How to reduce unnecessary polypharmacy, and when it makes sense to keep it going
Fresh from Mount Sinai we bring you the 8th commandment: Avoid unwise polypharmacy.
CHRIS AIKEN: Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief and the author of the Depression and Bipolar Workbook.
KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.
Before we venture into the weeds of polypharmacy, here’s a preview of the CME quiz for this episode.
Recently a large randomized controlled trial tested the antipsychotic brexpiprazole as an add-on to stimulants in ADHD. What was the result?
A. Brexpiprazole improved symptoms of ADHD
B. Brexpiprazole worsened symptoms of ADHD
C. Brexpiprazole did not change symptoms of ADHD
D. Brexpiprazole reduced stimulant side effects
Enter your answer, and earn CME credits through the link in the show notes.
CHRIS AIKEN: Long ago, there were only 5 classes psychiatric medications. Antipsychotics, mood stabilizers, antidepressants, stimulants, and gaba-ergic sedatives like barbiturates and benzodiazepines. That was in the 1950’s, but today psychiatry has advanced…. Well, maybe not. For the most part, we are still laboring with the same 5 classes. What has changed is that these psychotropics have become more tolerable, but not more effective. The barbiturates were stronger than Ambien or Restoril; Haloperidol is still valued for its potency in psychosis and mania; and it’s hard to argue that the SSRIs are more effective than the MAOIs and tricyclics.
We have many more options but few cures, and what that means in practice is…. More polypharmacy. If we’re not getting full recovery on one antidepressant, and have lots of tolerable options to pick from, why not add another, and another, and another.
Actually the rates of polypharmacy have remained high since the 1960’s. Somewhere between 30 to 50% of patients with chronic mood or psychotic disorders are taking more than one med. What has changed is the rate of quadrapharmacy. That’s when patients take 4 or more pills, and that rate has shot up in the past 15 years.
KELLIE NEWSOME: We don’t know why, but there was a notable shift in 2004. That’s when the combo pills returned to the market, starting with Symbyax – the fluoxetine/olanzapine combo for bipolar and treatment resistant depression – which was FDA approved on Christmas Eve 2003. More recently we’ve added Olanzapine/Samidorphan (Lybalvi) and bupropion-dextromethorphan (Auvelity), not to mention a slew of medications that are FDA approved to augment other meds, most of them antipsychotics.
But in the past we had had a long hiatus from combo pills, which were very popular in the 1950’s, 60’s and 70’s but went out of favor in the early 1980s. There were antidepressant / antipsychotic combos like Triavil, antidepressant-benzodiazepine combos like Limbitrol, and even a dextroamphpetamine-thorazine combo ThoraDex. These fell off the market as doctors realized there was little evidence to support their benefits, and evidence all around them of problems – particularly tardive dyskinesia in the mood disorder population who were taking tricyclics with antipsychotics. It just wasn’t worth it. And with their demise came a rise in articles that argued against the random, shot-gun approach of polypharmacy. Basically, psychiatrists got real to the fact that they didn’t know enough about what their meds were doing, and ought to proceed with caution.
CHRIS AIKEN: Today’s thought leaders are not as outspoken about polypharmacy. Maybe the Symbyax and Lybalvi dinner programs have loosened them up, but I wouldn’t jump to that cynicism. Most of the arguments against polypharmacy are that it’s not evidence based, which is no different from arguing that buspirone in PTSD or sertraline in bipolar disorder is not evidence based. But now we have more evidence supporting polypharmacy, at least for combinations of 2, and the thought leaders have turned their criticisms toward combinations of 4. The legendary Sheldon Preskorn, who helped develop many of the antidepressants we use today, expressed his biases about polypharmacy openly in a 2007 article criticizing the practice:
KELLIE NEWSOME: The reader should also be aware of our biases:
- Mono-drug therapy: the ideal
- Copharmacy: commonly needed
- Triple pharmacy: may be necessary
- Quadruple pharmacy: first consider that three drugs are not working.
CHRIS AIKEN: There’s a pretty good chance of that – that the drugs are not working. Most psychiatric drugs have a number needed to treat around 4, which means patients have a 1 in 4 chance of responding to it. So if a patient is on 4 meds, there’s around a 1 in 16 chance they are responding to all 4 of them.
NNT: Those figures seem awfully low, but they are actually the best case scenarios because they come from the FDA registration trials, where patients with pure, classic diagnoses are enrolled. We’d expect the “number needed to treat” to be much worse when the med is used off label in patients with complex comorbidities, major medical problems, active substance misuse, or suicidality. And those are more like the patients who end up on polypharmacy.
Dr. Preskorn does have a good point. When a patient is taking 4 meds the odds are very strong that at least one of them is not working. But on the other hand, when I see a complex patients on multiple meds, they often get worse – unless they feel ready to come off.
CHRIS AIKEN: Yes the number needed to treat does not include the placebo effect – which is about 50% of the benefit with most meds, and when you take away a med that the patient feels they need, you get the opposite – the nocebo effect. They get worse. Kellie you have a pretty high success rate with reducing polypharmacy. What’s your secret?
KELLIE NEWSOME: There are 4 steps to a polypharm makeover
1) Make sure they are ready. This isn’t going to work as well if it’s their first visit, you don’t have a therapeutic alliance, or they are doing really well on the combo.
2) Set the stage. The goal here is help them break free from any psychological attachment to the meds. Get them out of that passive, dependent mode of “I can’t exist without these meds” and into a more active stance where they are taking control of their treatment. Start with questions, like “Do you feel like these meds are actually working?” and as they realize that they aren’t really thriving on this combination go through the meds one by one and ask them what they take them for. Usually they don’t know, or they think it must be working because a doctor prescribed it, but they can’t say how it has helped. I’ll drop hints that the meds might not be such a good idea, like I’ll talk about side effects, I’ll single out ones that are experimental like Abilify for generalized anxiety disorder, or how the mechanisms cancel each other out – like a stimulant and an antipsychotic or – in a patient with rapid cycling – and antidepressant and a mood stabilizer.
3) Give hope. I’ll open up the possibility that they might feel better coming off the med. I’ll use patient examples, like for someone on a high dose benzo I’ll say “A lot of patients feel mentally sharper as they lower the dose. They have better balance and coordination, they’re quicker on their feet and not as groggy.”
4) Taper slowly. Nearly every psychiatric med has a withdrawal syndrome – of course serotonergic meds and benzos, but also antipsychotics, anticonvulsants, tricyclics, mirtazapine, and stimulants. We just don’t talk about those withdrawals much because they aren’t as serious, but they are for a fragile patient who might interpret mild withdrawal symptoms as a sign that they are getting worse. The longer they’ve been on the meds, the more likely their brain has accommodated and will experience some withdrawal, so unless there’s an urgent need to get them off I’ll taper slowly, inching down to the next available dosage about every 2 weeks.
CHRIS AIKEN: Some patients end up on polypharmacy because they never got to a therapeutic dose of the med that could work. Like a patient with bipolar disorder who is on 100 mg of quetiapine, 300 mg of oxcarbazepine, 50mg of sertraline, and 2 mg of clonazepam. Only one of those meds is actually effective in bipolar disorder – quetiapine – and if they brought it to a therapeutic dose of 300mg or higher the patient probably won’t need the others.
That was proven true in a classic study of lithium and paroxetine from the 1990’s, where paroxetine brought no benefit to bipolar disorder as long as the lithium was brought to an adequate dose, like a level at or above 0.6. In fact, lithium is one med that often eliminates the need for polypharmacy. About 1 in 3 patients with bipolar disorder are very good lithium responders, and after they recover on it they are able to taper off other meds. Another one that has that effect is clozapine in schizophrenia. Treatment resistant cases of schizophrenia often end up on multiple meds – 2 antipsychotics, an anticonvulsant, an antidepressant and a benzo. Once they get on clozapine the others can be slowly taken away.
KELLIE NEWSOME: Sometimes polypharmacy is totally rational and evidence based. Examples where we have evidence of additive benefits with two medications include:
- In bipolar disorder, an anticonvulsant-lithium combo, particularly Depakote, lamotrigine, or carbamazepine with lithium
- In bipolar depression or mania, adding an antipsychotic to a mood stabilizer
- In unipolar depression, adding an augmentation agent like lithium, and atypical antipsychtoic, thyroid, to an antidepressant
- In ADHD, adding an alpha agonists like guanfacine to a stimulant
- In schizophrenia, adding aripiprazole to clozapine
- To treat two separate disorders, like an SSRI for OCD along with methylphenidate for ADHD
- To provide temporary relief while waiting for another medication to work, like the temporary use of a benzo or hypnotic while starting an antidepressant. I’ll add to that one combo in particular – eszopiclone, but not other hypnotics – has evidence to improve not just sleep but anxiety when added to an SSRI.
- To treat a side effect, like propranolol for lithium tremor or metformin for antipsychotic weight gain
- Got other examples of legit polypharmacy? Send them to us at firstname.lastname@example.org
CHRIS AIKEN: Most polypharmacy is benign. The med isn’t doing any good, and it’s not doing much harm. But some combinations that pose real problems. We’ll cover those more in part II of this episode, but here’s a preview:
1. Benzos with stimulants
2. Antipsychotics with stimulants
3. Multiple serotonergics, particularly if one is an MAOI
4. Multiple hypnotics
5. Meds with additive side effects, like multiple anticholinergics, blood pressure lowering meds, or meds that prolong the QTc interval
6. Combinations that cause one of the meds to go to dangerously high levels due to a drug interaction. This is only a problem if the high level is dangerous, like when a tricyclic is raised by an SSRI, leading to arrhythmias and cardiac arrest.
Then there are a few combinations that aren’t dangerous, but are highly questionable, like
> Multiple antipsychotics, except when one of them is combined with clozapine – particularly aripiprazole – there is some evidence of benefit there
> A combination of a methylphenidate and an amphetamine stimulant
> Cholinesterase inhibitors like Donepezil and Galantamine with anticholinergics that will block any chance of the Cholinesterase inhibitor, like tricyclics and many antipsychotics with anticholinergic effects.
And now for the study of the day…. The antipsychotic-stimulant combo rides again.
We make our diagnoses thought language – the clinical interview – and we understand our medications through the names we give them. Those names have changed over time. In the 1950’s, stimulants were called antidepressants, and antipsychotics were called major tranquilizers. From that point of view, it makes sense to combine the two for anxious depression, which is how the combo drug Thora-Dex was advertised, a combination of dextroamphetamine and thorazine.
ThoraDex fell out of favor as the FDA started requiring actual proof of efficacy in 1963. Major tranquilizers had to be rebranded as antipsychotics, and the makers of stimulants were barred from advertising them as antidepressants, as they were unable to find proof that they improved depression for more than a week.
This month, Otsuka pharmaceuticals attempted to revive the antipsychotic-stimulant combo in for a new indication: Treatment resistant ADHD. In a large, industry sponsored trial, researchers from the University of Utah randomized 236 adults with ADHD who had failed to respond to a 5 week trial of a stimulant to augment with either a placebo or the antipsychotic brexpiprazole. It’s been speculated that brexpiprazole may improve cognition through 5-HT1A agonist activity, and it has weak dopamine blocking activity, suggesting it may not interfere with the therapeutic pathway of stimulants. Brexpiprazole has improved hyperactivity in rats – although we should point out the hyperactivity in those rats was caused by amphetamines – which doesn’t exactly speak to a mechanism favorable for ADHD. Brexpiprazole also antagonizes the alpha receptor, while the opposite – alpha agonists – treat ADHD.
So we weren’t surprised by the study’s conclusion. Brexpiprazole did not help ADHD, but it didn’t make it worse. That can’t be said for all antipsychotics, some of which have been known to dampen the cognitive benefits of stimulants, but it’s not enough to endorse this combo pill, which we’ll keep on our list of undesirable polypharmacy.
KELLIE NEWSOME: Join us next week for part 2 where we cover the riskiest polypharm combinations. Earn your CME for this episode from the show notes. Danny Carlat started the Carlat Report 20 years ago to cover all things psychiatric without the bias of industry funding. Your support helps us keep it that way.