Prevention, diagnosis, and management of postoperative delirium in older adults.

Full text

Page 1

COLLECTIVE REVIEW
Prevention, Diagnosis, and Management of
Postoperative Delirium in Older Adults
Denise R Flinn, MD, Kathleen M Diehl, MD, FACS, Lisa S Seyfried, MD, Preeti N Malani, MD
Approximately half of all operations performed in the
United States are in patients greater than 65 years of age.1
The aging of the US population is expected to result in an
increased need for surgical services among all subspecial-
ties.1Older adults represent a unique challenge to the sur-
geon, often presenting with multiple medical comorbidi-
ties and higher risk for postoperative complications.
Despite the increasing need for surgical procedures among
the elderly population, scant literature exists to address the
specific perioperative needs of older adults. Among the
many postoperative complications that result in increased
morbidity, mortality, and health care costs, delirium repre-
sents a vital concern for surgeons. But research targeting
prevention and management of postoperative delirium is
limited. In this review, we highlight the available evidence
for prevention and management of delirium. We also pro-
vide guidelines for optimal supportive care of the delirious
surgery patient.
Defining the problem
Delirium is a common cause of postoperative morbidity
and mortality. In one study of more than 1,300 patients
aged50andolderpresentingformajor,electivenoncardiac
surgery, 9% of patients developed postoperative delirium.2
Other studies report even higher rates of 15% to 53%, and
up to 70% to 87% among older patients in the intensive
care setting.3-5The development of delirium is associated
with increased mortality, increased length of stay, and an
increased rate of discharge to longterm care facilities.2De-
lirium is also associated with increased risk of major med-
ical complications including myocardial infarction, pul-
monary edema, pneumonia, and respiratory failure.2
In addition to immediate complications, postoperative
delirium is associated with increased risk of cognitive de-
clineinsucceedingyears.6,7Bickelandcolleagues6followed
200patientsundergoinghipsurgeryandfoundthat3years
afterhospitaldischarge,ratesofcognitiveimpairmentwere
more than 50% in the patients who had postoperative de-
lirium versus 4.4% in the nondelirious patients. Kat and
associates7and Kalisvaart and coworkers8followed more
than 100 patients admitted for hip surgery, who partici-
pated in a clinical trial of haloperidol prophylaxis. Rates of
mild cognitive impairment, dementia, and institutionaliza-
tionwereallhigherforpatientswithpostoperativedelirium.7
These studies suggest that postoperative delirium predicts
future cognitive decline and an increased risk of dementia.
Delirium is defined as a disturbance of consciousness
and cognition that presents over a short period of time and
has a fluctuating course.9Characterized by perceptual dis-
turbances and a reduced ability to focus, sustain, or shift
attention, delirium can be difficult to recognize, and may
be confused with other conditions common among the
elderlysuchasdementiaordepression(Tables1and2).10,11
Several clinical tools have been developed and validated to
assist clinicians in identifying patients with delirium, such
as the Confusion Assessment Method and a variation of
this tool, the Confusion Assessment Method -ICU, which
is more specifically tailored to improving diagnosis of de-
lirium among patients in ICUs12,13(Table 3).
Pathogenesis of delirium
Delirium is a complex phenomenon, often multifactorial
inoriginandlikelyaffectingmultiplespheresofthecentral
nervous system. The most recent emerging hypothesis in-
volves the cholinergic pathways, suggesting a deficiency in
these pathways may be one of the underlying factors caus-
ing delirium.14Factors that predispose or exacerbate delir-
iuminthenonsurgicalpatientcanaffectelderlypatientsin
the postoperative period as well.
Hypoxia, hypoglycemia, electrolyte imbalances, volume
depletion, infection, and drug interactions are all common
contributorstothedevelopmentofdeliriuminthepostop-
erative patient. Specific attention should be paid to the
patient’s overall volume status throughout the postopera-
tive period, with efforts to maintain euvolemia and overall
electrolytebalancebecauseevenmodestchangesinsodium
and water balance have been associated with delirium.15
Pain is a common postoperative complaint, and deliri-
ous patients may not be able to effectively communicate
with providers about pain. Both undertreatment of pain
Disclosure Information: Nothing to disclose.
Received January 21, 2009; Revised March 5, 2009; Accepted March 11,
2009.
From the Department of Internal Medicine, Divisions of Geriatric Medicine
(Flinn, Malani) and Infectious Diseases (Malani); the Departments of Sur-
gery (Diehl) and Psychiatry (Seyfried); the University of Michigan Health
System; and the Veterans Affairs Ann Arbor Healthcare System (Malani) and
the Geriatric Research Education and Clinical Center (GRECC) (Malani),
Ann Arbor, MI.
Correspondenceaddress:PreetiNMalani,MD,VAHealthcareSystem,2215
Fuller Rd, Ann Arbor, MI 48105.
261
© 2009 by the American College of Surgeons
Published by Elsevier Inc.
ISSN 1072-7515/09/$36.00
doi:10.1016/j.jamcollsurg.2009.03.008

Page 2

and overuse of narcotics can exacerbate delirium, making
postoperative pain management a challenge. Among post-
operative orthopaedic patients, Morrison and colleagues16
found that avoidance of opioids or limited use of opioids
increased the risk of delirium, presumably because of inad-
equate control of pain in those patients. Postoperative pa-
tients are frequently immobilized, with catheters and lines
in place that can contribute to discomfort. Finally, patients
with underlying dementia are at increased risk of delirium
with any hospitalization and this risk is increased even
more in the postoperative period.
Risk factors for delirium
Several investigators have attempted to determine risk fac-
tors associated with the development of postoperative de-
lirium. Marcantonio and colleagues2identified seven pre-
dictors that could be used preoperatively to stratify an
individual patient’s risk of delirium. These factors include
age greater than 70 years, self-reported alcohol abuse, poor
cognitive status, poor functional status, abnormalities of
serum sodium, potassium, or glucose, noncardiac thoracic
surgery, or abdominal aneurysm surgery. Litaker and asso-
ciates17found the incidence of delirium among 500 pa-
tients presenting for elective surgery was 11.4%; risk
factors included age greater than 70 years, preexisting cog-
nitive impairment, greater preoperative functional limita-
tions, and self-reported alcohol use. Patients who reported
narcotic use before operation were also at increased risk for
postoperative delirium. Litaker’s findings largely con-
firmed those of Marcantonio, suggesting these risk factors
may be useful in identifying high risk patients who might
benefitfromtargeteddeliriumpreventionmeasuresduring
the perioperative period. Robinson and coauthors18re-
cently published results of their study among veterans 50
years and older presenting for elective surgery requiring an
ICU stay. Their study confirmed once again that preexist-
ing cognitive dysfunction was the strongest predictor of
postoperative delirium. Additional studies among patients
undergoing spinal surgery, cardiac surgery, or gyneco-
logic surgery have identified other risk factors in those
populations, although across the surgical literature, the
risk factors most commonly reported include baseline
cognitive impairment, advanced age, and multiple med-
ical comorbidities19-22(Table 4).
Interventions to prevent delirium
A 2001 landmark trial conducted by Marcantonio and co-
workers23demonstrated the effectiveness of proactive geri-
atricsconsultationamongpatientsundergoinghipfracture
repair. In this study, implementation of the recommenda-
tions from a geriatrics consultation service reduced the in-
cidenceofdeliriumfrom50%inthecontrolgroupto32%
in the intervention group, and also reduced the incidence
of severe delirium from 29% in the control group to 12%
in the intervention group. Many institutions have focused
recent efforts on identifying high risk patients and de-
veloping interventions to prevent delirium among those
patients, such as prophylactic administration of atypical
antipsychotic medications or assigning patients to mul-
Table 1. Diagnostic Criteria for Delirium Based on the
Diagnostic and Statistical Manual of Mental Disorders,
4thEdition9
1. Disturbance of consciousness (that is, reduced clarity of
awareness of the environment, with reduced ability to focus,
sustain, or shift attention).
2. A change in cognition (such as memory deficit, disorientation,
language disturbance) or development of a perceptual
disturbance that is not better accounted for by a preexisting
established or evolving dementia.
3. The disturbance developed over a short period of time (usually
hours to days) and tends to fluctuate during the course of the
day.
4. The delirium is due to a general medical condition—there is
evidence from the history, physical examination, or laboratory
findings that the disturbance is caused by the direct physiologic
consequences of a general medical condition (further criteria
for specific forms of delirium caused by substance intoxication
or withdrawal).
Table 2. Distinguishing Characteristics of Delirium, Dementia, and Depression10,11
CharacteristicDelirium
Distinguishing
characteristicwith decreased attention
Onset Acute
Course Fluctuating
Consciousness and
orientation
Attention and memoryPoor short-term memory, marked
inattention
Psychosis present?Common (hallucinations are often
simple, fleeting)
DementiaDepression
Fluctuating level of consciousnessMemory impairment Sadness, decreased interest in
activities
Varies
Diurnal variation
Unimpaired
Gradual
Chronic, progressive
Clear until advanced stagesClouded, disoriented
Poor short-term memory without
inattention
Less common
Poor attention but memory
intact
Less common (hallucinations are
often complex)
262
Flinn et al
Postoperative Delirium in Older AdultsJ Am Coll Surg

Page 3

tidisciplinary teams of providers with experience caring
for elderly patients.
Clinical trials have addressed the potential benefits of
atypical antipsychotics in terms of reducing the incidence
of delirium. Kalisvaart and associates8studied 430 patients
undergoing elective hip replacement, randomizing to ei-
ther preoperative haloperidol continued for up to 3 days
postoperatively or placebo. Although it did not reduce the
overall incidence of delirium, haloperidol did reduce the
severity and duration of delirium and reduced overall
length of stay. Another randomized controlled trial among
patients undergoing cardiac surgery showed that a single
dose of risperidone postoperatively reduced the incidence
of delirium.24
More recent trials have evaluated the effectiveness of
donepezil in prevention of postoperative delirium. Samp-
son and colleagues25randomized 33 patients undergoing
elective hip arthroplasty to donepezil or placebo beginning
in the immediate postoperative period and continuing for
an additional 3 days. Although donepezil did not signifi-
cantly reduce the incidence of delirium or length of stay,
there was a consistent trend toward possible benefit in this
small trial. Liptzin and associates26also studied the use of
donepezilfor2weekspreoperativelytopreventdeliriumin
patients undergoing elective orthopaedic surgery. The
overall incidence of delirium was 18.8%, with no differ-
ence between the two groups. But the patients in this small
study were relatively young and cognitively intact. A pilot
study by Leung and coworkers27demonstrated promising
results with the use of gabapentin for postoperative pain
control. A larger confirmatory trial is underway.
Management of delirium
Despite the best preventive efforts, a certain percentage of
patients will become delirious in the postoperative period.
Optimal management of delirium requires eliminating
precipitating factors and providing supportive care. In ad-
dition to the general recommendations for managing de-
lirium in all hospitalized patients, there are several issues
specific to the postoperative patient (Table 5).
As noted previously, pain control is a vital issue, partic-
ularly in the elderly postoperative patient. Two studies
found that increased postoperative pain is associated with
higherratesofdelirium.28,29Methodsofpostoperativepain
control include neuraxial analgesia, patient-controlled an-
algesia, or orally administered opioids. Vaurio and associ-
ates28examined whether method of pain management was
Table 3. The Confusion Assessment Method Diagnostic
Algorithm12
Feature 1. Acute onset and fluctuating course
Is there evidence of an acute change in mental status
from the patient’s baseline? Did the (abnormal)
behavior fluctuate during the day, that is, tend to come
and go, or increase and decrease in severity?
Feature 2. Inattention
Did the patient have difficulty focusing attention, for
example, being easily distractible, or having difficulty
keeping track of what was being said?
Feature 3. Disorganized thinking
Was the patient’s thinking disorganized or incoherent
(eg, rambling or irrelevant conversation, unclear or
illogical flow of ideas, or unpredictable switching from
subject to subject)?
Feature 4. Altered level of consciousness
Overall, how would you rate this patient’s level of
consciousness (alert [normal], vigilant, lethargic, stupor
or coma)?
Diagnosis of delirium is confirmed by the presence of Features 1 and 2 and
either Feature 3 or 4.
Table 4. Risk Factors for Postoperative Delirium
Most commonly reported
Preexisting cognitive impairment
Multiple medical comorbidities
Age ?65 or 70 y
Elective noncardiac surgery2
Age ?70 y
Self-reported history of alcohol abuse
Telephone Interview for Cognitive Status (TICS)-score ?30
(suggestion of baseline cognitive impairment)
Specific Activity Scale (SAS) Class IV (severe physical
impairment)
Markedly abnormal preoperative sodium, potassium, or glucose
levels
Aortic aneurysm surgery
Noncardiac thoracic surgery
Spinal surgery21
Transfusion of ?800 mL
Hemoglobin ?10 g/dL
Cardiac surgery22,48
History of cerebrovascular accident
High medical comorbidity
Increased creatinine
Increased preoperative pain rating
Left ventricular dysfunction
Diabetes mellitus
Atrial fibrillation
Gynecologic surgery20
Age
Number of medications
Amount of narcotic medications
General risk factors for deliriumy49
Vision impairment
Severe illness
Cognitive impairment
Increased blood urea nitrogen/creatinine ratio
263
Vol. 209, No. 2, August 2009
Flinn et al
Postoperative Delirium in Older Adults

Page 4

associated with development of postoperative delirium.
Surprisingly, even though doses of opioid analgesics are
much lowerwhenadministeredneuraxially,therewasnodif-
ferenceintheratesofpostoperativedeliriumamongthosetwo
groups of patients; rates of postoperative delirium were lower
amongpatientswhoreceivedoralanalgesia.28Althoughover-
use of narcotics can certainly predispose a patient to delirium
and worsen cognition, inadequate treatment of pain also in-
creases the possibility of delirium.
Elderly postoperative patients may not be able to properly
operate patient-controlled analgesia, so low-dose scheduled
narcotics may provide better control of pain. Scheduled
acetaminophenisalsoacommonlyusednon-narcoticpain
management strategy in the geriatric population; this a
relatively safe and well-tolerated regimen. We recommend
up to 1 g of acetaminophen every 8 hours in all patients
who do not have a contraindication. If pain is not ade-
quately controlled, low dose scheduled IV morphine sul-
fate or oxycodone 5 to 10 mg every 4 to 6 hours (either
scheduledorasneeded)iscommonlyrecommendedbyour
consultation service. We prefer to avoid combination
agents (narcotic and acetaminophen) in the immediate
postoperative period because of a potential for acetamino-
phen overdose. Meperidine use should be avoided in the
elderly because it is not often effective oral analgesia and
may cause confusion.30
Postoperative nausea is another common symptom that
may require aggressive treatment to avoid prolonged peri-
ods without adequate nutrition. Although not specifically
studied among elderly postoperative patients, a recent
study by Jokela and coworkers31confirmed that ondanse-
troneffectivelyrelievednauseaandvomitingamongalarge
sample of patients. Because of its different mechanism of
action, this drug tends to be better tolerated among elderly
patients than promethazine or prochlorperazine, both of
which can have significant anticholinergic or extrapyram-
dial side effects.30
Review and adjustment of medications is another im-
portant management strategy for patients with delirium.
Since publication of the Beers criteria, the practice of
avoiding anticholinergics, antihistamines, and benzodiaz-
epines has become more commonplace.30Often order sets
for pain management include “prn” medications such as
diphenhydramine for itching or promethazine for nausea,
bothofwhichshouldbeavoidedinelderlypatientsbecause
of their anticholinergic effects, which can both precipitate
and exacerbate delirium. Conversely, in patients with un-
derlying dementia on cholinesterase inhibitors, abrupt ces-
sationofthesemedicationscanincreaseconfusionandpre-
cipitate delirium. Patients with Parkinson’s disease should
also continue on dopamine agonist therapy, receiving their
medication on the morning of surgery and resuming as
soon as possible postoperatively.32Centrally acting dopa-
mineantagonistssuchaspromethazineormetoclopramide
for nausea should be avoided in particular among patients
Table 5. Strategies for Managing Delirium23
Ensure adequate oxygen delivery to tissues.
Avoid hypoxia, hypotension, and severe anemia.
Maintain fluid and electrolyte balance.
Monitor fluid status to avoid volume depletion and fluid
overload.
Ensure adequate pain management.
Scheduled acetaminophen 1 g every 8 hours
Low dose scheduled oxycodone or morphine sulfate and/or as
needed
Review medication.
Discontinue unnecessary medications.
Discontinue or minimize use of anticholinergics, antihistamines,
and benzodiazepines.
Continue cholinesterase inhibitors.
Continue carbidopa or levo-dopa for patients with
Parkinsonism.
Maintain bowel and bladder function.
Scheduled senna tablets or sorbitol, polyethylene glycol.
Avoid magnesium hydroxide, which can cause
hypermagnesemia.
Discontinue urinary catheter by postoperative day 2 if possible.
Mobilize patients early.
Get out of bed to chair with meals, or at least daily.
Ambulate to toilet with assistance.
Initiate physical and occupational therapy as soon as patient is
able to participate.
Monitor and treat major cardiac and pulmonary postoperative
complications.
Optimize environmental stimuli.
Provide patients with glasses and hearing aids.
Reorient with clocks and calendars.
Try to maintain sleep/wake cycle.
Family members at bedside when possible, other familiar objects
in patient’s room.
Transfer out of intensive care to regular floor as soon as
medically appropriate.
Treat agitated delirium.
Low dose atypical antipsychotic such as quetiapine 12.5–25 mg
nightly or as needed.
If the patient is unable to take oral medications, consider low
dose haloperidol 0.25 mg
IV/intramuscularly as needed.
For patients in whom antipsychotic use is contraindicated, use
of lorazepam 0.25–0.5 mg
IV can be considered.
Avoid restraints.
264
Flinn et al
Postoperative Delirium in Older Adults J Am Coll Surg

Page 5

with Parkinsonism because of their potential to exacerbate
tremor and rigidity and cause confusion.32
Althoughconstipationisacommonpostoperativeprob-
lem among patients of all ages, aggressive management is
particularly important for the geriatric patient. Scheduled
senna tablets or osmotic laxatives such as sorbitol or poly-
ethylene glycol should be considered for elderly postoper-
ative patients. Magnesium hydroxide (milk of magnesia)
shouldbeavoidedintheelderlypopulation,particularlyin
those with chronic renal insufficiency, because prolonged
use can result in hypermagnesemia, which can itself lead to
ileus.33Discontinuationofurinarycathetersassoonaspos-
sible after surgery is recommended to decrease the inci-
dence of urinary tract infection and to facilitate mobility.
Early mobilization and initiation of rehabilitation is also
vital for postoperative elderly patients to prevent further
decline during hospitalization.
Elderly patients are at greater risk for postoperative car-
diac and pulmonary complications, which may present as
acute delirium and also prolong recovery. Typically, pa-
tients at high risk will be managed with perioperative elec-
trocardiogrammonitoringandcardiacenzymesforprompt
detection and treatment of ischemia. Use of incentive spi-
rometry is also the standard of care for many postoperative
patients, although elderly patients may require more assis-
tance or reminders about use of incentive spirometry.
To promote optimal interaction with an unfamiliar en-
vironment, elderly patients should be provided with their
eyeglasses, hearing aids, dentures, and other assistive de-
vices.Reorientationcanbefacilitatedwiththeuseofclocks
and calendars in the room, and placing familiar objects in
the room, such as using a blanket from home on the pa-
tient’s bed or pictures of family members at the bedside.
Frequentvisitsbyfamilymemberswillalsohelptoreorient
and reassure patients. Maintenance of a day-night sleep
cycleshouldbeobservedifatallpossible,withnaturallight
into the room during the daylight hours and lights and
television out at night.
Once medically appropriate, moving patients out of the
intensive care setting will eliminate the extraneous stimuli
of monitors and frequent interruptions that might further
disturbsleep-wakepatterns.Othersimplemeasuressuchas
ensuring that the patient’s call light is within reach and
maintaining consistency of staffing whenever possible are
also helpful.
Management of delirium-associated agitation
Patients who present with agitated delirium are often per-
ceivedashavingamore“severe”presentationofdelirium.34
But patients with hypoactive delirium (pronounced leth-
argy as opposed to agitation) have worse outcomes;
whether this is from fundamental differences in the sub-
types of delirium or differences in treatment has not been
determined.34Although identification and treatment of
underlying causes is the definitive treatment of delirium,
supportivecareisoftenneeded.Frequentreorientationand
reassurance, which can often be provided best by the pa-
tient’s family members, remains the first line of manage-
ment. Sitters on a short-term basis may also be an effective
nonpharmacologicmethodformanagingagitatedpatients.
Use of restraints should be avoided whenever possible un-
less patient safety is threatened. When these measures fail,
low dose antipsychotic medication may be used on an as
needed basis for control of severe agitation, although evi-
dence for their effectiveness is admittedly limited while
evidence of their increased risks continues to emerge.35,36
The most widely studied medication is haloperidol, a
first-generationantipsychoticmedication.Haloperidolcan
be administered IV. We recommend starting with a very
low dose, such as 0.25 to 0.50 mg every 4 hours as needed.
Doses may initially be repeated every 30 minutes for severe
agitation, and for intractable pain, continuous infusions of
haloperidol have been used.37Routine electrocardiogram
monitoring to follow QTc interval is recommended by the
American Psychiatric Association (APA), with consider-
ationofdiscontinuingtherapyorcardiologyconsultationif
QTc reaches ? 450 milliseconds or ? 25% above base-
line.38Extrapyramidal side effects are less likely to occur
with the IV form of haloperidol compared with oral or
intramuscular forms39and with less than 4.5 mg/day dos-
ing, although there is a slightly higher risk of cardiac ar-
rhythmias with IV administration.40But higher doses of
haloperidolareassociatedwithincreasedincidenceofextrapy-
ramidalsideeffects.41Ofnote,haloperidolisnotapprovedby
theFDAforIVuse.Nonetheless,severalconsensusstatements
haveidentifiedIVhaloperidolasthetreatmentofchoicefor
delirium in the critically ill42and postcardiac surgery
population.43
Newer atypical antipsychotics such as olanzipine and
risperidonearebeingusedfortreatmentofdeliriumaswell,
with potential for fewer extrapyramidal side effects, but a
potentialforanticholinergicsideeffects.37Controlledstud-
ies of these second-generation antipsychotics are lacking.
A Cochrane Database review looking at the use of anti-
psychotics in delirium found no differences in efficacy
or adverse effects between the atypical antipsychotics
risperidone and olanzipine and low dose haloperidol.41
Information on the use of quetiapine is limited. Scheduled
lowdoseatypicalantipsychoticscanbeusedwithasneeded
dosing of haloperidol for breakthrough agitation.
Antipsychotics should be used with caution in the elderly.
Concerns about the safety of atypical antipsychotics in treat-
ing patients with dementia-related behavioral disorders
265
Vol. 209, No. 2, August 2009
Flinn et al
Postoperative Delirium in Older Adults