Screening EKGs for Geodon

[hamstergang]

There's a current thread in the EM forum about getting CTs and EKGs on all their psych patients: http://forums.studentdoctor.net/threads/ekg-and-ct-on-all-psych-pts.1145693/

Someone commented that they felt EKGs were reasonable as antipsychotics can prolong the QT and so having that baseline would be important before prescribing such a medication. I responded that, as far as I was aware, the evidence did not support actually getting an EKG. Specifically:

"QTc is only a surrogate marker for what we really care about, and there isn't a direct correlation between QT prolongation and incidence of TdP."

"I just now rechecked the prescribing information from Pfizer with ziprasidone, and I don't see them recommending an EKG prior to treatment initiation. The way I would cover myself would be by following the evidence: Geodon does tend to increase the QTc more than other antipsychotics, but it's not clear that this is clinically relevant. I would avoid use of Geodon or get a screening EKG if it really is the best choice in patients with personal or family history of prolonged QT, uncompensated heart failure, recent MI, bradycardia, electrolyte abnormalities, or on other medications known to prolong the QT interval. I might be more careful in females since they seem to be at increased risk.

I know that there are many doctors and sources out there that will disagree and insist on an EKG before using Geodon, but that's not coming from the evidence that I've been able to find."

Is there more to this than I'm aware? How many of you do get EKGs, and why?

---

Members don't see this ad.

 

[birchswing]

The CT scan seems very concerning, especially when every hospital has an MRI machine available. It's hard to understand the cost argument when you're talking about a potentially carcinogenic fishing expedition.

 

[heyjack70]

The CT scan seems very concerning, especially when every hospital has an MRI machine available. It's hard to understand the cost argument when you're talking about a potentially carcinogenic fishing expedition.

Not arguing for it, but CT over MRI has the advantage of being much faster and much cheaper. I would be impossible time wise to get an MRI on each emergency patient.

 

[Armadillos]

This is an interesting thread in general, curious what everyone's institutions "require" workup wise before admitting from the ED?

 

[Armadillos]

The CT scan seems very concerning, especially when every hospital has an MRI machine available. It's hard to understand the cost argument when you're talking about a potentially carcinogenic fishing expedition.

Getting a CT on everyone is absurd but getting an MRI would be even more ridiculous. The entire medical system would probably implode if we MRI'd every psych pt.

 

[Wilf]

Getting a CT on everyone is absurd but getting an MRI would be even more ridiculous. The entire medical system would probably implode if we MRI'd every psych pt.

The poster who wrote that is not in the medical field.

 

[heyjack70]

I support getting a baseline EKG on psych patients coming through the ED. We do prescribe a lot of meds that effect QTc and additive effects are concerning so it's good to have a baseline on no meds (or fewer meds). If you get a baseline EKG and QTc is 430 and 2 weeks later on 30 mg Zyprexa and 20 mg Celexa and repeat QTc is 429, I would feel much better. Getting an EKG after 2 weeks with no baseline and having QTc at 485 is more concerning; has the QTc always been this long? Is this due to the meds?

We get lots of screening tests like cbc, chemistry, tsh, vitamin levels, RPR!, why not add an EKG. I also think hospitals should have rapid pharmacogenetic testing to assess for rapid or slow metabolism of meds so we can have that information before we even start trying meds. Right now we only check pharmacogenetics if the patient is not responding or seems to have exaggerated side effects.

 

[birchswing]

Getting a CT on everyone is absurd but getting an MRI would be even more ridiculous. The entire medical system would probably implode if we MRI'd every psych pt.

I agree—that's why I referred to it as a fishing expedition. I was saying it's even worse that it's a fishing expedition when it's potentially carcinogenic.

The time-sensitivity thing I get, and the not needing to use it for every patient I get—the cases I don't get are when it's not time sensitive and the patient does need it. If one is potentially carcinogenic and one isn't, it seems to me that cost shouldn't be the deciding factor. It also seems that the more MRIs are used the faster that they could be paid off and the cheaper the service could become over time—I guess insurance companies are the ones who want to see validity for an MRI over CT. For me, any case in which you can avoid ionizing radiation and achieve the same diagnostic results should be a valid enough reason, but that's obviously not the case currently.

 

[birchswing]

I support getting a baseline EKG on psych patients coming through the ED. We do prescribe a lot of meds that effect QTc and additive effects are concerning so it's good to have a baseline on no meds (or fewer meds). If you get a baseline EKG and QTc is 430 and 2 weeks later on 30 mg Zyprexa and 20 mg Celexa and repeat QTc is 429, I would feel much better. Getting an EKG after 2 weeks with no baseline and having QTc at 485 is more concerning; has the QTc always been this long? Is this due to the meds?

We get lots of screening tests like cbc, chemistry, tsh, vitamin levels, RPR!, why not add an EKG. I also think hospitals should have rapid pharmacogenetic testing to assess for rapid or slow metabolism of meds so we can have that information before we even start trying meds. Right now we only check pharmacogenetics if the patient is not responding or seems to have exaggerated side effects.

I have a card in my wallet than any doctor can use to see my most important pharmacogenetic info on the front and can pull up the entire profile by entering a code from the card along with their NPI. I give it to all my doctors to photocopy. I used the YouScript service. I like it a lot. It has a great interface for checking for drug-drug and drug-genetic interactions, but there are probably others like it. Any doctor with the code on the front of it can log onto a website and immediately see my profile.

 

[birchswing]

The poster who wrote that is not in the medical field.

I bet I've had more CT and MRI head scans than you.

 

[heyjack70]

I have a card in my wallet than any doctor can use to see my most important pharmacogenetic info on the front and can pull up the entire profile by entering a code from the card along with their NPI. I give it to all my doctors to photocopy. I used the YouScript service. I like it a lot. It has a great interface for checking for drug-drug and drug-genetic interactions, but there are probably others like it. Any doctor with the code on the front of it can log onto a website and immediately see my profile.

sounds fantastic, no need to repeat testing when the results will never change

 

[hamstergang]

I support getting a baseline EKG on psych patients coming through the ED. We do prescribe a lot of meds that effect QTc and additive effects are concerning so it's good to have a baseline on no meds (or fewer meds). If you get a baseline EKG and QTc is 430 and 2 weeks later on 30 mg Zyprexa and 20 mg Celexa and repeat QTc is 429, I would feel much better. Getting an EKG after 2 weeks with no baseline and having QTc at 485 is more concerning; has the QTc always been this long? Is this due to the meds?

We get lots of screening tests like cbc, chemistry, tsh, vitamin levels, RPR!, why not add an EKG. I also think hospitals should have rapid pharmacogenetic testing to assess for rapid or slow metabolism of meds so we can have that information before we even start trying meds. Right now we only check pharmacogenetics if the patient is not responding or seems to have exaggerated side effects.

How much of this is serious?

 

[heyjack70]

How much of this is serious?

All of it. If you take issue with something you might try addressing it specifically instead of asking a lame question.

 

[Jester25]

QTc prolongation is a risk factor for torsades. EKG isn't super helpful b/c it's a small picture, the numbers can vary within 30 seconds of getting another EKG. Still, it's better than nothing.


Geodon is pretty useless anyway, I barely use it. If I did, I'd def get a baseline EKG. I don't give the IM unless I have EKG and electrolytes.

 

[wolfvgang22]

Doctors do a lot of things to cover their arses that aren't evidence based. I've known some ER docs that insisted on giving benztropine to every single patient that got promethazine. I don't do that, and most ER doctors don't either.

 

[Jester25]

Doesn't have to be in the guidelines for it to make sense. It's hard to prove a 50 yr old on Geodon drops dead from TdP unless you have him on cardiac monitoring...which is going to be highly unlikely. Add to it the rampant polypharmacy and the risks go up.

 

[hamstergang]

We get lots of screening tests like cbc, chemistry, tsh, vitamin levels, RPR!, why not add an EKG. I also think hospitals should have rapid pharmacogenetic testing to assess for rapid or slow metabolism of meds so we can have that information before we even start trying meds. Right now we only check pharmacogenetics if the patient is not responding or seems to have exaggerated side effects.

Ok, so my problem with this post is that the reasoning in it seems to go against everything I remember being taught in med school. "Why not?" is a terrible reason for getting a test. Each and every test we do should be well reasoned, with potential risks and benefits weighed against each other based on whatever evidence is available. With this in mind, I don't understand getting an EKG prior to starting Geodon.

The pharmacogenetics is another thing that really bothers me because I haven't yet had anyone here or in real life (unless I blocked it out of my memory) explain why this is useful. For one thing, it's the phenotype, not genotype, that we actually care about, and I haven't ever heard of someone wanting to test for that (even though you can). Secondly, outside of this genetic testing, I never heard people talk about one's metabolic rate of a drug affecting whether that was a good or bad drug for that person; instead, the metabolic rate should influence dosage and dosing schedule. So where's the actual value in this pharmacogenetic testing? And can you actually justify the cost in getting it preemptively?

 

[heyjack70]

Ok, so my problem with this post is that the reasoning in it seems to go against everything I remember being taught in med school. "Why not?" is a terrible reason for getting a test. Each and every test we do should be well reasoned, with potential risks and benefits weighed against each other based on whatever evidence is available. With this in mind, I don't understand getting an EKG prior to starting Geodon.

The pharmacogenetics is another thing that really bothers me because I haven't yet had anyone here or in real life (unless I blocked it out of my memory) explain why this is useful. For one thing, it's the phenotype, not genotype, that we actually care about, and I haven't ever heard of someone wanting to test for that (even though you can). Secondly, outside of this genetic testing, I never heard people talk about one's metabolic rate of a drug affecting whether that was a good or bad drug for that person; instead, the metabolic rate should influence dosage and dosing schedule. So where's the actual value in this pharmacogenetic testing? And can you actually justify the cost in getting it preemptively?

Well it's not "why not" get the test. It's the drugs we use cause QTc prolongation which is a risk factor for torsades de pointes, and we get sick people in the hospital on multiple psychiatric medications, and also non psych meds that can lengthen QTc, so in this setting getting a baseline EKG and periodically checking seems reasonable. The benefits seem to far outweigh the risks with screening EKGs.

As far as genetic testing, the concept of testing for phenotype vs genotype is news to me. How is this phenotypic testing done? And I hate getting into these semantic debates, but I never said genetic testing would tell if a drug is a good drug vs bad drug for a person, but it does tell you if the person is going to require an ultra low starting dose and may still have excessive side effects, or may need a much larger than typical dose than average. If you knew this going in you could avoid meds with potential problems from the outset and save time in the hospital, and save the patient time struggling with unresolved symptoms. The testing companies charge something around $200 for the basic cytochrome screening. If this saved one day in the hospital it would more than pay for itself.

 

[Psychobabbling]

Gonna have to come to heyjack's defense on this one.

"Ok, so my problem with this post is that the reasoning in it seems to go against everything I remember being taught in med school. "Why not?" is a terrible reason for getting a test. Each and every test we do should be well reasoned, with potential risks and benefits weighed against each other based on whatever evidence is available. With this in mind, I don't understand getting an EKG prior to starting Geodon."

Risks/benefits of getting an EKG. Getting an EKG - no risk. Benefit - establishes baseline, improves monitoring, etc. With this in mind, I'm not sure what the problem is?

 

[Armadillos]

Gonna have to come to heyjack's defense on this one.

"Ok, so my problem with this post is that the reasoning in it seems to go against everything I remember being taught in med school. "Why not?" is a terrible reason for getting a test. Each and every test we do should be well reasoned, with potential risks and benefits weighed against each other based on whatever evidence is available. With this in mind, I don't understand getting an EKG prior to starting Geodon."

Risks/benefits of getting an EKG. Getting an EKG - no risk. Benefit - establishes baseline, improves monitoring, etc. With this in mind, I'm not sure what the problem is?

Obviously an EKG isn't itself dangerous, but if they are interpreted by providers who don't do dozens of them day in and day out a ton of unnecessary cardiology follow ups are generated. The stress of waiting on that appointment isn't trivial for patients.

Edit- If the place your ordering has a setup where cards fellows or someone end up reviewing all the EKGs eventually, then this obviously lessens that concern.

 

[heyjack70]

Obviously an EKG isn't itself dangerous, but if they are interpreted by providers who don't do dozens of them day in and day out a ton of unnecessary cardiology follow ups are generated. The stress of waiting on that appointment isn't trivial for patients.

Edit- If the place your ordering has a setup where cards fellows or someone end up reviewing all the EKGs eventually, then this obviously lessens that concern.

This is a problem in the outpatient setting. In the hospital all EKGs are likely read by a cardiologist, or at least internists who interpret them regularly.

 

[hamstergang]

Well it's not "why not" get the test.

Well, that is what you said ("why not add an EKG").

It's the drugs we use cause QTc prolongation which is a risk factor for torsades de pointes, and we get sick people in the hospital on multiple psychiatric medications, and also non psych meds that can lengthen QTc, so in this setting getting a baseline EKG and periodically checking seems reasonable.

I already mentioned in my original post that there are certain populations for which an EKG makes sense. But as a general rule to get an EKG, there's just not data supporting the practice as it doesn't actually give us the information that matters. We want to predict if our patients will get TdP, and even if you see an increase in the QTc after administering a medication, it doesn't really help you do that (again, at least as far as I've been able to see, and I've searched a lot on this particular topic).

As far as genetic testing, the concept of testing for phenotype vs genotype is news to me. How is this phenotypic testing done? And I hate getting into these semantic debates, but I never said genetic testing would tell if a drug is a good drug vs bad drug for a person, but it does tell you if the person is going to require an ultra low starting dose and may still have excessive side effects, or may need a much larger than typical dose than average. If you knew this going in you could avoid meds with potential problems from the outset and save time in the hospital, and save the patient time struggling with unresolved symptoms. The testing companies charge something around $200 for the basic cytochrome screening. If this saved one day in the hospital it would more than pay for itself.

You can test the phenotype by giving the patient a specific dose of a substrate of the enzyme you're interested in, and then after a specified time test for the metabolite (or ratio of metabolite to substrate) and do some math, and the computer gives you the rate of that enzyme. Not having heard of this isn't unique to you. In discussions I've had on this in residency, it seems almost no one knows about this, and it seems to be because no one cared about such things until the companies doing the genetic testing started trying to convince us that it was important. At least, that's the best explanation I've been able to come up with and haven't heard a refutation yet.

You are correct that you didn't say good or bad drug -- that comes from what I've heard from others when having this debate. Apparently one of the companies (is there only one?) gives you a list a drugs labeled in green, meaning it's 'good' to use, another list in yellow, meaning 'be cautious' or something, and another list in red, meaning something worse. It seems you've avoided this pitfall, which I think is obvious but somehow one of my most senior attendings hasn't.

Yes, if it saved every patient you test one day in the hospital, it would pay for itself. If it saved only one patient in a thousand a day in the hospital, it would not be cost effective. So where's the actual break-even point, and where's the evidence that this gets us there? I'll admit I haven't looked for this evidence yet because the concept never made sense to me, but your idea of not outright avoiding medications based on the pharmacogenetics but instead using it to inform dosing seems more reasonable.

Risks/benefits of getting an EKG. Getting an EKG - no risk. Benefit - establishes baseline, improves monitoring, etc. With this in mind, I'm not sure what the problem is?

On top of what Armadillos said, EKGs cost money, and depending on the setting can prolong one's stay in the ER. Even with cardiologists reading the EKGs, there are false positives that wind up with costs/harms from follow up tests. Also, you may get an incidental finding that can keep you from using a medication that would actually be safe and effective. So the risk isn't absent.

As for the benefit you mention of improving monitoring, that's exactly what I've been arguing about -- I haven't seen any evidence that this is actually true. It certainly comes from a logical place: prolonged QTc is a risk factor for TdP, we would like to avoid TdP, Geodon can prolong the QTc, so it seems reasonable that by monitoring the QTc we can help prevent TdP. The problem is that such logic doesn't always hold in medicine because the links in each step isn't quite so clean.

 

[heyjack70]

Well, that is what you said ("why not add an EKG").


I already mentioned in my original post that there are certain populations for which an EKG makes sense. But as a general rule to get an EKG, there's just not data supporting the practice as it doesn't actually give us the information that matters. We want to predict if our patients will get TdP, and even if you see an increase in the QTc after administering a medication, it doesn't really help you do that (again, at least as far as I've been able to see, and I've searched a lot on this particular topic).


You can test the phenotype by giving the patient a specific dose of a substrate of the enzyme you're interested in, and then after a specified time test for the metabolite (or ratio of metabolite to substrate) and do some math, and the computer gives you the rate of that enzyme. Not having heard of this isn't unique to you. In discussions I've had on this in residency, it seems almost no one knows about this, and it seems to be because no one cared about such things until the companies doing the genetic testing started trying to convince us that it was important. At least, that's the best explanation I've been able to come up with and haven't heard a refutation yet.

You are correct that you didn't say good or bad drug -- that comes from what I've heard from others when having this debate. Apparently one of the companies (is there only one?) gives you a list a drugs labeled in green, meaning it's 'good' to use, another list in yellow, meaning 'be cautious' or something, and another list in red, meaning something worse. It seems you've avoided this pitfall, which I think is obvious but somehow one of my most senior attendings hasn't.

Yes, if it saved every patient you test one day in the hospital, it would pay for itself. If it saved only one patient in a thousand a day in the hospital, it would not be cost effective. So where's the actual break-even point, and where's the evidence that this gets us there? I'll admit I haven't looked for this evidence yet because the concept never made sense to me, but your idea of not outright avoiding medications based on the pharmacogenetics but instead using it to inform dosing seems more reasonable.


On top of what Armadillos said, EKGs cost money, and depending on the setting can prolong one's stay in the ER. Even with cardiologists reading the EKGs, there are false positives that wind up with costs/harms from follow up tests. Also, you may get an incidental finding that can keep you from using a medication that would actually be safe and effective. So the risk isn't absent.

As for the benefit you mention of improving monitoring, that's exactly what I've been arguing about -- I haven't seen any evidence that this is actually true. It certainly comes from a logical place: prolonged QTc is a risk factor for TdP, we would like to avoid TdP, Geodon can prolong the QTc, so it seems reasonable that by monitoring the QTc we can help prevent TdP. The problem is that such logic doesn't always hold in medicine because the links in each step isn't quite so clean.

Excellent post. I'll play devils advocate here, just because being cynical and worried about lawsuits keeps me up at night. From the first post in this thread, where the Pfizer prescribing information is referenced: ..."I would avoid use of Geodon or get a screening EKG if it really is the best choice in patients with personal or family history of prolonged QT". Say you have a 30 year old man in front of you with schizophrenia and diabetes, but otherwise seemingly healthy. Geodon being less metabolically problematic is a good choice in this patient. How will you know if the patient does or does not have a personal history of prolonged QT?

 

[shan564]

I think it's reasonable to monitor EKGs on patients who are on QTc-prolonging drugs. As for the idea that QTc doesn't directly correlate with TdP - I'm not familiar with the evidence they're citing, but even if it's not a DIRECT correlation, I think it's pretty clear that there's a correlation. It's well-established that prolonged QTc is a risk factor for TdP, and that giving QTc-prolonging drugs to people with a prolonged QTc increases the risk further. I'm not a cardiologist, but even if the test isn't optimal, it's still useful. And if you scroll down in the thread cited in the first post, another ER doctor disagreed with the assertion that EKG is not useful in this situation, and cited pretty good rationale. It's a cheap and easy test.

I'm not saying that every psych patient should get an EKG, but it might be easier to have an all-encompassing rule than to rely on a busy ER doc to figure out which patients are at higher risk than others.

Here's an n=1 anecdote: I recently saw a consult in the ICU for a patient who had developed TdP on methadone. She was on a moderately high dose (but not super-high) of 90mg, and nobody was checking EKGs. She would have died, except that she happened to develop TdP while she was on a bus, and somebody on that bus happened to know how to do good quality CPR (she had some injured ribs, so CPR was probably done well). If she'd developed it while sitting at home, she would have died. She had previously visited our ER for something unrelated, but nobody checked an EKG because it wasn't necessary. If somebody had checked it, an adverse outcome could have been prevented. Yes, you'd have to do a lot of EKGs to save one person, but considering that it's a cheap and easy test, I feel like it could have benefited this patient. And I imagine that for every ICU patient who barely survived an episode of TdP, there must be at least one (or several) who died because they didn't happen to be in close proximity to somebody who knows CPR.

As for whether it's necessary for Geodon - I don't see why not.

As for the head CT, I agree that it seems kind of like a fishing trip. I understand the rationale that a CT is way cheaper/easier than an MRI and it'd be silly to do an MRI on everybody, but what would a non-contrast CT find? Are you looking for a brain bleed? I can't imagine what would cause isolated psychiatric symptoms and would show up on a non-contrast CT. Maybe an old stroke?
I feel like it'd be more reasonable to do an MRI if there's any clinical suspicion for something weird, but it'd take some convincing to make me believe

 

[Jester25]

You need to do a lot more reading. It is firmly established that QTc prolongation is an independent risk factor for torsades. Prolonged QTc is a contraindication for Geodon use.

Would you get electrolytes before prescribing Geodon? If so, by the same logic, you should get an EKG.

Well, that is what you said ("why not add an EKG").


I already mentioned in my original post that there are certain populations for which an EKG makes sense. But as a general rule to get an EKG, there's just not data supporting the practice as it doesn't actually give us the information that matters. We want to predict if our patients will get TdP, and even if you see an increase in the QTc after administering a medication, it doesn't really help you do that (again, at least as far as I've been able to see, and I've searched a lot on this particular topic).


You can test the phenotype by giving the patient a specific dose of a substrate of the enzyme you're interested in, and then after a specified time test for the metabolite (or ratio of metabolite to substrate) and do some math, and the computer gives you the rate of that enzyme. Not having heard of this isn't unique to you. In discussions I've had on this in residency, it seems almost no one knows about this, and it seems to be because no one cared about such things until the companies doing the genetic testing started trying to convince us that it was important. At least, that's the best explanation I've been able to come up with and haven't heard a refutation yet.

You are correct that you didn't say good or bad drug -- that comes from what I've heard from others when having this debate. Apparently one of the companies (is there only one?) gives you a list a drugs labeled in green, meaning it's 'good' to use, another list in yellow, meaning 'be cautious' or something, and another list in red, meaning something worse. It seems you've avoided this pitfall, which I think is obvious but somehow one of my most senior attendings hasn't.

Yes, if it saved every patient you test one day in the hospital, it would pay for itself. If it saved only one patient in a thousand a day in the hospital, it would not be cost effective. So where's the actual break-even point, and where's the evidence that this gets us there? I'll admit I haven't looked for this evidence yet because the concept never made sense to me, but your idea of not outright avoiding medications based on the pharmacogenetics but instead using it to inform dosing seems more reasonable.


On top of what Armadillos said, EKGs cost money, and depending on the setting can prolong one's stay in the ER. Even with cardiologists reading the EKGs, there are false positives that wind up with costs/harms from follow up tests. Also, you may get an incidental finding that can keep you from using a medication that would actually be safe and effective. So the risk isn't absent.

As for the benefit you mention of improving monitoring, that's exactly what I've been arguing about -- I haven't seen any evidence that this is actually true. It certainly comes from a logical place: prolonged QTc is a risk factor for TdP, we would like to avoid TdP, Geodon can prolong the QTc, so it seems reasonable that by monitoring the QTc we can help prevent TdP. The problem is that such logic doesn't always hold in medicine because the links in each step isn't quite so clean.

 

[Doctor Bagel]

At the university hospital where I did my residency, we started getting EKGs on all patients in the ED prior to admission. The hospital where I moonlight just adopted a similar policy as well. We also get annual screening EKGs for all patients in the methadone clinic. From what I understand, there aren't clear guidelines recommending this because as mentioned above torsades is a multifactorial type of thing. However, I get the why not approach, too, especially if you think you're going to be giving someone emergent IM antipsychotics. I certainly find it reassuring when someone arrives on the floor, and I can confirm that their QTc is only 425. The sad thing is I have no idea how much it costs the system to do all these EKGs, which would be the argument against them. A study about iatrogenic harm from excessive EKGs would also be interesting to see. I'm not up on the data on this. Ideally we shouldn't do excessive screening, but as a practitioner, I get why we do it, especially dealing with acutely ill patients.

So real life EKG ridiculousness -- repeating the EKG until you get an answer you like, which works because one's QTc can vary a lot within a short period of time.

 

[Doctor Bagel]

Hmm, even with a huge NNT, maybe a screening EKG is at least cost-effective. Just read the abstract though ---

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0127213

 

[milesed]

We routinely check Ekg's on anyone on Geodon, Orap, etc. as a baseline and every 6 months with other labs. My arnp recently had one come back at 499 on geodon. It's rare, but can be deadly, so why not?

 

[hamstergang]

Hmm, even with a huge NNT, maybe a screening EKG is at least cost-effective. Just read the abstract though ---

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0127213

I appreciate this, as it actually attempts to answer the question raised in this thread. The conclusion seems to put an end to this thread: "The results suggest that one death may be prevented by screening 3 000 patients at admission and thus identifying and treating patients pro-arrhythmic propensity. Despite the relatively high number of patients needed to screen, the LQT detection appears cost-effective"

But, I'm not so sure about the methods, in large part because I don't really understand them enough or the implications of them. For example:
1) "The study population was derived from 6790 men and women adults... Because the incidence of TdP resulting from LQT is low, a prospective comparative study would need several thousands of patients: we used a de novo decision analytic model with 10 000 cohort simulations, each of them having 100 000 patients; we extrapolated the results over entire life."

2) "As the risk of TdP in patients with LQT is poorly characterized in the literature, it was derived from an experts’ elicitation... we approached 13 experts ... including 11 cardiologists (of whom 7 were electrophysiologists), and 2 psychopharmacologists. Each expert was individually and independently asked to estimate the risk of TdP for patients with the three categories of LQT within a one-year period after the occurrence of LQT. We used the histogram method by presenting plausible ranges of values partitioned into intervals of 1%, and asking experts to estimate the probability of TdP risk for each interval. Experts’ opinions were then combined by taking the arithmetic mean of individual assessments"

It all sounds reasonable, though. Anyone think this article is actually misleading in some way?

 

[hamstergang]

As for the idea that QTc doesn't directly correlate with TdP - I'm not familiar with the evidence they're citing, but even if it's not a DIRECT correlation, I think it's pretty clear that there's a correlation. It's well-established that prolonged QTc is a risk factor for TdP, and that giving QTc-prolonging drugs to people with a prolonged QTc increases the risk further.

It is firmly established that QTc prolongation is an independent risk factor for torsades. Prolonged QTc is a contraindication for Geodon use.

As mentioned in the study I quoted in my most recent post, "the risk of TdP in patients with LQT is poorly characterized in the literature." There just isn't enough information to be able to say that someone with a QTc in a certain range or a certain QT prolongation from baseline has x% chance of TdP. If drug A prolongs the QT interval more than drug B, it isn't necessarily so that drug A confers a higher risk of TdP. While Geodon prolongs the QTc more than other antipsychotics currently in use, I haven't seen any evidence (and not from a lack of reading, despite what you may believe, Jester) that Geodon is associated with a higher incidence of TdP or sudden death. The prescribing information for Geodon doesn't even recommend an EKG prior to medication initiation.

So the problem is that, while there is a correlation between QT prolongation and TdP, getting the value of the QTc or prolongation of QTc doesn't do much to help you determine if your patient is at appreciably increased risk of TdP. You get the EKG, but then what?

Would you get electrolytes before prescribing Geodon? If so, by the same logic, you should get an EKG.

In patients who are at risk for electrolyte disturbance, yes I would check them. That's in the prescribing information. Similarly, in patients at risk for cardiac issues, I would check an EKG prior to starting Geodon. But for a healthy patient without significant family history, I don't know that I'd check either (only now changing my mind due to the above study by Doctor Bagel).

Excellent post. I'll play devils advocate here, just because being cynical and worried about lawsuits keeps me up at night. From the first post in this thread, where the Pfizer prescribing information is referenced: ..."I would avoid use of Geodon or get a screening EKG if it really is the best choice in patients with personal or family history of prolonged QT". Say you have a 30 year old man in front of you with schizophrenia and diabetes, but otherwise seemingly healthy. Geodon being less metabolically problematic is a good choice in this patient. How will you know if the patient does or does not have a personal history of prolonged QT?

Thanks, I post this question here for debates such as this. Otherwise, how can I learn?

For this patient, I would ask him. If he gives me information about any sort of cardiac issues in his family, I would consider that he may be talking about prolonged QT. I would also ask about sudden unexplained deaths, which I would assume to be cardiac in nature unless given sufficient evidence otherwise (but then they wouldn't be 'unexplained'). This isn't perfect, but I don't know that it has to be.

Would you check an EKG before starting any antipsychotic, or only Geodon?

 

[splik]

this thread is misleadingly titled as it has nothing to do with screening EKGs for geodon but screening EKGs for psych patients in the emergency department. the referenced thread talks about children specifically.

personally i think there are many good reasons why it is eminently sensible to do an ECG on all adult psych admissions:

- pretty much every pt admitted anywhere else in the hospital would get an admission ECG - should it be different for psych patients?
- it gives a baseline for comparison if events do occur
- patients may have electrolyte abnormalities causing cardiac changes found on EKG
- patients abusing substances are prone to arrhythmias and other cardiac abnormalities that may throw up something on the EKG (including a grumbling endocarditis)
- many psychotropic drugs do increase QTc and like it or not this is something we get all up in arms about and does influence selection of neuroleptics
- inpatients more likely to receive rapid tranquilization which has significant cardiac risk
- psychotropics may cause other cardiac problems or have other contraindications (for example lithium and sick sinus syndrome, clozapine may cause myocarditis, TCAs and MAOIs may also be contraindicated in various conditions, and antipsychotics can often cause other more benign tachyarrhythmias through their effects on alpha-1 receptors)
- psychiatric patients are at increased risk of cardiac disease (ECG may reveal LVH for example)
- ECG changes may provide support for specific toxidrome or overdose in a patient
- ECG may very rarely suggest or support a medical cause for a psychiatric presentation (SLE, sarcoidosis, hemochromatosis, lyme, syphilis etc)
- stress can cause a broken heart (Takutsubo's cardiomyopathy anyone?)
- it is a cheap and relatively simple, non-invasive investigation
- it is easier to do in the ED than on a psych unit

 

[birchswing]

EKG is going to become more and more common among consumers. I already use AliveCor which is FDA approved for consumers for detecting A-fib (got a good deal on it--only $49). The algorithm is approved for detecting A-fib or normal sinus rhythm. For other undetectable rhythms, you can pay a very small fee to have a board-certified cardiologist read it as well or e-mail it to your doctor. A three-channel wearable, continuous EKG device will be coming out soon (QardioCore) for consumers, also FDA approved. ZioPatch is also a great alternative to Holter monitors. It has 14-day recording, is small, disposable, and waterproof.

With sudden cardiac arrest being such a large killer, I think that these devices will become more and more common in consumer electronics. You also see people buying their own AEDs. Costco sells an AED for only $999. People using Apple Watch are already wearing what are essentially pulse oximeters on their wrists (although the FDA hasn't yet approved the oximetry part, so it only reports heart rate). I think once you get people into measuring these things, the metrics they'll want to keep track of will only increase.

This doesn't relate to the thread a lot, but I guess it does to the extent that the trend of doing an EKG on all psych patients in an ED is part of a growing ubiquity of EKGs and even AEDs in the consumer world. It's also part of this push to get more and more "big" data from which trends could be determined (AliveCor for example uses the data you give it in aggregate). Life-logging has been a thing for a while—it's kind of the opposite of intuition. It's trusting numbers to optimize your life. I think people want to do "life-logging" at an aggregate level, as well, so these hospital numbers might be interesting. Why trust a PI sheet over a census EKG survey of every patient who comes into a hospital on a particular drug? I mean that's the idea, at least.

 

[NickNaylor]

At my medical school, it was routine to get EKGs on ANYONE that we wanted to start an antipsychotic on - even if we were planning on using an antipsychotic that was known to have a minimal impact on QTc (Abilify, Zyprexa, etc.). This was on a C/L service, though, where many of the patients are anywhere from somewhat ill to seriously ill, and the attendings typically didn't want to take any degree of risk.

In contrast, the in-patient unit at an affiliated hospital didn't get EKGs routinely on anyone prior to starting an antipsychotic. However, most of those patients were otherwise young and healthy.

 

[hamstergang]

this thread is misleadingly titled as it has nothing to do with screening EKGs for geodon

Actually, this thread is perfectly titled as it is precisely about screening EKGs for Geodon. I made the thread and that's what I wanted to talk about even though it came from another thread on a slightly different topic.

personally i think there are many good reasons why it is eminently sensible to do an ECG on all adult psych admissions:

I don't know if you've read this thread yet, but several of your points have already been addressed. Others are specific reasons why specific patients should get EKGs, but certainly aren't generalizable (eg that overdose patients should be EKGs does not imply all patients should get EKGs).

 

[whopper]

Based on who is running the department, local practices could vary. One place could make an EKG mandatory for all patients (I guess an exception would be if the patient refused).

I agree with several of the viewpoints above. It's not a bad idea to get one before starting an antipsychotic though it's not mandatory by any professional standard such as an AMA/APA.

The most stringent guidelines I've seen for monitoring are the ADA/APA guidelines for antipsychotics but they don't include an EKG.

If one were to make a mandatory EKG prior to admission for a psych patient (I'm not saying it's a good idea), I'd also recommend that later EKGs shouldn't necessarily be done unless there's specific reason to do so. As we all know there are frequent-flyer patients. IF an EKG was done every single time it's cost-benefit would likely be not worth it. I often tell residents that if a patient is a frequent flyer, there's no need to do a TSH every single time. The TSH from the last visit 2 weeks ago should be enough. Other labs, however, such as a UDS should always be done every single time.

 

[Jester25]

I think people above me have all provided good reasons for why you should get an EKG before prescribing Geodon. If you see a patient for the first time and they are already on an antipsychotic, there is a 99% chance that someone got an EKG somewhere. If that EKG showed a prolonged QT and you didn't get an EKG before changing medications to Geodon, that's a lawsuit if there is a bad outcome. The warnings on clear on the package insert. Also, it's not only for determining QTc, but also discovering any other potential cardiac abnormalities which would increase the risk of the medication.

I worked at a major academic hospital. We were next to another major academic hospital. An ER physician gave IM geodon to a patient without any labs or EKG. Patient ended up in cardiac arrest and dying. They no longer use Geodon in that ED.

I am a child psychiatrist. Our practice parameters recommend a baseline EKG for patients about to be placed on Geodon. If you had a bad outcome and you tried to defend yourself, all the plaintiff would have to show is what a reasonable psychiatrist would do in that position (standard of care). 9/10 times you will lose as getting a baseline EKG has not only become the standard in outpatient clinics, but also at most hospitals around the country.(whopper can correct me on that).

The risk/benefit is way too low not to get an EKG, both for the patient and to protect yourself.

 

[MacDonaldTriad]

There are risks for getting an EKG and looking at the QTc and not having it read by someone who is good at reading EKGs. I understand an ER not using IM Geodon after a bad outcome, but QT problems are a class effect and getting EKGs for Geodon, but not for Mellaril doesn’t make sense. Don’t you love corrective action plans. We had a nurse get stabbed with a pencil a while back. This produced a policy that forbids #4 pencils in the ER. Pens and #3 or #5 pencils are OK.

 

[Jester25]

Haha, if I used Mellaril, I would get an EKG before that too. It's worse.

We have cardiologists read our EKGs.

 

[Doctor Bagel]

My residency hospital had very strict EKG monitoring rules (like 3 daily ekgs) that only applied to Haldol which was silly because haldol isn't even the worst Qtc offender.

The cardiologist read does bring up a good point. Getting an EKG, only looking at the QTc and then missing some other major finding could expose you to risk too.

 

[Psychotic]

The cardiologist read does bring up a good point. Getting an EKG, only looking at the QTc and then missing some other major finding could expose you to risk too.

My thoughts exactly...

 

[Jester25]

I'm certainly not advocating for getting EKGs if you can't read it or don't have someone look over it that can read it. That doesn't make any sense.

 

[Doctor Bagel]

I'm certainly not advocating for getting EKGs if you can't read it or don't have someone look over it that can read it. That doesn't make any sense.

I'm expecting to be shamed for this, but I can't read an EKG with any level of sophistication anymore, especially certainly not with the time constraints that I have when doing inpatient work. Getting it in the ED makes sense because I think ED docs are pretty good (or should be) at looking at/interpreting EKGs. Getting it on the psych unit, which happens, is another story. In some systems, it's not clear a cardiologist will be looking at them, at least not anytime soon.

Editing to add that while I'm anticipating being shamed, I actually don't feel shame about this.

 

[hamstergang]

I think people above me have all provided good reasons for why you should get an EKG before prescribing Geodon. If you see a patient for the first time and they are already on an antipsychotic, there is a 99% chance that someone got an EKG somewhere. If that EKG showed a prolonged QT and you didn't get an EKG before changing medications to Geodon, that's a lawsuit if there is a bad outcome. The warnings on clear on the package insert. Also, it's not only for determining QTc, but also discovering any other potential cardiac abnormalities which would increase the risk of the medication.

I started this thread by saying that I recognize many sources and doctors say that an EKG should be obtained prior to starting Geodon. What I was looking for, and only Doctor Bagel provided to some degree, was evidence that that is correct. The package insert warns that Geodon use was associated with some seizures, but I don't hear anyone calling for baseline EEGs. That's because we all recognize that it wouldn't really be a useful or cost effective test. So just because Geodon is associated with cardiac issues doesn't mean an EKG is warranted -- there needs to be some evidence that it will actually help us.

It's worth noting again that while the package insert warns about QTc issues, it does not recommend a screening EKG.

I am a child psychiatrist. Our practice parameters recommend a baseline EKG for patients about to be placed on Geodon.

Yes, despite no evidence that Geodon is associated with more cardiovascular events than other antipsychotics, the practice parameters for C/A recommends a screening EKG for Geodon but not other atypical antipsychotics (absent personal or family cardiac history). But if we trust whopper:

it's not mandatory by any professional standard such as an AMA/APA.

The most stringent guidelines I've seen for monitoring are the ADA/APA guidelines for antipsychotics but they don't include an EKG.

screening EKGs are not recommended for healthy adults.

 

[Jester25]

I was referencing whopper for liability issues regarding no EKGs/electrolytes prior to Geodon with a poor outcome. He's forensic trained.

I wasnt entirely clear on your question until your last email. That's a valid question. Geodon is relatively new, there haven't been enough studies done. Time will tell one way or another. I would just practice standard of care for now.