番茄社区

In this Instagram Live clip, Jeremy Faust, MD, editor-in-chief of 番茄社区, and Ziyad Al-Aly, MD, chief of the Research and Education Service at the VA St. Louis Health Care System in Missouri, discuss the latest research on long COVID treatments and how physicians should weigh the risks and benefits for their patients.

The following is a transcript of their remarks:

Faust: A lot of the questions from the audience have to do with treatments with risk factors. Let's talk about risk factors.

Obviously there seem to be some risk factors for developing long COVID, but again, as you said before, it's more than one disease. The way I look at the literature is that the biggest risk factor for long COVID is how sick you were at first. For really sick people early on, especially pre-vaccine and pre-immunity, risks of having symptoms 3, 6, or 9 months later were a lot higher.

Beyond that, what do we know about risk factors?

Al-Aly: Sure. For that, I really think the devil is always in the details. I like to classify long COVID into maybe two types.

The prototypical long COVID -- the fatigue, brain fog, post-exertional malaise parts. Those tend to happen more in females and in younger people [rather] than older adults. That's sort of the prototypical type of long COVID.

And there are the other types that involve kidney dysfunction, new-onset diabetes, heart disease, strokes. Those are actually happening in older adults in the post-acute phase of the disease and in people who have existing comorbidities or existing risk factors for those diseases, for heart attacks, etc.

But having said that, I think the one unifying thread for all of this, and actually, almost all the literature that you read is exactly what you just told the audience: that severity of acute disease really, really matters. That the more severe the disease in the acute phase, if it required hospitalization or, even worse, required ICU stay, those people tend to have the highest risks of postviral problems.

That does not mean at all -- I don't want the audience to take it this way -- that doesn't mean that mild disease cannot lead to long COVID. As a matter of fact, more than 90% of people with long COVID in the world have had mild disease, just because most people in the world had actually mild disease, most people in humanity.

Faust: Yeah. It's a base rate.

Al-Aly: Exactly.

Faust: One thing that I struggle with is -- I'm like middle age, right? I'm not young. I don't think I'm old, but --

Al-Aly: You're young!

Faust: But like people come to me in their 30s and they say, "Oh, this long COVID is a vascular disease, right?" And "I just got COVID, should I be on blood thinners because I'm a 35-year-old who just got COVID? Because that paper from the VA said they got clots." And I'm like, "No, no, no, no. I promise you, you don't need to be on blood thinners." Even though people in that risk cohort do need to be on blood thinners for a number of reasons.

I mean, do you worry that people read a VA paper and their eyes bug out and they're prescribing things to 25-year-olds who could get long COVID, and I think we should be giving them treatments if we can find them, but they might be over-treating them?

Al-Aly: Yeah. I mean, this is very, very important. It's always risk versus benefit, right? I see people on triple anticoagulation for no indication, really. I mean, they are really buying all the risk and there are really no benefits.

I think what complicates long COVID treatment is that we don't really have enough trials to help us understand what works or what doesn't and evaluate the risks/benefits. So I'm very familiar with the push to use anticoagulants in this setting because of presumption there's emphysematous, that they're more prone to clotting -- which is true that their definitely is macro-clotting and micro-clotting in the setting of long COVID -- but I think we need to be careful about treatment.

Especially with those [treatments] that could be associated with significant side effects like anticoagulants that are absolutely not free of side effects, and some side effects could be catastrophic. People could have hemorrhagic strokes and even die from it. I mean, these are serious side effects.

So I think we need to be careful about treatment and not generalize from one study to extrapolate that we need to all be on anticoagulants because we got SARS-CoV-2.

Faust: I do think that people do underestimate the risks of some things -- which is, by the way, why I thought metformin was such a positive development. This is a medication that is very well-tolerated, has very low risks. I can count on one hand, one finger, the number of metformin-associated lactic acidosis cases I've seen in my life, and really that was just an overdose.

Versus patients who I see who are on blood thinners because they're kind of borderline for indications for other reasons. They come in [because] they fell down the stairs, and now they've got a subdural hematoma. But I don't think the public always necessarily understands the severity of risk in some of these drugs.

Another thing that comes up a lot is the unknown. The big question is Paxlovid [nirmatrelvir/ritonavir], right? I mean, I think that Paxlovid is an important medication for high-risk people in the acute phase. For long COVID, though, I don't think it's played out yet. I don't know yet. I've done some research that shows maybe a signal, but I'm pretty skeptical, even though I'm involved with that. We don't really know whether it's causing rebound or whether that's background.

Where do you stand right now on Paxlovid research with long COVID?

Al-Aly: So there are two things. I think Paxlovid in the acute phase in people who qualify for Paxlovid -- basically they have to have a risk factor for progression to severe COVID-19 illness -- I think taking it is definitely better than not taking it, so better than placebo. It reduces the risk of hospitalization and of death in the acute phase and might reduce the risk of long COVID. That's when you take it in the acute phase.

For Paxlovid used in the post-acute, let's say somebody already got COVID and maybe has long COVID, there is no evidence, as you know, there are no trials in that space. The presumption there is that there is viral persistence and the virus is replicating for Paxlovid to work.

How does Paxlovid work? For the audience, it actually blocks viral replication. So if there is no virus that is persisting and it's actually not replicating, Paxlovid is unlikely to have any material effect or material benefit. So the presumption there is that in those cases there is virus that is persisting and is actually replicating.

There's not a whole lot of empirical evidence for that. That doesn't mean that a remnant of the virus, like RNA or a protein, cannot be persistent or persistently present in "immune-privileged sites" that are provoking chronic inflammation and subsequent sequela. But Paxlovid is not going to do anything for remnant RNA. If it's not replicating -- Paxlovid blocks replication, it's not going to do anything for RNA just sitting in a lymph node that is doing nothing.

Faust: It's like a bee sting, right? The way I think about it is like if you have a bee sting and you're allergic to bee stings, you're going to keep having that anaphylaxis until the bee sting is removed. It's not replicating, it's just the presence of it is what's causing the immune system to go crazy. And so, by extension, something is left over that's not replicating, that's the hypothesis, and so an antiviral would be kind of irrelevant.

Al-Aly: Correct. But that needs to be tested. We would need to test it. We need to see the pro and con arguments, for and against it, that need to be tested. We would need data to help us evaluate whether this thinking is correct or not.

The only way to solve this is we need those trials to come out, and more than one of them obviously, to help elevate our confidence that this is really the right answer, that Paxlovid really works or not.

For it to work, again, the virus has to be replicating, has to be present and replicating for months after the acute illness, right? And there's very, very little evidence to convince me of that. But I'm a scientist; I'm learning all the time.

Faust: And so we've got good news on metformin on the prevention side. The Stanford Paxlovid study was stopped, we don't exactly know why. We also know that at Yale they're doing this decentralized study of Paxlovid. What else are you looking at in 2024 for studies coming down? Where's your lens for like, "I can't wait for this study." Or what data are you waiting for right now?

Al-Aly: There are a lot of people who are interested in more antivirals other than Paxlovid. Again, those will only likely work if that viral persistence hypothesis is actually true and the virus is replicating.

People are actually trying low-dose naltrexone to try to see if those may ameliorate symptoms.

There was some interest in also thinking about SSRIs [selective serotonin reuptake inhibitors]. There was a paper not long ago that suggested that intestinal absorption of tryptophan, the precursor for serotonin, is impaired and [there might be a] subsequently low level of serotonin in the blood. That might impair serotonin vagal nerve signaling and [cause] impaired cognition, sort of explaining some of the brain fog and maybe the autonomic dysfunction that can happen in people with long COVID. So whether serotonin replacements like SSRIs would be helpful, I don't really know. But that also is being thought about as a potential therapeutic avenue.

I'm aware of at least a couple applications sort of thinking about low-dose anticoagulants. Again, the risks and benefits need to be evaluated, and the only way to know is to do the science, to evaluate the benefit and evaluate the risk and determine if very careful low-dose anticoagulants really might ameliorate symptoms or symptomatology.

But I have to say, I'm glad you asked this question. In my view, the world of trials on long COVID is really thin. I mean, I pretty much just reviewed everything that's going on in the U.S. There are some microbiome studies in Hong Kong, but there need to be more. There need to be more. And most of these trials are also small and may not yield really definitive answers.

I think that calls for, hopefully, more energy and more enthusiasm and more resources in this space to do trials that are commensurate with the problem. Long COVID, I think, is a big problem. We need to solve it, and hopefully we can solve it earlier than later. The only way to really discover treatment is to do trials, and the scale of those trials need to match the magnitude of the problem.

Comment