Introduction
Delirium lies at the interface between medicine and psychiatry. It is generally characterized by acute disturbances of consciousness, cognition, and perception that are precipitated by an underlying medical condition. Delirium occurs at rates ranging from 10% to 30% of all general hospital admissions. It is a negative prognostic indicator, often leading to longer hospital stays and higher mortality. Delirium is also associated with increased functional decline and morbidity, particularly if it is unrecognized and untreated. The etiology of delirium is multifactorial and many causes have been suggested (Attard et al., Reference Attard, Ranjith and Taylor2008).
Antipsychotics are the mainstay of pharmacological treatment and have been shown to be effective in treating symptoms of both hyperactive and hypoactive delirium, as well as generally improving cognition. Haloperidol was for many decades considered to bethe first-line treatment. Potential benefits of second generation antipsychotics (SGAs) compared with typical antipsychotics include the lower propensity to cause over-sedation and movement disorder. Of the SGAs investigated in delirium, most data support the use of risperidone and olanzapine. Other SGAs have been evaluated but data are limited (Schwartz and Massand, Reference Schwartz and Masand2002).
Recently, 14 studies were analyzed focusing on the use of antipsychotics for treating delirium. Study medications included haloperidol, chlorpromazine, olanzapine, risperidone and quetiapine. Improvements in delirium severity were reported with all of these antipsychotic medications. The improvements in delirium tended to occur soon after initiation of treatment, and most of the studies examined used relatively low doses of antipsychotic medication (Seitz et al., Reference Seitz, Gill and van Zyl2007). Furthermore, in an effort to identify the antipsychotic with the greatest efficacy and safety when used for the management of delirium, prospective, randomized, controlled trials comparing the clinical effects of antipsychotic therapy with placebo or comparing two antipsychotic treatments in an acute care setting were analyzed. Four prospective trials were included in which the authors concluded that antipsychotic drugs are efficacious, when compared with baseline, and safe for the treatment of delirium (Lacasse et al., Reference Lacasse, Perreault and Williamson2006).
Risperidone is the most widely used SGA in the Western world. Thus, it is of particular importance to evaluate its use in the treatment of delirium. Generally, most reports have concluded that there are no significant differences in the efficacy or response rate between typical antipsychotics (including haloperidol) and risperidone in the treatment of delirium and that such treatment is safe (Horikawa et al., Reference Horikawa2003; Han and Kim, Reference Han and Kim2004; Mittal et al., Reference Mittal2004; Liu et al., Reference Liu, Juang, Liang, Lin and Yeh2004; Parellada et al., Reference Parellada, Baeza, de Pablo and Martinez2004).
Due to the “popularity” of haloperidol as the first drug of choice in patients suffering from delirium the introduction of risperidone as a first line treatment warranted comparison between these two compounds. One large comparison was conducted retrospectively with medical records of 266 Japanese delirium inpatients. The incidence of adverse events was 6.5% for risperidone and 31.4% for haloperidol, while the incidence of death during delirium was 3.2% for risperidone and 2.1% for haloperidol. The authors concluded that the use of haloperidol as a first-line drug in delirium patients is no longer based on relevant risk-benefit assessment (Miyaji et al., Reference Miyaji, Yamamoto, Hoshino, Yamamoto, Sakai and Miyaoka2007). These conclusions were echoed by a metanalysis undertaken by Rea and colleagues (Reference Rea, Battistone, Fong and Devlin2007). They supported the findings from other authors that atypical antipsychotics are as efficacious as haloperidol but cautioned that these conclusions are based on a very limited number of studies (Rea et al., Reference Rea, Battistone, Fong and Devlin2007). Finally, researchers at the Cochrane Collaboration compared the efficacy and incidence of adverse effects of haloperidol with risperidone, olanzapine and quetiapine in the treatment of delirium. Only three studies were found that satisfied selection criteria. These studies compared haloperidol with risperidone, olanzapine and placebo. There was no evidence that haloperidol in low dosage (<4.5 mg/daily) has a different efficacy in comparison with the atypical antipsychotics olanzapine and risperidone in the management of delirium or a greater frequency of adverse drug effects. Again the authors cautioned that these conclusions must be tempered by the observation that they are based on small studies of limited scope (Lonergan et al., Reference Lonergan, Britton, Luxenberg and Wyller2007).
The aim of the present study was to compare the outcome of delirium treatment between SGAs and classical antipsychotic medications in a large, university-affiliated general hospital. Particular care was taken to ensure that the analysis reflected “real-life” clinical conditions and treatment decisions.
Methods
We conducted a one-year retrospective analysis of all medical records of adult patients (18 years of age and older) suffering from delirium admitted to the Wolfson Medical Center, acute internal medicine ward. This study was approved by the Abarbanel Mental Health Center's Internal Review Board (IRB). The computerized database was searched for all admissions of patients whose admission or discharge diagnoses included delirium. Between January 2007 and December 2007, there were 191 admissions fulfilling the following inclusion criteria: (i) aged 18 years or older, (ii) International Classification of Disease, 9th edition, Clinical Modification (ICD-9-CM) diagnosis of delirium, and (iii) availability of data about placement following discharge.
It should be noted that the diagnosis of delirium in all cases was made by a board-certified psychiatric consultant. Treatment by antipsychotics was initiated by the consulting psychiatrist and duration and dose titration were undertaken according to clinical response at the discretion of the treating psychiatric consultant. The type of medication was randomly chosen – there were no specific criteria for use of risperidone, classical antipsychotic or not to treat with an antipsychotic.
Computerized patient files were reviewed and the following data extracted: age, gender, diagnosis (at admission and at discharge), residence prior to admission, comorbid psychiatric disorders, presence of physical comorbidity and present or past alcohol and drug abuse, type of treatment for delirium (medication, dose and duration), length of hospital stay (LOS), place of discharge and mortality.
Patients who had been treated with risperidone were defined as the index group. It should be noted that at the time the study was undertaken, only risperidone of the second generation antipsychotics was indicated for the treatment of delirium in Israel. The comparison groups were comprised of patients suffering from delirium who were either treated with a classical antipsychotic (CA) or who did not receive any antipsychotic treatment during the episode of delirium (NT).
Statistical analysis
Patients and comparisons were compared on all demographic, clinical and outcome variables. Data were analyzed using the independent-samples approach. The two-tailed t-test and non-parametric test (Wilcoxon) were undertaken to test for differences between the evaluations for quantitative parameters. Examination of differences between the categorical parameters was based on the Pearson and Fisher's exact tests. The analyses were a test of univariate association of independent variables. All tests applied were two-tailed, and a p value of 5% or less was considered statistically significant.
The data were analyzed using the Statistical Analysis System software (SAS Institute, 1990).
Results
During the study period 191 patients were admitted to an acute internal medicine ward and diagnosed as suffering from delirium either upon admission or during their inpatient stay. Mean age for the group was 78.8 ± 1.1 years (Standard Error; SE), ranging between 42 years to 97 years. There were 108 male (56.5%) and 83 (43.5%) female patients in the group. The great majority of patients (165, 86%) had been residing in the community prior to admission while only 26 (14%) were referred from an institution (mostly nursing homes). Psychiatric diagnoses were considered valid only if these had been established prior to admission by a psychiatrist. The most frequent comorbid psychiatric diagnosis was that of dementia in 106 (56%) of patients followed by depression in 10 (5.3%) patients and single case diagnoses of post-traumatic stress disorder, anxiety or schizophrenia.
Comorbid physical conditions and disorders were frequent and often several disorders were diagnosed in a single patient. The most common conditions were as follows: cardiovascular (75 patients), sepsis (24), malignancy (22), metabolic (22), diabetes (18), chronic obstructive pulmonary disease (16), anemia (14) and cerebrovascular (14).
Treatment and outcome
The majority of patients suffering from delirium were treated by an antipsychotic medication (147, 77 %). The distribution between risperidone (73 patients) and CA (74 patients) was similar. The most frequently used CA was haloperidol (59/74) followed by perphenazine (11/74) and sulpiride (4/74). The mean risperidone dose was 1.7 ± 0.4 mg/daily. The mean haloperidol dose was 7.8 ± 1.9 mg/daily. Mean duration of antipsychotic treatment (days) was 3.8 days for the CA group and 2.6 for the risperidone group (p = 0.04). Both groups were treated until complete resolution of the delirium episode. The duration of treatment was measured as the time between the first and last dose of antipsychotic or between diagnosis of delirium and its complete resolution in the untreated group. In the untreated (NT) group the mean length of the delirium episode lasted 5.6 days, which was significantly longer than that of either treated groups (p = 0.02).
Mean LOS (days) for all patients was 12.8 ± 1.4 (SE). In the risperidone treated group the mean LOS was 10.8 days while in the CA group it was 13.53 days and in the NT group it was 15.5 days (p = 0.008). Fourteen patients (7.3%) passed away during the delirium episode: ten in the NT group, three in the CA group and one in the risperidone treated group (p = 0.04). The majority of patients (123, 64.7%) were discharged back to the community while (57, 29%) were discharged to a long-stay institution. There were no differences in placement between the various treatment groups.
Discussion
The first successful case reports of risperidone in the treatment of delirious patients were published in 1997 (Sipahimalani and Masand, Reference Sipahimalani and Masand1997). Since these case reports several studies have examined the efficacy and safety of risperidone in delirious patients. Focus on the use of SGAs in the treatment of delirium has intensified even further in recent years. In 2005, 13 studies examined the use of risperidone, olanzapine and quetiapine, and two cases were reported focusing on ziprasidone. Risperidone, olanzapine and quetiapine were found to be similarly effective in the treatment of deliriumand their use to be safe and carry a low burden of side effects (Boettger and Breitbart, Reference Boettger and Breitbart2005; Onur and Cimilli, Reference Onur and Cimilli2005).
In the present study we have retrospectively analyzed the outcomes (LOS, duration of treatment, discharge target and mortality) of delirium treatment in a large general hospital essentially comparing risperidone treatment to haloperidol or no treatment. Our findings are similar to those reported by Miyaji and colleagues (Reference Miyaji, Yamamoto, Hoshino, Yamamoto, Sakai and Miyaoka2007). We found that risperidone treatment of delirium resulted in shorter time to resolution of delirium, shorter overall length of stay in the internal ward and less morbidity than haloperidol or no treatment. It should, however, be noted that one of the limitations of the present study was that the mean dose of haloperidol was in the range that the Cochrane reviewers define as “high” and thus may have biased the findings reported here. Other limitations are the study's retrospective design, the rather large proportion of untreated patients, and the lack of quantitative outcome measures. However, this work does have several advantages: the naturalistic settings reflect “real-life” patient management, all diagnoses and treatment were carried out by a board-certified psychiatristic consultant, and computerized medical records enabled tracking of discharge outcomes.
We may tentatively conclude that the present study, despite the above methodological limitations, adds to the growing body of evidence favoring risperidone as the drug of choice for the treatment of delirium. However, due to the limitations inherent in a retrospective analysis, prospective large-scale trials are needed to support this recommendation.
Conflict of interest
None.
Description of authors’ roles
I. Briskman designed the study, supervised the data collection and assisted in writing the paper. R. Dubinski and Y. Barak collected the data and wrote the paper.
