Episode 271: Rawlings–Thompson A-F Classification, Low Slow Titration & Real-World Management with Dr. Michael Cummings & Dr. Blaire Heath
By listening to this episode, you can earn 1.25 Psychiatry CME Credits.
Other Places to listen: iTunes, Spotify
Dr. Puder, Dr. Cummings and Dr. Heath have no conflicts of interest to report.
Introduction (00:00)
Puder:
Alright, welcome back to the podcast. I am joined by Dr. Blaire Heath, MD. She is a pharmacist, now a child psychiatrist, working in the forensic setting. Also with me is Dr. Michael Cummings, psychiatrist, psychopharmacologist, and a true expert that I've had on multiple times. Today, we'll be discussing some psychopharmacology. We'll be discussing different approaches to side effects. Blaire, do you want to introduce this topic before we kind of jump into it?
Heath:
Well, I really appreciate all of Dr. Cummings' assistance and guidance over the years, and there's so many difficult areas of prescribing that having a psychopharmacology expert to help give us some awareness on how to approach new patients, prescribing situations, and side effects [is helpful]. So I thought we'd start with a little bit of ways that we focus on prescribing and then going into some of the major side effect categories to help us kind of break that down.
Puder:
Yes. Great. Dr. Cummings, do you want to talk about your general approach to prescribing, and especially with a focus on adverse effects?
General Approach to Prescribing & Medication Trials (01:14)
Cummings:
Yes. I'm, first off, pleased to be back. In terms of prescribing, I think it's important for clinicians to conceptualize each prescription as essentially a medication trial. The first goal, of course, is to educate the patient about why you're recommending the medication you are, and what its potential benefits and liabilities are. And then to initiate the prescription and reach what is called minimum response threshold. That is the level of drug either dose or preferably plasma concentration at which at least 5% of the population will show a response. Then continuing titration to one of three endpoints. Either the person gets better, which is of course what we desire, or the person develops intolerable side effects that can't be managed, or you reach the point of futility for that particular medication, which is formally defined as the point at which a further increase in the medication has a less than 5% chance of producing the desired result.
Cummings:
I think if people have that structure in mind, it can save a lot of confusion and not getting through a medication trial. In doing consults, I come across a lot of cases where the trial just sort of stalled and the person never got an adequate trial of medication or they were started on a number of things, all of which were subtherapeutic. And one of the things I'm fond of telling people is that multiple subtherapeutic treatments don't add up to a therapeutic treatment. Did you have further thoughts about that, Dr. Heath?
Heath:
That was kind of a lot of it. I think one of the things that we talked about, like for antipsychotics, I think we've talked about, that the first two weeks you should see some response, but then usually, a clarified full trials, say before clozapine, would be like six to eight weeks. Getting to therapeutic levels.
Cummings:
Yes. There's a, it's not a formal rule per se, but there have been a number of studies that have observed that if you give a medication at a given dose, whatever response you have is about 80% of what you're going to get at the end of two weeks. So typically if someone's on a medication, say an antipsychotic or antidepressant, it's being titrated upward. You have an initial dose, there's not much change after two weeks. You don't need to wait further. You can titrate the dose further and eventually you'll either get a response or you'll reach the point of futility for the drug. I think sometimes people wait either not long enough and titrate too quickly and produce a lot of side effects, or they wait too long recalling that the drug won't, in many cases, reach its full effect for four to six weeks, but at two weeks the antidepressants and antipsychotics at least will produce about 80% of what they're going to.
Puder:
Is that the case for OCD and anxiety?
Cummings:
OCD has a longer response curve so that you could roughly double that and say at the end of four weeks at a given dose you've got about 80% of the response you're going to. I think the other issue with OCD is people sometimes mistakenly use the dose ranges for major depressive disorder, whereas the required dose ranges for OCD are typically higher. For example, I've treated a number of OCD patients where they required 300 to 400 milligrams of sertraline. Whereas for depression, the quoted dose range is 50 to 200 milligrams per day.
Heath:
I was going to say, I really like how, what he was talking about just a few minutes ago, where I think one of the biggest things that have been helpful is not being afraid to do a low and slow approach. Dr. Cummings, you're saying people go up too quickly, so especially as the receptors, we talked about them, upregulating, downregulating, and of course, the sensitivities based on metabolism. Can you talk a little bit about trying to address side effects and those high doses?
Managing Side Effects: Low & Slow Titration, Metabolizers & Receptor Upregulation (06:09)
Cummings:
Yes. We're touching on the Rawlins–Thompson Classification of Side Effects. It's well, currently it's A through F types. It started out as just A and B expected and idiosyncratic (Aronson & Ferner, 2003; Rawlins & Thompson, 1977). The A type is related to the known receptor activities or other activities of the drug. And these are the most common side effects that we see. For example, if you're giving somebody a drug that is a good alpha-adrenergic antagonist, if you go too quickly, you'll make the person hypotensive. They'll be dizzy when they stand up. If the drug also has antihistamine properties, you'll make them sleepy. And they'll complain of fatigue and lethargy. And if the drug has anticholinergic properties, of course, they can get everything from blurred memory, blurred vision, rather memory impairment, GI problems, urinary retention. All of those can be mitigated by titrating the drug more slowly. And indeed, one of the points of getting plasma concentrations for things like the antidepressants and the antipsychotics is if you have a slow metabolizer, you need to titrate the drug in smaller steps because a smaller change in dose produces a bigger change in plasma concentration. The opposite is true, if you have a rapid metabolizer. You may need to make bigger steps in order to get the same amount of increase in drugs.
Heath:
Do you identify a certain approach? Like, this is just for me, if someone seems like they've had a lot of side effects from a lot of things, to me that identifies, I don't know for sure, they're a slow metabolizer, but I'll go really low and slow to try to establish that rapport. And, we can, it may seem even, I don't want to say ridiculously slow, but for, you know, a person, and someone took a year to cross titrate from one to another.
Cummings:
Yes. It's always fair to ask people, “When you've taken medications in the past, are you the sort of person who generally tolerates medicines pretty well, or do they cause you a lot of problems? And if so, what kind of problems?” That may help you. Not only with the rate of titration, but also with medication choice. Because, in many cases, the person can tell you what kind of side effects they're sensitive to and you can pick something that is not a particularly bad actor in that area.
Heath:
And you talk about the bedtime dosing, not doing stuff more than twice a day. I saw this a lot. [Working] as a pharmacist, people do give stuff three times and four times a day. People always forget that middle dose.
Cummings:
Yes, most of the medications that we use in psychiatry, there are some that during the titration of the drug, you need to give it in divided dosing, essentially, so that you don't have very high peak levels and produce side effects that way. But once you get the person adapted to the drug, most of the drugs we use can then be consolidated to bedtime so that two things happen. One, the peak occurs during sleep when it's not likely to bother the person. Not likely to produce side effects. And frankly, as you point out, adherence is a lot better if somebody's taking something once a day. Most of us are pretty good at remembering to take something either when we get up or when we go to bed. I have frankly yet to see anybody manage to take all of the doses in a five time per day regimen of antibiotics, for example. Invariably in the middle of the day, it's all, “Did I take it or didn't I take it?”
Heath:
So I think that's a really good thing. But I like how you pointed out that we don't always think about the concentration of the medication correlating to side effects. And so, I know that's something we talked about with guanfacine IR [immediate release] and ER [extended release], even though the half-life technically of the IR is 17 hours and the ER is 18 hours, you talked about how it was, I think, coated; and how the side effects could be based on the peak times that we don't always….
Cummings:
Yes. For most of the drugs that have concentration-related side effects, if you can make the peak less steep, the drug becomes more tolerable. A good example of that, for example, is valproic acid. Divalproex immediate release produces about the same side effects, but produces them about twice as often as the extended release formulation of Divalproex ER [see episodes 71 and 214].
Heath:
Thank you. And one of the biggest things I know you get a lot of questions about, that's really helpful, is a lot of us will get a really difficult patient. Right, Dr. Puder? That they'll come to us on multiple controls [controlled medications]. They might be on one or even two benzodiazepines and a stimulant, and then they're, “This is this is my last dose,” and you're worried about withdrawal seizures or you don't want to take on this patient, but then you feel some degree of, I don't want to say, responsibility or even liability if you don't prescribe or continue the regimen. How do you approach that situation?
Managing Patients on Complex Med Regimens & Withdrawal Seizure Concerns (11:36)
Cummings:
Well, you know, if you're treating somebody, if you've accepted them as a patient, you do have a legal obligation to them. There is a legal concept called “abandonment,” meaning that you're not allowed simply to say, “Well, I'm out of here. I'm not going to see you anymore.” Now, that doesn't mean that you can be forced to prescribe things in the long term that you don't want to. If somebody is demanding a drug that you don't think is appropriate for them, or they're misusing the drug, you can certainly tell them, you know, “Unless you use the medication reasonably, I won't be able to continue to treat you.” Your obligation in that sense, if the person simply won't agree, is to say, “Well, I will manage you in the short term and I will provide you with referrals to other clinicians.”
Puder:
One of my thoughts right off the bat with that one, Dr. Heath, is sometimes they'll come in saying that they're taking meds that they haven't really been prescribed, lately. So say I am prescribed and I'm running out of Xanax 2-4 times a day; and, nowadays we can look up on CURES [Controlled Substance Utilization Review and Evaluation System], or various different PDMPs [prescription drug monitoring programs], and see what they've actually been given recently. And so, there's sometimes strong inconsistencies in what they say they've been on consistently and what's been documented that they've been prescribed or how frequently they've been prescribed.
Heath:
Now, if it's your first visit with a patient, do you feel like you've officially taken them on? Or is that first visit, the evaluation, whether or not you're going to take them on, is that responsible?
Cummings:
It need not be acceptance of them as a patient at that first visit, but you need to be clear that you're, if that's the case, that your first meeting is an evaluation and you'll decide whether you will take them on as a patient.
Puder:
I personally have that in my intake paperwork, that the first visit is a consultation. It's just, you know, we are trying to find mutually agreed upon goals and directives, and that it's not establishing a patient-doctor relationship. And often, the people who come in with a lot of meds when you tell them that your plan would be to decrease them over time or that this is not, you know, appropriate, they will not want to continue to see you again and maybe not even pay you, since I'm, at this point, cash pay practice. And that's just a reality.
Cummings:
Yes. Dr. Puder makes a very good point. Having that in your intake paperwork and having the patient acknowledge that in some cases by signing said paperwork it makes it clear that the treatment relationship has not yet begun and you're seeking to come to an agreement about working together.
Puder:
I had a patient the other day that they almost wanted me to agree in the first meeting that I was going to completely not change any meds, and they also were very guarded to give a release for me to talk to their previous psychiatrist, and they wouldn't. And I sent them a nice email and I said, “Hey, I don't think this is going to be a good fit for us.” And they said, “Yes.” They agreed. So I think that there needs to be, for me, I want to know why they were put on this cocktail of medications. What was the thought process? Were they actually on this? What the story is, especially if they come in very demanding, very sure that this is the only medication that has ever helped. So yeah. What do you think? Is that reasonable? Michael Cummings?
Cummings:
Yes, I think that's very reasonable. You know, basically the doctor-patient relationship is just that, it's a relationship. It's essentially a business relationship in which the patient or their insurance pays you for services, and you provide services that you've been licensed to provide by the state. But you have to, in order to have a productive working relationship, you and the patient have to have an agreement about what the goals and methods are going to be. You know, I'm very careful about educating patients about medications. That's an inherent part of prescribing. I know a lot of, well, I know from doing consults, a lot of physicians don't spend enough time talking about why a particular medication would be desirable or why another medication might be undesirable in this particular patient's case. And frankly, also with respect to side effects, a lot of side effects are much more manageable if you tell the patient up front what to expect.
Heath:
I know you've mentioned that before, like trying to have that open channel of someone [a patient] being able to come back and be able to adjust things with you.
Cummings:
You know, and I think sometimes because as physicians we tend to be familiar with the potential side effects of a given drug, we sometimes forget the patient may have no idea what to expect. And it can be very frightening if, you know, you take a drug and the next time you stand up, you feel dizzy, or you feel unusually sleepy the next morning after you take a medication at night. You know, “Am I going to die of this? Is this bad? What's happening?” But if they've been educated and say, “Well, when you start taking this drug, these things may happen, but they will typically get better. They're not dangerous.” That goes a long way toward making the medication more tolerable.
Heath:
Yes. I had a patient that was actually, I found here, in the hospital, when I called him. I'm like, “Oh dear, you're in the isolation room with TB [tuberculosis];” and I was going to get off. And the patient was really scared because their urine had turned red because it was one of the TB medications. I told him, “Oh, that's actually a side effect of medication.” And no one had told him. And so it was just kind of, you know, this person was in the hospital already. But it was like a really scary side effect. He was like, “Oh, okay. That's why I'm having red tears.” You know, that expectation can make a really big difference on how people handle things. So are there any other things, Dr. Puder, that you wanted to add or are you okay if we move on to those different types of side effects?
Rawlins–Thompson Side Effect Classification: Types A–F Explained (18:28)
Puder:
There was one thing Dr. Cummings said earlier that there were different types of side effects. Okay. So we're going to go through each one, one-by-one.
Heath:
And so, we had to kind of expand. So we just thought we'd go over the six categories. This is based on, I don't know if you want to mention, Dr. Cummings, but the Merck Manuals, there's the A through F (Aronson & Ferner, 2003; Lynch, 2025; Rawlins & Thompson, 1977).
Cummings:
These appear in the Merck Manuals. They're actually called the Rawlins–Thompson Side Effect Classification. You know, like everything else, there's more than one classification out there. This one is particularly popular though. Type A is an extension of what the drug does either as its primary effect or from other interactions. In other words, it's part of the known pharmacology of the drug. For example, I think in the Merck Manuals, the example to give is if you were to give somebody too much warfarin, bleeding could become a side effect. But same thing if you gave somebody too much antihypertensive, you might make them hypotensive. We mentioned briefly drugs…a lot of the drugs we give are alpha adrenergic antagonists, or they're anticholinergic, or they're antihistaminic. So they can make people orthostatic, make them dizzy, or they can give them anticholinergic side effects, or they can make them sleepy.
Cummings:
All of those are Category A side effects. Category B are idiosyncratic, unexpected side effects. A classic example of that is allergy. You know, the person takes the drug and no one expected, but they are allergic to it, and they break out in a rash and you know, they're itchy. All of the classic signs and symptoms of allergy, that's a category B. Category C are chronic exposure side effects. If you take the drug over many, many years. Probably, for us, a chronic side effect is development of renal insufficiency when a person's taken lithium for decades, as a good example. Or osteoporosis, if the person's been taking steroids for many years, is a category C side effect. Category D are delayed effects. These are things like you've been exposed to the drug, you're not necessarily taking the drug now, but the drug did something that causes a later adverse effect. Good examples of that are teratogenic risks with some medications.
Cummings:
And then, [Category] E is ending treatment. These are the drugs that can produce withdrawal syndromes. As we were alluding to, if somebody's been taking benzodiazepines for years, or they're taking an opioid over a long period of time and the drug is stopped abruptly they will of course have withdrawal signs and symptoms. That's category E, most easily avoided by tapering the drug. And then finally, there's Category F, which is an unexpected failure of the drug. The classic example of that is you give somebody an antibiotic for an infection and they have no benefit from the drug, the microbe has become resistant to that particular antibiotic, but you didn't know that going in. For the psych meds, I think an important issue with failure is it's a good signal for us that we need to go back and reconsider the diagnosis.
Cummings:
If we give somebody, you know, trials of several medications that should be effective for what we think the person has, and they get nothing, you know, no benefit, it should cause us to go back and say, “Well, so there's something in the diagnosis I'm missing here. Do they have something else that I missed?” And I may need to take a completely different training approach. We've discovered some things like that. That's how anti-NMDA antibody psychosis was discovered. They were mostly young women. Initially, they were diagnosed incorrectly as having schizophrenia, but they didn't particularly benefit from the antipsychotics. It just got worse. And then they started developing cognitive problems, and eventually people started saying, “You know, this isn't schizophrenia. This is something else.” And they finally figured out that indeed, that it was an antibody inflammatory encephalitis, and they needed steroids. They didn't need an antipsychotic (Dalmau et al., 2019; Kayser & Dalmau, 2011).
Puder:
That's a great point. When there's one or two med failures, you should think, “Am I getting the full picture? Is there something I'm missing? Am I missing personality disorder or style that may be influencing things, or secondary gains in the illness role?” Or, we reexamine if there are drugs or alcohol on board, or something, you know? I've had therapy patients where it's like years into treatment and then finally it comes out. They've been drinking. So it's like we do not know within one session if a person is using or how much they're using. Often shame keeps the truth from us. Right?
Cummings:
Yes. And you know, always if the treatment response is not what we expect, we need to be very careful not to forget that our diagnostic impression may not be right or may not be complete. So we need to be willing to go back and examine that. I've seen way too many cases where, because somebody, way back when, wrote down a diagnosis in the chart, everyone after that just sort of automatically believed that that was the diagnosis. Sometimes, to the patient's detriment. One consult I did for the Department of Corrections, now, the patient, when they were hospitalized, when they were imprisoned, rather, they maligned having psychotic symptoms. They got put on an antipsychotic and eventually developed tardive dyskinesia. You know, that patient unfortunately never had psychotic illness. They were malingering in order to get better housing because they were trying for a single cell, a single person cell. Unfortunately, to their own detriment, eventually. But nobody ever questioned whether they actually had a psychotic illness, even though people would write repeatedly that no psychotic symptoms were being observed. They were simply being verbally reported. You know, so we need to be careful about diagnosis.
Puder:
I remember a patient who had about 20 abdominal surgeries by the time I saw her. And in getting a close history, there was a psychosomatic flare to it, but she had developed real consequences of surgeries that were exploratory. And, you know, she had gotten addicted to trying to have another surgery for the benefit of having a relationship with a surgeon and this connectedness that she was getting from that. So real consequences came from this. I think that's what you're describing here, Dr. Cummings, in your example. It's like real consequences came from the side effect of the actual treatment. And this is where it gets complicated, as we're trying to pull the threads of the diagnosis and [determine] what happened first. Right? Was there malingering first, or was there factitious disorder or some sort of secondary gain that they were seeking first? Let's go ahead.
Heath:
That goes back to the education, when you first start the medication. It's like being very real: “You know, you have to report these symptoms. You might want this medication, which we have to determine whether or not it's appropriate, but realizing this is what you may be taking on as a side effect.” And I think sometimes people may backtrack a bit going, “Oh, okay.” You know, trying to find those true symptoms, because they may realize, “Oh, maybe it's not worth that secondary, that adverse effect because of the secondary, you know, the immediate gain.”
Cummings:
Yes. You know, a lot of the treatments we've prescribed, they're beneficial in the right context, but if they're overused or if they're used in somebody for whom that isn't the right medication, almost all medical treatments come with a risk.
Heath:
Yes. Stimulants, if they don't really have ADHD, have real risk and consequences. Something we get a lot of. Going back to the, so go a little bit more in depth in each category. The Type A, augmented, you mentioned the bleeding with warfarin, or Coumadin. And you touched upon the orthostatic hypotension. I wonder if we could go a little bit more into that. I know it's something you get with a lot of medications. You mentioned the slow titration can help decrease the amount of dizziness (Bhanu et al., 2021). Can you explain?
Type A: Augmented/Dose-Related (27:31)
Cummings:
Yes. A lot of the drugs that we use in psychiatry, a lot of the antidepressants, the antipsychotics, in addition to the thing we're wanting them to do their alpha adrenergic antagonist. So when you take them and they block alpha receptors, it makes the blood vessels expand. Well, you know, pressure is how much fluid you've got and how small a vessel. If the vessel gets bigger, the pressure goes down. And of course, if you're standing upright, you have to have enough pressure to get the blood from your heart up to your head, otherwise you'll faint. And you know, the way to mitigate that, in most cases, is to titrate more slowly. Don't give the person more than they can handle until their alpha receptors upregulate and can resist the effect of that particular effect of the medication. And people who you really need to give them the medication, but they're incredibly sensitive,
Cummings:
There are other things you can do. You can use fludrocortisone to make the person retain more fluid. That's essentially adding to the volume of their blood and filling up the blood vessel so their pressure stays up. There are some people who are very prone to orthostasis, with or without medication. Some of those people get treated with an alpha agonist like midodrine to constrict their blood vessels. But in the vast majority of people, the trick to not causing dizziness is simply to slow down the titration and give the receptors time to catch up. Same is true of a lot of the other effects, like drugs that block histamine receptors and make people sleepy. If the histamine receptors upregulate, that sleepiness tends to go away. You can also help that by the timing. Don't give a person a major dose of antihistamine drug first thing in the morning. You know, they might be better at night, when it can maybe actually help them sleep. Anticholinergic effects, most often the issue there is be cautious about not adding multiple anticholinergic medications. Sometimes people forget that a lot of the medications we use are anticholinergic, and if you're giving one that's anticholinergic, try to avoid others. In other words, avoid adding to the problem. [great episode on this topic https://www.psychiatrypodcast.com/psychiatry-psychotherapy-podcast/anticholinergic]
Heath:
Thank you, that's really helpful. But a lot of times people might be on, you know, olanzapine and trazodone and, you know, on several things that can cause that orthostatic or then have ….
Cummings:
Yes, a lot of these because these are off target receptors. It's important to think about, “Well, if I'm giving this drug and it's a good alpha blocker, or it's a good antihistamine, or it's an anticholinergic drug, maybe if I need to add a second drug, I need to be looking at things that don't share those properties, so I don't just add to that particular side effect burden.”
Puder:
When I think about this specific issue of orthostatic hypotension, I think about trazodone, clozapine, Seroquel, risperidone. And probably the biggest issue I've had is with clozapine, where we really have to add in a secondary medication to raise the blood pressure. Like midodrine.
Cummings:
Yes. Yes. Clozapine, despite being the gold standard for treatment of treatment-resistant psychosis, unfortunately it comes with a whole host of side effects. It's an excellent alpha blocker, it's very anticholinergic, and it's also a very potent antihistamine. So, if you go too quickly, you've got somebody who has no blood pressure, they're lethargic, and they have all of the panoply of anticholinergic side effects.
Puder:
Here would be a good board question: “[The] patient is on Clozapine, but has some PTSD, as well, and is having nightmares. [The patient is] already having some orthostatic hypotension. Do you treat with prazosin?” What do you think, Blaire?
Heath:
I would say, I would avoid the antihypertensive, but I [would] maybe use some Topamax that has some off-label PTSD [applications] (Berlant, 2004), and with the decreased appetite…a lot of them, the prazosin, the clonidine, guanfacine, all our blood pressure ones. So I'd probably go a little bit off the beaten track for maybe concerning something like that, which I've had that benefit in some of my patients [who] actually, they can't do the antihypertensives.
Cummings:
Yes, that's a good point. You want to, if you can, avoid drugs that will add to the problem. For example, drugs that are alpha antagonists, or that are alpha-2 agonists [which] cause sedation and lower blood pressure. If you're giving the person clozapine that may not make a great combination.
Heath:
And sometimes, people put an alpha-2 agonist with an alpha-1 antagonist. Like, I've seen the patient that came to me on two of them together, like prazosin and clonidine, or something. It's like, well, you're kinda, it's really one or the other. Right, Dr. Cummings?
Cummings:
Yes. Well, and frankly, prazosin has taken a hit, you know, the VA (Veterans Affairs) did a very large scale study of prazosin in combat PTSD and found that it had zero effect on nightmares (Schrader & Ross, 2021).
Puder:
So initial studies were more positive than more recent studies. Is that what you're saying, Dr. Cummings? Were initial studies on prazosin more positive, and then…?
Cummings:
The initial studies were positive. In fact, a lot of the literature started at the University of Washington and the VA in Washington state. The VA, finally though, did a large scale national study with literally thousands of veterans. Some of them are well randomized to either be on prazosin or not. And they found that essentially prazosin did not achieve a significant difference over placebo. Now people have looked at that and said, “Well, maybe it's such a large study that it may have buried some subgroups that might be responders to prazosin.” But it certainly suggests that prazosin was not nearly as effective in treating nightmares in PTSD as people had expected.
Heath:
Sometimes having a creative, you know, looking at that overall picture and seeing is it really effective, or…that can be kind of a tough one to approach. So, thank you. Another one that we, I know we've gotten a lot of questions about, is a common side effect that would go under that Type A category, is sweating. I know I've heard a lot with SSRIs [selective serotonin reuptake inhibitors], especially SNRIs [serotonin and norepinephrine reuptake inhibitors]. Do you want to talk a little bit about that? Maybe how we can approach that?
Cummings:
Yes. As you know, the SSRIs block the reuptake transporter for serotonin, which of course, means that they increase the amount of serotonin available in synapses. And of our sweat glands are modulated in part by well, they're modulated by acetylcholine, serotonin, and norepinephrine. Norepinephrine will tend to decrease sweating, because it decreases blood flow. Serotonin, it opposes that and actually increases blood flow to the sweat gland. And then, some people that will produce a huge amount of sweat, people develop hyperhidrosis; and I've seen one or two patients who had such a severe reaction to increased serotonin that they literally were just dripping. You can, in milder cases, treat that with an anticholinergic medication most often. But it can be a bit ineffective in really severe cases, and you have to look for something that is not an SSRI to treat their depression (Cheshire & Fealey, 2008).
Heath:
I'd say that glycopyrrolate has been really effective. I know that's why you talk about more of the peripheral [action], but not with central [nervous system effects].
Cummings:
Yes. Glycopyrrolate is an anticholinergic that does not effectively cross the blood-brain barrier. So it's often used when we're wanting to give somebody something that's anticholinergic that doesn't affect the acetylcholine in their brain. Another one that's currently in use is trospium chloride. As you know, it is currently being used in the combination medication Cobenfy, which is xanomeline, an acetylcholine agonist and trospium, an antagonist. Trospium is being used to block the xanomeline peripherally to try to make the drug more tolerable.
Heath:
So the serotonin, is it also the norepinephrine that plays a part in this, the sweating?
Cummings:
Norepinephrine? It tends to decrease sweating because it decreases, it causes vasoconstriction.
Heath:
Oh, okay.
Cummings:
Which, you know, sweat is basically, when you sweat that the liquid is being derived ultimately from blood plasma. So if you have decreased blood flow to the sweat gland, you'll have decreased sweating.
Heath:
I think one of the most interesting points that I had never thought about is that you talked about the location of the glands and that the sweat might even be… it was so funny…I had a patient [who] actually said it was “stinky sweat” versus maybe, “night sweats,” and that actually, the location and the type of sweating may even vary. I thought that was an interesting point you made before.
Cummings:
Yes. Well, you know, apocrine sweat glands produce not only essentially what amounts to saline, they also produce lipids, as well. And it's frankly, bacterial action on the lipids that causes people to smell bad when they sweat.
Puder:
That's why a good tea tree, essential oil, is good. It kills the bacteria and fungus a little bit. Blaire, good questions. I really appreciate you.
Heath:
These [questions] are ones that have come over time, [from] different providers, and these are really common side effects that are really hard. And I think, on the other one, that was a really good discussion that we've had. I mean, there's so many coming, so helpful, but a really interesting one that I don't know if you've come across this also, Dr. Puder, is the menstrual irregularities with the SSRIs? Dr. Cummings, I know we've talked about that before, having the hormonal fluctuations being impacted by that (Weatherly et al., 2025).
Cummings:
Yes. You know, well, late luteal phase dysphoric disorder, also sometimes called premenstrual syndrome, is an actual deficit in serotonin neurotransmission. Which is why if somebody has a very regular period, that is, you can predict when it's going to occur, they can actually treat their late luteal phase dysphoria by taking an SSRI for only a, well, usually for about 10 days before their menses. And then, once the menses starts, they can stop the drug and wait until the next month and do it again. Of course, the problem for people who are not very regular is if you can't predict, then you wind up having to take the SSRI on a more ongoing basis, even though you really only need it for about a week to 10 days before the menses.
Heath:
So if someone is taking a SSRI, say every day, and a female reports menstrual irregularities, there really is a level of the SSRI impacting the hormones and not just their mood. So taking it on a regular basis, would you be concerned about that or just say that can be part of the medication acting on that level?
Cummings:
Basically the SSRIs, well, the hormone, it's a two-way street. The hormones alter receptor sensitivity in the brain for both dopamine and serotonin. And in turn, the neuromodulators, the monoamines alter the secretion of some of the hormones. So that you can, in some cases, if the modulation is not maintaining homeostasis, you can then help regulate it by using drugs that give the person a more consistent level of serotonin.
Heath:
Like oral contraceptives if it concerns them, basically.
Cummings:
You can see this more broadly. There's a reason that women during their reproductive years are four to five times more vulnerable to episodes of depression than are age-matched men. Interestingly, either before menstruation starts or after menopause, the rates of major depression tend to be one-to-one between the sexes. You know, the up and down cycle of the menstrual cycle is, frankly, from the standpoint of the functioning of the brain, is kind of hard on brain function.
Puder:
I actually wanted to go back to sweating. Because I did an episode on sauna [see episode 221], and one of the things that I found was that some people actually are on medications that decrease the ability to sweat. And they become more heat intolerant.
Cummings:
Yes. If they're taking anticholinergic medication, they can become dangerously overheated either in a sauna, or certainly here, in southern California, where I am, just summertime can be dangerous to them.
Puder:
So, anticholinergic medication. I'm thinking amitriptyline, oxybutynin, benztropine, tricyclic antidepressants, clozapine, olanzapine, quetiapine.
Cummings:
Yes. Anything with significant anticholinergic properties will decrease the person's ability to sweat and therefore, to lose heat. And if they're in a very hot environment, that can lead to either heat exhaustion or heat stroke.
Puder:
Also, thinking about antipsychotics, especially first generation ones, with their high affinity to the D2 receptors (D2R) can cause hyperthermia through neuroleptic malignant syndrome and potentially through non-NMS mechanisms by affecting the hypothalamus. Any thoughts on that?
Cummings:
Yes. Basically, the very potent dopamine antagonists interfere with D2 receptors and the hypothalamus and can make people more poikilothermic. That is, they tend to then drift toward the environmental temperature, either hot or cold. There have been cases where people living in cold climates, who’ve taken a potent dopamine antagonist, may become hypothermic more easily. Or, if they're living in a very hot climate, they may become hyperthermic.
Puder:
Yes. So, if you're getting into sauna [utilization], like I am, and you're on an anticholinergic medication, tread carefully and maybe start slow. Monitor your internal temperature.
Heath:
You can't use one to offset the other, Dr. Cummings?
Cummings:
I don't think that would be a very safe approach. It would be far better to be sure that you're well hydrated. And as Dr. Puder suggests, monitor your internal temperature.
Cummings:
Be sure you're not becoming hyperthermic.
Puder:
Put like an anal thermometer in there. You know, have some doctor oversight. Okay. So this was all talking about type A, which is augmented dose related, along with the drug’s known pharmacology. And now let's talk about type B.
Heath:
Which is bizarre.
Puder:
Idiosyncratic.
Heath:
Yes. And, a really good discussion we were just having recently was talking about some of the paradoxical effects we don't expect. And one of the things we're talking about, were caffeines, ephedrine-like stimulants, causing people to get tired. Which, on the flip side, something like Benadryl, making people more active, hyperactive and actually what you wouldn't expect. Can you expand on those, Dr. Cummings?
Type B: Idiosyncratic/Bizarre (44:58)
Cummings:
Yes. Probably everyone has observed that if somebody drinks too much coffee too quickly, they may complain of feeling tired, sleepy, lethargic afterward, which is not what you would expect from caffeine, which blocks adenosine receptors. Adenosine is a minor inhibitory pathway in the brain for decreasing arousal. If, however, you kind of poke that system a little too hard, you'll cause a GABAergic response. The major inhibitory neurotransmitter will be activated and it will shut down whatever increase in alertness you were having from the lower levels of caffeine. So you can essentially, it's an example of poking the bear and the bear fights back. The same thing is true with other stimulant drugs. There tends to be a counter response by GABA, and GABA's very effective at turning off brain activity. You can also see this with antihistamines, where if you block histamine receptors, most people, they'll get drowsy.
Cummings:
Though, particularly in children and the elderly, if you block histamine receptors, you may turn off their cortex because that's where histamine receptors reside in the brain, for the most part. And their subcortical systems, their basal ganglia, and the more inferior posterior parts of their limbic system are now free of any regulation by the cortex that you've put to sleep, and the person becomes idiosyncratically excited. And in the case of kids, they kind of become, they look like they're on the sugar rush of all time. They've, you've created hyperactivity. Fortunately, as it wears off, they will calm down. Same thing in the elderly, you'll see somebody become very agitated. The children are usually easier to contain just because they're physically smaller. Occasionally, in the elderly, it can become dangerous because they may have become agitated to the point where they're at risk of harming either someone else or themselves.
Heath:
So that's something we don't always expect. But I know, especially as a child psychiatrist, the way that children respond to medications is not always what, it's not a mini adult with the same response.
Cummings:
Yes. No, children are not just small adults.
Heath:
We see some more of that paradoxical agitation and things that we don't always expect as much. So thank you. That's an interesting thing that I think, a lot of times, we don't think about what causes it.
Puder:
But, can I ask a question here? Because we're talking about Type B, bizarre, idiosyncratic. But Dr. Cummings is talking about a mechanism for how to make sense of it. Is that bizarre and idiosyncratic then, or is that…?
Cummings:
It fits into the idiosyncratic in that it is not the expected response. You know, human beings have consumed coffee and tea, other caffeine-containing beverages, mostly to produce an increase in alertness. So, it's idiosyncratic in that if you push too hard, you can produce a countervailing response in the brain. The histamine response is also, it's not bizarre, but it is unexpected, in the sense that in the vast majority of people, if you give them an antihistamine, they'll get sleepy. But in a few cases, they'll become very hyper-alert, very agitated. And that tends to show up mostly in children and the elderly. I don't think I've ever actually seen a case in an adult, although I've read that there have been adults who've had a similar response to antihistamines.
Puder:
And what's your mechanism for the antihistamine in kids, for why they would have that response?
Cummings:
The antihistamine, basically, kids’ brains and elderly brains are not as well integrated in terms of cortical and subcortical systems. And in the case of H1 histamine receptors, those are mostly distributed in the cortex. They're, you don't really find a lot of histamine receptors in subcortical systems. So if you give somebody an antihistamine, you're basically decreasing neuronal activity in the cortex in a more integrated brain, that results also in a decrease subcortical via other communication systems. But if those subcortical systems are not all that well connected to the cortex yet, or they've lost some degree of connectedness, the subcortical systems may sort of run amuck because they're no longer being modulated by the cortical systems.
Puder:
Very good. And did you want to bring up alopecia?
Heath:
Yes. This is, I think, something I've definitely heard [about] from different medications; and we've talked about it, Dr. Cummings, is how medication can disrupt normal hair growth during the growth phase (Alhanshali et al., 2023). I think it was anagen, telogen—the resting phase. That hair loss is really disturbing to a lot of patients.
Cummings:
Yes. You know, probably the most common drug that we use again, it's not a bizarre response, but certainly unexpected, is valproic acid is great at causing alopecia. And in the case of this drug, it's because, well, frankly, we didn't realize for many, many, many years that valproic acid, aside from being an antiepileptic and a mood stabilizer, it's also a deacetylase inhibitor, a histone deacetylase inhibitor (HDAC). Histone is the protein that wraps around your DNA and you have to get it to unfold in order for the DNA to be replicated. If you give a deacetylase inhibitor, like valproic acid, it prevents the histone from unwinding and therefore slows down mitosis. Turns out, valproic acid essentially is a chemotherapy agent. And like many of the chemotherapy agents, it can cause alopecia because if the cells in your hair follicle can't undergo mitosis and produce adequate hair protein, you go bald.
Heath:
Some of the other ones talked about lithium I think, acutely, with the hair loss.
Cummings:
Lithium has some similar effects in terms of decreasing the rate of protein synthesis in hair cells and the follicle, and therefore, it can produce alopecia.
Heath:
How do you know if hair loss is from a medication? And this may be too broad of a question versus from a different cause or from more pattern, you know, it's more 'cause of the pattern balding or thinning.
Cummings:
For a lot of adverse side effects, at least the Category A and Category B, essentially, there's a fairly clear temporal association between the presence of the drug and the adverse effect. When you get into some of the more chronic adverse effects and/or the delayed effects, that basically depends a lot on knowing that that's a property of those drugs. You know, it took us decades to figure out that lithium can cause renal insufficiency, and it took even longer for us to figure out that the best way to prevent that was to be sure that we never prescribed lithium in a divided manner. In fact, to this day, if you go back and read the package insert for lithium, it says to give lithium in a divided dosing schedule, which is exactly wrong because all you do is increase the risk to the person's kidney (Castro et al., 2016).
Heath:
How would you…? So it sounds like a lot of time looking at when the agent was started, and I think we also talked about the pattern. Like, would it be more of an overall hair loss versus maybe a certain pattern of it? Like more like male pattern balding, for example.
Cummings:
Most medications cause patchy hair loss.
Heath:
Okay. Okay. That's helpful.
Cummings:
Of course, if it goes on long enough, and it's severe enough, then the person becomes completely bald. As in the actual, the more robustly chemotherapeutic agents, used to treat cancer, most of those people go completely bald. Although, if you watch their baldness unfold, it's initially patchy and then it becomes contiguous.
Heath:
Okay. Good point. How would you treat the hair loss? Would you decrease the agent, change it, or how could…?
Cummings:
There are a couple of things you can do. Selenium tends to be protective in both valproic acid and lithium cases. So you can try supplementing the person with selenium. These days, you also have, well, what was developed originally as antihypertensive, minoxidil. It causes vasodilation. So it improves the blood flow to the hair follicles when applied topically. Some of the dermatologists are also now using laser stimulation of the scalp. Again, the underlying mechanism is to increase blood flow to the follicle. And that can improve or reverse hair loss.
Heath:
A real concern for some patients. Dr. Puder, did you want to lean in, at all?
Puder:
No, this is great. Any other common medications outside of lithium, valproic acid, GLP agonists, cytotoxics, biologics?
Cummings:
Certainly, since they're a hot topic at the moment, the GLP-1 agonists. If the person has too much weight loss, too rapidly, it can cause a number of things: hair loss, loss of bone, and loss of muscle, as well. You know, we've had a couple of cases where the person went from being obese to being underweight. And, they were essentially too responsive to their GLP-1 agonist, and they needed…. Fortunately, in both cases, they needed, and got, their doses lowered so that they were not going to starve to death. The last patient we had, who had to be taken off the GLP-1 agonist for a while, basically, she had no appetite. She would get up in the morning, she had one piece of toast and a glass of water and felt full for the whole day. Of course, you can't live longterm on one piece of bread and a glass of water a day.
Heath:
Very important. Thank you. And moving on to the chronic, dose related, and time related. you mentioned the osteoporosis and steroids, which is kind of a pretty well-known long-term effect.
Type C/D: Chronic & Delayed Effects (56:24)
Cummings:
The lithium and renal insufficiency is a good example of a chronic effect.
Heath:
Can you expand a little bit more on that and how we can, you mentioned the once a day dosing.
Cummings:
Once a day dosing, yes. Lithium traps itself fairly effectively in the distal principal cells of the nephron. The best help for that is to give the kidney time to clear the lithium out of those cells. Lithium gets into them very easily but doesn't get out very easily. So it takes much longer for the lithium to clear than it does for it to enter. So the longer the trough time you can give the person between lithium doses, the better. The good news is lithium also traps itself in the brain. So the brain half-life of lithium is substantially longer than the plasma half-life, so we don't really need divided dosing. The brain half-life of lithium is 24 to 48 hours. So, you know, you don't have to take lithium several times a day in order to maintain your brain level of lithium.
Heath:
Makes more sense for the loading. Right? The loading doses, that if has longer in the brain….
Cummings:
Yes. Basically, lithium, even at non-toxic concentrations, slows down the sodium-potassium pump at the cell membrane, and that's the principal way it exits cells. Well, slowing that pump down means that it tends to get retained in cells longer than it will get retained, say, in plasma.
Heath:
One of the articles you sent me was about stratifying the risk of lithium, and it talked about even the 0.5 [mEq/L], like it was actually, I know we aim usually for 0.6-0.8 [mEq/L] with say, bipolar, but that was kind of interesting. It was talking about the risks at like 1.2 [mEq/L] versus, you know, 1.0 [mEq/L]. Like those long-term risks with the levels.
Cummings:
Yes. The higher the level, the bigger the risk; and the more often it's dosed, the bigger the risk. The recommendation, it used to be that it was recommended that lithium could be pushed up to as much as 1.4 millimoles per liter [Some laboratories report the same value as millimoles per liter (mmol/L) because lithium is a monovalent ion, so 1 mEq/L = 1 mmol/L for lithium]. The max recommendation now is 1.2 [mEq/L]; and it's really encouraged that it really shouldn't be allowed to go over 1.0, or at least not stay there very long, under two weeks. Because having lithium levels greater than 1.0, or giving lithium in divided dosing essentially worsens the prognosis for renal function in the long term.
Heath:
So the doses, the long-term exposure, some of it can be mediated. Another long-term risk is tardive dyskinesia (TD). Do you have any thoughts about the long-term use of antipsychotics and tardive dyskinesia?
Cummings:
Yes. I think people had hoped that when we moved from first generation to second generation antipsychotics that the risk would be greatly reduced. Well, it was reduced. It went from about an incident rate of 5.5% per year of exposure to 3.8%. Not as much of a decline as people had hoped for. I think the key is we need to monitor carefully for emergence of involuntary movements. We do need to, while we need to use antipsychotics at effective doses, we need to be careful not to push the dose beyond the point of futility for the drug, because oversaturating the receptors is one of the setups for causing upregulation of the receptor numbers, which seems to be a real risk factor for developing tardive syndromes like TD and tardive akathisia. You know, all the things you really don't want to have.
Heath:
Thank you, those are really helpful. And moving on to the Type D, delayed time release, appearing after long exposure. The teratogenic effects. I know we've kind of touched on this before, but there are the ones of course, in pregnancy, but also for men. Like you've been talking about the registries and sperm, even for a few months, so that long-term exposure can be kind of buried and what that looks like. Do you want to expand on that?
Cummings:
Yes, well, again we're back to valproic acid. People have been aware for quite a while that valproic acid does very bad things during pregnancy in terms of risk of neural tube defects. Also, risk of loss of intelligence in the offspring. Turns out, men don't escape unscathed. Valproic acid alters the methylation of DNA in sperm so that if a man participates in fertilization within three months after the last dose of valproic acid, there is an increased risk of neurobehavioral disorder in the offspring. It's not a huge increase, but it's there (Karakis, 2025; Sakai et al., 2023).
Heath:
Yes. Good thing to educate about. You know, because we moved more towards, I know we're picking on Valproate, today. I was going to say, moving on to, tremor is a really big one I know I hear a lot about. Dr. Puder, I feel like I hear that from a lot of different antidepressants. Depakote, lithium, the dystonic tremor. You mentioned antipsychotics. Dr. Puder, if you want to jump in, but do you want to talk a little bit about, there's a lot of different types of tremor from a lot of different things. Plus, the essential tremor.
Tremor Management: Dystonic, Parkinsonian, Intention & Essential-Like (01:02:01)
Cummings:
Yes. Tremor. Well, there are a number of types of tremor. The two most common types you see are from drugs like lithium or valproic acid, where you're interfering with the firing of neurons (Baek et al., 2014; Canning et al., 2012). And in order to, you know, have your hand at a particular point and space, you have to be able to process the sensory input of where your hand is versus where you want it to be. And if the motor and sensory systems are a little bit out of phase with each other, the system overcorrects in both directions. So you get an intention tremor. The tonic tremor is exactly what it sounds like. It's a dystonia that produces a rhythmic contraction of particularly the extensor muscles and produces a tremor in the hands.
Heath:
This is from antipsychotics, the dystonic tremor? Is it from antipsychotics?
Cummings:
Yes. Dystonic tremor comes from a dopamine antagonist. Antipsychotics. Of course, you can also get the Parkinsonian tremor from the same mechanism. Except, in this case, direct effect on the basal ganglia.
Heath:
How do you distinguish? So would a dystonic tremor, from an antipsychotic [present] right away, or would that be, would that become a sooner one or the Parkinsonian?
Cummings:
The dystonic tremor tends to be immediate. The Parkinsonian tremor tends to be after a longer exposure.
Heath:
Okay. Would you treat them with amantadine?
Cummings:
Yes. You could treat both of them with amantadine.
Heath:
I know we've talked about avoiding anticholinergics, especially long term. Would a beta blocker…? What do you think about a beta blocker for essential tremor? I know neurologists…I remember hearing that neurologists didn't like valproic acid. I mean, they may use it for seizures, but the lithium and the tremor from valproic acid.
Cummings:
Yes. Lithium tremor and valproic acid induced tremor resemble essential tremor in the underlying pathophysiology. And they will respond to a beta blocker. Not so much in terms of eliminating the tremor, but in reducing it.
Now, you can also get there by, if the person's illness permits, if you lower the dose of valproic acid or lithium, that also will reduce the tremor.
Heath:
The tremor actually was kind of a tricky one, because it actually, a medication can have different categories. It overlaps, right? Not just the one category. It might be delayed, but then it can be dose related.
Cummings:
Yes. This classification system, you know, is intended to talk about having essentially classes to put side effects in, but different patients can fall into different categories depending on the timing and development of their side effect.
Heath:
It might exaggerate a physiological tremor, potentially. I think that was one of the things that I was reading about.
Cummings:
Yes.
Puder:
So, one thing about the lithium is that usually it happens early on, but what if someone on lithium gets one later? Like what would you think? Like let's say all of a sudden their right hand has been shaking and they can get it to stop by moving, but it starts without them even noticing it.
Cummings:
It certainly, well, I would be suspicious of a couple of things. I would want to know if their lithium level has changed. Because if it's dose related or concentration related if you give any of us enough lithium, we'll have a tremor. In fact, the development of coarse tremor is one of the signs of lithium toxicity. So either the lithium has changed or the person may be developing an underlying tremor that's being aggravated by the lithium, but may be independent of the lithium in terms of its coming from an evolving pathology. So, you know, both of those things would be worth considering.
Heath:
When would it warrant a neurology referral? I mean, because even with antidepressants, I've seen tremor. Would you try to adjust both? Try a beta blocker? When would you warrant a neurology referral for a tremor?
Cummings:
Basically, when you can't figure out what the tremor is coming from. And you know, this is when you ask for help, in particular, you're looking for a neurologist who specializes in diagnosis of movement disorders.
Heath:
Okay.
Cummings:
The reason for that being you can get pure movement disorders, but you can also get some interesting syndromes that are compilations of more than one type of movement disorder. And those can be incredibly complex and difficult to unravel.
Puder:
So let's say you had this patient and you were like, “I increased the lithium from 900 milligrams at night to 1200 milligrams at night. Now they're having this tremor.” You check the blood level. The blood level went from 0.6 [mEq/L] to 0.9 [mEq/L]. And they're on other medications, but those haven't really changed. So what would be your next step?
Cummings:
Next step would probably be to lower the dose and see if it changes the tremor.
Puder:
And then let's say you really need them on the dose though, because like they're currently manic and you don't want them to go back into…
Puder:
You're treating the mania
Cummings:
The tremor is going to respond very quickly. So I would still probably lower the dose for at least one day. They
Puder:
Oh, okay.
Cummings:
And see what the tremor does. If it changes, it tells me, “Yes, it's definitely the lithium.” If I have to keep the lithium, that means I'm going to have to give them a beta blocker to treat the tremor. Unless, you know, depending on what other factors, whether other mood stabilizers that they respond to or not. But if I'm going to keep the lithium then at least I know for sure that the lithium is the source of the tremor.
Heath:
Which beta blocker would you pick?
Cummings:
Propranolol.
Heath:
Propranolol crosses the blood-brain barrier?
Cummings:
Yes. Selective. A peripherally acting beta blocker is not going to address a problem that's coming from the person's basal ganglia.
Heath:
Would they have problems with asthma? Would there be a different [option], like primidone or….
Cummings:
You could use metoprolol. Because it also crosses the blood-brain barrier.
Heath:
Okay.
Cummings:
You just can't use a beta blocker, like atenolol, that doesn't effectively get into the brain because your target, if you will, is the basal ganglia and it's inside the blood-brain barrier.
Heath:
Okay.
Puder:
Let's move on to Type E, types of side effects, which are end of use withdrawal occurring with stopping the medication. Like benzodiazepine withdrawal.
Puder:
Anticholinergics.
Type E: Withdrawal (01:09:22)
Cummings:
Yes. Anything that, basically anything that causes chronic arousal or anything that is a CNS depressant. If you give either category to somebody long term, their brain systems will adapt so that the countervailing systems upregulate. And then if you suddenly take away, you know, either the stimulant or the brain depressant you'll get the opposite response. For example, if it's a depressant, the person will have, you know, benzodiazepine withdrawal, alcohol withdrawal, opioid withdrawal, you know, activation symptoms, basically. On the other hand, if you suddenly take somebody off of a stimulant they are going to be incredibly lethargic until their system comes back to homeostasis, which is why it's not good to, if somebody's on either category, the only way you can really avoid withdrawal is to taper the agent or put them on a cross tolerant agent and then taper that. I mean, you know, for alcohol withdrawal, that's basically what we do when we put the person on a benzodiazepine and then taper it. You could stair step them down using alcohol itself, but alcohol, it would be very hard to work with unless you had an IV alcohol drip.
Cummings:
Now, you know, the major mistake I've seen made is with people who were serious drinkers, and were going to develop delirium tremens, as people overestimate how effective the benzodiazepines are going to be. I've seen orders like, well, way back in the day, 25 milligrams of Librium Q 12 hours, PRN for alcohol withdrawal. Okay, you know, 25 milligrams of Librium or a milligram of Lorazepam is equivalent to about one shot of whiskey. Well, when I worked at the VA, I had patients who would tell me, “Oh, I drink two-fifths of whiskey a day, every day. So, you know, giving this person a tiny dose of benzo and expecting it to treat their alcohol withdrawal was nonsense.
Heath:
One of the really good things you were talking about recently was the anticholinergic medication and someone on Cogentin 4 milligrams. Or you talked about changing the clozapine too, maybe tapering the Cogentin slowly.
Cummings:
The major thing where the acetylcholine system's been involved is if somebody has, for example, severe neutropenia with clozapine and their clozapine is then necessarily abruptly stopped. They're at real risk of cholinergic rebound. You know, 50 milligrams of clozapine is roughly equivalent to a milligram of Benztropine. So if you've got somebody on 500 milligrams of clozapine, they're taking the equivalent of 10 milligrams of Benztropine. If you suddenly stop that, you can put the person into cholinergic rebound, which can include delirium, nausea, vomiting, diarrhea, cramping, sweating. So if you have to stop the clozapine like that, or any other really anticholinergic drug, you need to put them on an anticholinergic, like benztropine, and then taper it (Desmarais et al., 2014).
Heath:
You said a half milligram. Every….
Cummings:
Typically a half milligram per week.
Heath:
We've talked about the benzodiazepines. I know we worry about any dose for the withdrawal seizures, but I think you've kind mentioned for lower doses, if they're only on maybe one milligram of clonazepam, or maybe if they're only on the Benztropine one milligram or something, or two milligrams would be be less concerned about those.
Cummings:
The lower the dose and the less time the person's been on the medication, the less likely withdrawal symptoms are. You know, the higher the dose and the longer they've been taking it, the more likely it is that they will have significant withdrawal. Because, again, the underlying mechanism is the countervailing systems have upregulated to compensate for the presence of the drug, and then suddenly the drug is gone and there's nothing opposing those systems.
Heath:
How slowly would you taper off a benzodiazepine? Like do you have a certain…?
Cummings:
Depends on how long the person's been on it and how high the dose. I think the longest it's taken me to get a person off of benzodiazepine was 18 months.
Heath:
Which, and I think you recommend going with the longer acting.
Cummings:
Yes. It's easier to taper a longer acting drug like clonazepam or previously, diazepam. You know, we went through a period, fortunately it's not as prevalent anymore, where people would come in, and I've had patients who came in taking 40 milligrams of diazepam/Valium a day. And in their history, as well, “How long have you been taking it at this dose?” “Oh, three decades.”
Heath:
Yes. Oh my gosh.
Cummings:
That person is going to take a, it's gonna take them a while to get off of the benzodiazepine. You know, that was, that was the case after that. Took me a year and a half. Because the person couldn't tolerate going down by more than about two and a half milligrams every couple of months. They eventually got off though and did much better in terms of cognitive and memory performance.
Puder:
What about serotonin? Serotonin reuptake inhibitors? I know there's been a lot of talk in the, some new articles on the withdrawal of that. What's your thoughts on…?
Cummings:
Yes, the antidepressants, including the SSRIs, do have a withdrawal syndrome. I think until recently, psychiatry largely ignored it because the vast majority of such cases are relatively mild. And the person, in the sense that the person doesn't feel well, kind of flu-like, malaise symptoms for a few days to two or three weeks, and then it gradually gets better and goes away. You know, so it's not quite as dramatic as some of, certainly not as dramatic as say benzodiazepine withdrawal or opioid withdrawal. But the antidepressants do come with withdrawal syndrome and the truth is we should, if we're going to stop an antidepressant, we should be tapering it. There is also data suggesting that abruptly stopping antidepressants may encourage a relapse of the depression. So we should be tapering people off of them, unless there's some urgent medical need to take them off of the antidepressant.
Heath:
And dopamine antagonists? The hyperkinetic and the slow taper, just seeing how they tolerate it, perhaps?
Cummings:
Yes. It's a bit like the titration and taper. I think historically we've tended to go a little too fast in both directions. You know, in fact, we've been talking quite a bit about clozapine. Jose De Leon (de Leon et al., 2022a; de Leon et al., 2022b; de Leon, 2023) just published a number, a series of papers actually pointing out that we've probably been titrating clozapine too rapidly until very recently.
Heath:
Yeah.
Cummings:
Particularly, in some populations that don't metabolize the drug very well.
Heath:
Which I also think that kind of leads us into the Type F: failure and unexpected. I know we're closing, losing time really fast. Only a few more minutes, but I wanted to, if you could briefly, at least go over pharmacodynamic versus pharmacokinetic failures. Because I know we talked a little bit earlier about metabolizers.
Type F: Unexpected Failure (01:17:32)
Cummings:
Pharmacodynamic failure means that the activity of the drug, whether you're talking about inhibition of serotonin reuptake or antagonizing dopamine receptors, a pharmacodynamic failure means that the person's illness does not respond to what that drug does as its intended mechanism of action. A good example of that is, there've been some recent magnetic resonance spectroscopy studies where they've found that in treatment-resistant schizophrenia, those people actually have normal or decreased dopamine turnover in the ventral mentum. So, it's not surprising that if you give them a dopamine antagonist, it doesn't treat their psychosis.
Cummings:
You're basically giving them something that they don't need. You're off target in terms of your treatment. So that's a good example of pharmacodynamic failure. Your drug's mechanism of action isn't what the person's illness needs. Pharmacokinetic failure means basically that the drug is not present in a sufficient amount to get above the minimum response threshold and to achieve the threshold that this person needs. Now, that can be for a number of reasons. It can be because they don't take the drug. They're non-adherent. It can be because they don't absorb the drug. Some people, some drugs require active transport in the GI tract and they don't get absorbed very well. And there are also ultra-rapid metabolizers for drugs. And if the person, basically their liver chews up the drug and eliminates it before it can reach its target organ. Those are all examples of pharmacokinetic failure.
Heath:
Maybe unusual, but you mentioned a patient on a dose of Haldol [haloperidol] 80mg and their level is only 2 [ng/mL] .
Cummings:
Yes. We had that patient. A younger, African American male, and people were very puzzled by him because he was titrated up to eight zero (80) milligrams of haloperidol in the mornings, and he took it. But his plasma level got to a grand total of two nanograms per milliliter [2.0 ng/mL is the minimum response threshold, 18ng/mL is the point of futility [see episode 127]. The trick in his case was to switch him to an antipsychotic that didn't principally go through CY450 2D6 [(CYP2D6) pathway].
Heath:
Thank you.
Puder:
Great. Well, this is an important discussion. Maybe not the most entertaining of late. Delusions is, I don't know, much more fascinating to me. But I think this is the practical stuff of sometimes a lot of what we do in outpatient psychiatry is manage side effects of medication patients get better on a certain medication and then have some side effect. And it's like, “What do we do next?” Right? And it's like a lot of what we do is think about how we minimize side effects. And a lot of our decision making is around do we decrease the med? Do we try to treat the side effect another way? And I think it's important as providers to think that our medications can cause side effects, and getting them off of medications can cause other side effects. So this is a great discussion. Thank you so much guys. Thank you.
Cummings:
Okay.
References
Alhanshali, L., Buontempo, M., Shapiro, J., & Lo Sicco, K. (2023). Medication-induced hair loss: An update. Journal of the American Academy of Dermatology, 89(2 Suppl.), S20–S28. https://doi.org/10.1016/j.jaad.2023.04.022
Aronson, J. K., & Ferner, R. E. (2003). Joining the DoTS: new approach to classifying adverse drug reactions. BMJ (Clinical research ed.), 327(7425), 1222–1225. https://doi.org/10.1136/bmj.327.7425.1222
Baek, J. H., Kinrys, G., & Nierenberg, A. A. (2014). Lithium tremor revisited: Pathophysiology and treatment. Acta Psychiatrica Scandinavica, 129(1), 17–23. https://doi.org/10.1111/acps.12171
Berlant, J. L. (2004). Prospective open-label study of add-on and monotherapy topiramate in civilians with chronic nonhallucinatory posttraumatic stress disorder. BMC Psychiatry, 4, Article 24. DOI: https://doi.org/10.1186/1471-244X-4-24
Bhanu, C., Nimmons, D., Petersen, I., Orlu, M., Davis, D., Hussain, H., Magammanage, S., & Walters, K. (2021). Drug-induced orthostatic hypotension: A systematic review and meta-analysis of randomised controlled trials. PLOS Medicine, 18(11), e1003821.https://doi.org/10.1371/journal.pmed.1003821
Canning, J. E., Burton, S., & Hall, B. (2012). Lithium and valproate-induced tremors. Mental Health Clinician, 1(7), 174–176. https://doi.org/10.9740/mhc.n92093
Castro, V. M., Roberson, A. M., McCoy, T. H., Wiste, A., Cagan, A., Smoller, J. W., Rosenbaum, J. F., Ostacher, M., & Perlis, R. H. (2016). Stratifying risk for renal insufficiency among lithium-treated patients: An electronic health record study. Neuropsychopharmacology, 41(4), 1138–1143. https://doi.org/10.1038/npp.2015.254
Cheshire, W. P., & Fealey, R. D. (2008). Drug-induced hyperhidrosis and hypohidrosis: incidence, prevalence, and management. Drug Safety, 31(2), 109–126. DOI: https://doi.org/10.2165/00002018-200831020-00002
Dalmau, J., Armangué, T., Planagumà, J., Radosevic, M., Mannara, F., Leypoldt, F., Geis, C., Lancaster, E., Titulaer, M. J., Rosenfeld, M. R., & Graus, F. (2019). An update on anti-NMDA receptor encephalitis for neurologists and psychiatrists: mechanisms and models. The Lancet. Neurology, 18(11), 1045–1057. https://doi.org/10.1016/S1474-4422(19)30244-3
de Leon, J., Schoretsanitis, G., Ruan, C. J., De Las Cuevas, C., & Kane, J. M. (2022a). An international clozapine titration guideline to increase its safety and move forward on the route started by German-speaking psychiatrists in the 1960s. European Archives of Psychiatry and Clinical Neuroscience, 272(3), 407–416. https://doi.org/10.1007/s00406-022-01407-7
de Leon, J., Schoretsanitis, G., Smith, R. L., Molden, E., Solismaa, A., Seppälä, N., Kopeček, M., Švancer, P., Olmos, I., Ricciardi, C., Iglesias-Garcia, C., Iglesias-Alonso, A., Spina, E., Ruan, C., Wang, C., Wang, G., Tang, Y., Lin, S., Lane, H., Kim, Y. S., Kim, S. H., Rajkumar, A. P., González-Esquivel, D. F., Jung-Cook, H., Baptista, T., Rohde, C., Nielsen, J., Verdoux, H., Quiles, C., Sanz, E. J., De Las Cuevas, C., & Cohen, D. (2022b). An international adult guideline for making clozapine titration safer by using six ancestry-based personalized dosing titrations, CRP, and clozapine levels. Pharmacopsychiatry, 55(2), 73–86. https://doi.org/10.1055/a-1625-6388
de Leon, J. (2023). Titrating clozapine amidst recommendations proposing high myocarditis risk and rapid titrations.
Psychiatry Investigation, 20(3), 183–195. https://doi.org/10.30773/pi.2022.0293
Desmarais, J. E., Beauclair, L., Annable, L., Bélanger, M.-C., Kolivakis, T. T., & Margolese, H. C. (2014). Effects of discontinuing anticholinergic treatment on movement disorders, cognition and psychopathology in patients with schizophrenia. Therapeutic Advances in Psychopharmacology, 4(6), 257–267. https://doi.org/10.1177/2045125314553611
Karakis, I. (2025). In the name of the father: Risk to the offspring from paternal exposure to valproate during conception. Epilepsy Currents, 25(2), 116–118. https://doi.org/10.1177/15357597251317286
Kayser, M. S., & Dalmau, J. (2011). Anti-NMDA Receptor Encephalitis in Psychiatry. Current psychiatry reviews, 7(3), 189–193. https://doi.org/10.2174/157340011797183184
Lynch, S. S. (2025, April). Adverse drug reactions. In Merck Manual Professional Edition. Merck & Co., Inc.https://www.merckmanuals.com/professional/clinical-pharmacology/concepts-in-pharmacotherapy/adverse-drug-reactions
Rawlins, M. D., & Thompson, J. W. (1977). Pathogenesis of adverse drug reactions. In D. M. Davies (Ed.), Textbook of adverse drug reactions (p. 10). Oxford University Press. https://openlibrary.org/books/OL4610966M/Textbook_of_adverse_drug_reactions?
Sakai, K., Hara, K., & Tanemura, K. (2023). Testicular histone hyperacetylation in mice by valproic acid administration affects the next generation by changes in sperm DNA methylation. PLoS ONE, 18(3), e0282898. https://doi.org/10.1371/journal.pone.0282898
Schrader, C., & Ross, A. (2021). A Review of PTSD and Current Treatment Strategies. Missouri medicine, 118(6), 546–551. https://pubmed.ncbi.nlm.nih.gov/34924624/
Weatherly, S., Garazi, C.D., Woldenberg, E., Ravigopal, T., & Costin, J. (2025). Unraveling the Cycle: A Scoping Review Exploring the Impact of Antidepressants on the Female Reproductive Cycle. Cureus, 17(11), e97360. DOI: https://doi.org/10.7759/cureus.97360