Yale Psychiatry Grand Rounds: March 18, 2022

March 18, 2022

Information

"The Mystery of Central Nervous System Complications of COVID-19: What Do We Know and What Do We Need to Learn?"

Serena Spudich, MD, Gilbert H. Glaser Professor of Neurology, Yale School of Medicine

ID7576

To CiteDCA Citation Guide

00:00Nelly, and for that really wonderful
00:03introduction and for a persistence
00:05in inviting me to to give this talk
00:07for the Department of Psychiatry.
00:09And this is a great honor, you know,
00:11I've been at Yale for 11 years and I
00:13over just the last year or two through
00:16interactions with Chris Pittenger
00:17and more interactions with other
00:19colleagues and psychiatry really,
00:20really realized what a missed opportunity
00:22that's being all this time for me to
00:24not interact more with your department.
00:25So again, I hope this is just the
00:28beginning of a long standing set
00:29of intersections that we can have,
00:31and I think.
00:32In particular,
00:33my research interests and my clinical
00:35interests are really intersecting
00:37with psychiatric disease, and again,
00:39the brain and the mind,
00:41both in terms of cognition but also in
00:43terms of mood disorders is something
00:45that's really relevant to my work.
00:47So as you heard from Professor O'Malley,
00:50I've been working for a long time
00:52studying the effects of HIV in the brain.
00:53And then of course, 2020,
00:56because we had this massive onslaught
00:58of concerns about the brain and COVID,
01:01this became an area that I've also.
01:03Started to investigate one thing.
01:06That's a really important part
01:08of my sort of portfolio is I'm a
01:10clinician and I see patients and so
01:12I'm going to talk about the issues
01:15of CNS effects of COVID-19 really,
01:19starting from a clinical perspective,
01:21and I think that I'm going to try to
01:23kind of go through what we've understood
01:25what we've learned over the last two years,
01:27and really,
01:28all of that begins with clinical
01:30observations and and, you know,
01:32people worldwide reports.
01:33Worldwide and then some of our
01:35own experience here at Yale.
01:36So I'm gonna I'm gonna anchor the talk
01:39and talking about the CNS sorts of
01:42clinical reports and acute COVID-19.
01:43I'm going to then talk about what's
01:45being learned over the last two
01:47years about the neuropathic genesis
01:48of that acute COVID infection.
01:49So really,
01:50what's happening in the brain and
01:52people during acute COVID-19?
01:54Primarily people who were pretty
01:56severely ill is where most of
01:58our information has come from,
02:00and this is going to be sort of the
02:02jumping off point for the rest of my talk,
02:04which is talking about.
02:05Nervous system,
02:06long COVID or post acute sequelae
02:09of COVID-19.
02:10So the first thing I'm going
02:12to describe is neurologic and
02:13psychiatric reports in acute COVID,
02:15and I think that we as a globe
02:17first heard about the concern that
02:20there might be quote neurologic
02:22manifestations and people with
02:24COVID-19 from this report from Wuhan,
02:26China,
02:27and this was from one of the
02:28major hospitals that had,
02:29you know,
02:30thousands and thousands of patients in the
02:33first wave and what emerged from this.
02:35This hospital experience and actually was
02:37originally a preprint in March of 2020,
02:39was this.
02:41Fairly generalized,
02:42very clinically based report suggesting
02:45that 36% of people who were hospitalized
02:48with laboratory confirmed COVID-19
02:50had neurologic manifestations of
02:51disease and what you'll see here
02:54under the nervous system.
02:55Symptoms in this table is that
02:57it's pretty broad definitions,
02:58so CNS conditions,
03:00including things like dizziness,
03:02headache, and her consciousness,
03:04acute with vascular disease or stroke,
03:07and then there were also changes in smell,
03:10taste, and vision which were
03:11considered peripheral nervous
03:12system and also skeletal muscle.
03:13Injury now this is the type of report that
03:16was available early on in the pandemic,
03:19primarily because research investigation
03:20of these patients wasn't possible.
03:22These hospitals were overwhelmed as
03:24we also were at Yale with just an
03:27onslaught of patients and research
03:29collection of samples or data or
03:31even neuroimaging was essentially
03:33not available in most patients
03:35because of infection concerns.
03:37But this was the first hint that
03:38there might be a neurologic concern,
03:40and I think one of the most
03:42important findings of this initial
03:43report was the finding.
03:44Some of the patients came to
03:46the hospital with an primarily
03:48in neurologic manifestation,
03:49as they're preventing presenting syndrome,
03:51so that was probably the first
03:53clue that COVID is not necessarily
03:55only a respiratory disease,
03:57but in fact also has an impact
04:00in other organ systems as well.
04:04And So what we were able to do here
04:05at Yale because we had this sort of
04:07warning ahead of time that there were
04:09concerns coming out of both China and Europe.
04:11And also when we sort of knew that the
04:14wave was coming in the beginning of March,
04:16we sort of scrambled to put
04:18together what we called a neuro
04:20COVID inpatient console service.
04:21And this was essentially me working
04:23with a couple of neurology residents,
04:25primarily Lindsey McAlpine
04:26shown up here on the right,
04:28but also to have a garden is
04:30also coming into your program,
04:31and we essentially were whenever there was.
04:35Inpatient neurology konsult
04:36on on someone who was known to
04:38have COVID-19 in the hospital.
04:40We essentially got involved in the
04:42consultation so we mostly did this
04:44through virtual chart reviews and
04:45looking up data that was available
04:47in the patients and then advising
04:49on clinical care of these patients
04:50and a lot of this was running by
04:52the seat of our pants because
04:54there wasn't a lot of data about
04:55how to manage these patients,
04:57but we were trying to sort of sift
04:59through the emerging case reports
05:01that were coming out while we were
05:03running this console service to try to.
05:04Harmonize what we were trying to
05:06do for these patients with what
05:07was happening around the world,
05:09and what other people have learned.
05:10And this is a snapshot of our sort
05:12of first 100 consoles that we saw on
05:15this neuro COVID console service,
05:16showing you the sort of panoply of
05:18different kinds of conditions of the
05:20consoles were forced so there was a
05:22large chunk of people who had had
05:23seizures in the setting of acute COVID.
05:25Most of these people were people who
05:27had underlying seizure disorders,
05:28and then they were coming in and in
05:30the setting of a febrile illness
05:31had a probably an unmasking
05:33of seizures, which is common.
05:35In anyone who has a seizure
05:37disorder when they have some
05:38other kind of systemic illness.
05:40But interestingly, 26% of these people
05:42had acute stroke during COVID-19,
05:45and then we had another large
05:46swath of people who would have
05:48what we called himself allopathy.
05:49And most of these people were either
05:52people who presented initially
05:53with confusion as as a primary
05:56presenting symptom of COVID,
05:57or there were people who had what
05:59we sort of thought in the end
06:01was encephalopathy or difficulty
06:03waking up out of proportion to,
06:05for example,
06:06situation or other aspects of
06:08their systemic illness.
06:09That's a difficult thing to term,
06:10because of course a lot of people
06:12have just hospital delirium in
06:13the setting of acute illness,
06:15but these were people that seem
06:16to really have more of this
06:18than would have been expected,
06:19and you can see there are also some other.
06:21Some conditions,
06:22including neuromuscular diseases and
06:23a few people who had acute psychosis
06:26in the sort of Perry COVID setting,
06:28which I'll talk about more later,
06:29and what we determined was that
06:31there were sort of two major
06:33categories of conditions that we saw.
06:35One were cerebral vascular
06:37injury or stroke and and,
06:39and this was something that was
06:41known to be potentially associated
06:43with clotting factors that are
06:45elevated in people with COVID-19,
06:47but also due to additional
06:48research investigation we did
06:49in some of these patients,
06:51we found that this also
06:52seemed to be associated with.
06:53Endothelial inflammation and activation
06:55in the setting of acute COVID-19 and
06:59the rest of the conditions that we saw.
07:01Although they were quite diverse,
07:03we ended up considering them to
07:04fall mostly in the categories
07:06of immune mediated neurologic
07:09and psychiatric conditions,
07:10and that was mostly based on the
07:12types of presentations we're seeing.
07:14Things like key on beret,
07:15things like what look like
07:18autoimmune myositis and the
07:20neurologic conditions seem to be
07:22associated potentially with immune.
07:24Some inflammation.
07:27This is a really nice example of
07:28such an immune mediated course
07:30or something that we decided
07:31was probably immune mediated,
07:33so this was a 70 year old woman who
07:35presented to you in March of 2020.
07:37This was before we were screening
07:39everyone for COVID when they came into
07:41the hospital and she actually presented
07:42with confusion and a witness seizure.
07:44And so she she actually didn't have
07:46fever cough and a positive news or
07:49fringilla swab for COVID until Ford
07:51four days into her hospitalization.
07:53So prior to that she had an MRI and
07:55the MRI images are shown on the right.
07:57This is obviously a fairly abnormal MRI,
07:59but I think these are really this is
08:01Shane's brain atrophy and she has a
08:03really chronic microvascular changes
08:04and she had a history of risk factors
08:06for significant microvascular disease,
08:08but her LP was completely
08:10benign and two white cells,
08:12normal protein of PCR's for
08:14viruses were not detected.
08:16Once we found out that she had acute COVID,
08:18we also were able to do research testing
08:21and looking for COVID in her CSS.
08:23SRS could be 2 in her CSF by PCR
08:25metagenomic sequencing and viral culture.
08:28And the virus did not was not
08:30detected by any of these means.
08:32However, when we used more sort of
08:34research testing to look at inflammatory
08:37cytokines in the CSF of this patient,
08:39what we found was that compared to
08:42pre pandemic control samples that we
08:44had that were age matched controls.
08:46This patient, whose CSF cytokines
08:48are shown in the orange bars,
08:50had evidence of neuroinflammation
08:52and in fact some of this seem to be
08:56similar to the the patterns of CSF.
08:58Cytokines are similar to the plasma.
09:00But a couple of these markers,
09:02and in particular MCP one,
09:03was quite elevated in the CSF
09:05and this patient,
09:06although it was not elevated in the plasma.
09:09So this was the first clue that we had
09:11that made the underlying some of these
09:13encephalopathies that we were seeing.
09:14Potentially,
09:15it wasn't necessarily viral
09:17encephalitis in the common sense,
09:19where you see viral replication
09:21in the nervous system,
09:22but instead this was some kind of
09:24an immune mediated phenomenon,
09:25despite the fact that she had no
09:27increase in her CSF white cells.
09:29And this is a sort of busy looking slide,
09:31but it's just a reminder for
09:33me to point out that many,
09:34many of the reports that have come out
09:37that have been case reports clinically
09:39described patients during acute COVID-19
09:41that have neurologic and then also I
09:44think psychiatric conditions that can
09:46best be attributed to immune pairing,
09:48infectious or autoimmune kinds of ideologies,
09:52so acute necrotizing encephalopathy,
09:54Guillain Barre syndrome,
09:56acute flossin,
09:57my leitis and many,
09:58many cases that we've seen of.
10:00I'm small fiber sensory neuropathy or
10:03neuropathy's have all been described in
10:05the sort of Perry acute COVID period,
10:08and many of these cases have
10:11had normal workups.
10:13No CSF pleocytosis normal protein levels,
10:16but the clinical phenomena are are
10:18really what we see typically in
10:22postinfectious immune mediated disorders.
10:25I also wanted to touch on
10:26the issue of stroke,
10:27which was another major category that we
10:28saw and in in our hospitalized patients.
10:31And this was actually very
10:32very early on the pandemic.
10:34There was a report that was super
10:36frightening of young people under age
10:3850 who had a few or no risk factors
10:41for stroke presenting with large
10:44vessel in a major strokes in the
10:46setting of acute COVID for some people
10:49stroke was actually there presenting
10:51syndrome or symptom to the hospital.
10:53And there was this sort
10:55of avalanche of concern.
10:55There was a lot of press about these cases,
10:57thinking that we were going to see lots
10:59and lots of strokes in young people,
11:01and there are reasons to think that
11:03people with COVID-19 are at increased
11:05risk of stroke and this has to do
11:08with coagulopathies thrombocytopenia.
11:09Elevation of D dimers and other
11:12blood measures that are clearly
11:14reflected in the prothrombotic state.
11:16And the fact that there are ACE 2 receptors
11:19which are the receptors that are necessary
11:21for the virus to entry on the surface
11:23of endothelial cells and pericytes,
11:25that line the the blood vessels and
11:28cells theoretically could be leading
11:29to vascular inflammation and changes.
11:31So this was a major concern
11:34based on this first paper.
11:36But I think that what was really
11:37interesting to learn and I think,
11:39was borne out by our studies as well,
11:41is that in fact,
11:44in the the most sort of comprehensive
11:46studies have suggested that stroke in young
11:49people is not a big feature of COVID-19.
11:51Thank goodness.
11:52However,
11:52there are increased rates of stroke
11:54in people living with people who have COVID,
11:57so some of the best data that has
11:59really characterized in their logic
12:01and psychiatric complications of
12:02cute COVID has come out of the UK.
12:05And it's really because of course.
12:06They are able to do,
12:07you know they have a unified
12:09healthcare system and also a very very
12:12connected healthcare system where
12:13hospital records are shared between
12:15all of their different institutions
12:17and data is collected in uniform
12:20ways across their institutions,
12:21and so they're able to collect data
12:23in a very far standardized way.
12:24And this was one of the first reports
12:26that came out of what's called
12:28the Coro Nerve study in the UK,
12:29where essentially they surveyed many
12:31many hospitals in the UK system for any
12:34reported cases of people who had neurologic.
12:36Diseases and they saw sort of
12:38similar cases that were reported.
12:40Of course,
12:41there's bias is what gets reported,
12:42what gets looked at,
12:43but they also saw many cerebrovascular
12:45events in many cases of altered mental
12:47status and what was interesting here
12:49is that when they looked at the age
12:51ranges of who was getting these
12:53conditions in this particular study,
12:55what they found was that although
12:57cerebral vascular disease shown in
12:58these blue columns was actually much
13:00more predominant in older ages and
13:02picked between ages of 71 and 80.
13:04What they term neuro psychiatric.
13:06That really included altered mental status,
13:09peripheral nerve disorders,
13:10and then some cases of things like
13:13psychosis were actually happening,
13:14sort of throughout the lifespan,
13:16and in fact,
13:17one might even say that they were
13:19more common in people ages 21 to 60.
13:22So I think this is the first clue
13:24that there obviously are different
13:26types of ideologies underlying.
13:27They may just super vascular complications,
13:29and some of these other complications.
13:31And it was also our first clue that
13:33some of these major complications of
13:35COVID-19 may not only be affecting.
13:37Elderly people who are at highest risk
13:39of death from COVID but actually are
13:42also affecting people at younger ages.
13:44This is another beautiful study
13:45from the UK Koro Nerve group,
13:47though this UK surveillance study.
13:49So this was a slightly larger
13:51population that they published
13:53later where they again looked at
13:54people who had been reported to have
13:57neurologic or psychiatric conditions
13:58around the time of acute COVID-19,
14:01and they did something very clever,
14:03which is they tried to look to see
14:05when the onset of these different
14:07conditions were with respect to onset
14:10of respiratory symptoms of COVID.
14:12So on the Y axis in this figure.
14:14The zero point would be the onset
14:17of respiratory symptoms and then the
14:19widest point on the violin plot would
14:21be the the main time that the onset
14:23of this particular condition had
14:24risen in these different patients,
14:27and I think what this does is it gives
14:29us a little bit of insight into again
14:32potentially diverse pathophysiologies
14:33of these different types of conditions,
14:35and So what they found was that in
14:3829% of people there were actually
14:39neuro or sex symptoms that predated
14:42respiratory symptoms.
14:43And when you look at the figure,
14:44it actually is primarily this
14:46river vascular complications.
14:47Starting very early and then our preceding
14:51the relative respiratory symptoms.
14:53Then there were another group of
14:56symptoms that were actually seen after
14:58the initial respiratory conditions.
15:00Severe respiratory system
15:01condition improved again.
15:03These were hospitalized patients,
15:04so many of these people were coming
15:06in with shortness of breath.
15:07Oxygen needs that kind of thing,
15:09but a good proportion of people actually
15:12develop their symptoms up to two weeks
15:15after the respiratory symptoms had started.
15:18And this was interesting,
15:19because these seem to be in what they
15:22termed central inflammatory psychiatric and
15:24peripheral nerve types of complications.
15:26And again this is,
15:27I think,
15:28reflective of what we've seen clinically,
15:29and it suggests again that these are
15:32likely immune mediated phenomena that are
15:34somehow a response to the viral infection,
15:37but maybe not directly caused by the virus.
15:40So again,
15:40this is another clue to kind of help us
15:43understand the etiology of these disorders.
15:45But what I've been telling you were
15:47basically all clinical reports
15:48and this has to do with the fact
15:49that clinical data was collected.
15:51Of course,
15:51in all of these patients who were
15:53hospitalized but there weren't
15:55really systematic research studies
15:56done at that time.
15:58And what we have learned,
16:00though,
16:00are that even small numbers of of
16:03systematic research can be helpful
16:05in trying to understand HIV.
16:06I'm sorry, acute COVID-19 era pathogenesis,
16:09and I just wanted to do a really
16:12quick review of the virus.
16:13For those of you who are not
16:15completely tired.
16:15Is remembering exactly what
16:17the structure is of cart.
16:18Stars could be 2 and what the viral
16:20life cycle is probably most of us
16:21know this way more than we would ever
16:23want to know anything about a virus,
16:25but just as a reminder.
16:28We've known since very,
16:29very early in the pandemic that this is
16:32a very simple virus with a single RNA strand.
16:34Very,
16:35very few proteins encoded and the
16:38virus is surrounded by a.
16:41Declare capsid and then a.
16:44Sorry it's bye bye and envelope
16:46proteins and then there are spike
16:48trimers that emerge from the surface
16:49of the virus and these spike
16:51trimers are again what the most
16:53of our RNA antibodies are directed against.
16:56I'm sorry vaccinations are
16:58directed against these.
16:59Spike trimers are.
17:02Basically, what's needed to
17:03be recognized by a host cell,
17:05so the ACE 2 receptor on the surface
17:08of mammalian cells is recognized as
17:10the spike trimmer of the virus and it
17:12allows for entry of virus into a cell.
17:14So the presence of ACE 2 receptors
17:17and then other related receptors and
17:19other proteins that allow for binding
17:22and entry of the virus is really what
17:25determines the cellular tropism of this
17:27virus and determines what types of
17:29cells and tissues the virus can enter.
17:31And then I also wanted to
17:32remind you with a figure on the.
17:33Right, but the viral life cycle is for SARS,
17:36Co V2 and essentially this is an RNA
17:39virus that uses the host cytoplasmic
17:41cell machinery to replicate itself.
17:44So unlike HIV,
17:46another RNA virus which actually uses
17:48reverse transcriptase to integrate
17:50its genome into our host genome,
17:52which is why this is a chronic and
17:54essentially incurable disease,
17:55except for three people out of 75
17:57million people have had it in SARS, Co.
18:00V2.
18:00The virus does not become hopefully
18:03as far as we understand.
18:04Chronic illness.
18:05It's an acute illness that uses the
18:07host machinery to make more of itself.
18:09But eventually, when the immune
18:12system can kill the infected cells,
18:14then the virus will die out.
18:16So what do we know about how this
18:18virus might affect the nervous system?
18:20And some of it is indicated by some
18:23of it initially was indicated by what
18:25we knew by prior similar viruses.
18:27So SARS,
18:28Co V2 has a lot of nucleotide
18:31identity to prior coronaviruses.
18:33In particular SARS,
18:35Co V1 and the mayor's virus.
18:37This is important because there
18:39were actually known neurologic
18:40complications even though very,
18:42very few people worldwide contracted
18:44either of those infections.
18:45But some of what we knew originally
18:47about how.
18:48Starts Kaviti might affect the
18:49nervous system was learned from
18:51these earlier outbreaks and what
18:53was understood from them.
18:55And one of the early concerns was
18:57whether or not stars could be 2 might
18:59enter and infect cells in the brain.
19:02So why would we think that could be possible?
19:04Well,
19:04one thing that's known is that
19:05stars that ACE 2 receptors,
19:07which again,
19:07as I mentioned,
19:08allow for entry of virus into cells
19:10and into tissues are actually located
19:12on multiple cell types in the body,
19:15and although there are very very
19:17common in the nasopharynx and in the
19:20alveolar epithelial cells of the lung,
19:22probably explaining why most of those
19:24are these major major target organs,
19:27other target organs also include the kidney,
19:29the intestines, the blood vessels,
19:31and the nervous system.
19:33And as again, we all learn very,
19:35very quickly in the first
19:37months of the pandemic,
19:38all of these organs can actually be
19:40affected in people with acute COVID-19.
19:42And what we didn't know,
19:44however,
19:44is which actual cells
19:46of the brain express ACE
19:492 receptors and on the right you can see a
19:52very nice in silico analysis looking for the
19:55distribution of genes relevant to severe.
19:57So stars Kobe 2IN different cell types in the
20:01brain and what you can see is that ACE 2.
20:03Actually seems to have only
20:06modest expression in neurons.
20:08It does seem to have higher
20:10expression in oligodendrocytes,
20:11but some of these other proteins that
20:13are important for viral entry are
20:16also are highly expressed in neurons,
20:18suggesting that there may be some
20:20vulnerability of neurons to SARS, Co V2,
20:23but but we don't have strong expression
20:26of ACE 2 on the surface of neurons.
20:29How would the virus get to the
20:30brain in the 1st place?
20:31I think this is again something
20:32that that many of the positive sort
20:35of mechanisms were based on what
20:37we knew from prior viruses.
20:38So this is a model that we put together
20:40back again in March or April 2020,
20:42saying well,
20:43one possibility is the virus could
20:45potentially get into the brain by
20:47haematogenous spread and this model
20:48is based essentially on the model
20:50that we know for other viruses,
20:52including HIV where viral particles but
20:56particularly infect virally infected cells.
20:59May traffic through the vascular endothelium
21:01in the blood brain barrier into the brain?
21:04I think that this is something
21:07that was a potential hypothesis.
21:09It turns out that SARS Co V2 is not really
21:13readily detectable in the blood and
21:15many people who are very sick with COVID-19.
21:17So in fact it's there's not as much
21:20SARS Co V2 viraemia as there is
21:23for example collection of stars.
21:25Kobe 2 virus and things like
21:26the whole factory epithelium,
21:28which makes I think this.
21:29Hematogenous spread a little bit less likely,
21:31although still a potential in,
21:33especially maybe in people with a
21:35breakdown of the blood brain barrier on
21:37the right is another transcranial root,
21:39which I think is potentially sort
21:42of more likely if there really is
21:44viral entry into the brain that
21:46it could be entering through.
21:48There's robust infection of the system
21:50tacular cells and the whole factory
21:52epithelium in the news of pharynx.
21:54This may be related to the anosmia
21:57that probably at least half of people
21:59experience when they get COVID-19.
22:01And the sub position was is the
22:04potentially the olfactory nerves
22:05could then locally become infected
22:07and through retrograde spread
22:09through between the synaptic cleft.
22:12The virus could then be transmitted into
22:15the olfactory bulb and into the brain.
22:17This has been shown to be possible by
22:19concerns COVIE 1 using animal models,
22:21and so this was really a putative mechanism.
22:24But what's the evidence that stars could
22:25be to you actually gets into the brain?
22:28I think there's really not a
22:29lot of evidence so.
22:31There have been a number of studies that
22:33have used small samples of looking for stars.
22:35Kobe 2 for example in CSF fluid and
22:37all of you know that super spinal
22:40fluid is not the same
22:41as the brain. It's not telling us about
22:43what's happening in the deep brain
22:45parenchyma, but in many many acute
22:47viral infections of viral particles are
22:50readily detected in the spinal fluid
22:52in the setting of brain infection,
22:54acute HIV being something I've been
22:56studying for 20 years is a good example.
22:58We're about 80 or 90% of people during acute.
23:01Striking infection,
23:02even without any symptoms,
23:04have readily detectable HIV
23:05RNA in the spinal fluid.
23:07But interestingly,
23:09in people with acute COVID-19,
23:12particularly looking at people
23:13with neurologic symptoms,
23:14there have been very,
23:16very little a few reports
23:17suggesting that one can detect SARS.
23:19Kobe 2,
23:20So this was a small study of
23:22just six people by my colleagues
23:24in Sweden where they looked at
23:26CSF markers of inflammation,
23:27and they also try to detect SARS,
23:30Co V2 by PCR.
23:31And in this study they found very,
23:34very low level detection by PCR
23:36in half of the participants,
23:38but they essentially felt that
23:39based on the levels of detection
23:41and their subsequent work that this
23:44was probably just contamination
23:45and they later did a review article
23:48where they actually looked at all
23:50of the published reports to date,
23:51which had been almost 450 reports to date.
23:54Looking for PCR detection of SARS,
23:56Co V2 in CSF in people with acute COVID-19,
24:00mostly with severe disease.
24:02Almost all with neurologic symptoms,
24:04which should be the indication
24:05for lumbar puncture and starts.
24:07Kovi 2 was only detected in
24:08less than 3% of those cases,
24:10so I think this is very convincing
24:13evidence that direct viral invasion,
24:15main and and replication within
24:16the CNS may not be a major
24:19driver of neurologic symptoms.
24:20However,
24:21this study shows elevations of immune
24:24activation biomarkers in the CSF,
24:26so similar to the patient
24:28case that I showed you,
24:29they found elevation of CSF,
24:31neoprene and beta microglobulin.
24:34Metro biomarkers of myeloid inflammation in
24:37all of the individuals that they looked at.
24:39The dotted line here being the
24:41normal levels and healthy controls,
24:42and they also found a hint of
24:45neuronal injury in a few of
24:47the people during this time.
24:48And I'll come back to this in a minute,
24:50so this was really the first clue that
24:54that neuroinflammation was predominant.
24:55Even without detection in viral PCR.
24:58Now,
24:58this is a beautiful study led by actually
25:01today's neurology grand rounds speaker.
25:04Shibani Mukherjee and colleagues
25:05from MGH where they looked at
25:08larger numbers of people so they
25:10had 27 people and they collected
25:12CSF during acute COVID-19 again,
25:13people who had LP's for neurologic
25:15indications and so this is not a
25:18survey and everyone with COVID,
25:19but it's actually people who had
25:21distinct neurologic disease and they did.
25:23QT PCR and metagenomic sequencing
25:25in all 27 samples and did not
25:27find detection of SARS Co V2 in
25:30any of the samples.
25:31However,
25:31they did find elevation of CSF inflammatory.
25:35Markers in the COVID-19 patients
25:37that are shown with the red dots,
25:40particularly a couple of the markers,
25:42including MCP one and CSF mcglen beta,
25:49compared to healthy controls I wanna I'll 6.
25:53And interestingly, in this study they
25:55also looked at other viral infections
25:56and compared the levels here in West
25:58Nile virus was the only one that had
26:00higher levels in some of these markers.
26:02So again, this is suggesting and
26:03I should say that most of these
26:05people had normal CSF white counts,
26:06so they didn't have.
26:08Will work evidence of Pleocytosis,
26:10but they had subtle neuroinflammation.
26:14Interestingly, there has been
26:16look at other types of markers
26:19in CSF in acute COVID patients,
26:21and this paper was recently published
26:23in about a lot of attention because
26:25it demonstrates the fact that
26:27markers that are associated with
26:29neurodegeneration in Alzheimer's disease
26:31were actually abnormal in people
26:33who had COVID were hospitalized with
26:35acute COVID and neurologic symptoms.
26:37So again, here are the COVID
26:39patients are shown in the orange,
26:41and it's showing reductions in soluble.
26:44A panel of precursor protein
26:46and amyloid beta in people with
26:49COVID-19 during acute disease.
26:51This has been a little bit
26:53interpreted to say that maybe this
26:54is setting these people up for
26:56Alzheimer's type neurodegeneration,
26:57but I was interpret this
26:59to say that actually,
27:00in most acute infectious,
27:03acute viral conditions,
27:05you can see these kinds of changes
27:07that resolve after acute infection and
27:08this is something that we've also seen
27:10and published in acute HIV infection,
27:12so hopefully this doesn't actually.
27:14Mean that there are going
27:16to be Alzheimer's disease.
27:17You know Avalanche coming in our future.
27:19Although we can talk about that at the end,
27:21but I think that these kinds of
27:24perturbations and amyloid processing and
27:26neuronal processes maybe something better.
27:28Hint that there's been some
27:30injury during acute infection.
27:32So this is really elegant work and this
27:34is being led by my colleague here.
27:36Shelly Farhadian,
27:37who is a faculty member in infectious
27:40disease who's also really a
27:42stellar investigator and clinician
27:43in neuro infectious disease.
27:46And Shelly really took forward
27:48the opportunity to characterize
27:50in a research level,
27:51some of the clinical patients that we're
27:53seeing on the Neuro COVID console service.
27:55So when patients did have CSF
27:58collection for clinical reasons,
28:00Shelly scrambled to do, really.
28:02Real time studies on the CSF,
28:04including things like single cell RNA,
28:06seek on fresh samples to try to
28:08characterize in great detail what
28:10some of the immune perturbations
28:12might be in these patients.
28:14And so in this first six
28:17patients that were evaluated,
28:19they she and other colleagues
28:21here at Yale looked for stars.
28:23Kobe 2 by PCR,
28:24metagenomics and culture did
28:25not find any evidence of virus,
28:28but did find really intriguing
28:30alternate immune responses in the CSF.
28:32And what we did is that we looked
28:34at the patients who had COVID-19
28:35and again on the left.
28:37These are shown in purple compared
28:38them to pre pandemic CSF and
28:40blood controls that we had been
28:42collecting for other studies.
28:44Healthy people who were age matched
28:46and what she found was increased
28:48CSF cytokines in the CSF compared
28:49to the controls.
28:50But what was really interesting
28:52here is that she specifically found
28:54a distinct pattern between the CSF
28:56cytokine elevations and those in blood,
28:58suggesting that there was actually
28:59a compartmentalized response.
29:00This wasn't necessarily just
29:02a spillover of inflammation.
29:03Systemic circulation,
29:04but instead was actually potentially
29:07a local response related to local
29:11cells and local perturbations.
29:13Another thing that she detected
29:14by single cell RNA seek was a
29:16increased frequency.
29:17B cells in the CSF and so this led
29:20to an interrogation of antibody
29:23responses in the CSF and really,
29:26really interesting work.
29:27Again led by Shelly and other colleagues
29:29here at Yale and also collaborators at UCSF.
29:33Using some really novel ways to
29:35detect whole panels of anti SARS,
29:37Kobe 2 antibodies.
29:38They were able to describe the fact
29:41that anti SARS Co V2 antibodies were
29:43detected in CSF and all of these patients,
29:45but actually some of the antibodies
29:47were distinct from the patterns of
29:49the antibodies that were found in
29:50the blood in the same patients.
29:52And in fact,
29:53when some of these antibodies
29:55remain into monoclonal antibodies,
29:57they were found in a novel auto
30:00antibody discovery method to
30:02actually be autoreactive.
30:04To brain tissue.
30:05So these are again hints that
30:07not only is there a generalized
30:10neuroinflammation and antibody formation,
30:12but some of these antibodies that are
30:14developing maybe in response to SARS,
30:16Co V2 are actually auto reacting or
30:19autoimmune antibodies that could
30:20be attacking the nervous system.
30:24So what have we learned from autopsy tissue?
30:26And I think that this is again,
30:28it's come out and fix and
30:29starts over the last two years.
30:31Is these autopsy series looking at
30:33people who died with acute COVID-19?
30:36Always have of course,
30:37all the caveats of the fact that they
30:39will affect people who have really,
30:40really severe disease who died from COVID.
30:43And many of these people
30:45had prolonged ICU stays.
30:46Many of them had, you know,
30:48hypertension, heart failure, hypoxemia,
30:49and you can see here for example.
30:53This is a group of patients that
30:55were collected again at MGH.
30:5818 people.
30:59Most of these people did not have
31:02significant neurological symptoms,
31:03but they, but they died.
31:04All of them had hypoxic ischemic
31:06injury in the brain.
31:07But I think what's important in this
31:09study is that this study was particularly
31:12looking for detection of virus,
31:14and they did different methods of
31:16detection trying to determine whether
31:17or not there was actually SARS.
31:19Kobe 2IN the brains of these people
31:21who were very very severely ill with
31:23COVID-19 using immunohistochemistry.
31:24They did not detect virus in neurons,
31:28glial and ileum,
31:29or immune cells in any structures of
31:31the brain. In any of these individuals.
31:33When they used PCR and looking just for
31:36fragments of the nucleocapsid protein,
31:39they did find positive findings in
31:40six of the 52 different blocks that
31:42they looked at and sort of equivocal.
31:45Very low, low level findings,
31:47and in some this is consistent
31:49with other studies.
31:50There's also a really nice study
31:51out of the NIH that also looked at a
31:53number of people that again did not
31:55detect intact virus in any of the patients,
31:58so this would suggest again that
32:00if you had replicating virus,
32:02typical viral encephalitis.
32:03You find a lot of intact virus,
32:05and that's not being found.
32:07This is another study in another group.
32:09This is a group out of Germany that
32:12has published reports suggesting that
32:14they can detect intact virus in some people,
32:18but but what's really remarkable
32:19about this study is the the widespread
32:21inflammation that was reported in this study.
32:24So again,
32:24this is autopsy data from 43 patients.
32:27Each column in this figure is
32:30showing you the the demographic
32:32information for an individual donor
32:35who died of acute COVID-19.
32:38And then you can see on the left the
32:40different brain regions that were
32:42examined and what this large portion
32:44of the chart is showing is microglial
32:47activation widespread in the brains of
32:49almost every single one of these donors?
32:52So microglial activation there
32:54was also a very similar graph
32:56shown for CD8T cell infiltration.
32:58Now again,
32:59in the setting of acute illness and death,
33:02all sorts of hypoxic ischemic injury.
33:04Maybe we expect some microglial activation,
33:06but this seemed to be to.
33:08This pathology group seem to be
33:09way out of proportion to what
33:11they would have expected,
33:12and compared to other types of donors who
33:14also of course died with acute illnesses
33:17and everything in this graph
33:18that is not blue is shown to be
33:21abnormal microglial activation in
33:22the bottom part of the figure.
33:24I'm showing you in purple the few
33:27brains in which they could base.
33:29They identified stars Kobe 2 RNA or in
33:31a few more brains of positive SARS.
33:35Co V2 proteins such as nucleocapsid
33:37or spike protein.
33:39And what's really remarkable here
33:40is that the high levels of neural
33:43inflammation do not necessarily
33:44cluster only in those people
33:46who had detectable virus,
33:48but in fact is widespread even in
33:50people with no detectable virus.
33:52So again, what's been emerging is
33:55high levels of neuroinflammation.
33:57In the absence of major viral replication.
34:00And this is one more study that
34:01I just revealed.
34:02One more thing,
34:02I think that's truly important about
34:04the neuropathology of COVID-19.
34:06Again, an autopsy study.
34:08Many of these people died with sudden death,
34:11so they were cases of people
34:13who were corner cases,
34:15so they weren't necessarily hospitalized.
34:19These individuals did not have any other
34:22kind of major respiratory findings.
34:24They did not have any overt strokes,
34:27but when brain tissue from these
34:30individuals was examined both by
34:3211 Tesla and Mirai autopsy tissue,
34:34as well as by histopathologic examination.
34:37They were found to have microvascular injury,
34:40so if fibrinogen leakage from
34:44blood vessels microhemorrhages.
34:45And large collections of perivascular
34:48macrophages so the the hint here was that
34:52even in people who don't have overt strokes,
34:55microvascular injury microvascular
34:56changes may be also underlying some of
34:59the neurologic types of consequences that
35:02we're seeing in people with acute COVID-19.
35:05Again,
35:05this study did not detect SARS
35:07Co V2 in any of their brains
35:09segments looked at and this group.
35:11This is from a group of based
35:13primarily at NIH.
35:14They continue to do studies and do not.
35:16Find any virus in the brain.
35:19So I I think the final thing to talk
35:21about in terms of pathology are some
35:24of these neuronal injury markers.
35:25So I mentioned neuroinflammation.
35:28I mentioned the fact that there doesn't
35:30seem to be robust active replication,
35:32but there may be microvascular injury,
35:35but is there actually brain injury
35:37in these patients during acute COVID?
35:39And this is a little bit of
35:40a of a concerning study,
35:42I think we have a lot more information
35:44that we need to look into,
35:45but this is a group that had large
35:48numbers of people who had acute COVID-19.
35:51So 246 people who were hospitalized with
35:54acute COVID-19 this is from NYU in New York.
35:57And samples were collected from the
35:59blood for these individuals and then
36:01highly sensitive assays were used
36:03to measure neuronal injury markers
36:05that are leaking into the blood.
36:07So these include neurofilament
36:09light chain was,
36:10which is a biomarker of axonal injury.
36:12Glial fibrillary acid protein,
36:13which is a marker of astrocyte
36:15injury and ECHL one which is also
36:18a neuron specific protein and what
36:20this group found was that NFL and
36:23GF AP were quite elevated in the
36:25blood in these acute
36:27COVID patients.
36:28These were not necessarily people
36:30who had neurologic conditions,
36:32they just took all comers and what
36:34was interesting was that these
36:36measures were actually quite a bit
36:38higher than even those found in
36:40people with Alzheimer's disease.
36:42So again, what the implications
36:44of this might be for people
36:46who survive acute COVID-19,
36:48I think is a really important question,
36:50but it's a really clinically unknown
36:52and I will say that during acute
36:54HIV infection we also see elevations
36:56of things like NFL and GFAP.
36:58Ubiquitously in people who do not
37:01have any neurologic symptoms,
37:03and we haven't seen an avalanche
37:04of things of nerve degenerative
37:06disease such as Alzheimer's and
37:08people who've had HIV for 40 years,
37:10so I think what this means
37:12clinically is is unclear,
37:13but it does suggest that there's
37:15some neuronal injury during
37:16the time of acute COVID-19.
37:17So all of this leads to this sort
37:20of very oversimplified figure,
37:23which is actually a beautiful figure put
37:24together by the science illustrators.
37:27That summarizes,
37:28I think what I've told you so far,
37:29so we think in the upper left that
37:31there's really generalized inflammation
37:33and generalizing neural inflammation.
37:35Immune cells,
37:36including monocytes and CD8T cells and CD.
37:39Four cells,
37:40probably traffic into the nervous system.
37:43You don't necessarily have to
37:44have a broken down blood brain
37:45barrier for cellular trafficking,
37:46of course,
37:47but there may be some blood brain
37:48barrier injury as well,
37:49and then cytokines and antibodies
37:51also seem to be diffusing into the
37:54nervous system and into the brain.
37:56There is also antibody production.
37:58Which may be exacerbating
37:59the nerve inflammation,
38:00and there seems to be production
38:02of antibodies both to SARS,
38:04Co V2 that are entering the brain and
38:06persisting potentially in the brain,
38:08and then also autoantibodies.
38:09There's limited presence of
38:11the viral proteins or a viral
38:13particles in the neurons based on
38:14all the current detection methods,
38:16although you know more research has
38:18to be done and there seems to be at
38:20least some level of microvascular injury.
38:22Some of this may be related
38:24to endothelial injury,
38:25some of it's related to micro clotting.
38:27And some of this may also
38:29be underlying disease.
38:31Which of these mechanisms are primarily
38:33leading to the signs of neuronal injury?
38:35I think are not certain.
38:37But I think that it's important
38:39to think that there are also
38:41susceptibility factors such as genetic,
38:42pre-existing comorbidities,
38:44and immune status that may be contributing.
38:47So I'm gonna spend the rest
38:50of my talk really focused on.
38:52The mystery of long COVID,
38:54and I suspect that for most people,
38:55this is the part that they
38:57really have heard less about,
38:58and I'm I'm hoping that to show a
39:00little bit of light on this so very
39:02early on in the pandemic there's
39:03there was some kind of popular
39:06press reporting of people saying,
39:07you know,
39:08I really am having trouble with this
39:10feeling that I have dementia and
39:12this term brain fog has come out.
39:14There's also being more scientific
39:16reports talking about these issues,
39:18and now it's really emerged that there are
39:21persistent issues relating to mental health.
39:24Affective disorders psychiatric
39:26psychosis disorders, cardiovascular,
39:28etc that are really persisting after
39:32COVID-19 and so really what we have.
39:35What we have done is again as I mentioned
39:37for the first wave we initially have
39:40learned about this through a clinical
39:42response and along with a really
39:45fantastic partners associated with
39:47running this multidisciplinary COVID-19
39:49recovery clinic which is directed
39:51by Jennifer Plastics and pulmonary.
39:53And colleagues and psychiatry,
39:56Dr Klingensmith and Doctor Fisher.
39:59We have been involved in seeing
40:01people who have recovered from
40:02COVID-19 but have lingering symptoms,
40:04and by October 2020 it became
40:06clear that there was really a large
40:08demand for people who are having
40:10persistent neurologic issues,
40:12and so again with my wonderful
40:14fellow stone Lindsay Farhadian.
40:15I'm let Lindsey and McAlpine began
40:18to run this clinic that's focused
40:20on these disorders and what you can
40:22see here are the chief complaints
40:24of the people who the 1st 300 or so
40:26people who came into our clinic.
40:28Again, there's a separate.
40:30Psychiatry referral clinics.
40:31So we were not primarily seeing
40:33people who are having severe anxiety,
40:35depression, PTSD, psychosis.
40:36These kinds of things so you can
40:39see that 60% of these people are
40:41coming in with cognitive impairment,
40:42others with neuromuscular complaints
40:44and headache.
40:45And a few things have emerged
40:47that have been interesting.
40:48First of all, surprisingly,
40:49in the clinic,
40:50the median age is less than 50 years old.
40:53So rather than what we might have expected,
40:55which is people who would have the
40:57most severe course of acute COVID-19
40:59like being hospitalized had had
41:01prolonged severe illness related to
41:03their risk factors of being older
41:05and having a lot of comorbidities,
41:07we were actually seeing a lot of people
41:10who were young and the other really
41:11remarkable finding in this clinic,
41:13and this is completely echoing
41:14what's being found.
41:15Worldwide and people who are presenting
41:18lingering symptoms after COVID.
41:20Is that the vast majority of people
41:21who are coming into our clinic
41:23now with lingering symptoms,
41:24were never hospitalized.
41:26So while 19% did have ICU admission,
41:3067% of people completely convalesced at home,
41:33reflecting the fact that they had mild COVID.
41:36So what does that mean?
41:39What are some of the?
41:40What are some of the stories that
41:42we had with patients?
41:43Well, here are just some examples.
41:46So one of the types of syndromes
41:47that we've seen are people who
41:49are having cognitive problems,
41:50and this is an example of a patient.
41:53They they presented to us saying that they
41:56were having trouble remembering tasks,
41:58conversations needing a notepad
42:00to write things down during calls,
42:02writing emails and correspondence,
42:03taking two to three times as long as usual,
42:06and one of the things that brought
42:07them into our clinic.
42:08Finally is that they were unable
42:10to follow a recipe to bake bread,
42:11and this is a physician who of course
42:14you know all of these things would
42:16be second nature and unthinkable
42:18before COVID-19 another patient reflected
42:20sort of another category of people
42:22that we're seeing which is headache.
42:24And for some people this has been happening
42:26in people who had severe headache as
42:28part of their initial COVID illness.
42:29But headaches then lasting daily for months
42:32and months and fearing with their jobs.
42:36We've also had numbers of people who are
42:38having patchy sensory abnormalities,
42:41and this is something that unfortunately,
42:42we're continuing to see.
42:43We just have a patient who had a breakthrough
42:46infection of Omicron after complete
42:48vaccination and booster who's having
42:50patchy sensory changes after COVID-19.
42:53And finally,
42:53this is a case that's well known to
42:56many people in psychiatry because we
42:58share taking care of this patient.
43:00People such as this gentleman a 30
43:02year old man with no past psychiatric
43:04history who developed paranoia,
43:05delusions,
43:06hypersomnolence in a few weeks after COVID-19
43:08that was refractory to anti psychotics.
43:11So these are some of the sort of
43:13individual stories and I'll tell
43:14you that the more and more you hear
43:15from people around the world who are
43:17experiencing lung COVID the more and
43:19more you hear stories that are very
43:21very similar to these individual tales.
43:23So what do we know from research standpoint
43:26about this condition about long COVID,
43:28though some of the best data has
43:30actually come from the community,
43:31and this has really been a remarkable
43:33effort from a group called the patient
43:36LED Resource Consortium for COVID-19.
43:38Some of the people in this consortium
43:41are actually neuroscientists originally,
43:43many people or people who have
43:45been in researching other ways.
43:47But once they experienced kovid and
43:49then had a what they called long
43:51COVID symptoms and the term long
43:52COVID has really come from the.
43:54Community,
43:56they have really come forward to
43:57try to be advocates and involved in
44:00research and setting research agendas,
44:01so this is a study that has a lot of caveats.
44:04Online survey.
44:05Not everyone was documented
44:07to actually have COVID,
44:09but it is a survey that I think
44:11reflects what the current knowledge
44:13is of us long COVID worldwide.
44:15So this is something that revealed a
44:17couple of really important messages.
44:19First of all,
44:20this is people reporting their
44:21average number of symptoms over
44:23seven months after having COVID.
44:25And what you can see here is that
44:27while many many people have recovery
44:29overtime with gradual decrement in
44:31the number of symptoms over months,
44:33there's a large number of people in
44:35this reported study that actually had
44:38frequent and numerous symptoms lasting
44:40seven months and longer after COVID-19.
44:43And what really came out
44:44in this initial report,
44:45which I think is really extremely important,
44:47is that many,
44:49many of the symptoms people are
44:51having are considered what they
44:53consider to be neuro psychiatric.
44:55So you can see at the top what they
44:57call brain fog, memory issues,
44:59sensory changes in the middle,
45:02and then what they call headache
45:05and insomnia.
45:06Being major features of some
45:08of these most lasting symptoms
45:10after COVID and I think has
45:12really drawn attention.
45:13And what is a is an incredibly
45:16important problem for many,
45:17many people worldwide and trying
45:19to understand what may be the
45:21ideologies that challenges.
45:22There's really been very little
45:24systematic research into the
45:25ideologies of these conditions so far.
45:27There's major research efforts worldwide,
45:29of course, so this is a study
45:31that's a little interesting.
45:32It's a study that came out a few months ago,
45:35and there's actually been a follow-up
45:37PET study really finding similar
45:39findings by another group independently.
45:41This is a group that looked at 45 people.
45:44Had persistent cognitive symptoms
45:46or psychiatric symptoms and median
45:48of three months after COVID-19 and
45:51what they found was that there
45:54was regional hypometabolism.
45:55Looking at FDG pet compared to
45:57pretty well matched healthy controls
45:58and what was interesting in this
46:00finding is that the metabolic values
46:03actually associated with symptoms
46:04in this fairly small study,
46:06we don't know,
46:08of course,
46:08is whether some of these regional
46:10hypometabolism I'm from metabolism
46:12differences were actually
46:13present before COVID.
46:15And whether they were somehow people
46:17who were sensitive to having some
46:19of these syndromes after COVID so
46:21a really beautiful study that was
46:23actually just officially published
46:25last week that was actually reported
46:27as a preprint last summer gets around
46:29a little bit of that issue so this
46:31is a study that takes advantage of
46:33the absolutely fantastic resource
46:35of the UK biobank study.
46:37Many of you may know about UK biobank study.
46:39It's basically a long term surveillance
46:42population based study that collects.
46:44All sorts of samples and data from
46:46thousands of people throughout the
46:48UK over years and it's this has been
46:50an amazing resource for studying,
46:52for example,
46:53genetic diseases and rare disorders,
46:55and they collect things like
46:56echoes in a subset of people.
46:58They collect all sorts of blood
47:00and CSF and other data and other
47:03people limited neuro psych data in
47:05all of the participants and then
47:08in some subsets of participants
47:09they do serial MRI imaging.
47:11And so this is a wonderful study,
47:13but didn't manage of the fact
47:14that they identified that they
47:15had participants in the UK.
47:17Biobank study that had SARS Co V2
47:20positive tests between their prior
47:22scan in their usual surveillance
47:25system and then their follow up
47:28scan and they were able to do a
47:30really really deep chart analysis
47:32to identify what I think is a fairly
47:34convincing group of 384 controls
47:36that really do not seem to have had
47:38COVID based on lab tests and other
47:41data that's available in there.
47:42Charts which can be linked to the
47:45UK Biobank database and when they
47:46did this comparison and these people
47:48were also very well matched in terms
47:50intervals between scans and other factors.
47:52They found focal loss of Gray matter
47:54in a variety of brain structures in
47:57those people who would have COVID-19.
47:59Now the Gray matter loss was very small
48:02and and relative to Gray matter loss.
48:04All of the groups lost Gray
48:06matter over the two
48:07years. Both the healthy
48:09controls and the COVID group,
48:10but the COVID group lost more Gray matter.
48:13Maybe certain regions
48:14than the other group did.
48:15There were also some disruptions
48:17in DTI imaging metrics,
48:18suggesting some other structural changes.
48:20So what does this tell us?
48:21This is not an all title on COVID.
48:23We have no idea whether any of these
48:26patients have long COVID symptoms.
48:27There were some changes in Neuro
48:29Psych testing overtime in the COVID
48:30group compared to the other group,
48:32and it wasn't convincingly shown
48:33to be associated with these
48:35changes in brain structure.
48:36But I think what this is really
48:38useful to tell us is there is
48:40a footprint of COVID effects in
48:41the brain that can be measured,
48:43which maybe gives us some more
48:45targets and even more sort of
48:46galvanizes us even more to try
48:48to study this question with the
48:50kinds of tools that we can use that
48:52we've developed for neuroscience.
48:55I'll say that there are,
48:56you know,
48:56one hypothesis might be that long COVID
48:58is associated with neuroinflammation and
48:59this is being sort of put out by many people,
49:02including our groups.
49:03We don't have really any
49:05convincing data for that.
49:06This is a tiny study that looked at
49:08some patients with cancer and World
49:09Sloan Kettering and looked at people
49:11who had cancer and neurological
49:13symptoms after COVID that are
49:15shown in the red and then looked
49:17at other comparison patients who
49:18had cancer and neurologic symptoms.
49:20But no kovid and found some
49:22higher levels of CSF inflammatory
49:24markers and those who had COVID.
49:26Much,
49:27much lower than those people who were
49:28treated with CAR T cell therapies,
49:30so I think it's hard to know really
49:31how clinically significant this is,
49:33but it's a hint that looking for
49:35neuroinflammation after Kovac may be helpful.
49:37Another group has looked prospectively
49:39and individuals who had acute COVID-19
49:42during hospitalization and then
49:44were followed over three months,
49:46and in this group,
49:48those people who had higher measures
49:50of depression and neurocognition 3
49:52months after COVID had higher baseline
49:54measures of blood inflammatory markers.
49:57Again suggesting that maybe an
49:59inflammatory hit could have
50:00resulted in more outcomes later,
50:02but of course there's lots of
50:03caveats for this kind of study and
50:05one issue I think for all of these
50:07studies that are based in primarily
50:09hospitalized patients is that it's
50:11really uncertain to know how these
50:12results to people who have milder
50:14forms of coverage or in their acute illness.
50:17And how they relate to longer term symptoms.
50:19So I mean,
50:20and by talking about the COVID mine study.
50:23And again,
50:24this is just really a preliminary study
50:26that we started over the last year at Yale.
50:29This is in collaboration with
50:31Shelly Farhadian against stellar
50:33physician investigator here and in
50:35new infectious disease and neurology,
50:37and an ID.
50:38And this is a study that's using
50:39a lot of the types of tools that
50:41we've been using over the years to
50:43look for subtle abnormalities and
50:44people with acute HIV and
50:47long term treated HIV.
50:48And essentially using these same
50:50really sense of interrogations to
50:52look for changes in people who are
50:54having neurologic and psychiatric
50:56complications months after COVID
50:57and most of these patients are
51:00referred from our neural COVID clinic
51:02and we really try to screen them
51:04to identify people who are having
51:06central nervous system symptoms,
51:07primarily because those are the
51:09kinds of tools we're focusing on,
51:10so you can see that we're using MRI
51:12brain really kind of detailed neuro
51:14psychiatric batteries of cognition and mood,
51:16lots of histories and surveys,
51:18lots of information about their.
51:19Acute illness and their ongoing
51:22conditions and then doing very detailed
51:24blood and CSF studies that range from
51:28everywhere between immune profiling
51:30with through single cell studies
51:32to unique studies of antibodies,
51:34autoantibodies and really high
51:36level markers are looking for
51:38evidence of viral persistence.
51:39And I'll just say this is a this
51:42is a super complicated study
51:43with lots and lots of expertise.
51:45These are only some of the
51:48collaborators unsteady worldwide and.
51:50We'd be really lucky to have
51:52collaboration of some investigators
51:53in Department of Psychiatry.
51:54Whoever you really interested in
51:56working together with us on this.
51:58So I'm just going to show you
52:00some super preliminary data.
52:00We've presented 1 abstract
52:02based on this preliminary data,
52:04but we're really hoping to have
52:06about 100 patients to pull together
52:08for a more complete approach.
52:10I partly say that because this is
52:14a very heterogeneous condition and
52:15I think reporting on small numbers
52:17of people will not necessarily
52:19give us the answers that we need.
52:21But the first thirty four
52:23people enrolled in our cohort,
52:24our young age 5374% women 300
52:28days from symptom onset.
52:30So almost a year after acute COVID,
52:3374% of them convalesced at home,
52:35and the vast majority of them came
52:37to us and rolled in our study
52:40because of cognitive impairment.
52:4259% also have mood symptoms,
52:45and we're comparing them to pre pandemic
52:47controls in the data that I'm going
52:49to show you in the next couple of slides,
52:51who were of a similar age, but.
52:53Different gender because most
52:54of our controls are men.
52:56So what are the kinds of symptoms
52:58these patients had during acute COVID?
52:59You can see it's all across the
53:02books and some of them had.
53:04Most of them had cough,
53:05fever and headache.
53:07Many of them also had other symptoms.
53:09Many people ask me well,
53:10did everybody have loss of taste and smell?
53:12Know in our study only 40% of these
53:14people who are having long term
53:16symptoms actually had a nausea or
53:18or thusia during acute COVID and
53:20what are the symptoms that they
53:22are now presenting with?
53:22We do a very very detailed survey of what
53:25kinds of symptoms people are experiencing,
53:27so you'll see lots of things on here.
53:29I think we've actually left
53:31off psychosis which a couple of
53:32people have also had psychosis,
53:34but you can see that.
53:35The majority of people are now
53:37having what they describe
53:39as cognitive impairment, anxiety,
53:41depression, fatigue and headache,
53:43and then some of these patients also have
53:45some of these other types of symptoms.
53:49So what I'm going to show you were
53:50just some really preliminary findings,
53:52and in all of these figures,
53:53the post COVID participants again
53:55about a year after having COVID-19
53:58a median time are shown in the red
54:00and our pre pandemic control samples
54:02who are basically healthy controls
54:04that were recruited for each of these
54:07studies are shown in the orange and
54:08what you can see here is we've looked
54:10at sort of different categories of
54:12markers and preliminary studies.
54:13These are immune measures that we
54:15wondered whether might be abnormal
54:17in the people who had COVID.
54:18And you can see that the only one
54:21that's persistently abnormal a year
54:22after COVID seems to be the D dimer,
54:25which is only slightly elevated
54:27and seems to be driven by small
54:29numbers of people. Do you die?
54:31Mark is a sort of general
54:32acute phase reactant,
54:33but it's also of course a marker
54:35of clotting and microphone by,
54:37so that's potentially a little
54:38hint of what could be going on.
54:40And then we've also had a nice collaboration
54:42to look at neuronal and astrocyte markers,
54:45and what we were surprised to find is
54:47that although most of these neuronal.
54:49Matt astrocyte markers NFL and Tau were
54:52normal or similar between the groups.
54:55We did find a hint of astrocyte
54:57injury in the blood and these people
54:59with long term symptoms after COVID.
55:01What exactly that means and where
55:03this GF AP is coming from?
55:04Is it definitely coming from the brain,
55:06or could it be coming from
55:07peripheral nerves in some way?
55:09Or some other kind of cell
55:10tech that we're not aware of?
55:11Is it interesting question?
55:13But this would be interpreted as actually
55:15being injury to astrocytes in the
55:18brain that we're detecting in the blood.
55:20We've also looked at spinal fluid
55:22in all of these individuals,
55:23so again these are really generous
55:25participants who are coming
55:26in for lumbar punctures,
55:27and you can see here that the
55:28white cell counts are normal.
55:29This is consistent with all
55:30the clinical experience,
55:31even during acute COVID.
55:32The protein albumin ratio when marker
55:35blood brain barrier are normal.
55:37We did find a higher percentage
55:39of monocytes in the CSF and
55:41individuals that have are still
55:43experiencing symptoms after COVID,
55:44so we're investigating that with more
55:46detailed solely on their studies,
55:47but it may be a hint that there's still
55:49some cellular immune perturbation.
55:51And finally, looking at the neuronal
55:54and astrocyte measures,
55:55we do see elevations in GF,
55:57AP and NFL in a couple of the people,
56:01but we're not seeing significantly
56:03different levels in the CSF.
56:05In these people,
56:06we're seeing a hint to elevations
56:09and CSF S tream,
56:10which is a microglial activation marker,
56:13but again, you can see why.
56:15I feel strongly that before we
56:17report any of this to the world,
56:19we really need more people.
56:20You can see there's a lot of heterogeneity.
56:22Here and we need to understand what
56:24are these patterns are persisting
56:26in larger numbers of people.
56:28Interestingly,
56:28we're seeing persistence of SARS, Co.
56:31V2 antibodies in the CSF and almost
56:33everybody that we're looking at.
56:35You expect the anti spike antibody
56:37on the top to be positive and
56:40everyone who's vaccinated and it
56:42does seem to be positive but the
56:44anti nucleocapsid antibody is not
56:46something that turns positive with
56:48vaccination and that is persistently
56:50positive in the CSF and the blood.
56:52And and most of the people
56:54that we've looked at,
56:55so it's kind of interesting to see that
56:58these antiviral antibodies are persisting,
57:00and we're doing now additional studies
57:02to look for evidence of autoimmunity.
57:04So I'm going to end by a comparison
57:06between two viruses that now I've
57:08been studied for a long time.
57:10Just to give you a sense of
57:11how stars could be,
57:122 might be different than other
57:13viruses that we understand.
57:15So in general acute HIV,
57:17which is really something that
57:18we've been studying for many years,
57:19looking at the first weeks and months
57:22after initial HIV entry into the body,
57:25is associated with robust viral invasion.
57:28This can be detected in the CSF.
57:30This can be detected in the brain,
57:32whereas servers could be 2 does
57:34not seem to have significant.
57:36Viral invasion,
57:36at least it does not seem to be
57:38something that's detectable at the level
57:40of CSF or most brain tissue studies.
57:43However,
57:43both viruses seem to be associated
57:46with immune cell trafficking to
57:49the CNS and immune activation,
57:51so generalized neuroinflammation
57:52I mentioned the fact that
57:54there's changes in siding kinds.
57:56There's also microglial activation
57:59in brain autopsy tissues.
58:02Autoimmune responses do occur in
58:03some people with acute HIV and
58:05it's kind of one of those hallmarks
58:07that you teach residents to look
58:09for acute HIV and somebody with
58:10Jan Murray and emergency room.
58:12But in fact this seems to be something
58:15we're seeing frequently in acute COVID.
58:17And is that because we just have
58:18billions of cases of COVID worldwide,
58:20so we're seeing these conditions more?
58:22Or is there actually something
58:23about this virus that is more
58:25prone to trigger autoimmunity?
58:26And so I think it's a super
58:28important question.
58:29Acute thrombosis and stroke is
58:30a feature of acute COVID which
58:32is different than acute HIV.
58:33So it really relates to something
58:35particular about how this virus activates
58:38endothelial cells and clotting systems.
58:40Nerdy generation seems to be detectable
58:42in some people with acute COVID.
58:45We also see this in acute infection with HIV.
58:48So what are going to be the
58:49long term consequences of this?
58:50I think we don't know.
58:51Clearly HIV is a persistent
58:53infection as I explained the viral
58:55life cycle is one that you get
58:57long term infection of host cells
58:58because the virus integrates its
59:00genome into our host cells.
59:02This is hopefully not true.
59:04Instars kovi 2.
59:05We don't even see initial infection,
59:07so hopefully it's not persistent infection,
59:09but that doesn't mean there isn't persistent.
59:11Antigen presence and I think that's
59:13a really important distinction
59:15and important area for research.
59:16Long term neurologic suquilla is
59:18clearly something that happens when
59:20you set up infection and inflammation
59:22in reservoirs in acute HIV,
59:23but I think the question of whether
59:25or not this is going to be a long term
59:27problem and people with acute stars
59:28could be too still needs to be answered.
59:31So I'm going to end by just saying,
59:32you know, these are the gaps and questions.
59:34Why do we still need to learn?
59:35I think the really pressing questions
59:37now relate to long COVID and these
59:39lingering effects of people who are really
59:41having trouble getting back to work.
59:42I'm having trouble functioning having
59:44very little to turn in terms of therapies,
59:48so we need to understand the biological
59:50mechanisms of these long ongoing symptoms.
59:52One possibility is this is due to
59:55injury occurred during two infection.
59:57I will say I do not think this
59:58is the problem.
59:59I think many people who have severe
01:00:01disease during acute infection recover
01:00:04completely and then some people who have
01:00:06mild disease who don't seem to have had
01:00:08neurologic symptoms even during that
01:00:10time seemed to have persistent symptoms.
01:00:12But of course we need to
01:00:14be studying that more.
01:00:15Another possibility is ongoing perturbations
01:00:17and inflammation and immune responses,
01:00:19and I think this is one of the major
01:00:21hypothesis for long COVID overall,
01:00:23and I think as it relates
01:00:24to the nervous system,
01:00:25we have a lot of evidence
01:00:26that that might be true.
01:00:27Can there be a vascular inflammation
01:00:30or microvascular compromise that's
01:00:31continuing to have impacts on
01:00:33blood flow and brain function even
01:00:35after recovery from acute COVID?
01:00:37And finally,
01:00:38is there a possibility that there's
01:00:40low level viral antigen either
01:00:42in the endothelial cells?
01:00:43Wanting them to blood vessels in the
01:00:46brain or actually in brain tissue itself,
01:00:48and I think this is still an area
01:00:50that needs to be investigated.
01:00:52And I'll end by talking about,
01:00:53you know this is a.
01:00:54This is a pretty vulnerable population.
01:00:56I spent my career taking care
01:00:57of people with HIV.
01:00:58And you know,
01:00:59this is it's.
01:01:00It's been a highly recognized for a
01:01:02really long time that people living
01:01:04with HIV are vulnerable to lots and
01:01:07lots of different social and other
01:01:08kinds of challenges that make living
01:01:11with illness even more difficult.
01:01:12And I would say that there are ways
01:01:14in which long COVID also has a lot
01:01:16of personal challenges for people.
01:01:18There's stigma associated with
01:01:19this condition.
01:01:20There's a lack of understanding
01:01:21a lot of people are.
01:01:22Really frustrated and how a lack of hope
01:01:24that things are going to get better,
01:01:25so I think this is really why we
01:01:27have such a research mandate to try
01:01:29to understand this condition and
01:01:30hopefully see that it resolves over time,
01:01:32but if not to develop their therapeutic
01:01:34interventions that are targeted to
01:01:36and really writing these conditions.
01:01:38So end by thanking all of our colleagues.
01:01:41Thank you so much for the invitation.
01:01:43Thank you in participant to the
01:01:45general study participants who
01:01:46were coming in for these studies in
01:01:48the setting of all the uncertainty
01:01:50they're experiencing and to
01:01:51particularly to Shelly and Lindsey who
01:01:53are Co directing the study with me.
01:01:55I will stop sharing so that I
01:01:57can see people for questions.
01:01:59Thanks so much for the attention.