Why would you choose Zoloft over Lexapro as a first-line SSRI for MDD?

[futureapppsy2]

I thought Lexapro generally had a better side effect profile and less of a taper issue--not so?

---

Members don't see this ad.

 

[CalmAndCollected]

Zoloft is generally thought as the ideal SSRI for PTSD, so that is one reason why it’s picked.

Some physicians might be more familiar with it, or the patient may have a history of responding well to it.

 

[tr]

More calming, often helps sleep if taken at night, great for anxiety, marginally less placental passage in pregnancy. Very good for anxious depression with insomnia.

I usually select antidepressant based on the individual's level of arousal (vegetative hypersomniac to anxious insomniac) and, if any, history of activation vs sedation with past trials. Rough order of activation is mirtazapine < paroxetine < sertraline < escitalopram < citalopram < fluoxetine < duloxetine < venlafaxine < bupropion. If a prior drug was too activating, step down. If a prior drug was too sedating, step up. I find on average Lexapro is the neutral midpoint, with most people reporting neither activation not sedation, although there are always a few who report one or the other.

I don't usually see taper problems with any of the SSRIs except Paxil. SNRIs are the big offenders.

 

[clausewitz2]

More calming, often helps sleep if taken at night, great for anxiety, marginally less placental passage in pregnancy. Very good for anxious depression with insomnia.

I usually select antidepressant based on the individual's level of arousal (vegetative hypersomniac to anxious insomniac) and, if any, history of activation vs sedation with past trials. Rough order of activation is mirtazapine < paroxetine < sertraline < escitalopram < citalopram < fluoxetine < duloxetine < venlafaxine < bupropion. If a prior drug was too activating, step down. If a prior drug was too sedating, step up. I find on average Lexapro is the neutral midpoint, with most people reporting neither activation not sedation, although there are always a few who report one or the other.

I don't usually see taper problems with any of the SSRIs except Paxil. SNRIs are the big offenders.

This is basically my approach as well, with the exception of people who want to be taking something but really struggle with adherence, in which case fluoxetine's half-life means they can miss two or three days at a time once at steady state with minimal effects on blood levels. If I'm going to do an SNRI it's going to be in a situation where the patient is okay with pushing it to the higher dose range fairly directly because at lower doses they are pharmacologically SSRIs with more withdrawal problems. I'm relatively quick to bring up the idea of TCAs after a couple of failures, especially if there are a lot of classical melancholic features, but then I so rarely see anyone in my current settings who hasn't been on 2-3 SSRIs already. Mirtazapine is a first choice for people with CVS and functional dyspepsia in my mind (this comes up way more than I would have anticipated).

 

[Stagg737]

I haven't found a significant difference in side effect profile between the two overall and I've found the claims that Lexapro is "cleaner" to be a load of bunk. I mostly look at it as nearly the same as Celexa and generally avoid both in any patient with significant cardiac issues. Zoloft's big issue is the GI symptoms (which I do see somewhat frequently), but these are mostly transient and my patients that can tolerate the first week usually do well after that.

 

[clausewitz2]

I haven't found a significant difference in side effect profile between the two overall and I've found the claims that Lexapro is "cleaner" to be a load of bunk. I mostly look at it as nearly the same as Celexa and generally avoid both in any patient with significant cardiac issues. Zoloft's big issue is the GI symptoms (which I do see somewhat frequently), but these are mostly transient and my patients that can tolerate the first week usually do well after that.

Sertraline has a significantly wider range of drug interactions due to broader CYP inhibition that escitalopram, I always assumed that is what people meant by "cleaner." Sertraline does have the advantage of having as more of an affinity for DAT than venlafaxine has for NET, so it would not be insane to talk about it as an SDRI. If diarrhea is an issue sertraline is probably doing this partially via sigma-1 antagonism, whereas citalopram/escitalopram is a sigma-1 agonist so is definitely not contributing to excess gut motility that way.

How much this sort of theory impacts anything anyone treating patients really cares about is a different issue.

 

[calvnandhobbs68]

More calming, often helps sleep if taken at night, great for anxiety, marginally less placental passage in pregnancy. Very good for anxious depression with insomnia.

I usually select antidepressant based on the individual's level of arousal (vegetative hypersomniac to anxious insomniac) and, if any, history of activation vs sedation with past trials. Rough order of activation is mirtazapine < paroxetine < sertraline < escitalopram < citalopram < fluoxetine < duloxetine < venlafaxine < bupropion. If a prior drug was too activating, step down. If a prior drug was too sedating, step up. I find on average Lexapro is the neutral midpoint, with most people reporting neither activation not sedation, although there are always a few who report one or the other.

I don't usually see taper problems with any of the SSRIs except Paxil. SNRIs are the big offenders.

The Duloxetine vs Venlafaxine thing is weird to me because I feel the same way (venlafaxine more activating than duloxetine) but idk if it's just bias...since theoretically duloxetine inhibits both 5HT and NE reuptake pretty quickly at normal doses, while venlafaxine basically acts as an SSRI until you start getting to higher doses. If we assume the activation is more from the NE then it doesn't really theoretically make sense...

 

[Psychresy]

More calming, often helps sleep if taken at night, great for anxiety, marginally less placental passage in pregnancy. Very good for anxious depression with insomnia.

I usually select antidepressant based on the individual's level of arousal (vegetative hypersomniac to anxious insomniac) and, if any, history of activation vs sedation with past trials. Rough order of activation is mirtazapine < paroxetine < sertraline < escitalopram < citalopram < fluoxetine < duloxetine < venlafaxine < bupropion. If a prior drug was too activating, step down. If a prior drug was too sedating, step up. I find on average Lexapro is the neutral midpoint, with most people reporting neither activation not sedation, although there are always a few who report one or the other.

I don't usually see taper problems with any of the SSRIs except Paxil. SNRIs are the big offenders.

This is really helpful, thanks

 

[Stagg737]

Sertraline has a significantly wider range of drug interactions due to broader CYP inhibition that escitalopram, I always assumed that is what people meant by "cleaner." Sertraline does have the advantage of having as more of an affinity for DAT than venlafaxine has for NET, so it would not be insane to talk about it as an SDRI. If diarrhea is an issue sertraline is probably doing this partially via sigma-1 antagonism, whereas citalopram/escitalopram is a sigma-1 agonist so is definitely not contributing to excess gut motility that way.

How much this sort of theory impacts anything anyone treating patients really cares about is a different issue.

I should have clarified that I meant the idea of Lexapro being "cleaner" hasn't really translated into clinically significant (or even noticeable) differences over Celexa from what I've seen.

 

[Monocles]

Sometimes that hasn't been mentioned... family hx. If I see a kiddo with mdd that has 2 siblings and a parent all with positive responses, it's both an easier sell from an adherence perspective, but also biologically more likely to work.

 

[J ROD]

It's the one I pulled out the hat.......

 

[tr]

The Duloxetine vs Venlafaxine thing is weird to me because I feel the same way (venlafaxine more activating than duloxetine) but idk if it's just bias...since theoretically duloxetine inhibits both 5HT and NE reuptake pretty quickly at normal doses, while venlafaxine basically acts as an SSRI until you start getting to higher doses. If we assume the activation is more from the NE then it doesn't really theoretically make sense...

I dunno but I think that trying to infer clinical level responses from an on-paper receptor agonism/antagonism profile is pretty much just useless speculation. Many of these drugs have partial agonist or partial antagonist responses at multiple receptors, and someone would have to go test each receptor affinity specifically in a laboratory, which I don't think has been done in a systematic way for all the drugs we have. So available sources list some hodgepodge of activities based on whatever it was decided to test, and then on top of that throw in that everybody's individual rates of drug metabolism and receptor affinities are different.

 

[Basis]

I use way more TCAs than SSRIs, but reasons I'd lean towards using Zoloft instead of Lexapro would be: bad PTSD, or poor response or bad reaction to celexa/lexapro in the past. In residency I heard about some researchers doing trials pushing zoloft to 300mg and 400mg. I'd never go over the FDA max in practice but it was interesting to read about. I do find people complain of nausea as a side effect with zoloft more than prozac or lexapro. And as people mentioned above, zoloft has more CYP concerns (vs lexapro).

But I barely ever use SSRIs unless the person is completely drug naive or had a good response to it in the past. The exception would be prozac and OCD, I use that a lot. It seems like half of the OCD patients do well on prozac, and the other half end up needing to be on clomipramine.

 

[clausewitz2]

I use way more TCAs than SSRIs, but reasons I'd lean towards using Zoloft instead of Lexapro would be: bad PTSD, or poor response or bad reaction to celexa/lexapro in the past. In residency I heard about some researchers doing trials pushing zoloft to 300mg and 400mg. I'd never go over the FDA max in practice but it was interesting to read about. I do find people complain of nausea as a side effect with zoloft more than prozac or lexapro. And as people mentioned above, zoloft has more CYP concerns (vs lexapro).

But I barely ever use SSRIs unless the person is completely drug naive or had a good response to it in the past. The exception would be prozac and OCD, I use that a lot. It seems like half of the OCD patients do well on prozac, and the other half end up needing to be on clomipramine.

How are you treating OCD with sertaline if you are not going to 400 mg? That is pretty standard.

 

[CalmAndCollected]

I use way more TCAs than SSRIs, but reasons I'd lean towards using Zoloft instead of Lexapro would be: bad PTSD, or poor response or bad reaction to celexa/lexapro in the past. In residency I heard about some researchers doing trials pushing zoloft to 300mg and 400mg. I'd never go over the FDA max in practice but it was interesting to read about. I do find people complain of nausea as a side effect with zoloft more than prozac or lexapro. And as people mentioned above, zoloft has more CYP concerns (vs lexapro).

But I barely ever use SSRIs unless the person is completely drug naive or had a good response to it in the past. The exception would be prozac and OCD, I use that a lot. It seems like half of the OCD patients do well on prozac, and the other half end up needing to be on clomipramine.

Why not Luvox for OCD?

How many SSRIs do you trial before going to TCAs?

 

[hamstergang]

family hx. If I see a kiddo with mdd that has 2 siblings and a parent all with positive responses ... biologically more likely to work.

Everyone says this and the theory sounds reasonable, but is it actually true? Any studies to support this?

 

[hamstergang]

Why not Luvox for OCD?

Why would you? It's no better than the other SSRIs, yet is needed BID at higher doses. I'm not a fan.

 

[futureapppsy2]

Why would you? It's no better than the other SSRIs, yet is needed BID at higher doses. I'm not a fan.

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

 

[hamstergang]

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

Finding someone who actually does that and takes insurance around me is impossible. So usually no.

 

[tr]

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

I always counsel people that an effective psychotherapeutic intervention is available, and describe it briefly.

There are many reasons why people choose not to pursue ERP.

Some people have already failed it.

Some people need the med anyway for comorbid depression/anxiety, in which case they might as well take an OCD level dose and be done with it.

Some people find the idea of doing ERP aversive and prefer to just take a pill.

I treat a lot of postpartum OCD which tends to resolve spontaneously over time. Postpartum women don't have lots of spare time for labor-intensive therapies and the long-term advantage of ERP is not so relevant since the condition is often self-limited.

As a psychologist you probably don't see this much since people like this will self-select out of seeing you, but I see a *lot* of patients who are like, "I'm too busy for therapy, just give me the pill." The time/labor investment of therapy is a big barrier.

 

[Liquid8]

I always counsel people that an effective psychotherapeutic intervention is available, and describe it briefly.

There are many reasons why people choose not to pursue ERP.

Some people have already failed it.

Some people need the med anyway for comorbid depression/anxiety, in which case they might as well take an OCD level dose and be done with it.

Some people find the idea of doing ERP aversive and prefer to just take a pill.

I treat a lot of postpartum OCD which tends to resolve spontaneously over time. Postpartum women don't have lots of spare time for labor-intensive therapies and the long-term advantage of ERP is not so relevant since the condition is often self-limited.

As a psychologist you probably don't see this much since people like this will self-select out of seeing you, but I see a *lot* of patients who are like, "I'm too busy for therapy, just give me the pill." The time/labor investment of therapy is a big barrier.

Agree with all of this - the time and cost of ongoing therapy is often a major limiting factor. Have a few patients with pure obsessions and in these cases ERP is also not as effective (as there's no compulsions to prevent).

With SSRIs and OCD, one needs to typically use high doses. I have not found that any one is superior to another, so it usually depends on what the patient can tolerate.

I thought Lexapro generally had a better side effect profile and less of a taper issue--not so?

While I would generally use Lexapro over Zoloft as a first line in MDD, the initial Cipriani papers had both of them in the top 4 for effectiveness and tolerability so I don't see any real problem with starting one over the other.

Usually I would lean toward Zoloft in patients who are pregnant or considering this in the near future, or if there had been a family history of a positive response to Zoloft.

Sertraline has a significantly wider range of drug interactions due to broader CYP inhibition that escitalopram, I always assumed that is what people meant by "cleaner." Sertraline does have the advantage of having as more of an affinity for DAT than venlafaxine has for NET, so it would not be insane to talk about it as an SDRI. If diarrhea is an issue sertraline is probably doing this partially via sigma-1 antagonism, whereas citalopram/escitalopram is a sigma-1 agonist so is definitely not contributing to excess gut motility that way.

How much this sort of theory impacts anything anyone treating patients really cares about is a different issue.

In terms of CYP450 drug interactions, I recall Zoloft had a few more than Lexapro although not to a clinically significant degree in most cases. Of the SSRIs broadly speaking escitalopram, citalopram, sertraline have fewer drug interactions compared to fluoxetine, paroxetine and fluvoxamine which I think would influence how most decide to prescribe.

 

[birchswing]

This is basically my approach as well, with the exception of people who want to be taking something but really struggle with adherence,

Paxil is the best for adherence. It has a built-in feature where it zaps your brain if you forget to take a dose.

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

When you say cure, well . . . I'm wondering if you're using the word cure as I think of it. Do they continue to have these problems lifelong and keep them at bay using the ERP tools? Or it's just not even a thing they have to consider because it's permanently gone?

Does the age of onset of ERP matter? For example, if OCD has an onset at age 5 but isn't treated with ERP until age 40? I wonder because it seems the brain might become more entrenched.

I had always heard that OCD was one of the more neurological and intractable psychiatric disorders, and it often goes with Tourette's which is also (from my knowledge) fairly intractable. So the idea of a cure is interesting.

 

[AcronymAllergy]

Paxil is the best for adherence. It has a built-in feature where it zaps your brain if you forget to take a dose.

When you say cure, well . . . I'm wondering if you're using the word cure as I think of it. Do they continue to have these problems lifelong and keep them at bay using the ERP tools? Or it's just not even a thing they have to consider because it's permanently gone?

Does the age of onset of ERP matter? For example, if OCD has an onset at age 5 but isn't treated with ERP until age 40? I wonder because it seems the brain might become more entrenched.

I had always heard that OCD was one of the more neurological and intractable psychiatric disorders, and it often goes with Tourette's which is also (from my knowledge) fairly intractable. So the idea of a cure is interesting.

There are actually effective behavioral/cognitive-behavioral therapies for tic disorders. As for "cure" in reference to OCD, you could consider it cured in that ERP can significantly reduce symptom frequency and severity to the extent that the person may no longer consider them to be "problems." Are you going to 100% completely erase ever experiencing the thought and/or compulsive urge? I'm no OCD expert, but probably not.

 

[clausewitz2]

There are actually effective behavioral/cognitive-behavioral therapies for tic disorders. As for "cure" in reference to OCD, you could consider it cured in that ERP can significantly reduce symptom frequency and severity to the extent that the person may no longer consider them to be "problems." Are you going to 100% completely erase ever experiencing the thought and/or compulsive urge? I'm no OCD expert, but probably not.

Most OCD treatment is distinctly not trying to eliminate urges or thoughts, although marked reduction in their frequency and intensity is often a happy byproduct. A frequent failure state in OCD is the conviction that it is unacceptable to have certain thoughts so they must be eradicated by any means necessary. Helping people move away from the agenda of thought control to a stance of willingness and acceptance is a critical part of the work, getting folks to the point of "okay, I'm having this thought, it's possible I am a murderer/pedophile/infected, cool story, brain, moving on"

Pragmatically, how would you ever get someone to do E/RP voluntarily without at least part of then being willing to experience their thoughts/urges?

Cure is maybe an overstatement for most people but certainly a majority have fairly dramatic results.

 

[hebel]

I use way more TCAs than SSRIs, but reasons I'd lean towards using Zoloft instead of Lexapro would be: bad PTSD, or poor response or bad reaction to celexa/lexapro in the past. In residency I heard about some researchers doing trials pushing zoloft to 300mg and 400mg. I'd never go over the FDA max in practice but it was interesting to read about. I do find people complain of nausea as a side effect with zoloft more than prozac or lexapro. And as people mentioned above, zoloft has more CYP concerns (vs lexapro).

But I barely ever use SSRIs unless the person is completely drug naive or had a good response to it in the past. The exception would be prozac and OCD, I use that a lot. It seems like half of the OCD patients do well on prozac, and the other half end up needing to be on clomipramine.

What are your preferred TCAs?

 

[Monocles]

Everyone says this and the theory sounds reasonable, but is it actually true? Any studies to support this?

eh, unless a well designed study comes out and says otherwise, I'm happy to suspend my disbelief and just go with it. 😛 I am just your average practicing shmuck with little interest in the splitting hairs of academic psychiatry 😛

I feel like *most* of psychiatry feels like wizardry and 40% of antidepressants is placebo anyway, so why not tell patients a fairly scientifically plausible, widely accepted explanation since it's hard to dispute otherwise?

(although I am not advocating putting my fingers in my ear and shouting "lalala". If the green journal comes out with an article tomorrow that says this is harmful to practice, I am willing to change on a dime)

 

[calvnandhobbs68]

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

In addition to the above answers that many patients would prefer to try to take a pill rather than therapy (as with most conditions...) nobody is paying psychiatry to do ERP (especially for employed psychiatrists) and/or most psychiatrists don’t have training in this. So it’s not really practical to do it yourself.

 

[tr]

Agree with all of this - the time and cost of ongoing therapy is often a major limiting factor. Have a few patients with pure obsessions and in these cases ERP is also not as effective (as there's no compulsions to prevent).

The Brain Lock system (Jeffrey Schwartz) is pretty good for pure obsessions. Many patients can implement it effectively themselves if you just point them to the book.

 

[clausewitz2]

Agree with all of this - the time and cost of ongoing therapy is often a major limiting factor. Have a few patients with pure obsessions and in these cases ERP is also not as effective (as there's no compulsions to prevent).



.

While there certainly may not be external/observable rituals in response to the obsessions, if it is actually OCD there are compulsions of some kind - could be mental compulsions, like cancelling thoughts, mental reviewing, existential rumination, intense monitoring of limb positioning, researching something horrible for reasons why it could not possibly apply to you, thinking through specific lists in a specific order. They may also be less obviously ritualized and more subtle, like confessing or reassurance seeking or moving very deliberately or constantly assessing their degree of groinal arousal for evidence that they might really be gay or a pedophile or want to cheat on their partners.

This is why a really careful assessment by someone who has a fair amount of specific experience but more importantly has the time to be exhaustive can be really important.

Mentioning "pure obsessions" around an OCD specialist is a great way to get a rant much more intense than this. The consensus widely seems to be that if there was such a thing as Pure O, it wouldn't be OCD, and to the extent the DSM allows for OCD with only obsessions, so much the worse for the DSM.

 

[NJWxMan]

Sometimes that hasn't been mentioned... family hx. If I see a kiddo with mdd that has 2 siblings and a parent all with positive responses, it's both an easier sell from an adherence perspective, but also biologically more likely to work.

This has recently been disproven in the literature.

 

[NJWxMan]

I use way more TCAs than SSRIs, but reasons I'd lean towards using Zoloft instead of Lexapro would be: bad PTSD, or poor response or bad reaction to celexa/lexapro in the past. In residency I heard about some researchers doing trials pushing zoloft to 300mg and 400mg. I'd never go over the FDA max in practice but it was interesting to read about. I do find people complain of nausea as a side effect with zoloft more than prozac or lexapro. And as people mentioned above, zoloft has more CYP concerns (vs lexapro).

But I barely ever use SSRIs unless the person is completely drug naive or had a good response to it in the past. The exception would be prozac and OCD, I use that a lot. It seems like half of the OCD patients do well on prozac, and the other half end up needing to be on clomipramine.

TCA's over SSRI's is not the standard of care.

 

[Basis]

How are you treating OCD with sertaline if you are not going to 400 mg? That is pretty standard.

I saw plenty of patients in residency on 200mg for OCD with little improvement. I know there is a continued increase in receptor occupancy over 200mg but I'm not willing to push anything over the FDA max for medical-legal reasons for one, and it's pretty marginal as a drug compared to clomipramine for patients with OCD.

Why not Luvox for OCD?

How many SSRIs do you trial before going to TCAs?

Luvox is alright I suppose, but I had little exposure to it in residency, so I don't use it. I don't feel it would offer anything vs. prozac.

It entirely depends on the patient, and what they've tried before, and how bad their symptoms are. Sometimes I'll start a TCA right away. Usually at least 1 SSRI is worth trying first. Depending on their response it helps to guide if I use a more serotonergic TCA or not.

What are your preferred TCAs?

I use a lot of desipramine, imipramine, and nortriptyline. Also a lot of doxepin, but only at doses appropriate for insomnia and anxiety related to that (usually like 10-30mg). Even at very high doses doxepin is still basically just a central antihistamine with a short half life. I did have some older attendings who used it skillfully at doses over 100mg but I wasn't really convinced. I wish I had more opportunities to use amitriptyline as it is a favorite, but most of my high functioning private patients are super sensitive to any sedating medications (as compared to residency clinic patients who can't ever seem to sleep per their report). I do have a few patients on it. I'd like to get more experience with protriptyline, but haven't had the chance lately.

TCA's over SSRI's is not the standard of care.

While I appreciate your snarky reply, TCAs, MAOIs, and typical antipsychotics are the only reason psychiatry is its own separate field of medicine. If SSRIs came out in the 1950s we would still be lumped in with neurology.

The suicide rate in the US has been going up since the exact year that rate-per-person-year prescriptions for SSRIs eclipsed TCAs (around 1999). Additionally, the US suicide rate for middle aged white men had previously been dropping since 1961 (when amitriptyline came out and imipramine came into more widespread use).

While obviously not the only factor, I feel very strongly that the prescribing of marginally effective drugs for patients with moderate and severe mental illness is one of the driving forces behind this trend.

Who is supposed to be an expert in these medications if not a psychiatrist? If a psychiatrist won't use them, who will? There are definitely patients you see who have tried several SSRIs and you know if their depression does not truly improve with medication treatment that they will surely kill themselves in the next year or two.

Remember that big 2018 meta-analysis where they compared 21 antidepressants? Elavil was way on the right in terms of efficacy, better than every newer drug.

I would encourage you to look up the story of Dr. Kuiper who was a dutch psychiatrist who was very much against MAOIs, saying that they were dangerous, until he himself was hospitalized in the 1980s with depression and only improved after the use of parnate. He wrote a book about the experience, which you can find translated copies of online. He spent most of the rest of his career advocating for the use of MAOIs and often gave lectures and interviews to this effect.

 

[hamstergang]

The suicide rate in the US has been going up since the exact year that rate-per-person-year prescriptions for SSRIs eclipsed TCAs (around 1999). Additionally, the US suicide rate for middle aged white men had previously been dropping since 1961 (when amitriptyline came out and imipramine came into more widespread use).

Please don't interpret this post as saying there's anything wrong with TCAs, but it feels like you're manipulating the data a bit to support your conclusion. If you think SSRIs are contributing to suicides, why would it take until SSRI prescriptions eclipsed TCA prescriptions to show this? Shouldn't this have been seen sometime in the decade before that given that they were still very widely used?

Also, we can break down the suicide rate data into many trends -- by gender, by age group, by race. Why did you specifically pick middle aged white men to see the trend starting 1961? This is almost like p-hacking in that if you look at enough trend lines, one group will support your conclusion. If you instead look at the overall suicide rate in the US, you'll see that it was rising through the 60s.

Similarly, I could decide to look at the suicide rate for only 15-24 year olds. I don't know the rate of TCA use vs SSRI use for that segment of the population, but their suicide rate didn't start to increase in 1999. Instead, it was falling, held flat from 2004-2007, and then started to go up. I think it's hard to blame SSRIs for this.

If anyone's interested in seeing the actual numbers, here's a thread I did on it: Suicide trends in the US, 1900 - 2019

 

[calvnandhobbs68]

I saw plenty of patients in residency on 200mg for OCD with little improvement. I know there is a continued increase in receptor occupancy over 200mg but I'm not willing to push anything over the FDA max for medical-legal reasons for one, and it's pretty marginal as a drug compared to clomipramine for patients with OCD.



Luvox is alright I suppose, but I had little exposure to it in residency, so I don't use it. I don't feel it would offer anything vs. prozac.

It entirely depends on the patient, and what they've tried before, and how bad their symptoms are. Sometimes I'll start a TCA right away. Usually at least 1 SSRI is worth trying first. Depending on their response it helps to guide if I use a more serotonergic TCA or not.



I use a lot of desipramine, imipramine, and nortriptyline. Also a lot of doxepin, but only at doses appropriate for insomnia and anxiety related to that (usually like 10-30mg). Even at very high doses doxepin is still basically just a central antihistamine with a short half life. I did have some older attendings who used it skillfully at doses over 100mg but I wasn't really convinced. I wish I had more opportunities to use amitriptyline as it is a favorite, but most of my high functioning private patients are super sensitive to any sedating medications (as compared to residency clinic patients who can't ever seem to sleep per their report). I do have a few patients on it. I'd like to get more experience with protriptyline, but haven't had the chance lately.



While I appreciate your snarky reply, TCAs, MAOIs, and typical antipsychotics are the only reason psychiatry is its own separate field of medicine. If SSRIs came out in the 1950s we would still be lumped in with neurology.

The suicide rate in the US has been going up since the exact year that rate-per-person-year prescriptions for SSRIs eclipsed TCAs (around 1999). Additionally, the US suicide rate for middle aged white men had previously been dropping since 1961 (when amitriptyline came out and imipramine came into more widespread use).

While obviously not the only factor, I feel very strongly that the prescribing of marginally effective drugs for patients with moderate and severe mental illness is one of the driving forces behind this trend.

Who is supposed to be an expert in these medications if not a psychiatrist? If a psychiatrist won't use them, who will? There are definitely patients you see who have tried several SSRIs and you know if their depression does not truly improve with medication treatment that they will surely kill themselves in the next year or two.

Remember that big 2018 meta-analysis where they compared 21 antidepressants? Elavil was way on the right in terms of efficacy, better than every newer drug.

I would encourage you to look up the story of Dr. Kuiper who was a dutch psychiatrist who was very much against MAOIs, saying that they were dangerous, until he himself was hospitalized in the 1980s with depression and only improved after the use of parnate. He wrote a book about the experience, which you can find translated copies of online. He spent most of the rest of his career advocating for the use of MAOIs and often gave lectures and interviews to this effect.

I also don't have anything in general against TCAs (amitriptyline/nortriptyline can be very helpful for chronic headaches) but it's interesting that you're so risk averse to pushing Zoloft above 200mg but are willing to prescribe TCAs as first line or after 1 SSRI trial. You don't feel liability for that would be high when you're prescribing something that's clearly many times more deadly in an overdose after no/weak SSRI/SNRI trial? I feel you can make an argument for TCAs after a patient has clearly failed multiple other antidepressants (or in OCD perhaps since clomipramine tends to get lumped in with SSRIs as first line treatment with guidelines) but if a family decides to sue after a patient dies from TCA overdose prescribed for regular old MDD, you're going to spend quite a bit of time trying to defend your reasoning of prescribing a medication known to be very deadly in overdose to a patient with an elevated risk of suicide over baseline population in general.

 

[hamstergang]

This has recently been disproven in the literature.

Do you have a citation? I've been wanting some evidence on this question.

 

[clausewitz2]

Please don't interpret this post as saying there's anything wrong with TCAs, but it feels like you're manipulating the data a bit to support your conclusion. If you think SSRIs are contributing to suicides, why would it take until SSRI prescriptions eclipsed TCA prescriptions to show this? Shouldn't this have been seen sometime in the decade before that given that they were still very widely used?

Also, we can break down the suicide rate data into many trends -- by gender, by age group, by race. Why did you specifically pick middle aged white men to see the trend starting 1961? This is almost like p-hacking in that if you look at enough trend lines, one group will support your conclusion. If you instead look at the overall suicide rate in the US, you'll see that it was rising through the 60s.

Similarly, I could decide to look at the suicide rate for only 15-24 year olds. I don't know the rate of TCA use vs SSRI use for that segment of the population, but their suicide rate didn't start to increase in 1999. Instead, it was falling, held flat from 2004-2007, and then started to go up. I think it's hard to blame SSRIs for this.

If anyone's interested in seeing the actual numbers, here's a thread I did on it: Suicide trends in the US, 1900 - 2019

Not defending the position necessarily but I took them to be saying that SSRIs are ineffective and that moving to them instead of TCAs explains the increase in suicides, rather than SSRIs directly causing them. They are welcome to correct me of course.

 

[hamstergang]

Not defending the position necessarily but I took them to be saying that SSRIs are ineffective and that moving to them instead of TCAs explains the increase in suicides, rather than SSRIs directly causing them. They are welcome to correct me of course.

I think this was just sloppy wording on my part. I still don't know why it would be expected, a priori, to see the rise in suicide rates only once SSRI prescriptions became more common the TCA prescriptions.

 

[clausewitz2]

I think this was just sloppy wording on my part. I still don't know why it would be expected, a priori, to see the rise in suicide rates only once SSRI prescriptions became more common the TCA prescriptions.

It might only become detectable statistically with a large enough proportion of the relevant demographic on one class rather than another. If TCAs were more effective in this specific way, you might expect not to reliably see more overall change with 100% TCA v 95% TCA, but maybe once TCAs are in the minority the signal is not swamped by random fluctuations.

Again, many missing intervening steps here, but not a formally illogical argument.

For the child people: my C&A trained attendings in residency would talk about a study showing TCAs lacked efficacy in adolescents for depression but I cannot for the life of me remember the authors of year of publication. Were they confabulating this or is this a familiar piece of the literature?

 

[hamstergang]

It might only become detectable statistically with a large enough proportion of the relevant demographic on one class rather than another. If TCAs were more effective in this specific way, you might expect not to reliably see more overall change with 100% TCA v 95% TCA, but maybe once TCAs are in the minority the signal is not swamped by random fluctuations.

Again, many missing intervening steps here, but not a formally illogical argument.

I get the logic but this still feels like trying to fit a conclusion to the data and then making up the reasoning. Why would we not see any effect when the TCA:SSRI use was 60: 40? And wouldn't you then think the suicide rate would be rising faster when the ratio became even more divided? We could think of a bunch of reasons why and why not for all of these questions but I don't know that the data can really prove this either way so how can we really know that the conclusion that SSRIs are worse than TCAs is valid?

For the child people: my C&A trained attendings in residency would talk about a study showing TCAs lacked efficacy in adolescents for depression but I cannot for the life of me remember the authors of year of publication. Were they confabulating this or is this a familiar piece of the literature?

This is correct. I know where I can find a citation no one does it quicker.

 

[Monocles]

This has recently been disproven in the literature.

Link please! 🙂

 

[hamstergang]

For the child people: my C&A trained attendings in residency would talk about a study showing TCAs lacked efficacy in adolescents for depression but I cannot for the life of me remember the authors of year of publication. Were they confabulating this or is this a familiar piece of the literature?

Tricyclic drugs for depression in children and adolescents - PubMed

Data suggest tricyclic drugs are not useful in treating depression in children. There is marginal evidence to support the use of tricyclic drugs in the treatment of depression in adolescents.

pubmed.ncbi.nlm.nih.gov

Hazell et al. Tricyclic drugs for depression in children and adolescents.

"Fourteen trials (590 participants) were included. No overall difference was found for the primary outcome of response to treatment compared with placebo (risk ratio (RR) 1.07, 95% confidence interval (CI) 0.91 to 1.26; 9 trials, N = 454). ... Subgroup analyses suggested a small reduction in depression symptoms among adolescents (SMD -0.45, 95% CI -0.83 to -0.007), and negligible change among children (SMD 0.15, 95% CI -0.34 to 0.64)."

 

[Stagg737]

Are you all trialing exposure response prevention therapy for OCD (either from you or via a referral) before trying meds, given that ERP can often be curative and even be curative with a relatively short course of tx?

Similar to above posters, good ERP can be tough to find in the community if the patient is willing to participate at all. Even if we do get them in with a therapist, I'd rather start them on meds now and attempt a trial with the understanding that we can always d/c them if they have adverse effects, do not feel they're efficacious, or complete therapy and are doing significantly better. I also ensure that my patients understand that starting a medication does NOT mean they will be on it the rest of their lives, as this is probably the most common medication misconception I encounter with my patients.

I treat a lot of postpartum OCD which tends to resolve spontaneously over time. Postpartum women don't have lots of spare time for labor-intensive therapies and the long-term advantage of ERP is not so relevant since the condition is often self-limited.

Is there much of a body of research on this? I've never seen this and would be curious if you'd even start most of these women on meds if it's that much of a self-limited disorder.

This has recently been disproven in the literature.

Disproven is pretty strong language. I'd also be interested in this research as most of my attendings still mention the efficacy of meds which have worked in the family and anecdotally I've seen the same thing (though I'll admit I've had very few long-term patients at this point).

 

[tr]

Is there much of a body of research on this? I've never seen this and would be curious if you'd even start most of these women on meds if it's that much of a self-limited disorder.

Do you treat a lot of perinatal women? I don't know if lots of people outside the field are paying attention but I sure see it a bunch. There's some documentation.

Onset and exacerbation of obsessive-compulsive disorder in pregnancy and the postpartum period - PubMed

Findings from this study provide additional evidence that pregnancy and childbirth are frequently associated with the onset of OCD or worsening of symptoms in those with preexisting disorder. In addition, there appears to be continuity between OCD onset and/or exacerbation across the...

pubmed.ncbi.nlm.nih.gov

You bet I'm going to treat it if the patient is spending most of her day fielding recurrent intrusive images of violent harm coming to her baby, or spending so much of her time in washing rituals that her husband and mother have to do 100% of the baby care. Self-limited meaning it usually goes away or back to baseline within a year postpartum. That's a long time to be untreated.

 

[Stagg737]

Do you treat a lot of perinatal women? I don't know if lots of people outside the field are paying attention but I sure see it a bunch. There's some documentation.

Onset and exacerbation of obsessive-compulsive disorder in pregnancy and the postpartum period - PubMed

Findings from this study provide additional evidence that pregnancy and childbirth are frequently associated with the onset of OCD or worsening of symptoms in those with preexisting disorder. In addition, there appears to be continuity between OCD onset and/or exacerbation across the...

pubmed.ncbi.nlm.nih.gov

You bet I'm going to treat it if the patient is spending most of her day fielding recurrent intrusive images of violent harm coming to her baby, or spending so much of her time in washing rituals that her husband and mother have to do 100% of the baby care. Self-limited meaning it usually goes away or back to baseline within a year postpartum. That's a long time to be untreated.

Not a lot, but I've had a few, mostly just PPD though and I actually do screen for OCD symptoms. If it's as severe as above, then I understand treating. How often are you treating less severe OCD symptoms/in those women? For example, when obsessions are concerning ability to breast-feed and compulsions involve washing rituals where mom is still able to appropriately care for child but symptom frequency is high.

 

[tr]

Not a lot, but I've had a few, mostly just PPD though and I actually do screen for OCD symptoms. If it's as severe as above, then I understand treating. How often are you treating less severe OCD symptoms/in those women? For example, when obsessions are concerning ability to breast-feed and compulsions involve washing rituals where mom is still able to appropriately care for child but symptom frequency is high.

I think it's a question of level of distress and functional impairment, as well as patient preference and tolerance for adverse med effects, just like anything else right?

 

[Stagg737]

I think it's a question of level of distress and functional impairment, as well as patient preference and tolerance for adverse med effects, just like anything else right?

Was just curious if you were more hesitant to start an SSRI in these patients since OCD typically requires higher doses, and pushing it in a breastfeeding/perinatal individual with significant hormonal changes for what you’re saying seems to be largely self-limiting condition.

Obviously if it’s severe enough that you’d need to consider hospitalization starting something would likely be imperative, but for those you do start on meds how fast are you titrating up? Especially in those who are breastfeeding?

 

[psyguru]

Luvox is alright I suppose, but I had little exposure to it in residency, so I don't use it. I don't feel it would offer anything vs. prozac.



\

Fluvoxamine also has antagonist properties at sigma 1 receptors which are unique and may be relevant to OCD. I prefer trying this before going to clomipramine. This property may help with other symptoms. For more: Understanding the Role of Sigma-1 Receptors in Psychotic Depression

 

[tr]

Was just curious if you were more hesitant to start an SSRI in these patients since OCD typically requires higher doses, and pushing it in a breastfeeding/perinatal individual with significant hormonal changes for what you’re saying seems to be largely self-limiting condition.

Obviously if it’s severe enough that you’d need to consider hospitalization starting something would likely be imperative, but for those you do start on meds how fast are you titrating up? Especially in those who are breastfeeding?

Most SSRIs have negligible lactational transfer. Ranging from undetectable to max 2-3% of weight adjusted maternal dose. The exception is Prozac which can get to 10% of weight adjusted maternal dose in young infants. No clinically detectable adverse effects of lactational exposure to SSRIs have been documented. I don't consider it a big barrier. A functional parent is way more important to the baby's health and well being than a very minor exposure to a safe and well tolerated medication.

 

[Ironspy]

Most SSRIs have negligible lactational transfer. Ranging from undetectable to max 2-3% of weight adjusted maternal dose. The exception is Prozac which can get to 10% of weight adjusted maternal dose in young infants. No clinically detectable adverse effects of lactational exposure to SSRIs have been documented. I don't consider it a big barrier. A functional parent is way more important to the baby's health and well being than a very minor exposure to a safe and well tolerated medication.

My limited experience with these patients is that by the time they finally see a psychiatrist their symptoms are severe enough to be impairing and they virtually all warrant treatment.

 

[clausewitz2]

Fluvoxamine also has antagonist properties at sigma 1 receptors which are unique and may be relevant to OCD. I prefer trying this before going to clomipramine. This property may help with other symptoms. For more: Understanding the Role of Sigma-1 Receptors in Psychotic Depression

The pursuit of an OCD indication for fluvoxamine was motivated entirely by the manufacturer's feeling that the SSRI antidepressant space was too crowded. No one has ever demonstrated any advantage for luvox compared to any other SSRI. Only advantage in reality is that it is much harder for insurance to deny the preauth for OCD doses, whereas 80 mg of Paxil or 40 mg of Lexapro daily is the hill some local insurers have chosen to die on.

Edit: quick review of literature confirms my recollection that numerous SSRIs antagonize sigma 1, although fluvoxamine has the highest affinity. For the curious sertaline is next highest.