Appendix D, Table 1Characteristics of Pharmacologic Prevention Studies

Author, Year, Country, Study Design, Funding SourcePrevention Strategy Used, ControlsInclusion and Exclusion Criteria, Recruitment MethodPatient Characteristics (expressed in means unless otherwise noted)Outcomes EvaluatedStudy Quality
Randomized trials
Al-Aama, 201137 Canada

Study Design: randomized controlled trial (RCT)

Funding Source(s): Division of Geriatric Medicine, Department of Medicine, Schulich School of Medicine at The University of Western Ontario
Prevention Strategy Used: melatonin 0.5 mg orally prior to sleep (n=72)

Controls: placebo (n=73)
Inclusion Criteria: at least 65 years of age and admitted through the emergency department to Internal Medicine in-patient services

Exclusion Criteria: an expected stay or life expectancy of less than 48 hours, were unable to communicate in English or to take oral medications, had an intracranial bleed or seizures, had a markedly non-therapeutic international normalized ratio (INR) less than one or more than four while on warfarin, or had a known allergy to the study compounds

Recruitment method: patients were approached directly in the emergency room or in their rooms by one of the three study clinicians within 24 hours of admission (up to 48 h was allowed on weekends)
N=145

Mean age (yrs): 84

Gender, male (%): 43

Race/ethnicity (%): NR

Medical unit: Internal Medicine
Incidence of delirium (diagnosed within 6 days post-operatively with the Confusion Assessment Method (CAM))

Delirium severity (Memorial Delirium Assessment Scale)

Use of sedatives

Use of restraints
Allocation Concealment: adequate (pharmacy controlled)

Blinding: double and outcomes assessment

Intention to Treat Analysis (ITT): no, 23 patients excluded

Withdrawals adequately described: yes
Larsen, 201038

US

Study Design: RCT

Funding Source(s): New England Baptist Hospital Research Department
Prevention Strategy Used: olanzapine 5 mg (oral) (n=243), administered perioperatively

Controls: placebo (n=252)
Inclusion Criteria: history of postoperative delirium who were scheduled for elective total knee- or total hip-replacement surgery; ability to speak English; and ability to provide informed consent

Exclusion Criteria: a diagnosis of dementia; active alcohol use; a history of alcohol dependence or abuse; allergy to olanzapine; and current use of an antipsychotic medication

Recruitment method: NR
N=495

Mean age (yrs): 74

Gender, male (%): 46

Race/ethnicity (%): white 98, non white 2

Medical unit: orthopedic teaching hospital
Incidence of delirium (defined using the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition (DSM-III)

Duration of delirium

Delirium severity (Severity of delirium according to the highest value of the DRSR-98)

Time-to-onset of delirium
Allocation Concealment: adequate (pharmacy controlled)

Blinding: double and a independent data and safety monitoring committee evaluated all potentially serious adverse events

Intention to Treat Analysis (ITT): no

Withdrawals adequately described: yes
Sieber, 201039

US

Study Design: RCT

Funding Source(s): NA
Prevention Strategy Used: deep sedation with propofol (sedation depth using bispectral index (BIS) of approximately 50) (n=57)

Controls: light sedation with propofol (BIS, ≥80) (n=57)
Inclusion Criteria: 65 years or older, undergoing hip fracture repair with spinal anesthesia and propofol sedation

Exclusion Criteria: preoperative delirium (determined by CAM); contraindications to spinal anesthesia (e.g., clinically important aortic stenosis, coagulopathy, anticoagulant use, spinal cord disease, refusal of spinal anesthesia), prior hip surgery, severe congestive heart failure (New York Heart Association class IV), severe chronic obstructive pulmonary disease (Global Initiative for Chronic Obstructive Lung Disease guidelines, stage III-IV), or mental or language barriers that would preclude data collection

Recruitment method: NR
N=114

Mean age (yrs): 82

Gender, male (%): 27

Race/ethnicity (%): NR

Medical unit: multidisciplinary hip fracture service
Incidence of delirium (DSM-III)

Delirium duration

Time from surgery until discharge

Mortality (during hospitalization)
Allocation Concealment: unclear

Blinding: double

Intention to Treat Analysis (ITT): yes

Withdrawals adequately described: yes
Gamberini, 200940

Switzerland

Study Design: RCT

Funding Source(s): Novartis (partial support)
Prevention Strategy Used: rivastigmine (oral) 1.5 mg × 3/day (n=59), starting one day prior to surgery and then post-op for 6 days

Controls: placebo (n=61)
Inclusion Criteria: age 65 or older and elective cardiac surgery with cardiopulmonary bypass

Exclusion Criteria: urgent or emergency surgery, previous cardiac surgery, cardiac surgery combined with non-cardiac procedures (typically carotid endarterectomy), insufficient knowledge of German or sensory impairment interfering with neuropsychological testing, a preoperative Mini-Mental State Examination (MMSE) <15, psychiatric illness necessitating regular use of antidepressants or antipsychotics, preexisting neurologic deficits, previous or ongoing treatment with cholinesterase inhibitors, and known contraindications for rivastigmine

Recruitment method: patients screened for eligibility based on the operation schedule for the following day
N=120, demographic information for 113 patients

Mean age (yrs): 74

Gender, male (%): 68

Race/ethnicity (%): NR

Medical unit: cardiac surgery
Incidence of delirium (diagnosed within 6 days post-operatively with the CAM)

Rescue medication use
Allocation Concealment: adequate, (hospital pharmacy using identical bottles)

Blinding: double

Intention to Treat Analysis (ITT): no, 7 excluded from analyses

Withdrawals adequately described: yes
Hudetz, 200941
US (Veterans)

Study Design: RCT

Funding Source(s): National Institutes of Health, United States Public Health Service, Medical College of Wisconsin Institutional Grant, departmental funds
Prevention Strategy Used: ketamine 0.5mg/ kg intravenous bolus (n=29)

Controls: placebo (n=29)

ketamine or placebo administered during anesthetic induction in the presence of fentanyl and etomidate
Inclusion Criteria: at least 55 years of age, provided written informed consent before the initiation of any study-related procedures, scheduled for elective coronary artery bypass graft surgery or valve replacement/repair procedures with cardiopulmonary bypass; patients receiving antidepressants, stimulants, mood stabilizers, anxiolytics, or depressants were eligible

Exclusion Criteria: history of cerebrovascular accident within 3 years of randomization, permanent ventricular pacing, previously defined cognitive deficits, patients receiving psychoactive drugs for the treatment of psychosis, hepatic impairment, chronic renal insufficiency, other pre-existing diseases deemed by the investigators to place the patient at an increased risk of perioperative complications

Recruitment method: NR
N=58

Mean age (yrs): 64

Gender, male (%): 100

Race/ethnicity (%): white 90

Medical unit: cardiac surgery
Incidence of delirium (Intensive Care Delirium Screening Checklist)

Length of stay
Allocation Concealment: unclear (“sealed envelopes”)

Blinding: double and outcomes assessment

Intention to Treat Analysis (ITT): yes

Withdrawals adequately described: no withdrawals
Maldonado, 200942
US

Study Design: RCT

Funding Source(s): none stated
Prevention Strategy Used: dexmedetomidine (loading dose: 0.4 μg/kg, followed by a maintenance drip of 0.2 μg/kg/hr–0.7 μg/kg/hr) (n=40)

Controls: propofol: 25–50 μg/kg /min (n=38)

midazolam: 0.5–2 mg/hr (n=40)

All administered postoperatively
Inclusion Criteria: patients undergoing cardiac-valve operations with cardio-pulmonary bypass

Exclusion Criteria: preexisting diagnosis of dementia or schizophrenia, the preoperative use of psychotropic medications, active or recent substance abuse or dependence, age less than 18 or older than 90 years, documented stroke within the last 6 months, evidence of advanced heart block, pregnancy, or anticipated intraoperative deep hypothermic circulatory arrest

Recruitment method: NR
N=118

Mean age (yrs): 58

Gender, male (%): 64

Race/ethnicity (%): NR

Medical unit: cardiac surgery
Incidence of delirium (DSM-IV)

Length of stay (hospital and ICU)

Rescue medication use (management of delirium)

Mortality
Allocation Concealment: unclear

Blinding: open-label

Intention to Treat Analysis (ITT): no, 28 patients excluded (24%)

Withdrawals adequately described: yes
Mouzopolous, 200943
Greece

Study Design: RCT

Funding Source(s): NR
Prevention Strategy Used: fascia iliaca compartment block (FICB) (n=102) - bupivicaine (0.3 mL/ kg) 0.25 mg dose of on admission and repeated daily every 24 h until delirium occurrence or hip surgery was performed; 24 hours after hip surgery the same dose of FICB was re-administered and repeated daily every 24 h until delirium occurrence or discharge

Controls: placebo (n=105)
Inclusion Criteria: Age 70 years and older admitted for hip fracture

Exclusion Criteria: Delirium at admission, metastatic hip cancer, history of bupivicaine allergy, use of cholinesterase inhibitors, severe coagulopathy, Parkinsonism, epilepsy, levodopa treatment, delay of surgery of more than 72 h after admission, and inability to participate in interviews (profound dementia, respiratory isolation, intubation, aphasia, coma or terminal illness)

Recruitment method: potentially eligible patients identified by systematically screening new admissions to one orthopedic ward
N=219, demographic information for 207patients

Mean age (yrs): 73

Gender, male (%): 26

Race/ethnicity (%): NR

Medical unit: orthopedics

Risk classification based on 4 predictive risk factors: (1) severity of illness, measured using acute physiology age and chronic health examination; (2) cognitive impairment, measured using the mini-mental state examination score; (3) index of dehydration, measured using the ratio of blood urea nitrogen to creatinine; and (4) visual impair-ment, measured using the standardized Snellen test

High risk defined as presence of three or more risk factors
Incidence of delirium (DSM-IV and CAM)

Delirium severity (Severity of delirium according to the highest value of the DRSR-98)

Delirium duration
Allocation Concealment: adequate, placebo identical in appearance to the active drug and was administered at the same site and in the same way as the FICB

Blinding: patients blinded

Intention to Treat Analysis (ITT): no, 12 excluded from analyses

Withdrawals adequately described: yes
Prakanrattana, 200744
Thailand

Study Design: RCT

Funding Source(s): Non-industry (Siriraj Grant for Research Development)
Prevention Strategy Used: risperidone 1 mg (sublingual) when regaining consciousness post-op (n=63)

Controls: placebo (n=63)
Inclusion Criteria: aged 40 years or older undergoing elective cardiac surgery with cardiopulmonary bypass

Exclusion Criteria: patients undergoing emergency surgery, admitted to intensive care unit, tracheal intubation before arriving to operating room, patients experiencing preoperative delirium, history of psychiatric disorders

Recruitment method: NR
N=126

Mean age (yrs): 61

Gender, male (%): 59

Race/ethnicity (%): NR

Medical unit: cardiac surgery ICU
Incidence of delirium (CAM)Allocation Concealment: possibly (no identical sublingual placebo but nurses taking care of the patient and assessing delirium left patient bedside to ensure blinding)

Blinding: double

Intention to Treat Analysis (ITT): yes

Withdrawals adequately described: none reported
Sampson, 200745
UK

Study Design: RCT

Funding Source(s): Pfizer Esai, UK
Prevention Strategy Used: donepezil 5mg (n=21), following surgery and every day post-op × 3 days

Controls: placebo (n=15)
Inclusion Criteria: patients undergoing elective total hip replacement

Exclusion Criteria: patients with mini-mental state examination (MMSE) scores of < 26; patients with sensory impairment who could not undertake neuropsychological testing and those with known hypersensitivity to donepezil or piperidine derivatives or contraindications to the use of donepezil

Recruitment method: all patients undergoing elective total hip replacement and attending the pre-admission assessment clinic, who were able to give informed consent, were invited to participate
N=50; demographic information for 33 patients

Mean age (yrs): 68

Gender, male (%): 52

Race/ethnicity (%): NR

Medical unit: orthopedics
Incidence of delirium (as indicated by the Delirium Symptom Interview)

Length of hospital stay
Allocation Concealment: adequate, by the hospital pharmacy

Blinding: double, and data were analyzed blind to randomization code

Intention to Treat Analysis (ITT): no; 14 withdrawn after randomization; 3 excluded after treatment allocation

Withdrawals adequately described: yes
Kalisvaart, 200546
The Netherlands

Study Design: RCT

Funding Source(s): NR
Prevention Strategy Used: haloperidol 1.5 mg/d (n=212), started preoperatively and continued for up to 3 days postoperatively

Controls: placebo (n=218)
Inclusion Criteria: aged70 and older admitted for acute/ elective hip surgery and were at intermediate or high risk for postoperative delirium

Exclusion Criteria: delirium at admission, no risk factors for postoperative delirium present at baseline, history of haloperidol allergy, use of cholinesterase inhibitors, parkinsonism, epilepsy, levodopa treatment, inability to participate in interviews, delay of surgery > 72 hours after admission, or a prolonged QTc interval of 460 ms or higher for men and 470 ms or higher for women on their electro-cardiogram

Recruitment method: a research team of geriatricians and nurses in a single 915- bed teaching hospital identified potentially eligible patients by systematically screening new admissions to two surgical and three orthopedic wards
N=430

Mean age (yrs): 79

Gender, male (%): 20

Race/ethnicity (%): NR

Medical unit: Surgical and orthopedic wards

Risk classification based on presence of four predictive risk factors: (1) Visual impairment (binocular near vision worse than 20/70 after correction); (2) severity of illness, measured using the Acute Physiology Age and Chronic Health Examination (score of 16 or higher indicating increased severity); (3) cognitive impairment (MMSE score ≤ 24 on a scale of 0–30); and (4) index of dehydration (ratio of blood urea nitrogen to creatinine of ≥18)

Intermediate risk -presence of 1 or 2 risk factors

High risk - presence of ≥ 3 risk factors
Incidence of post-operative delirium (DSM IV and Confusion Assessment Method criteria)

Delirium duration

Delirium severity (measured using the Delirium Rating Scale (DRS), revised version-98, range 0 (no severity) to 45 (high severity)).

Length of stay
Allocation Concealment: adequate (hospital pharmacist had prepackaged)

Blinding: double and members of the research team not involved in the clinical care of the patients performed all baseline and outcome assessments

Intention to Treat Analysis (ITT): yes

Withdrawals adequately described: yes
Liptzin, 200547
US

Study Design: RCT

Funding Source(s): Pfizer Corporation
Prevention Strategy Used: donepezil 14 days before and after surgery (n=39)

Controls: placebo (n=41)
Inclusion Criteria: scheduled for elective total knee or hip arthroplasy and aged 50 or greater; able to give informed consent

Exclusion Criteria: evidence of gastroesophageal reflux disease or sick-sinus syndrome; currently taking donepezil or previously intolerant of it; did not speak English; already in another trial

Recruitment method: recruited from pts scheduled for elective total hip or knee arthroplasty
N=90 (Baseline info for 80; 58 completed trial)

Mean age (yrs): 67

Gender, male (%): 43

Race/ethnicity (%): white 97.5, other 2.5

Medical unit: orthopedic surgery
Incidence of delirium (DSM-IV)

Mean duration of post-op delirium

Number with post-op subsyndromal delirium

Mean duration of subsyndromal delirium

Mean length of stay
Allocation Concealment: adequate (by pharmacist)

Blinding: double

Intention to Treat Analysis (ITT): no, 10 not operated on were excluded

Withdrawals adequately described: yes
Papaioannou, 200548
Greece

Study Design: RCT

Funding Source(s): European Commission's BIOMED2 program BMH4-98-3335 and Greek Ministry of Health.
Prevention Strategy Used: regional anesthesia (epidural or spinal) (n=25)

Controls: general anesthesia (n=25)
Inclusion Criteria: aged at least 60 years, scheduled for elective surgery that could be performed under regional or general anesthesia and who had agreed to be randomly allocated to receive either type of anesthesia

Exclusion Criteria: illiteracy, severe auditory or visual disturbances, central nervous system disorders, alcoholism or drug dependence, treatment with tranquillizers or antidepressants, Parkinson's disease and a preoperative MMSE score ≤ 23 points, indicative of dementia

Recruitment method: NR
N=50 (Baseline info for 47)

Median age (yrs): 68

Gender, male (%): 64

Race/ethnicity (%): NR

Medical unit: surgery (orthopedic and vascular)
Incidence of delirium (DSM-III)Allocation Concealment: unclear

Blinding: none stated

Intention to Treat Analysis (ITT): no, 3 patients were excluded

Withdrawals adequately described: yes
Aizawa, 200249
Japan

Study Design: RCT

Funding Source(s): none stated
Prevention Strategy Used: delirium free protocol, post surgery; diazepam 0.1 mg/kg IM at 20;00; funitrazepam 0.04 mg/kg and pethidine 1 mg/kg continuous IV infusions for 8 hours × 3 nights (n=20)

Controls: usual care (n=20)
Inclusion Criteria: patients aged over 70 but less than 86 years of age who underwent resection of gastric or colorectal cancer through an open laparotomy under general anesthesia

Exclusion Criteria: liver cirrhosis or liver dysfunction, renal dysfunction, respiratory disturbance, other poor risk factors, mental disorders, visual impairment, or patients who required extensive resection of other organs or emergency surgery

Recruitment method: NR
N=42 (Baseline info for 40; 2 excluded due to incomplete administration of agents)

Mean age (yrs): 76

Gender, male (%): 65

Race/ethnicity (%): NR

Medical unit: surgery
Incidence of delirium (DSM-IV)

Mean length of stay
Allocation Concealment: unclear

Blinding: outcomes assessor

Intention to Treat Analysis (ITT): no, 2 were excluded

Withdrawals adequately described: yes
Williams-Russo, 199213
US

Study Design: RCT

Funding Source(s):

National Institute of Aging
Prevention Strategy Used: continuous epidural bupivicaine and fentanyl infusions (n=26). Initiated post-op at first complaint of pain

Controls: continuous intravenous fentanyl infusions (n=25). Initiated post-op at first complaint of pain
Inclusion Criteria: scheduled for a bilateral knee replacement, speak English as a primary language, and have no serious hearing or vision impairment which would preclude cognitive testing

Exclusion Criteria: none stated

Recruitment method: bilateral knee surgery patients were approached
N=51

Mean age (yrs): 68

Gender, male (%): 45

Race/ethnicity (%): NR

Medical unit: Urban referral hospital specializing in elective orthopedic surgery
Incidence of delirium (DSM-III)Allocation Concealment: unclear

Blinding: physicians and nurses administering care not aware of purpose of study; study personnel not involved in patient care/treatment decisions

Intention to Treat Analysis (ITT): no

Withdrawals adequately described: yes
Kaneko, 199950
Japan

Study Design: RCT

Funding Source(s): Not reported
Prevention Strategy Used: intravenous haloperidol (5 mg in 1.0mL daily) from 1st to 5th post-operative day (n=38)

Controls: equal volume of normal saline injection (0.9%) (n=40)
Inclusion Criteria: patients scheduled for elective gastrointestinal surgery, admitted to High and Intensive Care Unit before scheduled surgery

Exclusion Criteria: none stated

Recruitment method: iInterviewed after admission
N=80 (2 patients excluded, unclear if excluded before or after randomization)

Mean age (yrs): 72.8

Gender, male (%): 63

Race/ethnicity (%): NR

Medical unit: high and Intensive Care Unit for gastrointestinal surgery
Incidence of delirium (DSM-III-R)Allocation Concealment: unclear

Blinding: none reported

Intention to Treat Analysis (ITT): unclear

Withdrawals adequately described: no – unclear when 2 patients were excluded
Berggren, 198751
Sweden

Study Design: RCT

Funding Source(s): Swedish Medical Research Council, No. 12x-5664, King Gustav V's 80th Birthday Foundation, and the Urnei University Research Foundation
Prevention Strategy Used: epidural anesthesia (n=28)

Controls: halothane anesthesia (n=29)
Inclusion Criteria: patients admitted to the orthopedic wards for femoral neck fractures and were fully lucid
Exclusion Criteria: none stated

Recruitment method: NR
N=57

Mean age (yrs): 78

Gender, male (%): 19

Race/ethnicity (%): NR

Medical unit: orthopedic wards
Incidence of delirium (DSM-III)

Mean length of stay

Mortality
Allocation Concealment: unclear

Blinding: outcomes assessor

Intention to Treat Analysis (ITT): yes

Withdrawals adequately described: yes
Non-randomized trials
Katznelson, 200952
Canada

Study Design: prospective observational study

Funding Source(s): University of Toronto
Prevention Strategy Used: statins (n=676)

Controls: no statins (383)
Inclusion Criteria: patients undergoing cardiac surgery

Exclusion Criteria: patients undergoing congenital or redo surgery, or requiring circulatory arrest, were excluded

Recruitment method: NA
N=1059

Mean age (yrs): NA

Gender, male (%): 71

Race/ethnicity (%): NR

Medical unit: cardiovascular ICU
Incidence of delirium (diagnosed with CAM), presented as an odds ratio and also stratified by age groups (age <60 years and ≥ 60 years)Allocation Concealment: Not applicable (NA)

Blinding: single blinded (nursing staff)

Intention to Treat Analysis (ITT): NA

Withdrawals adequately described: NA
Del Rosario, 200853
Spain

Study Design: retrospective comparison

Funding Source(s): NR
Prevention Strategy Used: patient-controlled femoral nerve analgesia (n=49)

Controls: intravenous analgesia (n=50)
Inclusion Criteria: ≥ 50 years old; underwent hip fracture surgery with intradural anesthesia

Exclusion Criteria: received general· anesthesia or epidural analgesia, presented failure of femoral analgesia, or had localized infection or coagulopathy

Recruitment method: NA, chart review
N=99

Mean age (yrs): 81

Gender, male (%): Intervention: 20, Control: 38, p=0.08

Race/ethnicity (%): NR

Medical unit: orthopedics
Incidence of delirium (documentation of altered mental status (confusion, disorientation, changes of level of consciousness, changes in the sleep-wake cycle))

Delirium severity (classified into two degrees of severity, low or severe, according to the need of prescription of any antipsychotic drug)

Rescue medication use
Allocation Concealment: NA

Blinding: NA

Intention to Treat Analysis (ITT): NA

Withdrawals adequately described: NA
Dautzenberg, 200412
The Netherlands

Study Design: Retrospective cohort study

Funding Source(s): NR
Prevention Strategy Used: rivastigmine chronic users (n=11)

Controls: non-rivastigmine users (n=29)
Inclusion Criteria: patients who were treated by the geriatric consultation team and had the appearance of a delirium or were considered to be at high-risk of develop delirium by their treating physician

Exclusion Criteria: NR

Recruitment method: group of 366 hospitalized patients, treated by the geriatric consultation team from January 2002 to June 2003, chronic rivastigmine users compared with randomly selected subgroup of all patients not treated with rivastigmine.
N=40

Mean age (yrs): 79

Gender, male (%): 40

Race/ethnicity (%): NR

Medical unit: Non-geriatric wards
Diagnosed delirium during the time of hospitalization of the patient (based on DSM-IV criteria, and recorded in the medical record)

Length of hospital stay

Mortality
Allocation Concealment: NA

Blinding: NA

Intention to Treat Analysis (ITT): NA

Withdrawals adequately described: NA
Savage, 197854
US

Study Design: non-random comparison

Funding Source(s): NR
Prevention Strategy Used: physostigmine (n=45)

Controls: No physostigmine (n=68)
Inclusion Criteria: randomly selected pts who underwent elective surgery and were either Status I or II (American Society of Anesthesiologists)

Exclusion Criteria: bradycardia, bronchial asthma, obstructive pulmonary disease, pregnancy, Parkinson's

Recruitment method: NA
N=113

Mean age (yrs): NR

Gender, male (%): NR

Race/ethnicity (%): NR

Medical unit: surgical
Subjects were evaluated, post-surgery, on the following scale: 1) restless, thrashing, a score of 1; 2) mumbling, incoherent, a score of 2; 3) reacting, quiet, but nonverbal, a score of 3; 4) and appropriate verbal responses, a score of 4Allocation Concealment: NA

Blinding: nurses who graded delirium, were blinded to intervention

Intention to Treat Analysis (ITT): NA

Withdrawals adequately described: NA

From: APPENDIX D, EVIDENCE TABLES

Cover of Delirium: Screening, Prevention, and Diagnosis – A Systematic Review of the Evidence
Delirium: Screening, Prevention, and Diagnosis – A Systematic Review of the Evidence [Internet].
Greer N, Rossom R, Anderson P, et al.
Washington (DC): Department of Veterans Affairs (US); 2011 Sep.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.