These are the medications you don’t want to combine.
Polypharmacy, love it or hate it. Whichever side of the line you stand on, these are the medications you don’t want to combine.
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.
CHRIS AIKEN: “The green reed which bends in the wind is stronger than the mighty oak that breaks in a storm.” Said the Chinese philosopher Confucius some 2500 years ago. And it’s good advice for the budding psychopharmacologist. Study the practice guidelines, learn the evidence base, but your patients will come in with very different ideas about where their treatment should go, and you need to meet them where they are at. Unless, of course, they are asking you to break one of the 10 Commandments. Granite does not bend with the wind, but it lays a firm foundation that will keep your patient safe. Today we are on Commandment #8: Avoid unwise combinations of meds, but first, Kellie can you read out the previous 7?
1. Do not worsen mental illness with psych meds, we’re talking about benzodiazepines in opioid use disorders, psychostimulants in psychosis, and antidepressants during mania.
2. Avoid stopping meds abruptly, particularly benzodiazepines, serotonergic antidepressants, and lithium.
3. Stop lamotrigine if any rash develops in the first 3 months of treatment with it.
4. Watch out for lithium toxicity by staying on top of drug interactions, dehydration, and your patient’s age and renal function.
5. Do not give benzodiazepines to patients who have an elevated risk of opioid overdose deaths – benzos increase the overdose risk 4-fold.
Rewind to our Sept-Nov 2022 episodes for more details on the first 5, and we picked up in March with the rest:
6. Honor thy MAOI interactions.
7. Controlled substances shall be controlled – by you
8. Avoid unwise medication combinations
CHRIS AIKEN: OK we started with that last week, and we kinda put the cart before the horse, talking about how to reduce polypharmacy before singling out the medication combinations that keep us up at night. That’s what today’s episode is all about, but first a preview of the CME quiz. Click on the link in the show notes to get CME for this episode.
1. Although benzodiazepines are best avoided with psychostimulants, which benzo has the lowest risk of causing substance misuse when taken with a stimulant?
Now we’ll walk you through the top combinations to avoid
1. Benzos with stimulants
Avoiding this combination will protect your patient and your license. In 2018, around 1 in 15 patients who were prescribed a benzo were also prescribed a stimulant, and it’s only increased since then, particularly the combination of Xanax and Adderall. The DEA has alerted pharmacists and physicians that it is watching out for excess use of benzos and stimulants, particularly in patients who take opioids.
Most surveys find excessive polypharmacy in areas where doctors are in short supply, like the Southeast, parts of Texas and the Northwest. But a recent study found the Xanax-Adderall combination followed the money – high networth private pay clients were more likely to get it. Make what you will of that.
Both medications activate the reward center, and the epicurean philosophers warned us long ago about the dangers of revving up the hedonic drive all day long. Modern research bears that out: benzos and stimulants have synergestic neurotoxic effects, meaning that high doses of either med are neurotoxic, and that neurotoxicity more than doubles when the two are taken together.
The two also cancel each other out, but not in the sense of “uppers and downers.” The sedative effects of benzos tend to wear off over the first 2 weeks. More worrisome is the cognitive effects. A decade of benzo use impairs cognition with the same effect size that stimulants improve it: a large one, around 0.8. When I read the list of cognitive domains that were affected one stuck out in particularly – problem solving ability. Isn’t that why we give benzos in the first place, to help patients who get overly stressed in the face of everyday problems?
Here’s another danger that is poorly understood but duly noted by the DMV. The combination of benzos and stimulants lead to more traffic accidents than any other recreational drug. This includes people with recreational use as well as prescription use. We’re not sure why, but it appears worse with amphetamines than methylphenidate. Something about that dopaminergic confidence boost and that coordination impairing benzo is just not good for highway safety.
If you do encounter patients on the benzo-stimulant combination and need to taper them, consider changing the medications first. Methylphenidate is less risky than amphetamines, and modafinils are safer still – Modafinils like Provigil and Nuvigil actually have neuroprotective effects. Among the benzos, oxazepam is the safest here. It has the lowest abuse risk, the lowest overdose risk, and we found one study where oxazepam actually lowered the risk of stimulant overuse – in contrast most benzos made people take more stimulants in these experimental designs. But warn your patient to bear with it if you switch to oxazepam – the drug takes 30-60 minutes to kick in while most benzos take 20 minutes. It’s a lot less rewarding.
KELLIE NEWSOME: You don’t always have to take it away. A benzo-stimulant combo might be appropriate to continue when the patient has legit disorders for which the two are FDA-approved (ADHD and Panic Disorder), lacks evidence of abuse/misuse, and uses the benzo rarely. Dr. Aiken may have said a lot of scary stuff, but consider how often your patients on Concerta have a glass of wine, or more? We don’t know of any studies of synergistic neurotoxicity there, but we did find this from Dr. Markowitz’s lab at MUSC: Alcohol raises methylphenidate levels by 25-40%.
To learn more about how to manage this dicey combo, check out our September 2022 article in the Carlat Psychiatry Report.
3. Antipsychotics with stimulants
This pairing sounds questionable on the face of it – a dopamine agonist with a dopamine blocker; a med that can cause psychosis with one that treats it. But is there any reason to combine the two?
CHRIS AIKEN: The only evidence based rationale is in children with ADHD who continue to have major problems with aggression after the ADHD is treated. There we have a two small controlled trial where adding risperidone reduced aggression at a dose of (0.5-3.5 mg/d). The combination is much more common than you’d expect for a risky med with such small empiric support. In the Medicaid population, anywhere from 1 in 20 to 1 in 5 children on stimulants also take an antipsychotic.
But I spoke with Josh Feder, our child editor, and he pointed out that this is a treatment approach of last resort. Besides creating a pharmacodynamic paradox of blocking and enhancing dopamine, children are more prone to metabolic side effects on antipsychotics. And depakote and lithium both reduce aggression and have studies in the ADHD and conduct disorder population, with fewer side effects in children.
KELLIE NEWSOME: Are there studies in adults with ADHD and aggression?
CHRIS AIKEN: No treatment studies. Most of the research is on follow-up – what happens to these children with ADHD and conduct problems when they grow up to be adults? Not surprisingly, they are at greater risk for antisocial personality disorder, criminal behavior, and alcohol use disorders, as well as depression and anxiety. But how to treat these adults is not well understood.
KELLIE NEWSOME: But a lot of adults with ADHD have emotional lability – they are irritable, reactive, even if they don’t have aggression or criminal behavior. Would antipsychotic help there?
CHRIS AIKEN: Last week we presented a new study where the makers of brexpiprazole looked at whether their antipsychotic could augment stimulants in adult ADHD. It failed to make any difference, but they also looked at symptoms of emotional dysregulation – and surprisingly it failed there as well. Now, if you read the study that part might look positive – it says there was a statistical trend in a secondary outcome of emotional dysregulation. Here’s how to interpret that. It’s like if I came home from a Yankee’s game and you asked who won, and I said, “Well, the Braves won by the main outcome – making it to home plate – but the Yankees hit the ball more often. Not a lot, but a little more often.” This wouldn’t make you think the Yankees were the better team, and if they came with serious side effects – metabolic syndrome, tardive dyskinesia – you probably wouldn’t want to use them – that is, brexpiprazole – for emotional lability in ADHD. Psychotherapy or perhaps omega 3 fatty acids – which have several positive controlled trials in that area – is a better option.
KELLIE NEWSOME: Antipsychotics can also dampen the cognitive benefits of stimulants. These studies tested the combo in a mixed populations of ADHD and health adults, and it doesn’t look good. While stimulants made people respond faster and persist longer with boring, repetitive tasks, antipsychotics like haloperidol blocked those benefits. It’s possible that this effect is not as strong with antipsychotics that are only partial blockers at the dopamine D2 receptor, like aripiprazole, brexpiprazole, and cariprazine, but we don’t know. At least in the brexpiprazole study it didn’t make ADHD better OR worse. On the other hand, pretreatment with antipsychotics do make people less likely to abuse stimulants, and the do treat stimulant-induced psychosis.
CHRIS AIKEN: Outside of ADHD, we worry about people who start a stimulant while taking antipsychotics for mania or psychosis. We’re going to cover this more in a future issue of the Report, but here’s the bottom line. Stimulants can definitely worsen the main problem – mania and psychosis. For a person with a psychotic disorder, even a single dose of a stimulant is dangerous – it causes psychosis 30% of the time when the patient is in remission and worsens psychosis 50-70% during an active episode. The effect is worse at higher doses, and worse for the amphetamines than the methylphenidates. In bipolar disorder, amphetamine is the main animal model for mania, so prescribing an amphetamine is a bit like giving haloperidol to treat Parkinson’s disease. Methylphenidate, again, looks a little safer, but isn’t safe in bipolar I disorder either. There are lots of even safer ways to treat ADHD in bipolar disorder, like clonidine, and we covered them in our November 2021 issue of the Carlat Report.
KELLIE NEWSOME: Antipsychotics and benzos don’t exactly have a dangerous interaction, but in a lot of patients they violate the first psychopharm commandment – don’t worsen mental illness with psychiatric meds. Our next combo pill is dangerous, and should never be used together: Serotonergics and MAOIs. It’s the MAOI here that matters. I mean, lots of people take two serotonergic drugs – an SSRI and lithium or an SSRI and tramadol – without problems. But MAOIs are much more dangerous here – the serotonin syndrome that can erupt is more common and more deadly. Check out page 145 of our textbook Prescribing Psychotropics for a full list of the meds you need to avoid.
CHRIS AIKEN: Then there are combinations that cause pharmacokinetic interactions, slowing down or revving up the enzymes that metabolize each other in the liver. Most of the time, the result is not a dangerous one, but if the medication that shoots to high levels has known toxic effects, beware. Clozapine and the tricyclics are classic examples. Simply adding a SSRI can slow down the enzymes that metabolize these drugs, and high levels of clozapine and tricyclics can cause arrhythmias and seizures. I know of a case where clomipramine was started for OCD. Not an unreasonable choice. It’s FDA approved, and the clinician used the titration schedule that was in the PDR. But the otherwise healthy patient died of a heart attack shortly after starting it. Why? The patient was taking fluvoxamine for OCD when the clomipramine was started, and fluvoxamine significantly raises clomipramine levels. These doesn’t mean you can’t combine the two – the combo has been tested in treatment resistant OCD and it did show merit, including in a small, placebo controlled trial were clomipramine successfully augmented fluoxetine in OCD – but unless the SSRI is escitalopram or citalopram you can’t use the standard dosing. Otherwise, I would not go higher than 50mg of clomipramine when these drug interactions are involved, and I would check the blood level before raising higher (they should be below 450 ng/dL).
KELLIE NEWSOME: Wait a minute, what about serotonin syndrome with that combo – isn’t clomipramine one of the strongest serotonergic meds out there?
CHRIS AIKEN: Yes that is a risk, but surprisingly clomipramine doesn’t cause serotonin syndrome any more than a regular SSRI –again, it’s the MAOIs that we about because of their unique mechanism, blocking the MAO-A enzyme that people need to break down the increased serotonin that an SSRI or clomipramine will put out.
KELLIE NEWSOME: We have a list of drugs that can cause serious problems when drug interactions make their levels go to high. It’s on page 105 of Prescribing Psychotropics. They are: Benzos and z-hypnotics, tricyclics, carbamazepine, Depakote, lamotrigine when starting it, and trazodone – although in the case of trazodone it’s a metabolite – mCPP – which can cause dysphoria, psychosis, and self-harm when its metabolism is blocked and its levels shoot too high. Lithium is also on the list, but here’s a blessing: We know of no psych meds that raise lithium levels.
CHRIS AIKEN: Finally, we have 3 that made the honorable mention list for problematic polypharmacy. These 3 are not dangerous, but they are highly questionable, like
> Multiple antipsychotics, except when one of them is combined with clozapine – particularly aripiprazole – there is some evidence of benefit there
> Cholinesterase inhibitors like the dementia medications donepezil and galantamine when combined with strong anticholinergics. Here one medication – the anticholinergic – is going to block the therapeutic mechanism of the other – the cholinesterase inhibitor. True, psychiatrists don’t prescribe dementia medications too much, but we do prescribe a lot of anticholinergics like paroxetine (Paxil), hydroxyzine, tricyclics, and many antipsychotics in patients who might be taking them. These dementia medications are also used off-label for cognitive symptoms in schizophrenia – not a good idea if they are also taking an anticholinergic antipsychotic.
> A combination of a methylphenidate and an amphetamine stimulant. I see this in practice – patients who take Vyvanse with a methylphenidate either to augment or to add extra coverage at the end of the day. These stimulants have different mechanisms – amphetamine mainly pumps dopamine out of the cell and somewhat blocks its reupdate, while methylphenidate just blocks its reuptake. Who knows what happens when they are taken together? I have not found a single paper reporting on the two together – so this is unchartered territory. If you’re aware of any, please let us know.
And now for the study of the day…. Esketamine flashes in the pan.
Janseen pharmaceuticals, the manufacturer of esketamine (Spravato) just released a surprising study. Their drug failed to separate from placebo after 4 weeks of treatment in a large, randomized controlled trial in China. The company looked at whether pharmacokinetic differences in an Asian vs. Caucasian population explained the difference, but it did not. However, an earlier meta-analysis of all the esketamine trials arrived at a similar conclusion – finding that the benefits dampened down to near zero by day 28 – and that analysis was not industry sponsored. We need to think hard about which of our therapies work better short term, and which should be used long term. The jury is still out on esketamine.
KELLIE NEWSOME: Get daily research updates like this by following Dr. Aiken’s twitter or linkedin feeds @ChrisAikenMD
Earn CME for this episode from the link in the show notes, and get $30 off your first year’s subscription to the full journal with the promo code PODCAST. Your support helps us stay in the ranks of other publications that operate free of ad support like Consumer reports, Mad Magazine, and the Medical Letter.