Memantine is a N-methyl-D-aspartate (NMDA) receptor antagonist which has proven efficacy in AD and has been proposed to target the glutamatergic dysfunction described in LB disease. In case series, both cognitive improvement and worsening have been described with this treatment. Two studies have attempted to address the issue. In one double-blind, multicentre study 72 patients with LB disease (either PD-D or DLB) were randomized to memantine or placebo . The main outcome variable was improvement in the Clinical Global Impression of Change scale (CGIC), while changes in MMSE, NPI, modified UPDRS, disability and cognitive speed (measured by A Quick Test of Cognitive Speed, AQT) were secondary outcome measures. High attrition was evident in both treatment arms, but a significant difference in the CGIC was demonstrated between groups at week 24. No significant differences for secondary outcome variables were found between the memantine and placebo groups between baseline and week 24. Another study randomized 199 DLB or PD-D patients to memantine or placebo . In the DLB group, patients treated with memantine demonstrated significant improvement in the Alzheimer’s Disease Cooperative Study (ADCS)-CGIC and NPI scales, but no changes were found in other scales or there were no statistically significant differences in any of the scales in the PD-D group .
Given the high frequency of psychotic symptoms in both DLB and PD-D, antipsychotic agents are frequently considered for DLB and PD-D patients. However, antipsychotic medications are often accompanied by numerous and severe adverse effects, including parkinsonism, delirium, and dystonia . Only a limited number of studies have examined the effect of antipsychotic medication in patients with DLB and PD-D. One randomized controlled trial of an atypical antipsychotic in patients with dementia and PD indicated that quetiapine did not confer any greater benefit than placebo in the treatment of psychosis . Another randomized trial without placebo compared the effect of citalopram and risperidone over 12 weeks in 31 DLB and 66 AD patients. Patients with DLB showed no improvement or deterioration in neuropsychiatric symptoms. In contrary, AD patients did show an improvement, and early withdrawal was significantly lower in AD patients (50%) than in DLB group (68%) . Importantly, severe neuroleptic sensitivity reactions have been identified as a major clinical issue in DLB. An initial study in 1992  reported that 50% of DLB patients treated with neuroleptics experienced severe drug sensitivity, with symptoms that included marked extrapyramidal features, confusion, autonomic instability, falls, and accelerated mortality. An accumulating literature of case reports and case series, as well as subsequent larger and more systematic reports, shows that severe neuroleptic sensitivity reactions occur with a wide range of typical and atypical antipsychotics, rarely including clozapine, in DLB [119, 120]. There is also evidence of similar poor tolerability in PD-D, albeit from fewer studies. A comparative study examined neurolepetic sensitivity in 94 patients (15 with DLB, 36 with PD-D, 26 with PD, and 17 with AD). Severe neuroleptic sensitivity occurred in patients with LB disease (DLB 53%, PD-D 39%, PD 27%) but not in AD. Thus, severe neuroleptic sensitivity syndrome is common in both DLB and PD-D . There are other adverse effects of antipsychotic treatment in DLB and PD-D: together with the limited evidence of benefit it is recommended to prescribe antipsychotics only in patients with severe neuropsychiatric symptoms which cannot be otherwise managed.