Haloperidol prophylaxis for elderly hip-surgery patients at risk for delirium: a randomized placebo-controlled study

Haloperidol Prophylaxis for Elderly Hip-Surgery Patientsat Risk for Delirium: A Randomized Placebo-Controlled Study Kees J. Kalisvaart, MD,Ã Jos F. M. de Jonghe, PhD,Ã Marja J. Bogaards, PharmD,Ralph Vreeswijk, RN, MSc,Ã Toine C. G. Egberts, PhD,z Bart J. Burger, MD, PhD,ÃPiet Eikelenboom, MD, PhD,§z and Willem A. van Gool, MD, PhDk OBJECTIVES: To study the effectiveness of haloperidol 14.4 Æ 3.4 and 18.4 Æ 4.3, respectively (mean difference prophylaxis on incidence, severity, and duration of post- 4.0, 95% CI 5 2.0–5.8; Po.001); delirium duration was operative delirium in elderly hip-surgery patients at risk for 5.4 versus 11.8 days, respectively (mean difference 6.4 days, 95% CI 5 4.0–8.0; Po.001); and the mean number of days DESIGN: Randomized, double-blind, placebo-controlled trial.
in the hospital was 17.1 Æ 11.1 and 22.6 Æ 16.7, respec- SETTING: Large medical school–affiliated general hospi- tively (mean difference 5.5 days, 95% CI 5 1.4–2.3; Po.001). No haloperidol-related side effects were noted.
PARTICIPANTS: A total of 430 hip-surgery patients aged CONCLUSION: Low-dose haloperidol prophylactic treat- 70 and older at risk for postoperative delirium.
ment demonstrated no efficacy in reducing the incidence ofpostoperative delirium. It did have a positive effect on the INTERVENTION: Haloperidol 1.5 mg/d or placebo was severity and duration of delirium. Moreover, haloperidol started preoperatively and continued for up to 3 days post- reduced the number of days patients stayed in the hospital, operatively. Proactive geriatric consultation was provided and the therapy was well tolerated. J Am Geriatr Soc MEASUREMENTS: The primary outcome was the inci- Key words: haloperidol; delirium prevention; prophy- dence of postoperative delirium (Diagnostic and Statistical laxis; elderly; orthopedic surgery; risk factors; delirium Manual of Mental Disorders, Fourth Edition, and Confu- severity; duration of hospital stay; assessment of delirium sion Assessment Method criteria). Secondary outcomeswere the severity of delirium (Delirium Rating Scale, revisedversion-98 (DRS-R-98)), the duration of delirium, and thelength of hospital stay.
RESULTS: The overall incidence of postoperative delirium Delirium is a serious postoperative complication in was 15.8%. The percentage of patients with postoperative elderly patients.1–3 It is associated with high morbidity delirium in the haloperidol and placebo treatment condi- and mortality, increased length of hospital stay, and a high tion was 15.1% and 16.5%, respectively (relative rate of institutionalization after discharge.2,4–7 Incidence risk 5 0.91, 95% confidence interval (CI) 5 0.6–1.3); the rates for delirium of 5% to 45% in patients undergoing mean highest DRS-R-98 score Æ standard deviation was orthopedic hip surgery emphasize the need for primary andsecondary prevention.2,8,9 Although delirium can occur in any older patient, some From the ÃDepartments of Geriatric Medicine and Orthopedic Surgery, are more at risk than others. Many predisposing and Medical Center Alkmaar, Alkmaar, the Netherlands; Department precipitating factors have been identified (e.g., cognitive of Pharmacy, Reinier de Graaf Group, Delft, the Netherlands; zDepartment of impairment, sensory impairment, severity of illness, and Pharmacoepidemiology and Pharmacotherapy, Utrecht Institute forPharmaceutical Sciences, Utrecht University, the Netherlands; Departments dehydration).10 To counteract the effect of some of these of §Psychiatry and kNeurology, Academic Medical Center, University risk factors, previous studies on delirium prevention have of Amsterdam, Amsterdam, the Netherlands; and Free University of focused on nonpharmacological interventions such as Amsterdam, Amsterdam, the Netherlands.
reorienting the patient, modifying the hospital environ- A paper with the preliminary results was accepted for presentation at the ment, proactive geriatric consultation, pain treatment meeting of the American Geriatrics Society in Baltimore, Maryland, 2003.
programs, family education, early mobilization protocol, Because of illness, no presentation was made. Abstract was published in theJournal of the American Geriatrics Society, May 2003.
nutritional support, and infection control measures.11–15 Ameta-analysis revealed that, on average, nonpharmacolo- Address correspondence to K. J. Kalisvaart, MD, Medical Center Alkmaar,PO Box 501, 1800 AM Alkmaar, the Netherlands.
gical interventions reduce the absolute risk of delirium by a mere 13%.16 Further reduction of the incidence, severity, and duration of postoperative delirium could have an r 2005 by the American Geriatrics Society important effect on the burden of surgical procedures in elderly patients, but little is known about the effectiveness A research team of geriatricians and nurses in a single 915- of prophylactic pharmacological treatment strategies for bed teaching hospital in the Netherlands identified poten- tially eligible patients by systematically screening new The antipsychotic drug haloperidol is widely used for admissions to two surgical and three orthopedic wards.
the symptomatic treatment of delirium.17 In one small Men and women aged 70 and older admitted for acute or study, haloperidol prophylaxis proved to be effective in elective hip surgery were considered for inclusion in the reducing delirium in gastrointestinal surgery patients.18 haloperidol prophylaxis study, provided that they were at Haloperidol is a dopamine antagonist. Dopamine D2 intermediate or high risk for postoperative delirium. Risk receptor blockade is associated with enhanced acetylcho- classification was based on the presence of four predictive line release.19,20 Delirium is highly associated with choli- risk factors, as described elsewhere.10,25 Visual impairment, nergic deficiency. So it can be hypothesized that haloperidol defined as binocular near vision worse than 20/70 after may have an indirect beneficial effect on delirium. Indeed, correction; severity of illness, measured using the Acute some dopamine receptor antagonists, particularly antipsy- chotics, appear to treat delirium, including that arising from (APACHE II) scale of 0 to 70,26 with a cut-off score of 16 or higher indicating increased severity; cognitive impair- In other conditions that are associated with cholinergic ment (Mini-Mental State Examination (MMSE) score 24 deficiency, such as Alzheimer’s disease, haloperidol and on a scale of 0–30);27 and index of dehydration (ratio of physostigmine have a positive effect on delusions and blood urea nitrogen to creatinine of !18). Intermediate hallucinations, which are symptoms of delirium as well.23 risk for postoperative delirium was defined as presence of The documented therapeutic effects, as well as its pharma- one or two risk factors and high risk as presence of three or cological profile and a possible ‘‘priming’’ effect, suggest more risk factors. The low-risk patients were assessed daily that haloperidol could prevent the occurrence of delirium or according to the protocol for incident delirium but received reduce its severity or duration, but no controlled studies have evaluated the prophylactic effect of haloperidol on Patients were ineligible if they had delirium at postoperative delirium. There are potential side effects: admission, no risk factors for postoperative delirium hypotension (minimal), particularly with parenteral admin- present at baseline, history of haloperidol allergy, use of istration; sedation; altering of cardiac conduction; and cholinesterase inhibitors, parkinsonism, epilepsy, levodopa extrapyramidal symptoms. In addition, haloperidol has a treatment, inability to participate in interviews (profound lower potency of cholinergic blockade than other neuro- dementia, language barrier, intubation, respiratory isola- leptics. Keeping the total daily dose of haloperidol below tion, aphasia, coma, or terminal illness), delay of surgery of 3 mg may reduce the risk of extrapyramidal side effects.24 more than 72 hours after admission, or a prolonged QTc This was a randomized, placebo-controlled, double- interval of 460 ms or higher for men and 470 ms or higher blind, clinical trial of low-dose haloperidol prophylaxis for for women on their electrocardiogram.
postoperative delirium in elderly hip-surgery patients who Eligibility was checked against patients’ clinical notes were at intermediate or high risk for this complication. The and their own recall. Patients were randomized between aim was to assess the effectiveness of 1.5 mg of haloperidol daily versus placebo on the primary (incident delirium) andsecondary (deterioration of delirium) prevention of post-operative delirium in hip-surgery patients.
Measurements and ProceduresEligible patients were sequentially randomly assigned tostudy treatment (placebo or haloperidol 0.5 mg three timesdaily) from a block of drugs that the hospital pharmacist had prepackaged, according to a computer-generatedrandomization code. Placebo medication was identical in appearance to the active drug. The research team and all The study was undertaken in accordance with the Declara- participants were blinded to the treatment group, and tion of Helsinki and the guidelines on good clinical practice.
blinding was maintained throughout the study and checked Approval of the regional research ethics committee was obtained. All patients or their relatives gave fully informed Trial medication was started on admission and continued until 3 days after surgery. A maximum delayfor surgery of 72 hours was permitted. The haloperidoldosage was based on the average starting dose for treatmentof older patients with delirium in the department and recommendations by the American Psychiatric Associa- This was a randomized, placebo-controlled, double-blind, tion.17 All patients were assessed daily for efficacy and clinical trial, with a minimum duration of 1 day and a safety evaluations. Experienced geriatric nurses and geria- maximum of 6 days, depending on the onset of delirium.
tricians provided proactive geriatric consultation to all The study aim was to evaluate efficacy of 1.5 mg of patients. The consultation was based on a structured haloperidol daily versus placebo on the primary and multimodular protocol (geriatric medical attention; en- secondary prevention of postoperative delirium in elderly hancement of orientation and cognition; sensory and mobility-improving advice; attention to pain and sleeping problems; extra attention to fluid and food intake; and ary outcome variables were severity of delirium, delirium patient, family, and nursing staff education). If post- duration, and length of hospital stay. Delirium severity was operative delirium occurred, patients were treated accord- measured using the Delirium Rating Scale, revised version- ing to standard procedures (haloperidol three times per day, 98 (DRS-R-98, range 0 (no severity) to 45 (high severity)).35 lorazepam three times per day, or both in increasing doses, Daily patients assessments using the MMSE, DRS-R-98, and depending on symptoms of delirium) and assessed for Digit Span test (assessment of attention, range 0 (no attention) to 42 (good attention)36 were used to make the Code envelopes were stored in the pharmacy and at the DSM-IV and CAM diagnoses possible and to assess delirium investigation site. In case of emergency, an independent severity. CAM and DRS-R-98 assessments were continued physician could request unmasking of the treatment allocation. A statement had to be made in the Case ReportFormulary after breaking the seal. This happened with two patients in the haloperidol group and five in the placebo Statistical calculations were performed using SPSS for group. These patients do not appear in the protocol Windows, version 11 (SPSS, Inc., Chicago, IL). Calculation violation count (Figure 1). In all other cases, the treatment of the required sample size was based on the assumption that was blinded until the end of the total study.
haloperidol prophylaxis would reduce the incidence of The clinical staffFindependent of the research staffF postoperative delirium from 40% to 27%. These figures recorded the level of adherence to the intervention, with were based on a 40% incidence of delirium in a comparable reasons for nonadherence, daily. Adherence was complete if population in the pretrial period in the study hospital in 1999 the patient received all medication at the times it had to be and an absolute 13% median risk reduction as found in given. Partial adherence indicated that the patient received studies using nonpharmacological interventions.16 With a some but not all the medication or not at the scheduled two-sided test, an alpha level of 0.05, and a power of 80%, times. Nonadherence indicated that none of the medication the analysis required 206 patients per group. The analysis was undertaken as intention to treat at all levels. The primaryoutcome of the study was the incidence of postoperative delirium, defined according to the DSM-IV criteria.
Members of the research team not involved in the clinical For the primary analysis of the intervention, delirium care of the patients performed all baseline and outcome was considered a binary outcome (absent or present) assessments. Assessors had extensive training before the according to its earliest occurrence. Secondary outcomes study and followed standard procedures. All data were were the severity of delirium, which was measured using the collected on standardized patient record forms and under- DRS-R-98 and was expressed as the maximum DRS-R-98 went extensive checks of error and validity.
score during the delirium period, the duration of delirium, The baseline screening and assessments were com- and duration of hospital stay of delirious patients (the pleted before surgery and within 12 hours after admission number of days spent in hospital until patients were ready and included the MMSE, the Informant Questionnaire on for transfer to a rehabilitation unit or home).
Cognitive Decline in the Elderly (measures preexisting Proportions of patients were compared using the Fisher cognitive impairment),28 the standardized Snellen test for exact test. Two-tailed P-values o.05 were considered to visual impairment,29 chart review to determine APACHE II indicate statistical significance. Parametric and nonpara- score (range 0 5 no acute health problems to 70 5 severe metric values were tested using Student t test and the Mann- acute health problems), ratio of blood urea nitrogen to Whitney U test, respectively. The results are expressed as creatinine, Geriatric Depression Scale (15-item version self- relative risks with 95% confidence intervals (CIs) for the rating scale for depression, range 0–15, higher scores haloperidol group relative to the placebo group, with a indicating depression),30 and the Barthel Index (range relative risk less than 1.0 indicating a beneficial effect.
0–20, lower scores indicating more dependency).31 Safetywas monitored throughout the study and was based on analysis of adverse events, daily examination by the treating Of 681 individuals initially admitted to the orthopedic and surgeons, spontaneous reports from the patients, and surgical units, 603 entered the baseline phase (Figure 1).
specific assessments; the Barnes Akathisia Scale (range 0 Failure to meet the inclusion criteria was the most typical (no akathisia) to 14 (severe akathisia) was used to assess reason for not entering this phase (n 5 78); 36 refused to drug-induced akathisia.32 Electrocardiogram was per- participate, 13 were discharged without surgery, eight could formed on admission and in case of an adverse event, for not be tested, six had surgery before testing could take evaluation of QTc interval. Daily blood pressure measure- place, six were known to have parkinsonism, four were ments were taken to check for postural hypotension.
taking antipsychotic drugs, three were missed by failure of Patients were clinically assessed daily for signs of sedation the emergency department staff to report them, one had extreme liver failure, and one was delirious at admission.
One hundred twenty-one of 603 patients were not randomized because all four risk factors were absent (low The primary outcome was postoperative delirium. Diagnosis risk). Of the 482 entering the baseline phase, 52 refused to of the syndrome was defined using Diagnostic and Statistical comply after baseline screening, all because they or their Manual of Mental Disorders, Fourth Edition (DSM-IV) and caregiver refused treatment with the study drug. The Confusion Assessment Method (CAM) criteria.33,34 Second- remaining 430 eligible patients were randomized (Figure 1).
Patients not meeting inclusion criteria (n = 78) Discharged without surgery (n = 13) Parkinsonism (n = 6) On antipsychotic drugs (n = 4) Not testable (n = 8) Surgery before testing (n = 6) Intermediate risk (n = 181) High risk (n = 35) Adverse events (n = 8) Randomization violation (n = 3) Figure 1. Flow diagram of the study.
elective and fracture patients. Patients with an intermediate The characteristics at the time of admission of the 212 baseline risk for delirium and nonacute (elective) surgery patients randomized to haloperidol prophylaxis and the were overrepresented, indicating that, overall, the study 218 patients in the placebo group are shown in Table 1. The group was in relatively good clinical condition. The low groups did not differ significantly in terms of any of these APACHE II scores and the high Barthel Index scores are characteristics. The mean number of risk factors per patient was similar in the two groups. On average, both studygroups included elderly patients with minimal cognitive impairment, some visual impairment, and light dehydra- Intention-to-treat (ITT) analysis included 430 patients; tion. MMSE results at baseline did not differ between delirium occurred in 68 (15.8%). The incidence of delirium Table 1. Characteristics of the Patients on Admission According to Study Group/Intention-to-Treat Group Mini-Mental State Examination score, mean Æ SDÃ Acute Physiology Age and Chronic Health Examination II score, mean Æ SDz Blood urea nitrogen/creatinine ratio, mean Æ SD§ Geriatric Depression Scale-15 score, mean Æ SDk Note: Because of rounding, percentages may not total 100.
à Range 0 (severe cognitive impairment) to 30 (no cognitive impairment).
w Range 20/20 (no visual impairment) to 20/800 (severe visual impairment).
z Range 0 (no acute health problems) to 70 (severe acute health problems).
§ Ratio greater than 18 indicates dehydration.
k Range 0 (depression not likely) to 15 (depression very likely).
z Range 0 (severe disability) to 20 (no disability).
# Two patients with no risk missing from total 218 patients in placebo group; see Figure 1.
in the ITT haloperidol group of 15.1% (32/212) did not having delirium was significantly lower after haloperidol differ significantly from the 16.5% (36/218) in the placebo prophylaxis (Figure 3). The mean duration of delirium in group (relative risk 5 0.91, 95% CI 5 0.59–1.44) (Table 2).
the haloperidol group was 6.4 days (95% CI 5 4.0–8.0; The baseline characteristics of patients in the haloperidol Po.001) shorter than in the placebo group (haloperidol and placebo group who developed delirium on follow-up 5.41 Æ 4.91 days vs placebo 11.85 Æ 7.56 days). The mean did not differ significantly (Table 2).
difference of days spent in the hospital until patients were Forty-four of the randomized 367 patients with an ready for transfer to a rehabilitation unit or home was 5.5 ‘‘intermediate risk’’ for delirium developed delirium (12%, days shorter (95% CI 5 1.4–2.3; Po.001) in patients from 95% CI 5 8.7–15.3%), whereas 24 of 63 high-risk patients the haloperidol group (17.1 Æ 11.1) than in the placebo became delirious (38%, 95% CI 5 26.1–51.2%). Only five group (22.6 Æ 16.7) (Table 3). No episodes with recurrence of the 121 low-risk patients (4.1%, 95% CI 5 0–4.4%) of delirium were observed in this study.
developed delirium. Per-protocol analysis included 382 No drug-related side effects were seen during the study patients; delirium occurred in 55 (14.4%). The dropout period. The adverse events were never related to extra- incidence was 20 (9.4%) patients in the haloperidol group, pyramidal symptoms. (Values on the Barnes Akathisia Scale of which 11 were lost for follow-up for the per-protocol were 0 for all the patients in both groups.) There was no analysis and 28 (12.8%) patients in the placebo group, of sedation reported, other than related to the use of which 24 were lost to follow-up (Figure 1).
There was partial or nonadherence in two patients in the haloperidol group and patients in the placebo group The characteristics of the episodes of delirium that occurredwere markedly different in both groups. The severity ofdelirium characterized by the highest value of the DRS-R-98 (DRS-Max) during an episode with delirium in patients from the haloperidol group was on aver- Low-dose haloperidol prophylaxis was not effective for the age Æ standard deviation 14.40 Æ 3.5, versus 18.41 Æ 4.4 prevention of postoperative delirium in elderly hip-surgery in the placebo group (mean difference 4.0, 95% CI 5 2.0– patients at intermediate or high risk for this complication, 5.8; Po.001) (Table 3). During the first 3 days after the but haloperidol prophylaxis markedly reduced severity and onset of delirium, the severity as measured using the mean duration of postoperative delirium. As a result, the burden of the DRS-R-98 scores was significantly lower in patients of postoperative delirium was less, as was the number of who had received haloperidol preoperatively (Figure 2), days patients stayed in the hospital. No drug-related side and from Day 5 until Day 8, the proportion of patients still Table 2. Characteristics of Patients Who Developed Delirium, According to Study Group: Intention-to-Treat Group Mini-Mental State Examination score, mean Æ SDà Acute Physiology Age and Chronic Health Examination II score, mean Æ SDz Blood urea nitrogen/creatinine ratio, Æ SD§ Geriatric Depression Scale-15 score, mean Æ SDk à Range 0 (severe cognitive impairment) to 30 (no cognitive impairment).
w Range 20/20 (no visual impairment) to 20/800 (severe visual impairment).
z Range 0 (no acute health problems) to 70 (severe acute health problems).
§ Ratio greater than 18 indicates dehydration.
k Range 0 (depression not likely) to 15 (depression very likely).
z Range 0 (severe disability) to 20 (no disability).
These findings have important implications for the Perhaps the findings of the current study indicate a management of elderly patients at risk of delirium. Primary ‘‘priming’’ effect (i.e., therapeutic blood serum levels of prevention (proactive geriatric consultation) is an effective haloperidol were reached sooner once treatment of delirium strategy in preventing delirium, and even when delirium develops, there was still an additional effect of this The strengths of this study include the targeting of older pharmacological prophylactic intervention on the duration people at (intermediate or high) risk of developing delirium and severity of delirium. The results of this study are not in for prophylactic treatment only. Patients at low risk were accord with the results of one other study, in which not randomized. Postoperative delirium occurred in only haloperidol prophylaxis led to a reduction of postoperative five (4.1%) of them, which is much less than in the at-risk delirium in gastrointestinal patients and no data are available sample (15.8%). The findings corroborate the prognostic on the reduction of severity or delirium duration.18 The model of one study, which is an important finding in and of current study did not find an effect on the primary endpoint of itself and provides important validation of this risk system postsurgery delirium. Rather, there was a significant effect on for this particular population.10 This enabled the restriction secondary end points (duration and severity of delirium).
of the prophylactic pharmacological treatment to those Dissimilarities between samples, design, and type of surgery patients who needed it, thus maximizing the efficiency and may explain the apparent differences in study outcomes. One clinical relevance of the intervention. All patients were study consisted of a small (N 5 80) group of gastrointestinal assessed daily using standardized and validated instru- surgery patients randomized, not blinded, for 5 mg haloper- ments. Outcome data were relatively complete, and few idol per day or saline solution.18 Another found a significant patients were lost to follow-up. Moreover, haloperidol improvement in the symptoms of delirium and cognition in prophylaxis was well tolerated, and the extensive clinical patients with acquired immunodeficiency syndrome treated experience with this antipsychotic drug in combination with low-dose haloperidol and chlorpromazine for delirium.37 with its low costs enhances the extent to which this Table 3. Results of Patients Who Developed Delirium, According to Study Group: Intention-to-Treat Group Note: Because of rounding, percentages may not total 100.
implementation of proactive geriatric consultation, pro- vided to all the patients in both groups, may have stimulated the attention for predisposing and precipitatingfactors for delirium of the nursing and medical staff of theparticipating wards. In turn, it could have led to extra care for at-risk patients, thereby decreasing the likelihood ofincident delirium. Such an effect may have caused the results to tend toward the null hypothesis. Second, the studyincluded far more patients who were at intermediate risk than at high risk for delirium. This may have resulted infewer incident cases of delirium, decreasing discriminating power. Third, patients were treated with a low dosage of haloperidol prophylaxis, which may explain the absence of a clear difference in the incidence of delirium between Figure 2. Mean of Delirium Rating Scale, revised version-98 patients receiving haloperidol and placebo prophylaxis.
(DRS-R-98) scores during delirium for treatment groups: x- The choice of 1.5 mg/24 hours dosage of haloperidol was axis 5 days; y-axis 5 mean DRS-R-98 scores of patients with based on the average starting dose for treatment of delirious delirium. The mean of the scores on the DRS-R-98 and the 95% older people and the minimal chance of extrapyramidal side confidence intervals in patients with delirium from the haloper- effects with a total daily dose of less than 3 mg.17,24 Another idol (n 5 32) and placebo (n 5 36) groups, respectively. During study found ‘‘extremely low’’ prevalence of extrapyramidal each of the first 3 days after onset of delirium, the severity, as side effects with haloperidol and chlorpromazine treatment measured using the DRS-R-98, was significantly lower in of delirium in patients with acquired immunodeficiency patients who had received haloperidol perioperatively.
syndrome.37 Higher doses of haloperidol may perhapsbe necessary for primary prevention of delirium, but thefrequency and severity of side effects may increase, espe- prevention protocol can be applied in other settings. The cially in vulnerable patient groups. A dose titration and significantly lower severity and shorter duration of delir- duration study might have benefited this study. Fourth, the ium, as well as the fewer total number of days haloperidol effect of the haloperidol prophylaxis could be solely due to patients were hospitalized, than those receiving placebo reduction of certain symptoms of delirium (agitation or strongly suggests that haloperidol prophylaxis contributed hyperactive symptoms), implying that haloperidol does not to the effectiveness of the intervention strategy. These change the underlying nature or course of delirium.
findings support that, when applied to clinical practice, this However, delirium was diagnosed based on clinical judg- strategy combining nonpharmacological and pharmacolo- ment and CAM ratings. A shorter duration of delirium gical interventions may lead to smaller numbers of patients implies a treatment effect beyond the psychomotor symp- who will have delirium and that it may shorten the severity toms. The total score on the DRS-R-98 before the onset of and duration of this neuropsychiatric syndrome.
delirium was not different between the haloperidol and Some limitations of this study need to be addressed.
placebo patients. The effect of sedation on psychomotor This study was underpowered, given the relatively low symptoms from haloperidol is therefore highly unlikely.
delirium rates. The overall incidence of postoperative Fifth, there was no information about prehospitalization delirium was much lower than was expected based on the mental status, such as from proxy report, because most of literature and previous experience in the study hospital. The the informant questionnaires were not returned, so in theacute (fracture) patients, the study intake MMSE might nothave been reflective. Nevertheless, the MMSE results did not differ in the fracture population from those of theelective patients. Sixth, the study group consisted of elderly orthopedic patients only, which may limit generalizabilityto other settings. Seventh, factors such as medical compli- cations, pain, and use of medications that have central nervous system (CNS) side effects may have influenced thesecondary outcome measures that proved to be signifi- cantFseverity of delirium, duration of delirium, and number of days in the hospital. However, but this was a randomized, controlled study, and if there were confound- Figure 3. Proportion of patients suffering from delirium during ing variables, they would have been expected to be present the first 2 weeks postoperatively: x-axis 5 days; y-axis 5 propor- in both treatment conditions alike. Moreover, a retro- tion of patients with delirium. The flags indicate the upper and spective review of the medical charts did not show any lower limits of the 95% confidence interval of this proportion evidence of more complications, more pain, or the use of in the haloperidol (n 5 212) and placebo (n 5 218) groups, more painkillers or drugs with CNS side effects in any of the respectively. During the entire postoperative period, this proportion was lower in patients who had received haloperidol Despite all these effects, which may have tended to bias prophylaxis; this difference was significant from Day 5 until Day results toward the null hypothesis, the significant results support the effectiveness of the intervention.
A formal cost-effectiveness analysis was beyond the scope of this study, but it is likely that the intervention 1. Dyer CB, Ashton CM, Teasdale TA. Postoperative delirium. A review of proposed here has the potential to yield net savings in 80 primary data-collection studies. Arch Intern Med 1995;155:461–465.
addition to reducing the burden for patients and caregivers.
2. Galanakis P, Bickel H, Gradinger R et al. Acute confusional state in the elderly This clinical trial holds substantial promise for the following hip surgery: Incidence, risk factors and complications. Int J GeriatrPsychiatry 2001;16:349–355.
prevention and reduction of delirium in hospitalized, 3. Lindesay J, Rockwood K, Rolfson DB. The epidemiology of delirium. In: nonsurgical, older patients too. Further studies are needed Lindesay J, Rockwood K, Macdonald AJ, eds. Delirium in Old Age, 1st Ed.
to determine whether similar beneficial effects can be New York: Oxford University Press, 2002, pp 27–50.
observed in other settings. Moreover, the potential effect of 4. Marcantonio ER, Flacker JM, Michaels M et al. Delirium is independently associated with poor functional recovery after hip fracture. J Am Geriatr Soc haloperidol prophylaxis on other patient outcome mea- sures, such as morbidity, mortality, institutionalization, and 5. Marcantonio ER, Simon SE, Bergmann MA et al. Delirium symptoms in post- long-term cognitive functioning, deserves further study.
acute care: Prevalent, persistent, and associated with poor functional recovery.
J Am Geriatr Soc 2003;51:4–9.
6. McCusker J, Cole MG, Dendukuri N et al. Does delirium increase hospital stay? J Am Geriatr Soc 2003;51:1539–1546.
7. O’Keeffe S, Lavan J. The prognostic significance of delirium in older hospital patients. J Am Geriatr Soc 1997;45:174–178.
The principle investigators designed this study, which 8. Berggren D, Gustafson Y, Erikson B. Postoperative confusion after anaesthesia the Medical Center Alkmaar funded. Kees J. Kalisvaart in elderly patients with femoral neck fractures. Anesth Analg 1987;66:497–504.
had full access to all the data in the study and takes 9. Williams-Russo P, Urquhart BL, Sharrock NE et al. Post-operative delirium: Predictors and prognosis in elderly orthopedic patients. J Am Geriatr Soc 1992; responsibility for the integrity of the data and the accuracy 10. Inouye SK, Viscoli CM, Horwitz RI et al. A predictive model for delirium in We would like to thank Piet Bartels, PhD (conception, hospitalized elderly medical patients based on admission characteristics. Ann design, data analysis), Adrie Steenhoek, PhD (conception 11. Cole MG, Primeau FJ, Elie LM. Delirium: Prevention, treatment, and outcome and design), Tjeerd van der Ploeg, PhD (data analysis), studies. J Geriatr Psychiatry Neurol 1998;11:126–137.
Liesbeth Geiteman, RN (data acquisition), and Folkert K.
12. Inouye SK, Bogardus ST Jr, Charpentier PA et al. A multicomponent de Boer (data acquisition) for their work on the study and intervention to prevent delirium in hospitalized older patients. N Engl J Med the staff of the orthopedic and surgical units and emergency 13. Marcantonio ER, Flacker JM, Wright RJ et al. Reducing delirium after hip department for their constant alertness, help, and enthu- fracture: A randomized trial. J Am Geriatr Soc 2001;49:516–522.
14. Milisen K, Foreman MD, Abraham IL et al. A nurse-led interdisciplinary intervention program for delirium in elderly hip-fracture patients. J Am Author Contributions: Kees J. Kalisvaart, MD: con- 15. O’Keeffe S. The prevention of delirium. In: Lindesay J, Rockwood K, tent; concept and design; acquisition of data; analysis and Macdonald AJ, eds. Delirium in Old Age, 1st Ed. New York: Oxford Uni- interpretation of data; drafting of the manuscript; critical revision of the manuscript for intellectual content; admin- 16. Cole MG, Primeau F, McCusker J. Effectiveness of interventions to prevent istrative, technical, and material support; supervision; delirium in hospitalized patients: A systematic review. Can Med Assoc J1996;155:1263–1268.
statistical analysis. Jos F. M. de Jonghe, PhD: content, 17. Practice guideline for the treatment of patients with delirium. American conception and design, analysis and interpretation of data, Psychiatric Association. Am J Psychiatry 1999;156:1–20.
drafting of the manuscript, critical revision of the manu- 18. Kaneko T, Cai J, Ishikura T et al. Prophylactic consecutive administration of script for intellectual content, supervision, statistical haloperidol can reduce the occurrence of postoperative delirium in gastro-intestinal surgery. Yonaga Acta Med 1999;42:179–184.
analysis. Marja J. Bogaards, PharmD: content; concept 19. Ikarashi Y, Takahashi A, Ishimaru H et al. Suppression of cholinergic activity and design; acquisition of data; analysis and interpretation via the dopamine D2 receptor in the rat striatum. Neurochem Int of data; critical revision of the manuscript for intellectual content; administrative, technical, material support; super- 20. Ikarashi Y, Takahashi A, Ishimaru H et al. Regulation of dopamine D1 and D2 receptors on striatal acetylcholine release in rats. Brain Res Bull 1997;43:107–115.
vision. Ralph Vreeswijk, RN, MSc: content; acquisition of 21. Itil T, Fink M. Anticholinergic drug-induced delirium: Experimental modifica- data; analysis and interpretation of data; drafting of the tion, quantitative EEG and behavioral correlations. J Nerv Ment Dis 2004; manuscript; critical revision of the manuscript for intellec- tual content; administrative, technical, material support.
22. Platt MM, Breitbart W, Smith M et al. Efficacy of neuroleptics for hypoactive delirium. J Neuropsych Clin Noeurosc 2004;6:66–67.
Toine C. G. Egberts, PhD: content, concept and design, 23. Cummings JL. Physostigmine ameliorates the delusions of Alzheimer’s disease.
analysis and interpretation of data, drafting of the manu- script, critical revision of the manuscript for intellectual 24. Hassan E, Fontaine DK, Nearman HS. Therapeutic considerations in the content, supervision, statistical analysis. Bart J. Burger, management of agitated or delirious critically ill patients. Pharmacotherapy1998;18:113–129.
MD, PhD: content, concept and design, acquisition of data, 25. Inouye SK, Charpentier PA. Precipitating factors for delirium in hospitalized drafting of the manuscript, critical revision of the manu- elderly persons. Predictive model and interrelationship with baseline vulner- script for intellectual content, supervision. Piet Eikelen- boom, MD, PhD: content, concept and design, acquisition 26. Knaus WA, Draper EA, Wagner DP et al. APACHE II. A severity of disease classification system. Crit Care Med 1985;13:818–829.
of data, analysis and interpretation of data, critical revision 27. Folstein MF, Folstein SE, McHugh PR. ‘Mini-mental state’. A practical method of the manuscript for intellectual content, supervision.
for grading the cognitive state of patients for the clinician. J Psychiatr Res Willem A. van Gool, MD, PhD: content, analysis and interpretation of data, drafting of the manuscript, critical 28. Jorm AF, Jacomb PA. The Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE): Socio-demographic correlates, reliability, validity and revision of the manuscript for intellectual content, super- some norms. Psychol Med 1989;19:1015–1022.
vision, statistical analysis, drafting of the figures.
29. Hetherington R. The Snellen chart as a test of visual acuity. Psychol Forsch 30. Yesavage JA, Brink TL, Rose TL et al. Development and validation of a 34. Inouye SK, van Dyck CH, Alessi CA et al. Clarifying confusion: The confusion geriatric depression screening scale: A preliminary report. J Psychiatr Res assessment method. A new method for detection of delirium. Ann Intern Med 31. Mahoney FI, Barthel DW. Functional evaluation: The Barthel index. Md State 35. Trzepacz PT, Baker RW, Greenhouse J. A symptom rating scale for delirium.
32. Barnes TR. A rating scale for drug-induced akathisia. Br J Psychiatry 1989; 36. Cummings JL. Clinical Neuropsychiatry. Orlando, FL: Grune & Stratton, 1985.
37. Breitbart W, Marotta R, Platt MM et al. A double-blind trial of haloperidol, 33. Diagnostic and Statistical Manual of Mental Disorders, 4th Ed., Text Revision.
chlorpromazine, and lorazepam in the treatment of delirium in hospitalized Washington, DC: American Psychiatric Association, 2000.
AIDS patients. Am J Psychiatry 1996;153:231–237.

Source: http://umg.umdnj.edu/gim/consult_manual/Haldol%20prophylaxis%20for%20delirium%202005.pdf