Transcript edited for clarity.
After the webinar, clinicians should understand how ECT works in the following ways:
1. Readily identify a patient who is appropriate for ECT treatment
2. Differentiate techniques to decrease the cognitive side effects of ECT
3. Identify options for maintaining ECT response
Hello, and welcome to the Carlat Psychiatry webinar, where we're going to talk about electroconvulsive therapy (ECT) in clinical practice.
Thank you very much for having me here today. It's a pleasure and a privilege to be here talking about something that's important to me, which is ECT or electroconvulsive therapy. My name is Steve Seiner. I'm the Director of Psychiatric Neurotherapeutics at McLean Hospital and our program includes ECT, TMS, and ketamine/esketamine. We perform over 10,000 ECT treatments each year, which makes us one of the largest ECT services in the county. I have no disclosures.
I'm going to try and keep the talk today very clinical and very evidence based. I'll start by talking about indications and the efficacy of ECT to help you decide which patients to refer for ECT. Then we're going to talk a little bit about preparing pre-ECT, the treatment itself, and about some of the challenges of ECT, including memory loss and, how we keep people better once we get them better.
After the webinar, I'm hoping that we'll be able to readily identify a patient who's appropriate for ECT and differentiate between techniques that we use to try to decrease the cognitive side effects and increase the efficacy of ECT and identify options for maintaining the ECT response.
Case example 1. All my case examples are real patients. This gentleman was in his 70s. He had no history of psychiatric illness until he retired after which he became severely depressed. This progressed over the next several years and despite trials of multiple antidepressants and antipsychotics, he became increasingly depressed.
And over the last two years before he came to us, he became anhedonic and anergic. He had lost 57 lb and became quite paranoid. He also was unable to care for himself and would even soil himself rather than going to the bathroom, which his wife was very upset about. This is a great example of a patient with psychotic depression.
He responded very well to ECT. It took a while to fully get him better because he had been depressed for so long. but we were eventually able to get them into remission and keep him in remission and taper them off the ECT. And he did very, very well. I use this example right away because, I think in case of psychotic depression or catatonia, these are instances where you might think about ECT first line.
Now you don't necessarily have to do it first time, but it should pop into your head when you see a patient with true psychotic depression or catatonia. And there are a couple of reasons. The first is that, as you can tell by this example, these illnesses are awful, and this patient lost several years of his life as did his family trying to care for him. That’s a miserable place to be. And, secondly, these illnesses are very difficult to treat. As you can tell, he had multiple medication trials without response. Catatonia, typically, will often respond to benzodiazepines or Ativan but when it doesn't, there are not a lot of great options for treatment. You don't have a lot of choices other than ECT
Then finally, a third reason, is ECT almost always works. The response rates for psychotic depression and catatonia are between 80% and 95%. One of the wonderful things about ECT is that generally when you need it to work the most, when people are at their worst, the most acutely ill, is when ECT actually tends to work the best. We don't have a lot of times in psychiatry when we have 80-90% responses let alone for such severe potentially life-threatening illnesses.
Another place where we use ECT is for major depression. And I'm talking about unipolar treatment-resistant major depression. This is our bread and butter. We do a lot of this. No trial of any medication has ever been found to be superior in efficacy to ECT for major depression.
The response rates are somewhere between 40% and 90%. It's more dependent on who you're treating rather than how you're doing the treatments. By that I mean ECT works best when you have patients who've had an acute episode of depression. They were doing well and now they're not—that works better than patients who have had 20 years of chronic depression or decide they’ve tried everything else and now we're looking at ECT. The reason for that is if you think about what ECT is doing in the brain, it's kind of like hitting a reset button. It's bringing people back to their baseline. If at baseline you're a happy-go-lucky ne’er do well, well, that's what we're aiming for.
ECT doesn't change people's personalities, depression changes people's personality. We're bringing people back. One of the most important questions I have when I'm doing a consultation is, “What is your baseline and how long has it been since you've been there?”
When you hear somebody who says, “This is not me. I'm not like this. I'm way off my baseline,” that usually, portends a good response to ECT.
Case example 2. A woman in her 30s with bipolar-II is a high-functioning health care professional was admitted for suicidal thinking, guilt, poor energy, weight loss, and poor concentration and functioning. She currently was taking venlafaxine and ariprazole for years, had not had a lot of medication trials. There was a question about lithium or lamotrigine before trying ECT and as a patient, you clearly could try other medication options for her.
However, given the severity of her illness at the time, and she was quite anguished and suicidal, we opted for ECT. This was one of those great patients that responded very, very quickly to very gentle ECT and got better. She did well for a year and then paged me on one night and said, “I’ve got to get back in. It's coming back. I can feel it. I've got to get back in.” She was very proactive and we treated her again. She was able to improve without missing much work.
Again, this is an example of bipolar depression. There is pretty good evidence, for ECT and bipolar depression. We tend to treat a lot of bipolar depression in part, because it's so hard to treat otherwise. You have to be very careful with antidepressants and these depressions can be quite debilitating and go on, sometimes for years.
We do get a lot of referrals for bipolar depression. In my experience, about a third respond like the patient I presented who goes into remission very quickly. An even bigger group gets better, maybe even more than 50% better. But they don't quite get into remission and it takes a while to get them fully back into remission, sometimes months.
If you stick with those patients, we find usually they do get better. They usually get back into remission, but it can be a bit of a long slog. I don't know why that is. I don't know if it's because they've tended to be depressed for so long before they get to us or I think sometimes when patients come off a manic episode into a severe depression that has settled in, it feels like you're fighting the natural cycle of the illness.
And, like I said, you can speed things up and get people out of bed and back to work and no longer suicidal, but sometimes it takes a while to fully become their old selves. But usually they do.
Mixed or manic episodes with bipolar disorder. We don't treat a lot of mania with ECT and that's because we've got a lot of good treatments for mania. Usually, it's for the more dramatic cases and ECT does work for many; it works pretty well, actually. It’s usually reserved for those patients who are in restraints or court ordered or refusing medications or delirious/catatonic, the severe bipolar cases and those situations—in these cases, ECT tends to work very, very well.
We don't do a lot of ECT for schizophrenia, in part, because schizophrenia is a chronic illness and ECT is an acute treatment, but there are some situations where it can be very useful. The first is, again, going back to the idea that ECT is a reset button in the brain and bringing people back to their baseline, sometimes you'll see a patient who was living in a clubhouse or a group home and they're doing relatively okay. And then something happens—the clubhouse closes, the group home closes—or they stop taking their clozapine. Well, they ended up in the hospital, you put them back on the Clozaril but they don't get back to their baseline.
Patients still remain quite symptomatic, and in those situations, ECT can be very helpful to reset things. There was a nice study that came out by George Petrides in the American Journal of Psychiatry a few years ago that showed that for patients with treatment-resistant schizophrenia, adding ECT can make a difference and can make a meaningful, positive change.
You do have to do some maintenance treatments for a while because this is a chronic illness. We treated several patients from Tewksbury State Hospital, on a contract and many years ago. And we would see patients get significantly better, well enough that they could usually leave the hospital so it can have a meaningful effect on schizophrenia.
Preparation. I'm going to move on to preparing for ECT. Once we've decided that we're going to do ECT, what do we need to do to get a patient ready? First off, we have to make sure that they're medically clear. They’ll need an ECT clearance by an internal medicine doc familiar with ECT. And that's also a Joint Commission requirement, and an EKG.
We do lab work, and one of the main reasons we do that is a lot of times these patients are so sick, that a lot of these lab results will come back off. I recently worked with somebody who came back whose sodium was 159 and they had acute renal failure just because they had stopped eating and drinking. You want to correct those things before you go ahead and do ECT.
There are no absolute contraindications to ECT, but there are some situations or conditions that cause increased risk. When you think about what ECT does, it's basically increasing the heart rate. We're causing a little bit of a stress test, so if somebody has severe coronary artery disease or ventricular arrhythmias, that can be an issue and the same thing with cerebrovascular disease as well. Obviously, if they're at risk for stroke, we also increase the intercranial pressure in the brain transiently.
If somebody had a big mass in the brain, theoretically, they could herniate. Severe reflux could be a concern if patients aren't intubated because it can cause aspiration, and severe pulmonary disease is an airway issue. And of course, pregnancy, again, not a contraindication ECT, but you'd want to be doing it in a general hospital with the right type of backup care available to you.
I think the point is that these are all things that we're looking for. We can treat most patients despite these issues, but it can change the risk-benefit ratio and you want to treat them in an environment where you can manage these things. For instance, if somebody has severe arrhythmias or coronary disease, you might want to treat them in a general hospital when you control blood pressure and put in a line—those sorts of things you wouldn't necessarily do at a private hospital.
In terms of the medications for ECT, obviously we're trying to induce a seizure, so we don't want anything that will make that harder for us on board. Anticonvulsants and benzodiazepines can get in our way. Having said that, please do not taper patients off their benzodiazepines in the two weeks right before they come into ECT if they've been on them for 5-10 years, because we will sometimes get people who show up for the first treatment in full blown benzo withdrawal.
So don't do that. We can work around these things if we need to. For instance, with benzodiazepines, we can hold the evening dose, give it divided doses, and hold the evening dose.
If we must give it, we can also sometimes reverse benzos with flumazenil. It's not perfect, but it can help. With anticonvulsants, again, if they're not helping, it's good to get rid of them. But, for example, somebody that we're treating for mania, you don't want to take them off their Depakote because that's only going to make the mania worse. In those situations, we'll often give divided doses and hold the night before. We'll work with you. You don't have to figure all this out, but just to give you an idea of why we're doing these things. Also, when a patient has a seizure disorder, we leave them on the anticonvulsants and we just have to shock through it because you don't want those patients to have a seizure at home or on the unit, or, heaven forbid when they're driving to the supermarket. You don't want them to miss doses of their anticonvulsant.
Lithium is a little more complicated. Lithium mixed with ECT, can cause some confusion or encephalopathy. We don't fully understand why that is. Some people think it's because ECT disrupts the blood-brain barrier temporarily, and so the brain sees more lithium. than is in the blood. So, they look lithium toxic, even though their level may not be technically. So, for lithium it's a little more complicated because we all know that we'd like to take people off lithium if it's not doing anything.
On the other hand, there are some patients that lithium is very, very important to the regimen and, and bad things can happen when you taper it off. That's something we'll work with you on. We can, again, give it in divided doses, holding it the night before. That's on a case-by-case basis and certainly we will help with that. Most antidepressants and most antipsychotics are fine. If you refer a patient for ECT, we will go through this with you, but it's helpful to know why we're doing some of the things we're doing.
Treatment. When we do ECT, we use a couple of different electrode placements. The three most common are bilateral, unilateral, and bifrontal. There are some other choices, but these are the most common.
Bilateral is the original type of ECT. It has the advantage of being quicker, and sometimes gets people better more quickly, and has a little bit better efficacy on average than unilateral. The disadvantage is we see more cognitive effect. And I'll talk more about that in a minute. In terms of unilateral, it doesn't work quite as fast as bilateral, but it does have the advantage of not having as much cognitive effect and there, and there's a lot of different things you can do with unilateral, which I'll talk about again in a minute. Most of the time we start with unilateral, probably 90% of the time. The exceptions being when somebody is catatonic or, in a delirious mania in restraints, where you need to get people better, quickly. They're unstable so we choose bilateral because it works more quickly.
Once those patients are stable, you then have the option of switching to something like unilateral or bifrontal.
Bifrontal is kind of in-between unilateral and bilateral; there's an overlap between the two. We use it primarily when the unilateral doesn't seem to be working and the patient does not want bilateral or doesn't tolerate it.
How do we know what dose to use? How do we know how much power to use? Well, on the first treatment, we start at very, very low settings and work our way up until we get a seizure. Then we know that's the seizure threshold. And we know from the literature that for unilateral, you need to be further above this threshold, than you do with bilateral. You need to be at least four to six times seizure threshold, and dose does matter. You also do start to see a more side effect as you get to high doses. With bilateral, we tend to use lower doses—one and a half to two and a half times seizure threshold.
Bifrontal is a little more complicated. It's usually dosed more like bilateral. We do treatments three times a week. If we start to run into memory issues, we'll drop it to twice a week, and a typical acute course is 8-15 treatments but then usually followed by a taper or maintenance as I will talk about in a minute.
Side effects. I've talked about all the good things about ECT, the wonderful lifesaving effects it has. I do have to talk about the challenges that we have and the single most important one is memory loss. I think memory loss is greatly misunderstood with ECT, and I think there's a lot of confusion about it, but some of that's because it varies so greatly between patients.
Even though there are things you can do to minimize memory loss, like electrode placement, the frequency of treatments, the number of treatments, the power—all these things can influence memory. There is really an individual effect. Some patients are just very, very sensitive to the memory effects of ECT, just like with medications.
Then there are other patients you can do bilateral on every day and they'll beat you at Sudoku in the recovery lounge. They just tolerate it well. It makes it confusing for both clinicians and patients, to see that wide range. Typically, the memory loss is manageable. There are things we can do to make it less of a problem. It usually resolves once the ECT is finished.
Some patients may have gaps left over in the memory, most commonly for the course of ECT and for the period just preceding it. That makes sense because a lot of those memories right before the ECT were still being stored.
By definition, that period right before ECT is when people are the most depressed and when you're the most depressed you're not making great memories. you're not storing memories because you're distracted. I think those two things together make that period the most vulnerable to some gaps in memory.
We tell patients they should expect some short-term memory loss, and it's not a great time to make major life decisions, go on a job interview, go on a first date, do your estate planning. People still do, but we tell them not to.
What about long-term effects on memory? I think most people are willing to say they can live with it a few weeks, not being at the top of their game or being forgetful or losing a few memories. But what about longer term? A number of studies have looked at this, and to understand this, first you have to understand there are two types of memory loss that we see with ECT.
The first is anterograde memory. This is difficulty storing new memories. The good news about anterograde memory is that it's easy to study. We have good neuropsychiatric testing, which is objective. It is effort based, so it's not perfect, but it is very well validated. There have been more than 80 studies looking at anterograde memory and objective neuropsychiatric testing with ECT.
This was looked at in a famous meta-analysis in 2010 in Biological Psychiatry. What they found is that most of the anterograde memory issues started to improve after three days and by 15-30 days, you start to see processing speed, working memory, anterograde memory, and some aspects of executive function improved beyond the pre-ECT level.
That seems strange, right? It's not that we're making patients smarter. It's just that prior to ECT, they were very depressed, and I think we just underestimate how much depression affects memory and cognition. It can be devastating, especially in the patients we treat. When they’re no longer depressed, and you wait a couple of weeks for the ECT to clear, and now they're going to test better than they did before. So, there's no evidence of any long-term effect of ECT on anterograde memory.
What about retrograde memory or the ability to recall past memories? This is a little bit more of a problem, and this is what patients tend to complain about more. Now, this is much more complicated because we don't have great tests for this, right? There are some. Autobiographical memory scales measure this, but they weren't set up for ECT. We do use them. They have been used in studies, but they're not perfect, and they're a bit controversial.
Additionally, there are a lot of things that affect retrograde memory, including age which most of us know. Also, things like the time since you made the memory, whether you were depressed when you made the memory, and when somebody says to me, “I can't remember a trip I took to Aruba in 2007,” and I have no idea if that has anything to do with ECT or not.
Also—and I think we've all had this experience—where their spouse may say to them, “Oh yeah, don't you remember? We went parasailing and then they say oh yeah, we stayed at the Hyatt and ate at that Mexican restaurant.” Whether the memory gone or they're just having trouble accessing, at that moment, it is very difficult to test. Using the tests that we do have with their limitations, it does appear that some patients are left with some sustained gaps in their memory—not every patient, but some, and the highest risk, as I mentioned before—the few months right before ECT, and during the ECT, of course, but the few months before when those memories are not being well made because patients are depressed.
Those memories are still being stored when we do the ECT, but sometimes it can go further back. Sometimes some patients will complain of more remote memory loss. It's less common nowadays, but occasionally you will see it. Usually, it's something that somebody did once and then never think about again. You're not going forget who your daughter is, but you might forget a piano recital you went to six months ago. We don't always understand it, but it can happen. And we do try very hard to prevent that.
There is a large naturalistic study that looked at this and found using these scales that bilateral ECT was associated with more of this autobiographical memory loss.
That is why we try to use primarily unilaterally ECT, as I mentioned. For most patients, it's a tolerable amount of memory loss. It's usually more annoying than it is significant. Sometimes patients find it even a little humorous, but there are some patients were very, very bothered by it.
We try to minimize and to mitigate memory loss as much as we can. How do we do that? I mentioned using unilateral as opposed to bilateral when we can, but another change that's happened in the last 15 years with ECT is called ultrabrief pulse, and this has been a major advance.
When we induce a seizure, when we give an electrical stimulus during ECT, it's not like cardioversion where we're getting this shock and we're saying I'm clear, you're clear. It's nothing like that. We’re giving these tiny pulses of stimulation over about eight seconds.
When we used to do this, the pulse with these little pulses were about a millisecond. Sometimes even further back, it was two milliseconds and people at Columbia said. The optimal amount of time to depolarize a neuron in the brain is between 0.1 and 0.2 milliseconds while we are using these wide pulse widths and just hitting neurons and continuing to stimulate them even while they're in a refractory period.
And so that led to people trying what's called ultrabrief pulse, which is reducing these pulse widths down to 0.3 milliseconds, which is just over the optimal depolarization time. And what they found with this is that the ultrabrief, (and this was a famous study that was done in 2008 and led to us, everybody retrofitting their machines in order to do this), that ultrabrief pulse unilateral treatments worked as well as standard brief pulse with less retrograde memory loss - specifically, the retrograde stuff that people worry about. We have an extensive experience. We started doing this back in 2008. About 80 to 90% of the time, we are starting with ultrabrief pulse unilateral. So we literally treated thousands of patients with it. And in our experiences, I think that it is a milder form of ECT.
It sometimes takes longer, and it doesn't work for everybody. But probably 50, 60% of the time we can get away with just ultrabrief pulse unilateral. And that is significant. There's no question it causes fewer cognitive side effects and patients are going to be much clearer and tolerate it much better. It is a nice option for patients, particularly those who are worried about their memory.
The other challenge we have with ECT is once we get people better, how do we keep them better? One option is to treat their depression with medications.
There was a famous study that came out in JAMA in 2001. They looked at this and they took patients who had, severe treatment-resistant depression, or psychotic depression. They took the patients who responded to ECT, and then they followed them for roughly six months after ECT.
What they found is if you gave them just placebo, none of these patients got maintenance ECT, just medication. ECT was stopped and they were put on placebo. You had an 84% response or relapse rate. That seems awful, and it is awful because of the very sick patients. But again, what that implies is that 16% of patients were able to get by without any treatment after ECT and didn't relapse, which is for six months and actually probably a little higher than I would have thought. Those patients who just got nortriptyline had a 60% relapse rate, which is still high. However, those patients who got nortriptyline plus Lithium, the relapse rate dropped to 39%.
So that's without any maintenance ECT, more than half stayed well, without relapse. So that led to most studies in psychiatry, with ECT, any maintenance treatment, usually lithium was a part of it. And you'll see that in studies. We don't always use lithium.
It is a good option. A lot of patients have already been on lithium but it has a lot of side effects and so it can be difficult to use. But there is some data to support it post-ECT.
Maintenance. What about maintenance ECT? We break this down into two different types.
The first is continuation ECT. The idea here is that sometimes they've been depressed for months or even years, so to get them better and then just stop, you're going to run into a high relapse rate. What we will often do is taper the ECT, and so continuation ECT means you're tapering it usually over a period of a few months. We start out with weekly and go to every other week and then out to once a month. Usually if patients are doing well, we can stop. The idea of that is to give them patients time to put their life back together, find the right medications, engage in therapy, get back to the gym—all those things they haven't done in a very long time—and to build up some resiliency, before we stop ECT.
The analogy I use with patients is in the first month or two after ECT, you're walking on a tightrope and it's very easy to relapse over time. If you can stay well, that becomes a balance beam and then eventually a sidewalk. And when you hit the sidewalk, you don't need us anymore.
Maintenance ECT is for those patients who we try and taper off the ECT and they just relapse every time you taper them off sometimes right away, sometimes a month or two later. Those patients receive treatment every four to six weeks on an ongoing basis. That doesn’t mean forever, it just means until they reach a period of stability where we can try to taper the ECT off. Does this work? That's the question. This was looked at in the PRIDE study, which came out in 2016, where they took elderly patients who had responded to ECT and they had a 61.7% remission rate, which is very good.
They were using ultrabrief pulse, so it's a very good remission rate. They divided those patients into one group that got venlafaxine plus lithium. They replaced the nortriptyline with venlafaxine, which makes sense. It's better tolerated in older patients than nortriptyline. One group got back venlafaxine plus lithium; the other got venlafaxine plus lithium and flexible dose ECT continuation. They got a taper with ECT. What they found in this study only a 16.7% relapse overall. They had a very good rate of relapse.
The odds of relapsing were almost twice as high if you didn’t get the continuation ECT. It was 1.7 times higher without the ECT, but more importantly, the odds of being rated not ill at all were five times greater if the maintenance ECT or continuation ECT was added. It makes a big difference in cutting down on relapse, but also in improving the quality of life if you're able to taper the ECT, in addition to finding the right medication.
Maintenance ECT: Take home points
In terms of maintenance, pharmacologic treatment is required for most patients, to prevent relapse. If a medication failed prior to ECT, I would consider it switched to try an antidepressant or augmentation If it was just a breakthrough -In other words, if somebody who had done well for 10 years on Paxil and then lost their spouse and had a depression. If you get them better with ECT, you might put them back on the Paxil.
You might increase the dose for augmentation to try and give a little more resiliency to it. Continuation ECT can help prolong its benefit and help prevent relapse. A small percentage of patients may need longer term maintenance treatment, especially when nothing else seems to help them.
As a brief aside, I highly recommended psychotherapy post-ECT. A lot of these patients have lost so much because of the depression and if you don't support them as they try and put their life back together, they're more likely to relapse.
In conclusion, ECT can be lifesaving. Consider it first-line or at least think about it for psychotic depression and catatonia, or other times when you need a fast response. There have been advances in ECT that have made it safer, more comfortable, and with a better cognitive profile that allowed for its use other than just as a last resort.
Then of course, right from the beginning, you need to think about once we get them better, how are you going to keep somebody better? Thank you very much.
Earn CME for watching the webinars with a Webinar CME Subscription.
APA Task Force. The Practice of Electroconvulsive Therapy. Recommendations for Treatment, Training, and Privileging. Second Edition. American Psychiatric Association; 2001
Tew JD et al, Am J Psychiatry1999;156(12):1865-1870
Gormley N et al, Int J Geriatr Psychiatry 1998;13(12:871-874
Manly D et al, Am Int J Geriatr Psychiatr 2000;8(3):232-236
O’Connor M et al, Am J Geriatr Psychiatry 2001;9(4):382-390
Dierckx B et al, Bipolar Disord 2012;14(2):146-50
Ciapparelli A et al, J Clin Psychiatry, 2001;62(7):552-555
Lisanby S et al, N Engl J Med, 2007;357(19):1939-1945
Sackeim H et al, Neuropsychopharmacology, 2007;32(1):244-254
Sackeim H et al, Brain Stimul 2008;1(2):71-83
Sackeim H et al, JAMA 2001;285(10):1299-1307
Kellner C et al, Am J Psychiatry 2016;173(11):1110-1118