Antipsychotics and dopamine transporter gene polymorphisms in delirium patients


Dr Changsu Han, Department of Psychiatry, College of Medicine, Korea University, Ansan Hospital, 516 Gojan-Dong, Ansan City, Kyunggi Province 425-707, South Korea. Email:


Abstract  The main objective of the present study was to determine the relationship between treatment responses of delirium and genetic polymorphisms in the dopamine transporter. The optimal dosages of haloperidol and risperidone in the treatment of delirium were also investigated. Either haloperidol or risperidone was administered to delirium patients, and delirium symptoms were measured daily until remission. Variable number of tandem repeat (VNTR) polymorphisms of the dopamine transporter were determined using the polymerase chain reaction. Among 42 subjects, symptoms of delirium appeared a mean of 9.68 days after hospitalization. A majority of the subjects (83.3%) had the type 10/10 polymorphism. Dosages of haloperidol and risperidone at the day of recovery were 1.67 mg/day (SD = 1.32; range 0.5–2.5 mg/day) and 1.19 mg/day (SD = 1.14; range 0.5–5.0 mg/day), respectively. The mean drug response time was 8.5 days in the haloperidol group and 4.8 days in the risperidone group (no significant difference). The response rates at the 3rd and 7th days after medication did not differ with either the drug group or the dopamine transporter polymorphism. Relatively low doses of risperidone and haloperidol exhibited similar efficacies, and dopamine transporter polymorphisms do not appear to play a major role in the action of antipsychotics on delirium.


Delirium can occur in the presence of various clinical conditions, suggesting common neuronal pathways in the central nervous system. According to the neurotransmitter hypothesis, the principal monoamine transmitters in the brain involved in delirium are dopamine, norepinephrine, and serotonin.1 They are all implicated in the control of the sleep–wake cycle and arousal, both of which are disturbed in delirium.2 Increased dopaminergic release and neurotransmission are known to cause psychotic disturbances, and antipsychotic drugs that block dopamine receptors may relieve the psychotic symptoms frequently seen in delirium. Serotonin has also been implicated in the development of psychosis and delirium. Serotonergic neurons are involved in behavioral symptoms such as aggressive and impulsive behavior, mood, and motor activity that may be seen in delirium.3 Both dopamine and serotonin agonists can induce psychosis.4 These factors have resulted in antipsychotics usually being the first-choice drugs in the treatment of delirium.5 Haloperidol is used most frequently, with one report stating that 97% of hospitals use haloperidol as a first-choice drug.6,7 However, classical high-potency antipsychotic drugs such as haloperidol are associated with extrapyramidal side-effects,8 the presence of which interferes with the evaluation of delirium symptoms. Recently, many clinicians have started to use atypical antipsychotic drugs such as risperidone, olanzapine, and quetiapine in delirium patients.9–15 Risperidone, which blocks dopamine D2 receptors and serotonin 5-HT2A receptors, might have superior efficacy in the control of aggressive behavior and mood disturbances of delirium other than psychotic symptoms. Recently, the efficacy of risperidone on delirium was reported to be comparable with that of haloperidol.16 However, the optimal dosages of risperidone in delirium patients have not been determined.

The dopamine transporter (DAT) gene is located on chromosome 5 (5p15.3) and includes a variable number of tandem repeat (VNTR) polymorphisms: as many as 40 bp in the 3′-non-coding region.17 DAT plays an important role in controlling blood levels of do-pamine,18 supporting the assumption that the drug responses of delirium patients vary according to DAT function. It is not yet clear how polymorphisms of VNTR could affect human DAT function. Because the VNTR is in the 3′-non-coding region, allelic variants cannot result in structural or functional differences in the human DAT protein. Recently, Nakamura et al. suggested that VNTR can function as transcriptional regulators.19 The closely related serotonin transporter gene contains a VNTR within a non-coding (intron) region, and allelic differences in the number of VNTR copies are associated with susceptibility to anxiety disorders.20,21 Some authors have suggested that a VNTR of DAT could enhance gene transcription in a similar manner to the serotonin transporter gene.22 However, this awaits further direct experimental evidence. We investigated the optimal dosages of antipsychotics and the association between treatment responses to antipsychotics with DAT gene polymorphisms in delirium patients.


This study was performed at Korea University Ansan Hospital, which includes four medical wards, four surgical wards, two intensive care units, and two oncology wards. The Korea University Ethics Committee approved the research protocol, and subjects or their family members gave written informed consent after the nature of the study and its procedures had been explained. Researchers screened all patients with altered mental status who were referred to the consulting psychiatry division. Subject disease-related and demographic data were collected from their medical records. Screening and detection of delirium was conducted with the Confusion Assessment Method and Delirium Rating Scale-Revised-98 (DRS-R-98).23,24 Diagnosis was determined using the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders (4th edn; DSM-IV)25 according to DSM-IV criteria.26 Patients with any type of dementia or other psychiatric disorder were excluded, as were patients who already had received antipsychotics or benzodiazepines in the emergency room or intensive care unit for their disturbed behavioral problems before the arrival of the consulting psychiatrist. Among the 113 patients screened, 71 patients were excluded for the following reasons: family refused, n = 18; terminal illness, n = 20; prior benzodiazepine usage, n = 7; prior antipsychotics usage, n = 12; and de-mentia, n = 14. The remaining 42 subjects were aged 71.26 ± 7.22 years (mean ± SD, range 60–86 years), and 25 of them (59.5%) were male. Subjects took either haloperidol or risperidone in a flexible-dosage regimen, with the dosage adjusted depending on the severity of delirium throughout the study period. One of the researchers measured the delirium symptoms and side-effects of the drugs every day until remission. The response time was defined as the number of days until the DRS-R-98 severity score was <15.25.24

We amplified the 40-bp VNTR polymorphisms in the 3′ untranslated region of the DAT gene using the polymerase chain reaction as described by Sano et al.27 The primers used for genotyping the DAT allele were 5′-TGTGGTGTAGGGAACGGCCTGAG-3′ (forward) and 3′-CTTCCTGGAGGTCACGGCTCAAGC-3 (reverse).

Statistical analyses were performed using spss version 10.0 (SPSS, Chicago, IL, USA). Hardy–Weinberg equilibrium was analyzed for each sample. Paired and unpaired Student's t-tests, Fisher's exact test, Mann–Whitney U-test, Kolomogorov–Smirnov Z-test, and two-way analysis of variance with repeated measures were used as appropriate.


All subjects had medical and surgical problems (Table 1); 19 (45.2%) had received surgery, eight (19.0%) had undergone chemotherapy just before the onset of delirium, and 14 (33.3%) were smokers. Twenty-six subjects (61.9%) took benzodiazepine for sleep control during the study period. The delirium symptoms began 9.68 ± 11.15 days (range 1–62 days) after hospitalization.

Table 1.  Medical diagnoses
Diagnosisn (%)
Alcohol dependence1 (2.4)
Brain trauma2 (4.8)
Renal failure1 (2.4)
Cancer18 (42.9)
Cardiac insufficiency2 (4.8)
Cerebrovascular accident1 (2.4)
Diabetes1 (2.4)
Electrolyte imbalance1 (2.4)
Bone fracture 9 (21.4)
Infection 5 (11.9)
Osteoarthritis1 (2.4)

Among the 42 subjects, 24 were prescribed haloperidol and 18 (42.9%) received risperidone. There was no significant difference in age between the two groups (t = −0.57, d.f. = 40, P = 0.57). The mean numbers of medications were 5.21 and 1.97 in the haloperidol and risperidone groups, respectively (< 0.001). Subjects recovered from delirium 5.93 ± 4.80 days after beginning medication, and this did not differ statistically between the two drug groups (6.67 ± 5.47 days in the haloperidol group and 4.81 ± 3.43 days in the risperidone group; t = 1.20, P = 0.23). There also was no significant difference in the DRS-R-98 severity scores between the two groups, which were measured daily after delirium onset (Table 2; Figs 1,2).

Table 2.  Total DRS-R-98 scores vs treatment days
 GroupDRS-R-98 sum
(mean ± SD)
  • DRS-R-98, Delirium Rating Scale-Revised-98.

  • Recovery occurred at 6.67 ± 5.47 days in the haloperidol group and 4.81 ± 3.43 days in the risperidone group.

Day 1Haloperidol22.04 ± 3.740.340.73
Risperidone21.61 ± 4.20
Day 2Haloperidol16.79 ± 5.47−0.150.88
Risperidone17.06 ± 5.67
Day 3Haloperidol13.52 ± 6.01−0.320.75
Risperidone14.11 ± 5.86
Day 4Haloperidol12.65 ± 6.830.290.77
Risperidone12.06 ± 5.92
Day 7Haloperidol11.04 ± 3.80−0.250.80
Risperidone11.45 ± 7.32
Recovery dayHaloperidol 8.00 ± 3.02−1.410.17
Risperidone 9.72 ± 4.87
Figure 1.

Delirium Rating Scale-Revised-98 (DRS-R-98) scores in the haloperidol group according to dopamine transporter (DAT) polymorphisms (mean ± SD).

Figure 2.

Delirium Rating Scale scores in the risperidone group according to dopamine transporter (DAT) polymorphisms (mean ± SD).

Starting dosages of haloperidol and risperidone were 2.67 ± 2.71 mg/day (range 0.75–10.00 mg/day) and 0.97 ± 0.67 mg/day (range 0.5–5.00 mg/day), respectively. Dosages of haloperidol and risperidone on the recovery day were 1.67 ± 1.32 mg/day (range 0.50–2.50 mg/day) and 1.19 ± 1.14 mg/day (range 0.5–5.00 mg/day),  respectively.  The  dosages  in  the  two drug groups did not differ with the DAT genotype (Table 3).

Table 3.  Dosages of haloperidol and risperidone according to DAT polymorphisms
 DosageDATnDosage (mg/day)tP
  1. DAT, dopamine transporter.

HaloperidolInitial10/10192.90 ± 2.990.840.41
 9/10 51.75 ± 0.75
Recovery10/10191.71 ± 1.460.310.76
 9/10 51.50 ± 0.64
RisperidoneInitial10/1016 1.0 ± 0.710.480.64
 9/10 20.75 ± 0.35
Recovery10/10161.25 ± 1.200.570.57
 9/10 20.75 ± 0.35

With respect to DAT polymorphisms, 35 subjects (83.3%) were type 10/10 and seven (16.7%) were type 9/10; in the haloperidol group 19 (79.2%) were type 10/10 and five (20.8%) were type 9/10; and in the risperidone group 16 (88.9%) were type 10/10 and two (11.1%) were type 9/10. There was no significant difference in the response time between the two DAT genotypes (Mann–Whitney U = 77.5, P = 0.18). The response times did not differ significantly with the DAT genotype in the haloperidol group (t = 0.82, Mann–Whitney U = 35.5, P > 0.05) or in the risperidone group (t = 1.02, Mann–Whitney U = 4.00, P > 0.05). The presence of smoking, surgery, and anticancer treatment did not affect the response times.

The number of responders at the third day was evaluated to delineate the early response of the drugs. Fourteen subjects (58.3%) in the haloperidol group and 11 subjects (61.1%) in the risperidone group responded by the third day. The number of responders at the third day in the two drug groups did not differ significantly (χ2 = 0.51, d.f. = 1, P = 0.19). The number of responders did not differ significantly with DAT genotype at the third day (χ2 = 1.22, d.f. = 1, P = 0.41), in either the haloperidol group (χ2 = 0.25) or the risperidone group (χ2 = 3.54, P > 0.05). No significant differences were revealed by the Mann–Whitney U-test (P = 0.15).

By the seventh day, 14 subjects (58.3%) in the haloperidol group and 14 subjects (77.8%) in the risperidone group had responded (no significant difference between the drug groups). The number of responders did not differ significantly with the DAT genotype at the seventh day (χ2 = 0.86, d.f. = 1, P = 0.77), in either the haloperidol group (χ2 = 0.007) or the risperidone group (χ2 = 0.643, P > 0.05). Correlation analysis revealed no significant correlation between the response time and DAT genotypes in each drug group (haloperidol group: Pearson's R = −1.71, Spearman's ρ = −0.18, P > 0.05; risperidone group: R = −0.26, ρ = −0.42, P > 0.05).

No patient developed new-onset movement disorders (e.g. extrapyramidal symptoms) during the study period. Unrelated to extrapyramidal symptoms, one patient in the haloperidol group experienced mild drowsiness 2 days after medication, which subsided after 3 days.


Symptoms of delirium generally improve within 10–12 days after beginning drug treatment, but can persist for more than 1 month in approximately 15% of delirium patients.28 In the present study, delirium improved a mean of 6 days after starting medication, which agrees with a previous report.28 However, we found no significant differences between the efficacies of the two drugs tested, which is consistent with a previous double-blind study.16

Haloperidol at a dosage of 1–10 mg/day is recommended for delirium symptoms,29–31 but this dosage range is rather wide for clinical application. In the present study, delirium patients recovered a mean of 1.67 days after taking haloperidol and 1.19 days after taking risperidone. Some researchers have reported that olanzapine at 5–10 mg/day and risperidone at 1.5–4 mg/day are sufficient for controlling delirium,9,11 with the optimal level varying with ethnicity. Pharmacokinetic differences between Asian and Caucasian patients suggest that Korean patients require smaller dosages of antipsychotics.32,33 The mean dosages used in the present study were lower than the recommended dosage for Caucasian patients.30,34 Mittal et al. reported that low-dose risperidone (0.75 ± 0.11 mg/day, range 0.5–1.50 mg/day) was sufficient to control delirium in Caucasian patients.10 Low-dose risperidone (1.17 ± 0.76 mg/day, range 0.5–4.0 mg/day) was also effective in treating the hyperactive symptoms of Chinese delirium patients.35 Our results indicate that smaller dosages of antipsychotics had similar efficacy and produced no difference in the response times in the treatment of delirium.

The type 10/10 allele is the most frequent DAT polymorphism. Vandenbergh et al. described differences in the frequencies of allele types between Caucasian, black, and patients of other races.36 Mitchell et al. reported that 52% of Greek and 100% of South American subjects carried the type 10 allele (480 bp).37 In contrast, 83.3% of all subjects in the present study carried the type 10/10 allele, similar to a study involving Asian subjects (Japanese and Mongolian) in which more than 90% of the subjects carried type 10 alleles.38,39 The drug response rate and response time did not differ significantly with the DAT polymorphism in either drug group, which indicates that it might be difficult to predict the responses to antipsychotics based only on DAT polymorphisms. However, the number of medications was significantly lower in the risperidone group and Fig. 2 contains some visible differences, even though they are not significant. The possibility that pharmacokinetic factors nullified possible effects of the VNTR polymorphisms should be considered.

The present study was subject to a few limitations. First, variables that could influence the subjects’ general medical status were not controlled. However, this does not invalidate the study because the baseline DRS-R-98 severity scores did not differ significantly between the groups. It is necessary to confirm our results with more homogeneous subject groups. Second, because we wanted to determine the optimal dosage of each drug, dosages were not controlled, and so the antipsychotic dosage may have significantly altered the length of the symptomatic period. Third, the researchers did not use a standardized scale for drug side-effects despite measuring the subjects’ symptoms every day. Even though no serious adverse events occurred during the study period, a patient's self-report of side-effects in the presence of a delirium may underestimate the side-effects of the drugs. Fourth, a relatively small sample was used, and only the 9- and 10-repeat alleles were investigated, which makes it difficult to conclude categorically that DAT polymorphisms do not play a major role in the action of antipsychotics on delirium.

Despite these limitations, the results of the present study indicate that relatively low doses of haloperidol (1.67 ± 1.32 mg/day, range 0.5–2.5 mg/day) and risperidone (1.19 ± 1.14 mg/day, range 0.5–5.0 mg/day) might be sufficient to control delirium, and that the number of medications was significantly lower in the risperidone group. DAT polymorphisms might not influence the responses to antipsychotics in Korean delirium patients. More randomized controlled studies with larger samples and broader genetic epidemiological studies in Korea should be performed to confirm these results.