CGPR: What is atypical parkinsonism?

Dr. Flaherty: Atypical parkinsonism refers to a group of neurological disorders that present with parkinsonian symptoms such as tremors, rigidity, and bradykinesia, but do not respond well to typical Parkinson’s disease (PD) treatments like levodopa. Symptoms of atypical parkinsonism often include early dementia (preceding or co-occurring with parkinsonian symptoms), hallucinations, spasticity, ataxia, and orthostatic hypotension. The major subtypes include progressive supranuclear palsy, multiple system atrophy, corticobasal degeneration, and dementia with Lewy bodies (DLB).

CGPR: Why is it important for clinicians to know about atypical parkinsonism?

Dr. Flaherty: Conditions like DLB often present with psychiatric symptoms such as hallucinations and cognitive fluctuations before motor symptoms appear, meaning patients may see a psychiatrist before a neurologist. Additionally, many psychiatric medications can cause parkinsonian symptoms, leading clinicians to overdiagnose parkinsonism as atypical rather than the result of medications. Understanding atypical parkinsonism helps clinicians to accurately identify the root cause of symptoms and avoid using treatments that could exacerbate the condition, such as dopamine blockers. Although psychiatrists often try to guess the subtypes of atypical parkinsonism, it is not a worthwhile use of time. Even movement disorder specialists often misidentify the subtypes—they have overlapping symptoms and don’t have different treatments. Just keep in mind that atypical parkinsonian syndromes don’t respond much to dopaminergic drugs like levodopa, and that any kind of dopamine blocker is bad in both idiopathic PD and atypical parkinsonism.

CGPR: Can you tell us about Lewy body dementia (LBD)?

Dr. Flaherty: LBD includes both DLB and Parkinson’s disease dementia. In DLB, hallucinations, arousal, and cognition fluctuate almost like delirium. Early on, patients can appear to be in dream states. Psychosis in DLB is not like schizophrenic psychosis. It’ll often be complex visual hallucinations, sometimes with a dream-like plot. People may experience dream intrusions, as when they wake up and see an ancient warrior in the bathtub. Patients may develop more severe psychiatric symptoms with medications that boost dopamine, like levodopa, and they may experience worsening motor symptoms with medications that block dopamine, like antipsychotics.

CGPR: How long do you wait to reevaluate movement symptoms after stopping an antipsychotic?

Dr. Flaherty: At least two months. By then patients should be significantly better. But it can take four to five months to get completely back to normal. Getting back to normal doesn’t rule out the possibility that the patient also has early PD, which will show up in a year or two. Either way, you’re going to have to change your patient’s medications unless their symptoms are tolerable. Some people are able to tolerate tardive dyskinesia, but very few people will tolerate parkinsonian symptoms.

CGPR: How do you tell if patients are levodopa-responsive?

Dr. Flaherty: Make sure you’re not adding levodopa to a dopamine blocker. For patients not taking antipsychotics—and they may need to be off for more than a month—neurologists try a stepwise levodopa challenge. We start with a test dose of ¼ tab of carbidopa/levodopa (Sinemet) 25–100 mg to make sure it doesn’t make them nauseated. Then one tablet three times daily for a week, increase to two tablets three times daily for a week, and then three tablets three times daily for a week. If they don’t respond at this dose, we can be pretty sure they’re not going to respond and that they have atypical parkinsonism. If patients develop nausea, we titrate much slower and make sure they’re taking Sinemet with meals. Rarely, patients develop psychosis at this dose. But once you stop the medication, psychosis often resolves within six hours.

CGPR: When do you consider ordering a DaTscan?

Dr. Flaherty: Almost never. A DaTscan rarely increases diagnostic accuracy or changes management. It can’t distinguish idiopathic PD from atypical parkinsonism—it’s more useful to test the patient’s response to levodopa. Some physicians are attracted by a DaTscan’s ability to tell antipsychotic-induced parkinsonism from idiopathic PD and atypical parkinsonism. But if a patient has any kind of parkinsonism and is on an antipsychotic, you need to lower or change the antipsychotic if possible. A DaTscan can distinguish someone with parkinsonism from someone with essential tremor. But telling a rest tremor from an action tremor is very easy.

CGPR: Tell us more about tremors and evaluating for parkinsonism.

Dr. Flaherty: First focus on whether your patient has a tremor, then observe for rigidity and slowness, then test for parkinsonism. If the patient has a tremor, notice whether it’s an action tremor or a rest tremor. Rest tremors are very specific for parkinsonism. You can differentiate an action tremor from cerebellar dysmetria by finger-to-nose testing. An action tremor will be present throughout the movement, but dysmetria causes overshoot you will only see at the end. If there is no tremor, it’s easy to overlook the rigidity and slowness of parkinsonism. Psychiatrists may be used to seeing depressed patients and patients on antipsychotics move slowly, have flat expressions, and write with tiny handwriting. But depression alone won’t cause a rest tremor, or tiny handwriting that gets even smaller at the end of the line, or a shuffling gait. In video visits, you may not see your patient walk into the room, so it’s best to look for parkinsonism with two provocative tests: the lightbulb test (rapid wrist rotations) or the foot-tapping test.

CGPR: How do you do these over Zoom?

Dr. Flaherty: For the lightbulb test, ask your patient to put their hand up in front of their face so you can see it. Have them rotate their hand five times slowly, all the way around, as if they are screwing in a lightbulb. Then ask them to do the same motion as quickly as possible. Most depressed patients will do the quick movements well. PD patients will have low-amplitude, arrhythmic, incomplete rotations. This ability usually goes away early, even in someone who doesn’t otherwise look parkinsonian. For the foot-tapping test, ask patients to stamp their foot 10 times at least three inches in the air, as fast as they can. In parkinsonism, stamps will usually be very small and irregular. Even if stamps start big, they will slow down over the 10 movements.

CGPR: How do you treat apathy in parkinsonian syndromes?

Dr. Flaherty: In apathetic patients with parkinsonism, bupropion (Wellbutrin) or a dopamine agonist like pramipexole can help, just as they can with apathetic patients in the general population. Pramipexole is more likely to cause nausea and sedation than bupropion. Ropinirole (Requip) is my least favorite of the agonists, as it causes the most lightheadedness, nausea, and sedation, but sometimes its sedation is useful for treating restless legs syndrome. In patients with true idiopathic PD, sometimes levodopa alone is enough to help apathy or even depression. Some early-stage PD patients will present with a first symptom of depression before any motor symptoms. PD is truly a neuropsychiatric disorder that retards emotions and thoughts as well as movement. The basal ganglia don’t distinguish much between motion and emotion, and they’ll often get better together. Conversely, if you overdo treatment, you may see both dyskinesias and hypomanic activity.

CGPR: Tell us about using antidepressants in PD.

Dr. Flaherty: It’s not wise to stop selective serotonin reuptake inhibitors (SSRIs) in patients with premorbid melancholic depression when SSRIs have helped them. However, SSRIs can worsen PD symptoms like stiffness and tremor. They can also cause restless legs syndrome and an SSRI apathy syndrome. The serotonin “increase” suppresses dopamine, lessening patients’ libido and interest in their work. SSRIs do help treat excessive crying due to pseudobulbar affect, just as they do in melancholic depression. For depression, I often start with bupropion, which addresses a patient’s lack of motivation, although it may improve apathy more than mood (Vismara M et al, BMC Neurol 2022;22(1):169). So, SSRIs for tears, bupropion for pleasure and motivation. I also frequently prescribe tricyclic antidepressants (TCAs), as head-to-head studies suggest they may significantly improve PD depressive symptoms, whereas SSRIs on average do nothing in PD (Menza M et al, Neurology 2009;72(10):886–892; Liu J et al, PLoS One 2013;8(10):e76651). TCAs’ anticholinergic effects can help PD symptoms like insomnia, pain, muscle stiffness, and overactive neurogenic bladder. They also treat pain better than medications like duloxetine (Cymbalta). (Editor’s note: Keep in mind that the higher doses of TCAs used to treat depression may be intolerable for many older adults.)

CGPR: How do you treat anxiety in parkinsonian syndromes?

Dr. Flaherty: Anxiety in PD is often a low-dopamine anxiety, not the more standard serotonin-responsive anxiety. First try increasing levodopa. Many anxious PD patients, especially if it’s anxiety at times like early morning when PD meds are low, will get better with ropinirole or even bupropion, but not SSRIs.

CGPR: For PD patients, when do the motor benefits of dopaminergic meds outweigh psychiatric side effects?

Dr. Flaherty: Most patients prefer to be able to move and will put up with side effects like dyskinesias, compulsions, and even mild psychosis. When patients’ PD meds wear off, many get this aversive “off state” where they feel physically and mentally trapped. I’ve seen people go from talking normally to saying they feel like they are going to die in the space of five minutes. Many times, dopamine-driven compulsions are ego-syntonic and goal-directed, unlike fear-driven OCD compulsions. Patients report getting pleasure from compulsive hobbies caused by dopamine compulsions, but their families may feel differently, causing tension. As the patient’s advocate, I try to explain to the family why this is happening and urge them to worry only if the patient’s actions are illegal or dangerous. Patients feel they are choosing the activity and enjoy it. One of my patients, a physicist, told me he used to love physics but now wants to do it all the time. All of our interests are driven by dopamine; these patients just have a little more.

CGPR: Tell us more about impulse control disorders.

Dr. Flaherty: They can range from collecting fountain pens to gambling or severe hypersexuality. Gambling can make patients deplete their life savings in a few months. Impulse control disorders tend to be more severe and disruptive in men. Although I feel like I am perpetuating gender stereotypes, women with PD are more prone to obsessive gardening, shopping, and cooking. That being said, the obsessions often have little relation to previous interests. One man whose PD meds drove him to compose music obsessively said he hadn’t been very interested in music earlier in his life. When compulsions are significantly disruptive, first I try to lower dopamine agonists; I save levodopa for last. Unfortunately, lowering the meds often makes the motor symptoms intolerable. At that point I recommend antipsychotics that are selective enough to lower cognitive and emotional overdrive without affecting movement.

CGPR: How is psychosis different when it’s secondary to parkinsonian syndromes?

Dr. Flaherty: In idiopathic PD, psychosis often results from medications, with a clear time course between medication intake and hallucinations. Dopaminergic agonists are more likely to cause hallucinations than levodopa. Hallucinations may start with vivid dreams followed by dream intrusions. This is separate from REM sleep behavior disorder, in which patients act out their dreams but may not remember them vividly. Because dopamine is crucial in perception of animacy, dopamine-induced hallucinations typically involve seeing inanimate objects as moving or animated; for example, patients may see a fire hydrant as a little red dwarf. The hallucinations are often visual—motion in the corner of the eye, mice running across the floor, a man in the trees. Patients may also experience illusions of presence or touch and feel that someone is in the room with them. They often have insight, so it’s more hallucinosis than hallucinations. In dementia, they may see dead relatives or frightening hallucinations like intruders. Delusions are often more dangerous and disruptive than hallucinations. They can include financial paranoia and Othello syndrome, where the patient is convinced that their spouse is slipping out to have sex with the mail carrier. Patients and families rarely tell you about this unless you ask.

CGPR: If you think psychosis is due to PD medications, how do you treat it?

Dr. Flaherty: First, minimize dopamine agonists because they are more hallucinogenic than levodopa. Then try pimavanserin (Nuplazid). It’s an inverse agonist suitable for PD psychosis and does not usually interfere with other treatments. It’s pretty good, just weak. It takes more than two weeks to help most patients. If pimavanserin is not enough, then I recommend clozapine (Clozaril), which acts quickly. (Editor’s note: In psychosis due to PD, patients often respond at much lower doses; a usual nightly dose ranges between 6.25 mg and 50 mg. Clozapine monitoring for PD psychosis is the same as in schizophrenia.) Many will also try quetiapine (Seroquel) first, which is not a bad strategy for patients with minor hallucinations. It’s more affordable and less likely to affect motor symptoms at low doses. I’ve had patients who were comfortable on 25 mg because it’s easy to make early hallucinations go away; it’s not like schizophrenia. If they have severe PD, however, patients will get extrapyramidal side effects even on quetiapine. For those who cannot tolerate clozapine’s sedation, adding or switching to aripiprazole (Abilify) is another option. However, aripiprazole can exacerbate parkinsonian motor symptoms, especially facial ones, which may persist for months after discontinuation.

CGPR: Are there any additional downsides to using pimavanserin?

Dr. Flaherty: Pimavanserin is not mood-stabilizing, although there’s emerging evidence that it’s mood-elevating and can help anxious depression (Papakostas GI et al, Int Clin Psychopharmacol 2020;35(6):313–321). However, I’ve never heard patients say they are no longer depressed after starting pimavanserin, like they sometimes say after starting Sinemet or clozapine. Very rarely, I have seen agitation on pimavanserin, as it’s stimulating. People like it because they are more alert during the day and sleep better at night.

CGPR: How does parkinsonism affect psychiatrist prescribing in BD?

Dr. Flaherty: This comes up when clinicians switch patients from lithium to antipsychotics and patients don’t do well. Clinicians will then increase the dose, causing patients to shuffle, fall, or become delirious. They may think the patient has DLB and recommend a nursing home. I had one patient who was put in a nursing home after he was diagnosed with dementia and PD, but his symptoms were all due to side effects from his bipolar medications. He had a Depakote-Haldol tremor for which his psychiatrist prescribed primidone, which caused drowsiness. He then received antidementia meds to help the side effects of the primidone. He couldn’t feed himself, let alone do his taxes. I replaced his bipolar medications with clozapine, and he was able to move out of the nursing home. Clozapine is terribly underused in bipolar patients who develop extrapyramidal symptoms. It not only doesn’t cause them, but it also suppresses them. When patients find it too sedating, you can lower the clozapine and add a bit of a stimulating antipsychotic such as aripiprazole. The clozapine will actually suppress the extrapyramidal effect of the conventional antipsychotic.

CGPR: Which patients benefit from deep brain stimulation (DBS)?

Dr. Flaherty: It’s used for patients with dyskinesias and severe motor fluctuations; they go through moving too much and then being frozen. The main thing is that patients should have a very clear response to meds. If meds don’t help, DBS won’t help either. It’s of no use in atypical parkinsonism, and it’s not appropriate in patients with dementia.

CGPR: What else should psychiatrists know about DBS?

Dr. Flaherty: On average, people’s mood significantly improves with DBS, partly because they’re walking better and can eliminate medication side effects. However, there is an increased risk of suicide from DBS in people who are depressed (Costanza A et al, Front Integr Neurosci 2021;15:632249). It’s probably the same phenomenon as those who end their life after starting antidepressants—they regain motivation before their mood improves, and they may impulsively die by suicide.

CGPR: Thank you for your time, Dr. Flaherty.