Anxiety and depression
Anxiety and depression are common in DLB; they affect 27 % and 59 % of cases, respectively [
10]. Like many nonmotor symptoms of synucleinopathy, anxiety and depression can predate the onset of parkinsonism and dementia by decades [
10,
21,
34], possibly due to early pathology in serotonergic projection cells of the dorsal raphe [
35,
36]. Incipient synucleinopathy should therefore be considered in the differential diagnosis of late-onset anxiety and depression, particularly in patients without an obvious precipitant and/or subtle parkinsonism [
37,
38].
There are no controlled trials of treatments for anxiety in DLB or PDD [
33]. Depression was one of four neuropsychiatric symptoms in a composite measure that improved in trials of rivastigmine [
39] and olanzapine [
27,
40]. A single, uncontrolled trial of citalopram and risperidone in 31 patients found no improvement for either drug after 12 weeks [
41]. Otherwise, there is a paucity of evidence in the treatment of depression in DLB. Serotonin and serotonin/norepinephrine reuptake inhibitor antidepressants have mixed results in the treatment of PD-related depression [
20,
42,
43]. Electroconvulsive treatment and transcranial magnetic stimulation are both effective in DLB [
44,
45].
Hallucinations and delusions
Hallucinations occur in 60 to 70 % of DLB patients [
8]. Commonly, they begin in the first 2 or 3 years of the disease whereas they are a late phenomenon in Alzheimer’s disease [
46]. There may also be qualitative differences. In Alzheimer’s disease, hallucinations generally have a threatening or fearful quality [
47]. They are often accompanied by delusions of suspicion [
47]. In contrast, the hallucinations in DLB are often (but not always) nonthreatening misperceptions of ambiguous stimuli. For example, a patient may misinterpret a shadow to be a person or an animal. Tests of these misperceptions, termed pareidolias, accurately differentiate DLB from Alzheimer’s disease [
48]. Electing not to treat these symptoms is often appropriate, but cholinesterase inhibitor therapy is safe and effective [
39]. In one study of PDD, more than 90 % of patients reported reduced visual hallucinations with cholinesterase inhibitor use [
12]. The hallucinations can be minimized by regular vision correction and a bright light or no light policy, whilst minimizing the risk of falls. Medications that can exacerbate neuropsychiatric symptoms in DLB should also be stopped. These include anticholinergic medications, amantadine, dopamine agonists, monoamine oxidase inhibitors, catechol-
O-methyl transferase inhibitors, and levodopa, bearing in mind that abrupt cessation can trigger the neuroleptic malignant syndrome [
49].
The most fraught decision in the management of DLB relates to the use of antipsychotic medications. In Alzheimer’s disease and all-cause dementia studies, antipsychotic medications are rarely effective at reducing symptoms [
50] and they increase the risk of stroke and sudden cardiac death by at least 50 % [
51,
52]. Ceasing these medications decreases the risk of death by the same amount [
53]. The latter study counters the idea that the antipsychotic-associated mortality is due to confounding by indication. This theory states that end-stage dementia causes increased mortality and also causes the prescription of antipsychotics. Data from these pivotal studies led to a US Food and Drug Administration black-box warning for antipsychotic medication use in dementia and a substantial change in prescribing practice.
A major factor in the success or otherwise of treatment trials of antipsychotic medications in DLB and other dementia relates to the target symptoms. Hallucinations and delusions are more likely to respond to these medications than behavioral disturbance, as described below. Physicians should reserve the use of neuroleptics for symptoms likely to respond to their use, after focused behavioral interventions have been attempted. There is ample evidence of efficacy for these behavioral interventions [
54], but limited funds for implementation. Gitlin and colleagues provide an excellent description of such a program [
55].
A continued role for antipsychotic treatment remains for the short-term treatment of subjects at risk of harm due to their psychosis [
47]. Higher scores on the Neuropsychiatric Inventory may indicate a patient is more likely to respond to the treatment, particularly if the findings are within the domains most amenable to treatment with neuroleptic medication (see discussion below) [
56].
DLB patients are particularly at risk of antipsychotic medication morbidity and mortality. Severe neuroleptic sensitivity occurs in 30 to 50 % of patients [
57]. Typical antipsychotics (for example, haloperidol) are best avoided [
57], but reactions can occur after any neuroleptic and no differences in mortality were found between the atypical antipsychotics in all-dementia clinical trials [
50]. Profound sedation, confusion, exacerbations of parkinsonism, rigidity, dysautonomia, and death can occur [
4,
16,
57,
58] even after a single dose [
57,
59,
60]. These effects are associated with a threefold increase in stroke occurrence and a twofold to fourfold increase in the rate of cognitive decline [
50,
53]. There is also evidence to the contrary, however. Antipsychotic medications have been well tolerated in DLB trials [
61,
62], and a large observational study that controlled for cardiovascular risk and psychosis severity found no increase in mortality amongst Alzheimer’s disease patients [
63]. Irrespective of the controversy over the degree of risk, the large placebo response seen in trials of antipsychotics [
27,
50] and the data showing improved survival for those taken off long-term antipsychotics treatment [
53] dictate that all new prescriptions of antipsychotics should include a programmed trial of cessation [
64].
The choice of which antipsychotic to use is also a vexed question. Quetiapine and clozapine are equally effective in head-to-head PDD trials [
49,
65], although other data show mixed results for quetiapine in underpowered trials of DLB/PDD [
20,
61,
62,
66,
67]. Despite the paucity of evidence for its efficacy, many clinicians use quetiapine, reserving clozapine for second-line or third-line treatment because of its potential to cause agranulocytosis [
68‐
72]. More data are sorely needed to inform antipsychotic medication choice in DLB. In the interim, it is reasonable to select antipsychotic medications on the basis of their side-effect profiles. Patients at risk for diabetes or hyperlipidemia should avoid quetiapine, olanzapine and clozapine, whereas those with elevated cerebrovascular risk should avoid olanzapine and risperidone [
47]. Olanzapine is also associated with motor decline in PD patients with psychosis [
20]. The Movement Disorders Society advises against olanzapine use in PD [
20], advice that might reasonably be extended to DLB patients. Irrespective of which antipsychotic medication is used, prescribers should be wary of the cardiac denervation seen in the synucleinopathies [
22] and so consider monitoring the QT interval. This is especially important when cholinesterase inhibitor and neuroleptic medications are used together. Fortunately, several new agents without such side effects are under development. For example, a recent trial of pimavanserin, a selective serotonin 5-HT2A inverse agonist, shows great promise in the treatment of PDD, both in terms of treatment response and trial design [
73].
Agitation and behavioral disturbance
Agitation and behavioral disturbance often respond to simple measures such as caregiver training, removal of fear triggers, and increased social interaction [
74]. Many triggers for agitation are fleeting, and episodes of agitation are self-limiting, so watchful waiting is often preferable to antipsychotic prescription [
53]. In the late stages of disease, when patients have difficulty expressing their needs, pain is often a trigger for agitation: investigation for potential sources of pain and empiric treatment with simple analgesics such as acetaminophen should be first-line therapy [
75]. Antipsychotic medications have the same qualifications to their use as noted above. Furthermore, behavioral disturbances such as sleep–wake cycle disturbance, shouting, oppositional behavior, pacing, agitation, and aggression are not good targets of therapy for the neuroleptic medications. There are numerous carer training programs designed to decrease the disturbances, but only six medications have evidence for efficacy, three of which are readily accessible [
76‐
78]. Successful programs train caregivers to understand care situations from the perspective of people with moderate to severe dementia, and to adapt their approach to such encounters to encourage respect for the patient’s personhood [
55]. Extensive training is required for program success [
76‐
78].