Assessment and diagnosis
As insight and memory can be impaired in patients with PD-D, it is important to include an informed other in the assessment of NPS; patient self-completed assessments are not appropriate for assessing such symptoms in PD-D. The new version of the UPDRS [108, 109] has individual questions that can be used to screen for symptoms that are common in PD-D, including depression, anxiety, psychosis, apathy, and disorders of sleep and wakefulness.
The most commonly used global instrument for assessing the presence and severity of NPS in PD-D is the NPI. The NPI was developed to overcome difficulties associated with assessing behavioural symptoms in patients with dementia, such as inaccurate reporting of symptom severity or frequency. The original NPI comprised a series of questions concerning 10 behavioural symptom domains: delusions, hallucinations, agitation/aggression, depression, anxiety, euphoria, apathy, disinhibition, irritability/lability, and aberrant motor behaviour. A 12-item version was subsequently developed, which includes additional questions on nighttime behavioural disturbances and appetite/eating changes. Domain-specific interview questions are administered to an informant, who is asked to assess the patient’s behaviour in the past month. If a particular behaviour is endorsed, the severity (e.g. mild, moderate, or severe) and frequency (e.g. occasionally, often, frequently, or very frequently) are rated, and domain-specific scores are determined by multiplying severity and frequency. Subsequently, a brief caregiver-completed version of the NPI, the Neuropsychiatric Personality Inventory-Questionnaire (NPI-Q)  was also developed.
Management of neuropsychiatric symptoms
Depression and anxiety
There have been no controlled studies on the use of psychiatric medications for symptoms of depression and anxiety specifically in PD-D. Approximately 20-25% of PD patients in specialty care are taking an antidepressant at any given time, most commonly a selective serotonin reuptake inhibitor (SSRI) [111, 112], which is the recommended first-line antidepressant treatment type for depression in the elderly . Results of numerous open-label trials using SSRIs and other newer antidepressants in PD suggest a positive effect and good tolerability . Until recently, placebo-controlled studies with SSRIs reported negative findings [115-117]. However, a recent placebo-controlled study of SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs) showed promising results for treating depression in PD with paroxetine and venlafaxine XR, respectively. Both medications were found to improve depression without worsening motor function. However, patients with PD-D were not included in this study, so results should be interpreted cautiously for the PD-D population . In addition, although apparently rare, the combination of a SSRI and a monoamine oxidase inhibitor—used in the treatment of PD—can result in serotonin syndrome (i.e. development of symptoms such as mental confusion, hallucinations, agitation, headache, coma, shivering, sweating, hyperthermia, hypertension, tachycardia, nausea, diarrhoea, myoclonus, hyperreflexia, and tremor).
Two placebo-controlled studies with tricyclic antidepressants (TCA) were positive [117, 119]; however, TCAs can be difficult for PD patients to tolerate due to aggravation of PD-associated orthostatic hypotension, constipation, and cognitive problems , and they must be avoided in PD-D patients as they can cause cognitive worsening and increased confusion.
A commonly used dopamine agonist (pramipexole) for motor symptoms in PD was found to improve depressive symptoms as well. A 12-week placebo-controlled trial showed that participants (without dementia) taking 0.125-1.0 mg of pramipexole three times a day showed a 5.9 (+ 0.5) point decrease on the Beck Depression Inventory (BDI) as well as improvement on the Geriatric Depression Scale-15 (GDS-15). As was expected, patients also experienced an improvement in motor symptoms. Statistical analysis determined that 80% of the improvement was due to the effect of pramipexole on mood and 20% to improvements in motor functioning .
There have been no controlled studies on the treatment of anxiety in PD . For patients who experience anxiety as part of an ‘off’ state, adjustments can be made to the PD medication in an attempt to decrease the duration and severity of these episodes. Anecdotally, newer antidepressants are commonly used for anxiety disorders, whether or not co-morbid depression is present. However, anxiety in PD responds variably to antidepressants, and many patients require treatment with benzodiazepines (most commonly low-dose lorazepam, alprazolam, and clonazepam). Given that PD-D patients are cognitively and frequently physically impaired, benzodiazepines must be used cautiously due to their propensity to worsen cognition, sedation, and gait/balance.