Medical Lectures

The Right Antidepressant for the Right Person

Prescribing for depression can be a minefield. Learn how to select antidepressants for new patients or patients with complex medical comorbidities.
September 7, 2021
Robert P. Bright, MD


John is a 32-year-old man who came in with the first episode of diagnostic four criteria, major depression. He has no major medical comorbidities and no personal or family history of manic or hypomanic episodes. 

I never see anybody this straightforward, but if you did, which antidepressant would you choose for him? Prozac, Valiasr Donor hybrid. That's Venlafaxine Presti Dellucci 18 Cymbalta. It doesn't matter. I'm too burned out to care. 

Studies show that for a population, every single antidepressant works equally well to every single other antidepressant out there. There is no clear frontrunner. So if you have somebody like John walking in, and you're right out of the bat, it doesn't matter which one you pull out of your pocket. 

I think Fluoxetine probably would have been my choice for John. You're probably thinking what I'm thinking about that, and I'll come back as to why it would have been. But it actually doesn't matter if you're just looking at potential efficacy for your choice. 


Natasha's 39 years old. She's got depression, she's morbidly obese, and she has prediabetes. She's resistant to taking antidepressants because she doesn't want to gain any more weight. If you have a patient who is worried about taking the antidepressant for fear of weight gain, this comes up commonly. And it’s more than just Wellbutrin, which is the one we tend to think of. 


This is Jackie. She's 42 years old. She has a Cuneen score of 24 consistent with severe clinical depression. She has a history of bulimia, and her QT is 525 milliseconds. So bulimic, very depressed, and prolonged QT. 

Considering her QT, which of these is the safest antidepressant choice for her? Zoloft is out in front. Bupropion for Stic is actually your right answer here. Why does it matter? It seems like everybody we see in the hospital now has a prolonged QT interval, and it's just the bane of my day. 

But does it really matter? The reality is that all the antidepressants for most of them, and I'm going to show you the exceptions, do cause QT prolongation, but six milliseconds is about the amount that they prolong it. So it's not a lot compared to placebo, but it is a statistically significant difference. 

And that was a comprehensive review that showed that, by and large, there's little evidence that psychotropic drug use by itself is sufficient to predict that Assad in more than 90 percent of the cases is at least one additional risk factor. And in the cases that did have to decide, 20 percent of them actually had a QT less than 500 going into the whole thing. And 75 percent of them, the drugs were at a therapeutic dose. So a high dose didn't even predict the torture. 


These were the other risk factors. So if there's a congenital, prolonged QT, then you guys are used to looking at these things as primary care providers: female hypoglycemia, hyperkalemia, bradycardia. I look at it and take care of and address and ask about suggested things all day long, especially that bottom line, because we see a lot of transplant patients, and they're on five or seven different medications that can prolong the QT. 

Look at those and try to figure out which ones are essential. Are there alternatives for some of those, like the antidepressants, that may not cause acute prolongation that we could use instead? 

Citalopram, you are probably familiar with by now, has a black box warning for QT prolongation, especially at higher doses. All of these SSRI also cause QT prolongation, as do mirtazapine and venlafaxine. So most of the things you would reach for right off the bat are listed here. Cymbalta or plastic for Stik was the option I think I gave you back there, both of those or fat similar do not cause QT prolongation. All these others do, but are not clinically significant at therapeutic doses, such as DRL Wellbutrin or Treinta Alex. 


If we go back to Jacqui with bulimia and prolonged QT, which one of the antidepressants is contraindicated because of her eating disorder? Actually, Wellbutrin is contraindicated for patients with eating disorders because it lowers the seizure threshold and increases the risk of seizures in that population, specifically Mirtazapine or Imuran is the drug of choice many people treating eating disorders may go to to help with appetite, stimulation, and weight gain. But it does not have an FDA indication that sometimes people will choose it for that purpose. 

Fluoxetine has demonstrated efficacy and FDA approval for the treatment of bulimia. There are no antidepressants that have an FDA indication for the treatment of anorexia. 


LeBron has been diagnosed with major depression, he's read that genetic testing can help you pick the medication. That's right for him. 

Pharmacogenetic testing will not predict which medicine is going to most likely benefit the patient if you take a person who has normal pharmacogenetics. They're going to respond to whatever antidepressants that you give them based on just their clinical response. It will tell you what they're metabolizing and how they're metabolizing it. 

It can help predict medication tolerance. So if patients have poor metabolism of 2,006, for example, their tricyclic antidepressants are going to start backing up, and their tricyclic level is going to go up. Their clinical response will predict their tolerance level. It may predict which ones are not going to be helpful because they're metabolizing them too quickly. 

So they're never going to get to a therapeutic drug or level, but it's not actually going to tell you. It's going to say which ones are metabolizing, but it's not. We'll tell you whether that was going to work for them. It's just going to tell you if they've got Skari metabolism or that medication. I phrased it can help predict versus it will predict. 

It's very subtle, but it will not actually predict which medication will help any more than I could have predicted. Paxil is going to work for you, and Prozac is going to work for you, and Vibert is going to work for you. There's no test that is going to tell me which one is going to work.  


Gerard has end-stage liver disease and is listed for transplant, which of the following is contraindicated for Gerard? Wellbutrin, Cymbalta, Zoloft, Effexor, Pristiq. So actually, the one that is contraindicated for Gerard with advanced liver disease is Dulac Sateen. It's contraindicated in patients with advanced liver disease. 

We need a medicine that has minimal hepatic metabolism, so Cymbalta might be a really good choice for him. Which of these has the least hepatic metabolism that is going to be a really good choice for a patient with advanced-stage liver disease? The correct answer is DesVenlafaxine, which has very minimal hepatic metabolism in patients with very advanced liver disease. 


We've taken care of those folks, maybe I should try one of the new antidepressant medications, what's one? Should I pull out one of those? These are the newest ones. Until just recently, Vibert came out in 2011, 2013, and Fitzsimon 2013. 

So Vibert is an SSRI, but also a partial serotonin agonist. It has a high rate of gastrointestinal side effects, but its niche is in patients with major depression and comorbid generalized anxiety disorder. So consider that for anxious, depressed patients. But keep in mind that none of these work any better than the others SSRIs that also have FDA indications for the treatment of generalized anxiety disorder. And I know of no particular study that's showing that this is better. 

But if you're looking for when I would use this, I would think about this situation. 26 or 40 Occitan, the mechanisms not really well understood, is sort of playing poker at the serotonin receptor and lauded several different ways. 

It's possible that there were some findings that it may improve cognitive functioning in patients with major depression that was independent of his antidepressant effect. I'm truly kind of trying to find you a niche for these medications – when would you go for this one rather than another one? And this was a finding that came from this study. 

And Leavel Marzipan, which is Fitzsimon, is an enantiomer, M.A. Sopranos, Savelli, which is approved, as you know, for fibromyalgia, SAVOLA did not have any does not have FDA indication or proven efficacy for the treatment of major depression. 

But this enantiomer does. It's a potent SSRI and because it's an NRI, the noradrenergic reuptake inhibition increases motivation and energy. So that's going to be true of any norepinephrine medication. So I can't say that it's unique to this one, but that may be a niche if you're looking for a reason to use it. High incidence of nausea as well with this one. 


James is a 23-year-old. He's coming in with sexual dysfunction from the SSRI you prescribed for him. What? These are the antidepressants associated with minimal sexual dysfunction. Wellbutrin is probably the one that's going to come to your mind first and foremost, but Vibert and Intellichoice also have no sexual dysfunction or none were found in the studies. 

Mirtazapine is one that we will sometimes try and patients who are having sexual dysfunction associated with their antidepressants. It does seem better than the others, but there is a risk of it as well. But keep in mind again that this is generic, and these may be very expensive for your patient. 

Levomilnacipran was just approved three days ago. For postpartum depression, it will be released sometime in May. It looks like it is a synthetic allow pregnant alone, which is the metabolite of progesterone. It's Agaba, a receptor Alistaire modulator. 

It works within 48 hours of infusion. It requires 60 hours of continuous infusion in the medical setting. Its main side effects are dizziness and syncope. It only requires one treatment of 60 hours in the medical setting, but that one treatment costs $34,000. That does not include the cost of being in the hospital or the doctors or any of that stuff. And does anybody have any idea what it costs to be in the hospital for a day? Any ideas how many? $33,000 per day. That adds up quickly, but it is FDA-approved. 


Women did find that suicidal thoughts started clearing as early as 24 to 36 hours after starting the infusion. Their progesterone rises during pregnancy and plummets at the time of delivery. And that's one of the thoughts as to what's triggering postpartum depression, which works. So it's an FDA indication, but it's $34,000. There is a formulation similar to this compound that's being developed for oral administration that is not yet available, but it's in the pipeline. So that may be coming out. But this is just very much off the presses, as it is available.

Then there's S.S ketamine's bravado, and it's the first and this was just approved two weeks ago, it's the first and only NMDA receptor antagonist and it's approved as an adjunct to other antidepressant medications for treatment-resistant depression. 


Ketamine was a horse tranquilizer. It is a party drug nicknamed a Special K that gives people dissociative, floating, kind of out-of-it kind of experience, can have psychotic side effects with it. 

Ketamine was first studied in 1964 on prisoners in this compound. It became approved by the FDA in 1975 as an anesthetic. And in 2000, studies started to be done around its use as an antidepressant. It has a rapid onset antidepressant effect within hours, sometimes within 30 minutes. People's depression will clear, it's really amazing. 

There's something here working on the glutamate receptor, something novel and different about the way this is working and very promising about the future in treating depression. And it works even for treatment-resistant depression. So there were studies that have also been shown that it works very quickly and very effectively for suicidality patients coming into a psychiatric emergency room, given a single infusion of point, five milligrams per kilogram, intravenous ketamine had resolution of their suicidality within 40 minutes. 

It was approved on March the 5th of this year. It can now be prescribed. 


The dosing is 56 to 84 milligrams. You start with 56, and if it's not responsive, the next time you do 80. It’s approved for treatment-resistant depression. The administration of it is twice a week for four weeks, then you go weekly for four weeks, and then you can go every one or two weeks thereafter, usually going every two weeks thereafter. But some patients need it more. 

But then there's the punch line: $600 and $900 per session. And this is for the medicine, not including the cost of the doctor, the visit, and all of that stuff. Its side effects are dissociation, dizziness, and nausea.

And another catch: you can't drive for 24 hours after you've taken the medication, so if you're getting it twice a week, that's four days a week that you can't drive. 

There's a restricted distribution system, so only certain pharmacies will be able to get it. They have to sign up and register to be part of it. There is a special risk evaluation and mitigation strategy program, just like there is for Clozaril and some other medications that are high risk. And that's because this is a drug of abuse.

People get high off of ketamine and use high doses, even though this is just a little spray. You can only get so much out of it. You spray once and then, five minutes later, you spray again. And that's all there is to the whole kit. So you're not going to get very far with that little kid if you were to steal it or get it. 

But if you had a lot of it somewhere, you'd have a lot of street value and people would really want it. It was very highly controlled. It does have to be administered in a physician's office or hospital, and they have to have at least two hours of monitoring by a healthcare professional before they can go home because of the risk of dissociation, the risk of treatment, emergent psychosis, hypertension, and the other side effects I showed you. So it's not without risk. 

It has amazing potential, very powerful benefits, but it has its issues. 


We're going back to that first guy whose name I forget (John), where he was at. 

No medical history, no bipolar history, any of that stuff. 

How do you choose antidepressants for people in general? So which side effects do you want or not don't want? Do you want this patient to gain weight? Do you when you might use mirtazapine, do you want them to not gain weight? So you might use Wellbutrin or Villazon? Do you want to sedate them (so you may use mirtazapine), or do you want to activate them and give them some energy? 

One of the other considerations that you're looking at for drugs that don't cause sexual dysfunction is their liver disease, comorbid liver disease, liver-kidney problems, metabolism, drug interactions, all of that stuff. 

Does that patient have a personal history of taking something and responding to it? That's your choice. If they do, I took Prozac before, and it did great, and I stopped it. Why do you stop it? Because I was better. Well, let's go back to that. 

Is there a family history of response? This guy's coming in for the first time, he's never seen an antidepressant, but everybody in his family's had major depression, and they've all responded to whatever it was. So is it generically available? Prozac is about three dollars a month. Fluoxetine was your first choice there: it’s generically available, and it’s one of the best-tolerated antidepressants. 

Fluoxetine doesn't tend to be sedating. It's one of two or three that are approved by the FAA for pilots to take. I think it was a great choice you made right off the bat. There are a lot of things going forward that made it a reasonable first choice. I'm not plugging that medication. They all work equally well with each other but just looking for factors. 

And when would you choose one if the patient does have a known pharmacogenetic profile that's going to guide you. So if now he's walking in, and for whatever reason, he paid 23 and me or somebody, or he has his pharmacogenetic profile in front of you, and it shows he’s never been on an antidepressant. 

He has 2B6 poor metabolism, which is how Prozac is metabolized. I'm not going to go with that medication because I already know he's not going to metabolize it well. If he has an ultra-rapid metabolism that I know, I could go with that and really crank up the dose, or I could just use something from a different pathway that's not going to get in my way. And does the patient have a preference? 


I was training when Prozac came out, like the late eighties. And there was this whole Prozac hysteria at that time that Prozac was making people go crazy, and they were killing people and killing themselves. That was never true and it was proven to not be true. 

But to this day, people who were not even born at that time look at me and tell me, I don't want to take Prozac. I've heard about it, and it's just not true. But the patient may have a preference in a way that may or may not have any validity. Go with that. Listen to what they want.

To learn more, click here.

Watch Video


GIBLIB is the premier streaming service with a vast catalog of unrivaled continuing medical education content covering surgical videos, medical lectures, courses, and transcripts. Learn from the world’s best doctors, from the top leading institutions, including Mayo Clinic, UCSF, Cedars-Sinai, Keck Medicine of USC, and many more.