Four RCTs (n=158) were included. Two studies were reported to be double-blinded (in an additional study the investigators were blinded). The method of randomisation was described by only one study, but this method was inadequate.
One study (n=30) treating AIDS patients for delirium found statistically significant improvements in the DRS score for haloperidol (p<0.001) and chlorpromazine (p<0.001), but not for lorazepam (p<0.63). Significant improvements were also found in the MMSE score for chlorpromazine (p<0.001), but not for haloperidol (p<0.09) or lorazepam (p<0.40). There were no significant changes in ESRS scores for any of the drugs used. All participants in the lorazepam group experienced treatment-limiting adverse events including oversedation, disinhibition, ataxia and confusion.
One study (n=24) comparing the effects of risperidone and haloperidol for the treatment of delirium in oncology, medical and intensive care unit patients found a significant improvement in the MDAS score in both groups (p<0.05), but showed no statistically significant differences between groups (p=0.51) for treatment effect. One patient developed mild symptoms of akathisia in the haloperidol group; no other adverse events were reported.
One study (n=73) comparing haloperidol and olanzapine for the treatment of delirium in intensive care unit patients found that DI scores improved for both groups (p=0.02), but there were no statistically significant differences between groups for treatment effect (p=0.83). Six participants in the haloperidol group rated low on symptom testing for extrapyramidal symptoms, compared with no extrapyramidal manifestation for the olanzapine group.
One study (n=31) evaluating the treatment of delirium in medical and surgical patients reported an improvement in the DRS-R-98 for treatment with amisulpride (p=0.000) and quetiapine (p=0.001). Again, there were no statistically significant differences between the two groups for treatment effect (p=0.842). One patient in each group developed oversedation.