DELIRIUM REMAINS THE most common and devastating neuropsychiatric complication in patients with advanced cancer.1 Delirium is invariably multifaceted in nature, especially in the setting of advanced cancer. The standard management approach involves an investigation for potentially reversible precipitants. Although it has been noted that a delirium episode is reversible in up to 50% of cases with appropriate management,2 such reversibility depends on the nature of underlying eliciting factors. In the clinical setting, especially in the consultation liaison psychiatry arena, it is often difficult to identify the etiology of delirium in the patients with advanced cancer. Antipsychotic medications have frequently been regarded as the treatment of choice for delirium. Some studies show efficacy in terms of overall delirium score reduction.3,4 A review found that haloperidol, risperidone, and olanzapine are equally effective in treating delirium with few adverse effects.5 In studies of delirium, however, few investigations have assessed concomitant medical conditions. One of the present research questions was whether pharmacological interventions for delirium in cancer patients simply reduce agitation and/or perceptual disturbance while affecting a delirium episode indirectly by improvement in ‘general medical condition’. In this study we examined the clinical efficacy of risperidone for the treatment of delirium, combined with a repeated assessment of underlying medical severity levels. Also we compared the specific delirium symptoms changes before and after risperidone treatment.
Aim: Antipsychotic medications have frequently been regarded as the treatment of choice for delirium. This study examined the clinical efficacy of risperidone for the treatment of delirium in cancer patients, combined with a repeated assessment of underlying medical severity levels.
Methods: The study included consecutive referrals of 29 delirious cancer patients (mean age, 68.9 ± 12.5 years; male, 69%) to the psychiatric consultation service. Risperidone was given orally once per day (mean dosage, 1.4 ± 1.3 mg/day). Study participants were assessed using quantitative standardized scales of cognitive function, delirium, and physical impairment at baseline and at the end of the study (seventh day).
Results: Risperidone with routine clinical management was effective for the treatment of delirium: 48% of the patients responded and 38% achieved remission. The reduction of delirium severity occurred in 79% of the patients. Changes in delirium severity were unrelated to age, gender, general cognitive dysfunction, or to severity of attendant medical conditions. In addition to changes in agitation and perceptional disturbances, risperidone was also effective for other specific delirium symptoms.
Conclusions: Risperidone with routine clinical management is effective in the treatment of delirium in advanced cancer patients, independent of changes in the underlying medical condition.
The study included consecutive referrals of delirious cancer patients to the psychiatric consultation service of Tokai University Hospital, in Kanagawa, Japan. The diagnosis of delirium was made by staff psychiatrists with reference to DSM-IV-TR diagnostic criteria. Patients with any type of dementia were excluded. This study was approved by the Institutional Review Board at Tokai University. After obtaining informed consent from study participants (either when they regained mental capacity or by proxy), medication was implemented. Risperidone (oral solution) was given orally once per day (at bedtime). Risperidone was initiated at 0.5–1 mg. Subsequent titration was based on clinical judgment during the 7 days. Risperidone was allowed only for the control of agitation. Measurements were obtained at baseline and at the end of the study (seventh day).
Thirty-one patients were enrolled in this study. Among those, two patients were discharged before the end of this study; a total of 29 patients completed the study. The mean dosage of risperidone during the study period was 1.4 ± 1.3 mg/day. Patient demographic characteristics are presented in Table 1; most were advanced cancer patients. Variables that may be causally linked with delirium are also listed in Table 1. The mean number of possible causes was 1.9 (range, 0–4).
|n = 29|
|Age (years), mean ± SD||68.9 ± 12.5|
|Sex, male n (%)||20 (69.0)|
|Primary cite of cancer, n (%)|
|Stage IV or recurrence, n (%)||25 (86)|
|ECOG PS ≥3||26 (90)|
|Potential etiologies of delirium, n (%)|
|Opioids use||17 (58.6)|
|Renal failure||7 (24.1)|
|Respiratory failure||6 (20.7)|
|Respiratory failure||6 (20.7)|
|Brain metastasis||4 (13.8)|
|Post operation||3 (10.3)|
|Liver failure||3 (10.3)|
|Severe pain||2 (6.9)|
|Drug intoxication/withdrawal||1 (3.4)|
Study participants were assessed using the following quantitative standardized scales of cognitive function, delirium, and physical impairment at baseline and at the end of the study (seventh day) by a fully trained psychiatrist (YK, MK or TO). The raters in this study were already trained fully in the scale and were experienced in its use.
Delirium Rating Scale–Revised
The Delirium Rating Scale–Revised (DRS-R98) is a 16-item clinician-rated scale with anchored item descriptions corresponding to both symptoms and temporal aspects of delirium.6 The severity scale has 13 items each rated from 0 to 3; the sum has a maximum of 39 points and higher scores indicate greater severity of delirium. Three additional items (rated from 0 to either 2 or 3) capture temporal course and attribution to an underlying etiology, and when added to the sum of the 13 symptom items produce the DRS-R98 Total score that ranges from 0 to 46. This scale has been validated in Japanese subjects.7 The best cut-off score for differentiation of delirium and non-delirium for the DRS-R98 total score is 14.5 and for the DRS-R98 severity score, 10.0. In this study, the remission of delirium was defined as a DRS-R98 severity scale score ≤10 points. The treatment response was defined as a 25% reduction of DRS-R98 severity scale scores from baseline to the seventh day.8
Mini-Mental State Examination
The Mini-Mental State Examination (MMSE) is a widely used reliable and valid means of assessing generalized cognitive dysfunction as occurs in delirium or dementia, 9 and has been validated in Japanese subjects.10 MMSE scores range from 0 to 30, and a cut-off score of ≤23 indicates significant cognitive impairment.
Clinical Global Impression Severity scale for Delirium
Overall severity of delirium was also assessed using the Clinical Global Impression Severity Scale for Delirium (CGI-S-D) scale,11 which captures global severity based on the clinician's experience with delirious patients in general. The CGI-S-D is scored as a single overall impression of illness severity on a Likert-type scale ranging from 1 to 7, where 1 in normal or not ill at all and 7 extremely ill.
Acute Physiology and Chronic Health Evaluation
To assess the course of severity of medical illness, the vital signs and laboratory abnormalities score of the Acute Physiology and Chronic Health Evaluation (APACHE)-III was used.12 The APACHE-III acute physiology scores (APACHE-III APS) were calculated at baseline and at the end of the study (seventh day). The APACHE-III APS assessed pulse, mean blood pressure, temperature, respiratory rate, partial pressure of oxygen in the blood, hematocrit, white blood cell count, creatinine, urine output, blood urea nitrogen, sodium, albumin, bilirubin and glucose. Higher scores indicate greater severity of acute medical illness. It has been reported that the severity of medical illness, assessed by the APACHE system, is positively correlated with alteration of mental status including delirium.13–15
Eastern Cooperative Oncology Group Performance Status assessment
Patient daily living performance status was assessed using the Eastern Cooperative Oncology Group Performance Status assessment (ECOG PS).16 The ECOG PS ranges from 0 to 5, with higher scores indicating more trouble with activities of daily living.
Drug-Induced Extra Pyramidal Symptoms Scale
Severity of extra-pyramidal symptoms was evaluated by means of the Drug-Induced Extra Pyramidal Symptoms Scale (DIEPSS).17
Data were analyzed using spss 12.0 (SPSS, Chicago, IL, USA). T-tests (paired, 2-tailed) were performed as well as regression analysis of DSR-R98 change scores. The significance level was set at P < 0.05.
Table 2 lists the mean scores on each of the scales at the entry and at the 7-day follow-up. At the 7-day follow-up, the scores of the DRS-R98 and CGI-S-D were significantly improved and those of the MMSE had a trend toward a significant improvement. The APACHE-III APS had a trend toward a significant deterioration. The DRS-R98 scores were improved in 23 (79.3%) patients during the study period. The DRS-R scores in four (13.8%) and in two patients (6.9%) were no change and worsened during the study period, respectively.
|DRS-R98||19.8 ± 6.8||14.3 ± 7.8||t = 4.2, d.f. = 28, P < 0.001|
|MMSE||17.2 ± 7.8||18.8 ± 7.5||t = −2.0, d.f. = 28, P = 0.051|
|CGI-D||5.4 ± 0.8||4.8 ± 1.2||t = 3.5, d.f. = 28, P = 0.002|
|APACHE-III APS||16.8 ± 10.5||18.9 ± 9.3||t = −1.8, d.f. = 28, P = 0.078|
Table 3 lists the mean scores on each DRS-R98 items at the entry and at the 7-day follow-up. In addition to changes in agitation and perceptional disturbances, risperidone was also effective for other specific delirium symptoms except motor retardation and long-term memory disturbance.
|Sleep–wake cycle disturbance||2.0 ± 0.5||1.5 ± 0.7||t = 3.8, d.f. = 28, t < 0.01|
|Perceptual disturbances and hallucinations||1.8 ± 1.2||0.9 ± 1.0||t = 4.1, d.f. = 28, P < 0.01|
|Delusions||1.6 ± 1.1||0.8 ± 1.0||t = 4.4, d.f. = 28, P < 0.01|
|Lability of affect||1.5 ± 1.0||0.8 ± 0.8||t = 4.2, d.f. = 28, P < 0.01|
|Language||1.3 ± 0.8||0.9 ± 0.9||t = 2.5, d.f. = 28, P = 0.02|
|Thought process abnormalities||1.5 ± 0.9||1.3 ± 0.9||t = 1.8, d.f. = 28, P = 0.08|
|Motor agitation||1.2 ± 1.1||0.7 ± 0.9||t = 2.4, d.f. = 28, P = 0.02|
|Motor retardation||1.1 ± 0.9||1.2 ± 0.9||t = −0.5, d.f. = 28, P = 0.60|
|Orientation||1.3 ± 0.8||1.0 ± 1.0||t = 2.1, d.f. = 28, P = 0.04|
|Attention||1.8 ± 0.7||1.3 ± 0.8||t = 2.8, d.f. = 28, P = 0.01|
|Short-term memory||1.9 ± 1.0||1.4 ± 1.2||t = 2.4, d.f. = 28, P = 0.02|
|Long-term memory||1.7 ± 1.1||1.6 ± 1.2||t = 0.5, d.f. = 28, P = 0.66|
|Visuospatial ability||1.1 ± 1.2||1.0 ± 1.3||t = 0.3, d.f. = 28, P = 0.74|
Table 4 lists the scores at the entry and at the 7-day follow-up between responders versus non-responders and remitted versus non-remitted. The response rate of risperidone, measured by a 25% reduction of DRS-R98 severity scale scores, was 48.3% (n = 14). There were no significant differences between the responders and non-responders at entry in age, DRS-R98, MMSE, CGI-S-D or APACHE-III APS scores. Eleven patients (37.9%) achieved remission (DRS-R98 severity scale scores ≤10 points). At entry, the patients who later achieved remission had higher MMSE (P = 0.033), lower DRS-R98 (P = 0.053), and lower CGI-S-D (P = 0.053) scores than those who did not. There were no significant differences in age or the APACHE-III APS.
|Responders (n = 14)|
|DRS-R98||20.3 ± 6.8||9.4 ± 4.3||t = 7.4, d.f. = 13, P < 0.001|
|MMSE||18.3 ± 7.4||21.8 ± 5.5||t = −3.0, d.f. = 13, P = 0.010|
|CGI-D||5.4 ± 0.7||3.9 ± 1.0||t = 7.4, d.f. = 13, P < 0.001|
|APACHE-III APS||15.6 ± 11.1||18.4 ± 10.9||t = −1.2, d.f. = 14, P = 0.246|
|Non-responders (n = 15)|
|DRS-R98||19.3 ± 7.1||18.9 ± 7.7||t = 0.5, d.f. = 14, P = 0.657|
|MMSE||16.2 ± 8.2||16.0 ± 8.2||t = 0.2, d.f. = 14, P = 0.813|
|CGI-D||5.5 ± 0.8||5.6 ± 0.7||t = −0.4, d.f. = 14, P = 0.719|
|APACHE-III APS||17.7 ± 10.1||19.3 ± 7.9||t = −1.5, d.f. = 14, P = 0.162|
|Remitted (n = 11)|
|DRS-R98||16.6 ± 6.6||7.3 ± 2.1||t = 4.3, d.f. = 10, P = 0.001|
|MMSE||21.1 ± 7.3||24.4 ± 4.5||t = −3.0, d.f. = 10, P = 0.014|
|CGI-D||5.1 ± 0.5||3.6 ± 0.9||t = 4.7, d.f. = 10, P = 0.001|
|APACHE-III APS||16.3 ± 10.5||17.5 ± 8.3||t = −0.5, d.f. = 10, P = 0.657|
|Non-remitted (n = 18)|
|DRS-R98||21.1 ± 6.4||18.6 ± 6.9||t = 2.2, d.f. = 17, P = 0.039|
|MMSE||14.8 ± 7.3||15.4 ± 7.0||t = −0.6, d.f. = 17, P = 0.587|
|CGI-D||5.7 ± 0.8||5.4 ± 0.9||t = 1.2, d.f. = 17, P = 0.260|
|APACHE-III APS||17.1 ± 10.7||19.7 ± 10.0||t = −2.7, d.f. = 17, P = 0.016|
Table 5 lists the mean scores on each of the scales at the entry and at the 7-day follow-up among the patients with deteriorated medical conditions during the study period. Among the 29 patients, 18 had medical conditions assessed as deteriorated on APACHE-III APS. There were no significant differences in the response and remission rates between the patients whose medical conditions deteriorated compared to those without deterioration.
|DRS-R98||19.8 ± 7.1||15.4 ± 8.9||t = 3.0, d.f. = 17, P = 0.008|
|MMSE||16.7 ± 8.0||18.0 ± 8.4||t = −2.8, d.f. = 17, P = 0.012|
|CGI-D||5.4 ± 0.8||4.8 ± 1.3||t = 2.8, d.f. = 28, P = 0.012|
|APACHE-III APS||14.3 ± 9.8||19.6 ± 9.8||t = −6.15, d.f. = 28, P < 0.001|
To evaluate possible competing explanations for reduced delirium over the 7-day period, age, gender, MMSE, CGI-S-D and APACHE-III APS were regressed on the pre–post-change scores of the DSR-R98. This yielded a non-significant R2 of 0.142 (F(5,23) = 0.762, NS). Beta (β) weights are standard partial regression coefficients, meaning that they are in standard score units with the other independent variables held constant. They represent the relative contribution of the independent variables in the multiple regression equation (e.g. the direct contributions are derived by squaring the β coefficients). None of the β coefficients for the aforementioned variables was significant: age, −0.116 (P = 0.60); gender, 0.340 (P = 0.133); MMSE, −0.227 (P = 0.414); CGI-S-D, −0.420 (P = 0.155); and APACHE-III APS, −0.089 (P = 0.657). These data indicate that changes in delirium rating scores were unrelated to age, gender, general cognitive dysfunction, or to severity of attendant medical conditions. Correspondingly, greater tenability is rendered for the effects of risperidone in the attenuation of delirium symptoms.
No patient developed movement disorders as assessed on the DIEPSS. Mild sedation occurred in one patient.
The present findings suggest that risperidone was effective for the treatment of delirium, independent of changes in the underlying medical condition. In addition to changes in agitation and perceptional disturbances, risperidone was also effective for other specific delirium symptoms (Table 3). The MMSE scores related to cognitive function were also improved (P = 0.051).
It is important to acknowledge the methodological limitations of the study. First, because this investigation involved delirious patients referred for psychiatric consultation at one university hospital, the findings may not be applicable to advanced cancer patients in general. Second, we assessed medical severity changes using the APACHE-III APS. This scale was not specifically validated for participants of the type used in the present study, although APACHE scores have been shown to be correlated with alteration of mental status including delirium.13–15 Correlations between the DRS-R98 and APACHE-III APS calculated at entry and on the seventh day were not significant. This suggests that levels of delirium were not associated with severity of acute medical symptoms. An aim of the current investigation was to ascertain if reduction of delirium over time was related to possible improvements in the patient's general medical condition. The present findings suggest that delirium severity was unrelated to acute medical conditions stemming from cancer. We would contend that APACHE-III APS is a valid measure for cancer-related symptoms and that delirium changed independent of those symptoms. Third, we did not evaluate non-pharmacological management of delirium. Future research might address other types of therapy in interaction with pharmacological interventions for delirium. Fourth, we did not systematically measure electrocardiogram parameters and QTc duration as safety measures in this study. In real-world delirium practice, antipsychotics are considered the standard pharmacologic treatment for delirium.18,19 The Food and Drug Administration in the USA is, however, notifying health-care professionals that both conventional and atypical antipsychotics are associated with an increased risk of mortality in demented elderly patients, who are at high risk of delirium.20 The risk of cardiovascular death is the highest.21 Furthermore, a recent study showed that long-term exposure to antipsychotics has a subtle but measurable influence on brain tissue loss over time.22 It has also been shown that a significant, reversible decrease in striatal volume, powerfully predicting neurological adverse effects, can occur within 2 h after being treated i.v. with haloperidol.23 This finding suggests the importance of careful risk-–benefit review of off-label antipsychotic use for delirium treatment. Finally, we did not assess the motor subtypes of delirium as possible antecedents in recovery from delirium. Further studies using the validated delirium subtype scale24 should be conducted in the future.
Given these caveats, several important findings emerged. This study showed that 50% of the delirious patients responded to the risperidone treatment, and that 40% with advanced cancer achieved remission. The reduction of the DRS-R98 scores occurred in 80% of the patients. This reduction rate is essentially identical to other risperidone studies focused on delirium.25,26 Even in the patients with advanced cancer, this study shows that risperidone has response rates comparable to other medical illnesses. Furthermore, improvement in delirium occurred even in the patients with deteriorated medical conditions.
The results of this open-label study indicate that with routine clinical management risperidone is effective in the treatment of delirium in advanced cancer patients. Because we did not assess the effects of routine clinical management and did not have a control group taking another medication or placebo, randomized controlled studies are indicated to confirm the present results.
None of the authors has any conflict of interest to disclose.