill 12 months after hospital discharge was much better in planned surgical patients.

Interestingly,, median survival of elderly patients who were admitted after planned surgery

didd not differ from survival in the age- and gender-matched general population.

Survivall of unplanned surgical or medical elderly ICU patients was substantially lower

withh a median survival of 28 days and with more than 60% of patients dying before 12

monthss after hospital discharge. In these groups lif e expectancy is much lower than that

off the matched general population.

Resultss of the multivariate analysis, however, indicate that various factors determined

ICUU mortality, but not admission type. Factors reflecting initial severity of illness, such

ass GCS score, SAPS II score, urine production, body temperature, bicarbonate levels,

andd need for high F I02 are the prognostic factors for ICU mortality. In contrast, in

229 9 AspectsAspects of acute hospital admission in the elderly

S U M M E R Y Y

patientss who survived ICU these factors were not independent risk factors for hospital

orr 12-month mortality. Instead, admission type and renal function at admission were

foundd to be risk factors for hospital and/or 12-month mortality.

Itt can be concluded that very elderly ICU patients admitted after planned surgery

generallyy have a high likelihood of survival that does not differ substantially from that

off the general population. In unplanned surgical or medical patients prognosis is not

lesss good, but still approx. 35% of patients survive up to 1 year after hospital discharge.

Prognosticc models should be developed that are validated in (very) elderly ICU patients to

betterr predict outcome in subgroups and in individual patients. Ideally these models

shouldd not focus exclusively on short-term and long-term survival but also include cognitive

andd functional status and quality of living after hospital discharge.

Soo far, only few studies concerned long-term functional outcome and quality of life in

veryy elderly patients surviving medical and surgical ICU (6-11) and no studies addressed

long-termm cognitive status after ICU discharge in very elderly patients.

Thee aim of the study described in Chapter 9 is therefore to examine cognitive and

functionall state of very elderly patients at least one year after their ICU treatment. Our

resultss among intensive care patients aged 80 years and older at least one year after discharge

(meann time period since discharge 44.2 months) indicate that only one fifth experienced

(severe)) cognitive impairment and also one fifth had severe functional impairment.

Furthermore,, 88% of the patients with cognitive impairment had at least 4 disabilities,

comparedd to 48% of the patients without cognitive impairment. Interestingly, the largest

partt of this cohort, almost 75%, still lived at home, with a small medical consumption

ass expressed in hospital admission or consultation of their general practitioner. Quality

off life of these patients was not different from an age-matched British non-ICU population.

Mostt patients reported to be willin g to receive ICU treatment again if necessary.

2 3 0 0

AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

Inn conclusion, we found that ICU treatment in the very elderly can be performed with

goodd long-term outcome in the survivors, measured by both cognitive and physical

functioningg and by health-related quality of life. Long term survival of ICU patients

afterr unplanned surgery or for medical reasons is however poor and insufficient (pre

admission)) data exist on their long-term cognitive and functional prognosis. Estimating

prognosiss is an essential consideration in evaluating treatment options for very elderly

criticallyy il l patients. Good decision-making requires knowing a patient's likely outcomes

andd incorporating patient's preferences towards intensive life-sustaining treatments.

Futuree research should focus on prognosis of elderly patients, especially after urgent surgery

orr for medical reasons, and on the factors that determine outcome. Baseline cognitive

andd functional status may be major determinants in long-term functional status and

moree prospective studies are necessary to study this thoroughly.

Inn Chapter 10 we report the development of a prognostic model for elderly ICU patients

agedd 80 years and older, to predict mortality using data obtained within 24 hours after

ICUU admission. The model should reliably identify specifically subgroups at very high

riskk of dying. Finally, it should be "intuitive" or at least easily interpretable, clearly showing

thee factors that lead to a worse prognosis. In order to develop such a model we used the

techniquee of recursive partitioning analysis (RPA). RPA is a nonparametric technique

whichh iteratively subdivides a population in subgroups by creating mutually exclusive

subsetss according to a set of predictor variables. The process results in model, a so called

classificationn tree.

Wee compared this model with the original SAPS II model and a SAPS II model after

recalibrationn for a Dutch population of ICU patients aged 80 years and older. The overall

performancee of our classification tree model was similar to the performance of the original

SAPSS II model and the recalibrated SAPS II model. The classification tree identified

231 1 AspectsAspects of acute hospital admission in the elderly

SUMMERY Y

mostt patients with very high risk of mortality (9.2% of patients had a risk > 80% vs.

8.9%% for original SAPS II and 5.9% for recalibrated SAPS II) . Using a cut-point at risk

off 80%, the positive predictive value was 0.88 for the classification tree, 0.83 for original

SAPSS II and 0.87 for the recalibrated SAPS II .

Wee consider all three prognostic models to be able to reliably identify large sub-

groupss of patients with a high risk to die before hospital discharge. However, although

modelss based on logistic regression, such as the APACHE and SAPS models, may have

aa slightly better discriminating power in all patients, our classification tree model, based

onn recursive partitioning, appeared to better identify patients with a very high mortality

risk.. When compared with the SAPS II model and the SAPS II model after recalibration

forr Dutch very elderly patients, more very high risk patients could be identified with

thee classification tree model and positive predictive value was higher. Another potential

advantagee of classification tree based models is the fact that information presented in

classificationn trees may be more easily accepted by patients, their family members and

healthh care providers. In conclusion, our results show that prognostic models may reliably

identifyy subgroups of elderly patients with a very high risk of dying before hospital discharge.

Futuree research should focus on how prognostic models may support individual patients

andd their families in decision making to ensure that they receive care consistent with

theirr preferences.

Finally,, in the last chapter, Chapter 11, the general discussion was described.

232 2

AspectsAspects of acute hospital admission in the elderly