 Hello and welcome to noon conferences hosted by MRI online in response to the changes happening around the world right now and the shutting down of in-person events We have decided to provide free daily noon conferences to all radiologists worldwide today We are joined by dr. Brent little he is a fellowship trained in Cardiothoracic imaging a member of the division of thoracic imaging and intervention at Massachusetts General Hospital And is also an assistant program director for the radiology residency He is interested in a variety of research topics and has published articles on pulmonary infections long cancer screening Diffuse lung disease and cardiac and vascular imaging a reminder that there will be time at the end of this hour for a Q&A session Please use the Q&A feature to ask all questions and we will get to as many as we can before our time is up that being said Thank you so much for joining us today. Dr. Little. I will let you take it from here Okay, well, thank you so much and thank you to dr. Collins for inviting me to give this webinar today so I'll be talking about COVID-19 a topic that everyone is talking about right now and I just wanted to remind you the views that I have in my talk or my own they do not represent my own institution or any other national entity and I'd just like to start by saying that I'll talk about today COVID-19 and the virus that causes the disease of COVID-19 that is SARS coronavirus 2 interchangeably today so when I say a patient was positive I simply mean they tested positive for SARS coronavirus 2 and They may have a disease known as COVID-19 which is still poorly understood and could be a collection of different diseases involving different body parts and organ systems now, of course, this is a chart of the total number of cases in the world of COVID-19 And you can see that this little inflection point on February 11th. That's the time at which we thought that the epidemic in China might be leveling off but unfortunately because of Growth of cases around the world. This is the curve that we have now that continues to go up over 2 million cases Now there's been absolutely astounding media engagement of this particular topic obviously as a pandemic a very very devastating Pandemic and you can see that this is the total number of articles in the popular press that have been published 4.5 million articles since January 1st and you can see that each of these bars represents one day of articles and the scale is in 100,000 articles so you can see that on some days at the peak We have over 120,000 Articles published per day. These are the media social media engagements of those articles and you can see that that's measured in billions So potentially billions of people have been tweeting and posting first Facebook messages about these very articles So astounding media engagement. This is a friend of mine a thoracic radiologist By the name of Adam Bernheim. You may have heard of him. He's from outside I he was part of the first publication spearheaded the first publications on the imaging appearance on CT of At that time just the coronavirus novel coronavirus 2019 And I cannot remember a time when another radiologist was able to go on CNN live on the news Talk to Chris Cuomo and have a national audience talking about the appearance of anything on a CT scan So just astounding media engagement But just like any other in any other pandemic or important world event There's a lot of controversy about the role of various Modalities and that's been true especially of COVID-19. These are two recently published editorials by some of these Folks are friends of mine good friends of mine on the role of CT and being a little bit critical of some of the published articles and so these are good to reference and I'm going to present what I think is a very balanced account of recently published articles and of the role of CT So what I'm going to do and you can see that one of these is ironically entitled Chest CT detection of coronavirus disease 2019 don't rush the science and so you may argue that actually you do want to rush the science in the setting of a pandemic So today I'll be talking about the role of PCR and CT and about a little bit of the controversy of CT use About the role of chest radiographs as primary modes of imaging. We'll talk about the various CT appearances of COVID-19 and we'll talk about the use of CT particularly as a problem-solving tool and Primarily is not a primary mode of diagnosis Now why is COVID-19 such a difficult disease to deal with well in part because the symptoms can be very non-specific They can range from very typical symptoms like fever cough sore throat to less typical symptoms like fatigue GI symptoms neuro neurologic symptoms including confusion and Changes in mental status and even conjunctivitis. That's been recently reported and as we'll talk about a lot of patients can be Asymptomatic and have absolutely no symptoms as was evidenced by those Passengers and crew of the diamond princess cruise ship and other cruise ships and we'll talk about that in just a second But what is the role of PCR the polymerase chain reaction laboratory test? That is used to assess for the presence of the virus that causes That causes COVID-19 Well, let's explain it this way We can have patients with Presenting with no symptoms or being asymptomatic and then they can become symptomatic Some some never become symptomatic, but many do and then they can convalesce and symptoms can abate but PCR Let's put it this way can be positive with Asymptomatic patients and then can become negative or can be negative even and symptomatic patients For a first test so you can actually be tested and be negative on the PCR Even though you have the virus and your space you are symptomatic You can then test positive a couple of positives on PCR and then even while you're still symptomatic with a virus can test negative again for For SARS-CoV-2 and even when you convalesce when you're discharged from the hospital You can test negative on PCR and it's been well documented recently that you can test positive even days later Even when you're asymptomatic, so the role of imaging in theory if you had a test that was persistently positive throughout all these time points a test like for example CT Where we know that the findings once they appear don't take take weeks to resolve in some cases You could presumably have a test that turns positive at some point if it does turn positive and then it remains positive throughout all these time points so when you miss a Assessing the for the coronavirus with a negative PCR you could presumably still at that time point have a positive CT so that could be one role of imaging meaning we could pick up some sensitivity points throughout this entire time course and Presumably mitigate these missed opportunities at PCR because for example the PCR doesn't have an adequate sample or because the virus is not shedding during those time points And so I think that the interesting thing to learn about asymptomatic versus Versus symptomatic people is from the diamond princess and of course, that's the cruise ship Where the first passenger who disembarks on February 1st from that cruise ship who developed symptoms of COVID-19 caused the entire ship to be quarantined for several weeks and The interesting thing is that this is a so-called pure core board because it wasn't just a random testing of people here Everyone was tested Just about everyone was tested passengers and crew on this ship and for a while. It was actually Listed and it's still listed on the world a meter Site website count of coronavirus cases. It's listed adjacent to countries and it used to be listed as a As the first, you know most frequent site of infection because it had so many patients when other countries did not but now it's listed next to Cyprus and Latvia and Andorra and you can see that there were over 3700 passengers and crew on that ship and 712 of them tested PCR positive for the virus out of those 27% at time of presentation had absolutely no symptoms later that went down to about 18% because then Some percentage of those patients became symptomatic, but 73% showed no symptoms at all In the early period there were seven deaths and so that means that the case for Case fatality rate was only 1.1% which is well above that of the seasonal flu But it's well below that of the 3% that has been quoted by the WHO and by some other organizations So the interesting thing for us as imaging specialists is This what do you think these are three patients from that particular article in radiology cardiothoracic imaging With abnormalities on their chest CTs This one has quite a bit of ground glass opacity a little bit of consolidation is a peripheral distribution This one has just a little bit. You might wonder if this patient is asymptomatic This one might be symptomatic and then we have quite a bit of ground glass peripheral capacity that has a typical pattern here So which of these patients was symptomatic had symptoms Well, would it surprise you that all of these every single one of these patients was absolutely Symptom-free at the time of these chest CTs That's really remarkable when you consider how much abnormality there is on some of these chest CTs So as a chest cardiologist, it's interesting when somebody comes down or calls me in the reading room and says well You know given this sort of pattern and the absence of symptoms. Do you think this could be something else? Well, it could be a lot of things as we're going to talk about in a second It's not a specific pattern, but the one thing I cannot do I can't exclude COVID-19 simply because of the absence of symptoms What about this case? This is the same case. We saw a symptomatic woman And let's remind ourselves from this article. There were a hundred four patients to were included And let's look at the presence of symptoms and the presence of some sort of CT abnormality in these patients What about if you had symptoms? Well, 79% of those patients actually went on to have a positive CT and 21% actually presented with a negative CT What about if you had no symptoms? Well in that case astoundingly 54% of the patients who were asymptomatic without any symptoms actually had a positive chest CT But look at this. There were 46% who had no symptoms and had a negative CT So that means that a negative chest CT does not mean that you do not have COVID-19 And you do not carry the SARS coronavirus to What is the gold standard? Well, I think most people would accept that the gold standard is laboratory testing as PCR and several publications have studied the sensitivity for various forms of sampling of Samples for PCR and you see the results here for pharyngeal swabs nasal swabs and sputum samples And in general the nasopharyngeal swabs or nasal swabs are between 60 and 70 percent sensitive now There are over 300 different types of PCR testings testing going on right now and there's the recently promulgated Rapid testing and all these things have very different sensitivities But they're they're within this this range for many of the tests some are better some are a little bit worse And remember that BAL fluid that is even more sensitive and sputum samples are more sensitive And you can see that that is 93 and 72 percent sensitive in this particular study But we've heard a lot in the radiology literature about the sensitivity of chest CT So wouldn't it be good just to run all of our patients through the chest CT scanner and see if they have COVID-19 or not and a lot of these studies have quoted very high sensitivities this particular study in radiology By these authors is a famous one now and they What is quoted in the media is a 97 percent sensitivity and that's what people were really excited about I'd like to point out that that's not really the whole story of this article You can see that this 97 percent sensitivity figure This is actually out of the first PCR that was done and that was positive in a certain number of patients And that's in 580 patients. There were out also 21 patients who had a positive PCR with a negative CT So what that means is that only a very small number of this group who had a first positive PCR Actually had a negative chest CT That means there were very few false negatives in the chest CT group only 3% But that's not all that we want to know we want to know about the patients who had a first negative PCR and How did CT help us in those patients because that's what we're encountering in clinical practice We can see in this group. What happened was that they looked at this group again, but they didn't look with PCR They did a clinical assessment. They looked again at the imaging They looked at the symptoms and the course and they decided whether they thought it was highly likely the patient had the virus COVID-19 and 33% were possible cases 19% were deemed uncertain So if you look at this group though, remember that this group is not PCR based So PCR was not the gold standard in this particular group. So that's very interesting The other interesting thing and the thing that we really want to know about is What about the patients who had a first PCR that was negative and then had additional PCRs that were positive How does CT are CT do in those cases? Well in this case we had that group and we had only five patients We had 15 patients in that group But five of those about a third or actually one third had a normal CT Which means that a normal CT actually in this group cannot be used to reliably exclude the presence of the virus In fact one third of those patients we relied on CT would have been missed would have been called normal when in fact They had COVID-19. So that's very interesting. Here's another study that's quoted most often And this came out a little bit earlier and it quoted a 98 percent sensitivity for CT in picking up SARS-CoV-2 patients and we had 51 patients in that study. They all presented with symptoms They all had some sort of exposure to Wuhan or to someone who had had contact there And they were done with the PCR that was done within three days of the CT in question and they found That 50 of those CTs were actually abnormal and that's a very high rate of abnormality, right? The problem is that in that study the threshold for calling something abnormal a CT was very low And you can see that this was called abnormal now It is peripheral and it is ground glass and there's a little bit of consolidation So conceivably this could be COVID-19, but remember this could also be a pulmonary infarct. This could be some other type of pneumonia This could be a lot of other things. This could be organizing pneumonia from anything This one was called positive 2. Look at these just small Ground glass and solid mixed attenuation nodules here. These could have been anything ranging from malignancy to any other infectious or inflammatory cause What about this one? This one was also published from their article. This is a single opacity. It's a little sub-solid nodule And this could have been anything so the price that you pay for getting a lot of Sensitivity is a lack of specificity So in this case in this article a lot of things were called positive that probably in clinical practice may not have been thought to be Definitely positive Now I think that there are several points that we all we can all agree on and one of them is that PCR Is the primary method of diagnosis of COVID-19 and chest radiographs if any imaging is going to be done They are probably going to be the first form of imaging at many institutions But these things none of these things are a hundred percent sensitive We can also agree that CT from literature has a higher sensitivity than chest radiographs That's that's something that everyone is agreement with And also we can agree that CT is not the gold standard above PCR because it does miss cases They're actually positive cases and here's such a case This is a completely negative CT and you can see this patient had one of the other organ manifestations of COVID-19 That's been appreciated recently and that's conjunctivitis eye disease What does the American College of Radiology recommend well as most of you know The ACR has guidelines that say that CT should not be used for screening or first-line diagnosis of COVID-19 and that CT should be used sparingly and reserved for hospitalized symptomatic patients with indications for CT That portable radiography could be used when medically indicated and that radiologists should familiarize themselves with the appearance of COVID-19 and The reason for that is going to become even more clear throughout the rest of the talk Now I want to talk about some radiographic presentations Quickly and I'll talk to you about several of these cases This is a male in his 50s who presented and you can see this is a typical rather typical appearance of COVID-19 on chest radiography You can see that this can be used for quick confirmation if you clinically suspect these cases even before the PCR is done or is back It can be used to evaluate alternatives. What if this patient had a huge low bar pneumonia? What if they had a large pleural fusion? What if they had a pneumothorax that you didn't know about that could be used to evaluate alternative diagnoses or additional complications or other diagnoses also to evaluate severity You know is most or most of the lungs involved or only a small portion of the lungs involved that can be relevant for patient care So all those things are reasons why we might want to do radiography as a first step This is the typical peripheral appearance. It's typically nodular and most multifocal It's typically it can be but bilateral it can be lower zone And it's often bilateral Here's another case and in this case instead of bilateral pasties We have a peripheral opacity right here and we don't see much on the right side though On the chest CT that was done for different reasons later You see that this opacity corresponds to the one on the left and in this in addition We see multiple other pasties that were harder to see on the right on the chest radiographed and Again, this is not a specific appearance. It's reminiscent of an organizing pneumonia and many things can have this peripheral different Distribution in the differential diagnosis and some of those things are septic emboli eosinophilic pneumonia all types of other Organizing pneumonia is ranging from cryptogenic organizing ammonia to drug reaction To connected tissue related lung disease can present with this sort of peripheral pattern Here's another appearance that you'll see a lot, you know, you see these peripheral pasties. They're bilateral They're sort of multifocal multi lober. Here's another one. This is actually the same patient who progressed one day later So these are pasties can rapidly worsen over just a short time course Here is a another case that was given to me from my institution And on day one you see these typical bilateral peripheral opacity Some of these are quite round and then on day four they have worsen they become more diffuse You see not only the peripheral pasties, but also some central pasties and more diffuse Opacity so it's uncertain whether in some of these cases a lot of these central pasties could be due to something else like development of ARDS or whatnot This is the CT that corresponds. We have a lot of both peripheral pasties and also some central pasties in this particular case Here is though a variation that unfortunately you're going to see look on day one I'm a chest radiologist and I still don't see anything going on in the lungs here So this is normal on day two essentially normal again So this patient eventually tested positive had few days of symptoms respiratory symptoms and was Positive for the virus. So chest radiographs can be negative especially in the early setting But they can also be negative in well into the course of the disease Chest radiographs are not specific though You can see that two patterns here are very reminiscent of the patterns we've just seen This patient though had H1N1 pneumonia several years ago and has ARDS So other pneumonias can do it especially viral pneumonias and this one has cryptogenic organizing pneumonia or COP From an unknown cause so both of these can cause this organizing lung injury pattern Now what does the literature say about chest radiography before we start talking about CT? Well, this was a nice case a nice series and radiology nice study of 64 patients most of them were symptomatic and 91% of them had a first PCR that was positive. That's actually a nice Sensitivity for PCR. This is higher than some of the other literature on PCR But the gold standard in this was PCR in general So if they didn't have a first positive PCR they went on to do more testing second and third until they got their gold standard Which was PCR so all these 64 patients eventually had a positive PCR, but let's look at radiographs with that PCR positive out of all those patients only 69% had an abnormal first chest radiograph so if you're thinking about using chest radiographs to Absolutely exclude COVID-19 Pneumonia you can't really do that because you're going to miss about a third of the cases in 59% of the time both the PCR first PCR and first chest radiograph were both abnormal and Interestingly though, here's where we can pick up some cases that we might not have seen with the first negative seats PCR We picked up nine percent additional cases or six additional cases that had an abnormal chest radiograph even before that PCR was positive. So it can overall increase the sensitivity of testing, but let's look at the Distributions and appearances of the opacities on chest radiographs in that study Can we use them to deduce anything about what we should be looking at? Well, in a lot of cases we had consolidation We had some hazy ground glass opacities But notice how we didn't have these in all so that means that if you use these things to try to discern Whether something is COVID-19 or some other type of pneumonia, for example, if we take the peripheral predominance that I've talked about so far 51% of the cases had that but that then means 49% did not have that so even these typical findings It's very treacherous to use those to sort of discern whether you're dealing with COVID-19 or some other type of lung pathology or type of pneumonia even You know the bilaterality 63% had bilateral opacities, but then that means that some Relevance other percentage did not have them and then even lower low predominant only 63 were lower zone predominant So when someone asked me how certain are you that a certain chest radiograph represents COVID-19 or not I typically use the peripheral predominance that I talked about to assess whether I'm really really concerned about it But then remember that certainty scores don't mean much when you have such a large percentage that actually do not have this so-called classic or typical finding and This is actually one of those cases. This is a focal opacity It's quite round and I would start to think as a chest radiologist about could this be a malignancy? Is this a mass? Is this in a round area of other pneumonia sort of a segmental? Pneumonia from bacterial infection. This is actually a case of COVID-19 and we'll see the CT of this case in just a few minutes What about this chest radiograph? This is just diffusely abnormal and I see a little bit of cardiomegaly perhaps I'm thinking could this be edema or because could this be some other type of viral pneumonia? Could this be pulmonary hemorrhage? Could this be a lot of other things? It's really non-specific chest radiograph This is COVID-19. It's diffuse. It may even be central. It's COVID-19. So what about this one? This is also COVID-19 another case of just sort of diffuse hazy opacities Maybe some sepal thickening interstitial thickening bilaterally This could be a lot of things a lot of infections a lot of diseases could have this appearance What about this one? This one has almost a low bar occurrence down here This dense retrocardia capacity at the left lung base. This could be aspiration could be a bacterial pneumonia Other types of pneumonias could be low bar and lektosis And then we have some diffuse hazy opacities elsewhere. So is this a case of ARDS? This is kind of a non-specific appearance This is COVID-19 as well and this patient went on to have really a bad ARDS and the lungs are almost completely pacified after the initiation of more support measures and What is the role of CT? Let's look into that. We've talked about radiographs for just a few minutes But let's talk about CT the rest of the time. There are certain benefits, of course And the benefits are when you have a short as your PCR kits or it takes time to do the PCR time to get the results There are some people dealing with days or even weeks of delay It's it's better than that. It's better than nothing, right? It's also more sensitive than just a single PCR alone in some cases And it can be used as a problem-solving tool And it can also pick up unsuspected cases and we'll see a few of those cases in just a few minutes from my files But remember there are a lot of risks. There are patient risks from involving patients and being scanned in rooms that have had other patients who've had COVID-19 There are staff risks involving being exposed to patients with COVID-19 when they may not need to be and there are also false Negatives and false reassurances that one can get from a false negative CT when the patient actually has the disease There can also be false positives What if we call COVID-19 pneumonia on a case and we divert resources away in those cases that they need to be Diverted to some other place in the hospital. So it's not without harms but the ACR does acknowledge that CT should not be used for screening or first-line diagnosis and the reason is we see it We've said is that chest CT can be completely normal. This is an unpublished case from China and We have a negative CT and we know that this patient had COVID-19 56% of the time CT can be negative This is Adam Bernheim study from Mount Sinai Showing that in early symptomatic patients 56% of those can have a negative chest CT and CT as we are seeing is non-specific and we'll get into a few of those cases in just a few seconds But how can CT be used it can be used in a lot of different ways It can be used for primary diagnosis It can be used for screening in some cases if you don't have access to a lot of PCR kits It can be used as a triage measure even while you're waiting for the PCR to come back It can be used to assess complications in known COVID positive patients But it's important to remember that we don't really have studies of the outcomes of the use of chest CT in any of these situations And the reason is that we're in the middle of a pandemic We haven't had to time to study this nor would we expect anyone to be able to study all the factors involved and some of those things that could affect the study and the cohort Groups would be the prevalence of the virus the population health whether you have a lot of smokers or a lot of overweight in your population the medical resource including the The likelihood you have enough PCR kits or CT resources and then also just your imaging resources in general Availability of chest radiographs and CT in general So there's so many other factors No one has really been able to do this to control for all these factors yet and we wouldn't expect that to be done yet So we have to keep an open mind about the use of CT. I think in a balanced perspective Here's a case where I think CT does a really good job and this opacity was found on a radiograph here of a patient Who had a mild cough and so what happened with this patient is that there was a PCR? That was negative the patient continued to have a mild cough a second PS PCR was done a day later And that was also negative and so this patient went to CT to sort of see okay Well, are we dealing with something else entirely or what can CT show us now? I'll show you the CT These are the CT images and we see in the left upper lobe here We see this area of regional opacity and this opacity is Ground glass it has a few areas of consolidation in it It has some so-called septal thickening in it that you could call crazy paving when it's superimposed upon this ground glass This hazy abnormality and it really doesn't it has a little bit of a rounded appearance But other parts of it are just very indistinct. So all in all it was classified as Sort of an indeterminate appearance We said that you know, you need to really look into doing more testing for COVID-19 but knowing that this could also be another type of viral infection or another type of pneumonia or even You know less likely but still possible a malignancy What happened was this patient actually went on for sputum testing and the sputum was positive for COVID-19 this patient had COVID-19 So that's one way that CT can really serve as a problem-solving tool and can push You into doing more clinical testing even after you've had a couple of negative swab results that are PCR negative and so Acknowledging that and acknowledging that some in some settings even CT is even going to be used for primary diagnosis of COVID-19 the RSNA and Simpson Have published this really nice article on structure recording for chest CT and they break things down into the four categories here You can have a typical appearance for COVID-19 and we'll go over that indeterminate atypical or negative That means no features of pneumonia And so at my institution, we were currently using this to report These cases that are asked to be reported on suspected COVID-19 But it's important to know that there can be a distribution of these various appearances that we'll talk about and we Still do not know what percentage are we going to see there are going to be the so-called typical appearance in Determinate atypical and then even normal we may have a situation where depending on our cohort groups that are different hospitals We have mainly typical appearances of CT as it's defined by the RSNA ACR str consensus guideline We may have some typicals and a lot of indeterminates But we may have this situation where we have mostly atypical cases and indeterminate and just a few typical cases And again, this depends on okay What are we using for our first method of diagnosis? Are we sending all the patients who have typical chest radiographs? Or we just never seeing a CT on those patients because it's so typical on chest radiographs or because the first PCR Is already positive that we don't see many typical cases. Are we going to see that situation? Or are we going to see this situation? We just don't know it depends on prevalence of disease around the cashmere group. It depends on the host factors It depends on severity of disease when you see it whether you see it earlier late And it depends on the reason for doing a CT so a lot of variables go into what appearances we might see and So again certainly scores if somebody says well, can you sign a percentage certainly score to that? Are you 25 percent sure 80 percent sure 100 percent sure? I don't know how to answer that question I can just tell you that these are the typical appearances that we see both in the literature and in our practice and then the other ones that are less Specific for the disease So let's talk about some other findings here CT can also be used not only for describing disease but also in known cases of COVID-19 can be used to assess complications and associated diseases now some of those complications remember that Patients who are in the hospital have a higher risk of pulmonary emboli patients who have ARDS have a known higher incidence of pulmonary emboli and So we wouldn't be shocked to see pulmonary emboli in COVID-19 patients And here we have a combination of peripheral ground glass Apasties and consolidation some of these look like that like an organizing pneumonia So this would be a typical appearance for COVID-19, but look at these areas these look more wedge shaped down here So they could be due to pneumonia, but they could also represent pulmonary infarcts And that's actually what they represented on this pulmonary embolism CT that was done for suspicion Pulmonary emboli based on a change in clinical status So you see pulmonary emboli right here and we saw them in other lobes multiple other lobes on the chest CT So we're not doing chest CT in my institution as a routine method of diagnosis We're doing it as a problem-solving tool and to assess other suspected complications in some cases Now what are the patterns that the RSNA guidelines can sense this guidelines talk about? Well, let's start with the typical patterns because that's what I'm talking about right now That's what I've been showing you the typical patterns are peripheral Ground glass apasties and with or without consolidation They tend to be round they can be round ground glass apasties like this and You can also see reverse halos in some of these cases and this is a sign of some sort of organizing pneumonia lung injury in many cases and So how can CT help us in when we're talking about these categories? Well, this is a very confusing radiograph some of you May say it's relatively normal, but look in the right upper lobe This is a very hazy opacity very hard to pick up at the lung periphery, but something is abnormal here This needed clarification based on a confusing clinical symptoms based on confusing radiography And so a CT was done. Look at the CT shows bilateral peripheral ground glass apasties And look at the axial images. We have bilateral summer round very well defined Some have enlarged vessels within them the so-called thick vessel sign that has been described in covid-19 pneumonia and so this case is a typical case According to the our typical morphology according to the rsn a acr guideline Here's another case that would be classified as a typical morphology You have peripheral ground glass apasties band like apasties that can be seen in organizing lung injury And this is the case that I started the presentation with another case of covid-19. So a very typical appearance by those guidelines What about this case here highlights another role for CT? This patient came in for a routine cancer follow-up CT scan and my fellow called me up and said What do you think about this case? Are you concerned about covid-19 and she was the first one to suspect covid-19 on this particular case And we can see that we have band like apasties within the lung periphery We have ground glass apasties multifocal and this case would be classified as typical morphology for covid-19 on chest CT and We alerted the Physicians who referred the patient and the patient was diagnosed with covid-19 via PCR within a day And what about this case we have another typical appearance of covid-19 pneumonia We have multiple ground glass apasties. We have a so-called reverse halo sign here And the reverse halo is this it's a band of consolidation with ground glass apacity in the central portion of it So it's sort of ground glass that's within the central portion And surrounding that is a band of consolidation. So it's a reverse halo not a halo sign So this is in the rsna categories as a typical appearance Morphologically here's a rare case of a halo sign Sorry reverse halo within a reverse halo in a teenager courtesy of um michael chung and adam bernheim Here's another reverse halo sign or so-called atel sign very round apacity You have this linear consolidation around it. It corresponds to the radiograph. I've already shown you This is a case of covid-19 with a typical peripheral appearance with a Reverse halo sign Here though is another sign that's called the halo sign that can be seen as subset of patients And this is a nodule with surrounding ground glass apacity and remember this is not specific for covid-19 this can be seen in a lot of infections including other viral infections Including invasive frontal infections and other things But even the typical patterns are not specific for covid-19 These are patterns here that are actually not covid-19 pneumonia. This one is um h1n1 uh pneumonia so all sorts of Viral pneumonias with with or without rds can have this appearance And this one has an appearance that is on that of an adenocarcinoma and this is not a covid-19 patient Patient so the question has been asked in the Literature how well do radiologists do in distinguishing cases of covid-19 from other types of pneumonias? Well, um, there was one one uh article recently this one. There was a nice article They claimed a very high specificity of ct and what they did was they took Part pcr positive patients 219 Covid-19 patients and they matched these these cases with Other non covid viral pneumonias They chose those pneumonias from the medical record from reports of the radiology reports that had things that were consistent with pneumonia and the Impression or impressions that said highly concerning or or very concerning for pneumonia And they had a viral panel to go along with these where the viral panel was positive And that's those are the 205 cases they included Now they had a variety of seven different radiologists Read those cases and tried to discern which ones do you think are covid-19 or which ones do you think are non covid-19? Cases and they claimed a 97 accuracy and I can say that while You know, some of these cases are very very suggestive of covid-19 the real accuracy in Medical practice is probably lower than that and the reason is that some of the it's hard to prove that some of these cases were not Selected with some degree of selection bias because they were deemed very consistent or highly consistent with pneumonia in the reports and a lot of pneumonias viral pneumonias are not going to be very Maybe atypical On chest CT and so simply selecting for these patients may have led to a little bit of a selection bias in this case Also, I can show you that let's look at some of the reader Variation in some of the cases that were called covid or not covid this case I would have thought would have been a typical pattern for covid-19 Because it has this band-like peripheral pasty some grand glass It looks like an organizing sort of lung injury pattern And this was called non covid by one of the readers This case I would have thought well, this is more central on distribution We have a little bit of peripheral pasty here, but some of its central A lot of central sort of thought could this be indeterminate actually and this was called covid rather than non covid etiology and look at this case this has a Reverse halo sign and a reverse halo sign here and I would have thought well, you know This has been reported a lot in covid-19. So this is probably a Typical case, but this was actually called non covid In the case series. So well, this is a very good article and it exposes the fact that we Do fairly well when we look at cts and try to predict. What is a typical appearance? There's still more work to be done, but very good article What about indeterminate patterns? It's that second rsn a category Now this includes things that are not the typical appearances like talked about they can be diffuse ground glass non peripheral A few small round glass pasties that do not have rounded morphologies and things like this This is a case from my institution where you had ground glass Really just diffuse ground glass. We had some peripheral some central Perhaps most of its central in the upper lobe here and on the radiograph. It doesn't really look that You know, it could be a little mild edema. It could be some other viral pneumonia. It doesn't look that typical So that would be classified as indeterminate This one on the radiograph just looks like it could be a lot of Various things could be edema could be another pneumonia could be hemorrhage. This is the ct So instead of having a mainly peripheral predominance of pasties We have a pasties that really have no particular distribution some are central some peripheral I will say that there's some hints of band like Opacity here that could be viewed as an early organizing lung injury pattern. So maybe that's a tip off but still this case would probably be Graded as indeterminate by the rsn categories And here's another case that would have been graded as indeterminate And it improved just one or two days later and two pcrs were negative So presumptively this is a negative case with negative lab values and it rapidly changed Which probably says there was something else altogether And what's that third category? So the last category in the rsn a guideline is reserved for things that are usually not COVID-19 related so things like cavitation really not reported in pure COVID-19 involvement What about tree and bud well-defined central ovular nodules? Those have not really been reported low bar consolidation And segmental consolidation also septal thickening with a pleural effusion And what about lymph adenopathy and pleural effusions themselves? Well, these were were really termed as atypical in some of the early reports, but since then quite a few groups have published on these two things And a lymph adenopathy alone and a pleural effusion alone. They're so common Due to edema and due to reactive things And even due to COVID-19 that they're not deemed really so atypical in the rsn a guideline This is a case that I showed you already that had some mild lymph adenopathy And here's another case I showed you already of COVID-19 that has some mild lymph node enlargement So it can be seen. It's not incredibly unusual These though are cases that can be deemed atypical. You have this inter lobular septal thickening Fissural thickening looks like interstitial edema. You have pleural effusions bilaterally So this would have been graded atypical for COVID-19 pneumonia And here we have some well-defined central lobular nodules You can talk about some of these as tree and bud nodules. You have a little septal thickening So this would have been graded as atypical for COVID-19 Now this case is another case that you would say is atypical. You have this round area this mass-like area of consolidation And you also have tree and bud nodules around it And this is a patient who went on a vacation to California This is a case from a couple years ago and came back with a case of acute coxie So fungal infection there What about the exceptions though? There are some cases that do have those atypical features It's important to know that they cannot completely exclude the presence of COVID-19 and nor should anyone expect them to This is a case and from Seattle that was published online As a new and journal of medicine resident case where you had some central lobular well-defined tremba nodules And this case actually was thought at first to be something else like an aspiration or some other viral pneumonia and turn out to Be COVID-19 and not to say that this could not have been something else superimposed upon COVID-19 But this patient was a patient who had COVID-19 Now in those cases, why might you see a variety of patterns? Why might you see some patients with Both an atypical pattern In COVID-19 and also typical patterns. Well, one reason is that in the published and unpublished literature right now A lot of people are showing that there is not an insignificant percentage of other viral infections associated with COVID-19 and you can see that in this case Almost 22 of the group that had COVID-19 And were positive for SARS coronavirus 2 actually had other viral infections So some of these combination cases may be co-infection with other things But, you know, in spite of the value of these RSA categories, I think that we're going to uncover more and more cases that are very problematic Now this case almost looks like it should be deemed atypical for COVID-19 But you can notice that it's very focal. It looks like a round area of consolidation. Maybe even a mass here If they fall into patterns with very mild disease or blends of two or more different patterns I found that those are hard cases because the patients that have mild disease Are less likely to be symptomatic and they're more likely to be something else like either another type of pneumonia Or even just a small amount of aspiration or some something else that is infectious inflammatory Now the blends of two different patterns can be misleading because there can be an atypical pattern blended in with the typical patterns So what do you do? It's important to realize that in the rsa guidelines to qualify as atypical You really need to have the absence of typical features Or absence of indeterminate features in order to qualify as atypical So just because you have a blend of patterns if you think that you're seeing a typical pattern Technically that should be graded as the higher category, which is going to be typical rather than atypical And we went over the typical appearances just a few minutes ago Now what about this particular case I showed you? Well elsewhere It has things that look more like they could be in keeping with just a typical appearance. We had multifocal round ground-glass opacities with some areas of solid Opacity or consolidation. So that would sort of bump this up to a more typical appearance What about this one? This one's hard to gauge because this was the only thing This is courtesy of dr. Chung and dr. Bernheim who again were part of the first group to describe findings on ct of COVID-19, but this is the only thing that was on the ct and this is a COVID-19 patient What about this case? This case is confusing because the patient presented with three days of diarrhea and bloody stool and no respiratory symptoms no long symptoms whatsoever And you can see the major abnormality here is not going to be anywhere else But down here in the right lower low, but you can see that this is the predominant abnormality You have a area of nodularity solid nodule here with adjacent ground-glass It looks sort of like a halo sign And in retrospect, you can go back up and you'd see a little few areas of subtle ground-glass And nodular consolidation in some of the other areas of peripheral lung, but definitely confusing case because of its relatively confined nature to the right lower low and presentation with GI symptoms so gastrointestinal symptoms can present in up to 50 of these patients at presentation And just this is my last slide actually and just to see how well you're paying attention to The talk and the rsnc categories here today I'd like you just to look at these patients. So we have one patient two patients patient three patient four patient five And so five different patients and what categories of rsnc consensus guideline would you apply to these cases? So let's just start to look through them. So this one has peripheral ground-glass Has some so-called thick vessels or enlarged vessels within those peripheral predominance bilateral So I would say well, this is typical pattern, right? COVID-19 pattern. This one much the same It looks like a so-called typical pattern with this peripheral opacity Some consolidation some ground-glass Let's turn our attention to this one. This one has almost a Semi-low bar consolidation appearance here. That would be atypical We have some clustered central ovular nodules that look like they're tree and bud nodules Which would suggest that they're atypical for COVID-19 and so I would term that atypical What about this one again peripheral ground-glass a little bit of septal thickening? So this is great so-called crazy paving where you have septal thickening and ground-glass opacity This has been reported commonly in COVID-19 So this would be term typical and this last case well This one looks more atypical because you have this sort of central predominance You have some clustered central ovular nodules. Some of these look like they're tree and bud And so this looks like either an indeterminate or even an atypical pattern Uh, so how well did you do if you said that this patient has COVID-19? This one has COVID-19 This one has COVID-19 you would be absolutely Incorrect because I'll tell you none of these patients on this page Has COVID-19 pneumonia. These are all old cases. These are from a couple years ago This one is influenza pneumonia. This one is h1n1 pneumonia. This is para influenza pneumonia This is rsv pneumonia and this is hontovirus pneumonia Not a single one of these patients who has very typical patterns Of lung disease has COVID-19 pneumonia And so I think that underscores the fact that we have to be very humbled by this disease We don't quite understand it. It has a lot of appearances on ct and on chest radiographs that are atypical Or that are nonspecific that overlap with other diseases And we're still learning every day about the use of imaging and the disease And I encourage you just to follow up read the literature appreciate the imaging appearances of COVID-19 I've talked a little bit today about the role of imaging and role the major role of pcr as a gold standard And I hope that you appreciate some of the limitations and some of the benefits of the literature that's come out In the midst of this pandemic, so I'd like to thank you for your attention and I'll turn it back over for Any questions that we can try to address Perfect. Thank you so much for your time. Dr. Little today And thank you probably for joining us in this new conference before we move into our q&a section Just wanted to remind you that this conference will be made available on demand on mri online dot com In addition to all previous new conferences and on monday We will be joined by dr. Barbara hamilton for a new conference on trauma in the community Dr. Little if you could please open the q&a feature and answer a few of the questions that are in there that would be great Okay, um, let me just start from the top here and um, there's a question What is the method for confirming recovery? Is it possible to diagnose recovery without positive pcr test? Like in asymptomatic patients and how long does immunity last? um, well, I think those are the um, you know the million dollar questions and everyone Around the world has teams working on the question of how long does immunity last that's very relevant to Vaccine questions and as a reminder. I'm a radiologist. I'm not an immunologist and um, you know I know that a lot of very very smart people are working on that right now that question What is the method for confirming recovery? Well, Technic it is typically clinical and not imaging it is Both pcr and symptom recovery and time And so imaging does not really at the current moment play a role in assessing recovery Now, I need to say that one interesting topic that has not been addressed completely in the literature yet is um, is there lasting lung damage in patients who have recovered from COVID-19 and we know that there is some from some of the papers. There is some residual You know Reticular and band-like abnormality in patients who are several weeks out or a month out But we don't have long-term follow-up yet. So that's one thing to look for So let me Let me look at some other questions here How can we one of the questions is how can we differentiate between? Cop, which is cryptogenic organizing pneumonia and COVID-19 Well, I don't think you'll be shocked after the whole talk to know that there is really no way to distinguish between them imaging wise Because many of these cases look like an organizing pattern of lung injury That's really not specific and it can be seen in a drug reaction With organizing ammonia it can be seen in connective tissue related lung disease with an organizing appearance It can be seen in other viral pneumonias with an organizing parent. So Really the gold standard remains laboratory testing such as the pcr for the the virus So let me go to some other Fine questions here When there are chest CT findings, does it mostly imply the patient has pneumonia? Infection of the lower respiratory tract or ARDS And if the patient's symptoms are limited to Upper respiratory tract infection Shouldn't the CT be negative? Isn't it thus straightforward that CT can be negative in some patients with COVID-19 and shouldn't be relied on? I think that's a great set of observations and questions because It is in part what's going on here. Remember that the um COVID-19 is not just a single organ disease It's a multi organ disease So we shouldn't be surprised that there are no lung findings in some of these patients We should not be surprised from the literature if there are patients who predominantly have GI symptoms or upper respiratory symptoms who do not have findings on CT So that is very true. So we've got to look at I think this is a disease that we don't know enough about and it's multi organ So in some cases even though a patient is positive for SARS-CoV-2 There may not be lung findings Uh, let me look at another question here How do you distinguish? Symmetric posterior parallel bands from compressive atluxes Well, in general, I would not feel confident in suggesting A typical so-called typical pattern or suggesting that perhaps the patient may have COVID-19 Unless I saw, um, you know other findings within the lungs one band within the lung periphery Could be due to anything could be due to prior fibrosis could be due to atluxes compressive atluxes as you say could be due to other things. So I would want to see more to be confident about even suggesting That the pattern falls into a typical pattern. Let me go to the next one Let's see As the role of micro embolic disease becomes greater concern in COVID-19 infected patients Might the peripheral distribution be related to ischemic disease? Um, oh, this is from a friend of mine, uh, evan stein. Thank you for the question That is a good question and that is again a million dollar question Some of the early autopsy studies are showing Some interesting vascular findings in the lung periphery and showing some thrombogenic States throughout the literature in general. So a lot of groups are currently looking at that question So that's a very good question. It's unclear right now whether there are higher rates of Thromo embolic disease for example or thrombotic phenomena in COVID-19 then there are in general in other cases of ARDS Um, and then there's another question second quiz case has ground glass opacities in a central distribution Why is it typical then? Well, um, I may have misspoken, but I meant to say I thought my I mentioned there that that's a reason why I would not think it's a straightforward typical case When you have a more central predominance of opacities That sort of indicates that it's an that is a At least an indeterminate pattern. So that case was meant to show that That we probably shouldn't have suggested that that was COVID-19 A typical pattern Let me look at some other things Um Let's see in my routine CT reporting for suspicion of COVID-19 If the PCR is not done I am getting to see CTs where long practical changes are typical of COVID And associated with pleural fusions and lumped adenopathy How confidently can I report typical pattern in the presence of pleural fusions and lumped adenopathy? that's a very Good question and a very difficult question Just remember though that just having pleural fusions and just having lumped adenopathy That can be seen in a lot of different conditions So I think you have to look at in the setting even of COVID-19 So remember that these patients they they come to us with pre-existing cardiac conditions in some cases pre-existing diseases pre-existing lumped adenopathy They can even have some lumped adenopathy lumped enlargement from COVID-19 So I would just say if you're seeing Parental changes that you think Could be typical of COVID-19 Don't let that don't let the presence of pleural fusions completely dissuade you from Actually mentioning that and being concerned about COVID-19. We still don't have large-scale data on that on In our cohorts in the united states on the incidence of pleural fusion in patients because they have so many other Coexisting diseases Here's another one Most of our cases come from rheumatology departments where the Patients already have basal lung interstitial disease Is there any particular pattern to rule out COVID-19 and them? and uh, I would say the short answer is no Because um COVID-19 can have a pattern that is identical to Organizing pneumonia and connective tissue related long disease I will say though that in my experience. I've had um a few patients now who've had some pre-existing Printable findings that are peripheral that are due to known connected tissue related rheumatologic disease And then have come in with new capacities that are peripheral And in those patients when you have a patient has chronic disease and then suddenly has this new peripheral pattern Unfortunately, we've already had several cases in which we've we've had tests positive for SARS coronavirus 2 in those patients. So things that are new. I would say are important there So let me get on to some more questions Let's see. How do you diagnose COVID-19 in elderly patients with underlying interstitial lung disease if there are no prior studies? So that sort of is a similar question It can be harder in the setting of background lung disease to make Diagnoses, but again, I would say that if you're doing comparison scans if you have a lot of scans for comparison Just compare. Are you seeing a lot of new? Epacities that are concerning that so other peripheral or or that are even not peripheral and that just looked like pneumonia That's how you would diagnose any sort of new disease, right? So you want to compare older ones you want to discern what is old from what is new And then go from there and use your clinical judgment Let's see. Is there a radiographic lag between clinical improvement and imaging clearance? Well, that's another question That we need to know more about but in general there can be some dissociation, right between clinical symptoms and imaging As I showed in the cases that were published in radiology cardiothoracic imaging from that nice paper On the diamond princess cases so It's unclear whether serial imaging is actually a good thing to do or not In fact, some people would say do not do serial imaging unless there's a really good reason to To do it. So it's really unclear right now How much of a lag there is between clinical improvement and imaging clearance? It's a very interesting question for more research Let's see Let me look at some more questions here Okay, so here's a good one. Do you use typical or atypical terms in your chest? X-ray reports So that's a very good question If you look at the data there that I presented from chest radiographs and how many turn out to be peripheral and basal predominance and bilateral you find that a lot of them are actually not a significant percentage Approximating 50% do not actually have what you would call a classic appearance on those radiographs So I really I hesitate To actually put typical and atypical in in my chest radiograph reports because I fear that People may use when they don't see that they may use that as an indication that I'm not thinking as much about COVID-19 when in my experience with now looking at hundreds of chest radiographs with COVID-19 patients patients with COVID-19 I've seen quite a few quite a few both normal presentation radiographs and also very atypical radiographs that look like they're central Or very subtle capacities or just a single opacity. That's unilateral Even things that look like masses or nodules and when you when you add up all those atypical appearances I really hesitate to dissuade someone from thinking about COVID-19 Pneumonia, but on the phone I can I can you know, I sometimes say well, you know, this is a Appearance we see a lot in COVID-19 and my reports I can say There are capacities which have been described in COVID-19, but I I hate to use to my knowledge. There is not a multi Organization consensus yet on using the terms typical or atypical in chest radiographs yet Let me look at some more cases here What about what about other system involvement cardiac cns involvement and etc So that's one area that has been investigated. It's being increasingly investigated now. There are reported cardiac findings. There have been reported myocarditis cases There have been reported arrhythmias. There have been a lot of other Multi organ involvement cases cns cases have been reported I have not been specializing in looking at those areas of involvement But you can look to the literature and you're going to find some early cases there in some case series And some studies that are looking at those other findings Here's another question. Are there other conditions that demonstrate halo signs or reverse halo signs? Yes, there are and that's one of the problems here because the imaging is not Specific, so let's take the halo sign first. So the halo sign Can be seen in other infections including other viral infections Um, it's been described well described in fungal pneumonias including invasive astrogelus and ucor mycosis In other pneumonias Halo signs can also be seen in things like malignancies adenocarcinomas There are a lot of different things that halo signs can be seen in and and That again is a central solid nodule with surrounding ground glass and also the reverse halo, which is the consolidation on the outside and the ground glass on the inside that was classically Described in organizing pneumonia that was actually not due to infection and can be seen in organizing pneumonia due to drug reaction due to connected tissue disease related disease And can be seen and also in other infections including invasive fungal infection And so these signs are not Specific the halo sign or the reverse halo sign. They can be seen in a lot of different things and Let me see. I'm just going down the Oh, um, do you do you recommend point of care ultrasound for first line imaging? That's a topic that is being studied a lot right now and some groups have published on that I haven't been following that literature as well as the ct and and radiographic literature But I know there's a lot of excitement about that However, let me just give you a caveat about point of care ultrasound Remember that whatever is true of ct is likely going to be true of point of care ultrasound Because if you have a negative ct You can't exclude the presence of at least the virus and the patient So you can't by therefore you can't really use point of care ultrasound to exclude COVID-19 if you're if you're amused to do that so but I know there's a lot of excitement about the findings on ultrasound of of this disease Let me get to another question. How long do imaging findings persist after Recovery would you expect them to persist after clinical recovery? This could make the utility of ct limited to assess recovery Especially when false negative pcr is suspected If I understand that question, it's just asking, you know, the time force of findings. Well, we know that Some findings can persist into recovery and the question is How many of them persist? What's the severity? Is there any lasting lung injury? Those questions are not completely understood. The answers are not completely known right now But we do know, you know, as the cases I've showed you you can have patients who are asymptomatic who have ct findings So it wouldn't be shocking if there were a significant lag in complete resolution of the ct findings There's a lot of there are a lot of questions here about pulmonary embolism associated with COVID And the jury is still out about that Remember that any hospitalized patient has Especially those with Acute lung injury There are studies showing in other pneumonias that the incidence can be elevated of pulmonary emboli Now I can tell you that some of this strategy here Obviously the D dimer and other inflammatory markers can become quite elevated in a lot of these patients. So our teams in some cases are doing some ultrasound of these large remedies You know by the the bedside and trying to not to do as much PECT because of you know utilization of those rooms and all the Issues that go along with that and also because You may pick up some dbt's and other Thrombotic phenomena that you can pick up in the and by the bedside and not have to move the patient to the scanner So but the word is still the word is still out on that. So it's a very It's a it's a good topic for investigation right now Um, okay. I think I may just I'm kind of running out of breath here So I may kind of wrap this up. Oh, let me just look at this one other question Any reason for the pasties to be rounded in this case? I think the word is still out on that as we get more Pathologic proof of what is actually going on in terms of the lung injury that's going on I think that we're going to know more about why some of these pasties are rounded Um, you know, again, uh, a lot of this is an organizing lung injury pattern So whatever the reasons for organizing mode to look the way it does. They're likely the same reasons And a lot of that has to do with the actual histology of the small airways involved and The lobular structures of the lungs and that peri lobular structure of the capacities Um, and but I would I want to take this question to actually note that the reverse halo signs, you know, those Areas of peripheral consolidation with internal Ground glass a lot of those are actually incomplete reverse halos They're only a half halo or part of a halo So that's an interesting finding in these cases as well But it still looks like an organizing pattern. So but anyway, I think I will wrap that I'm sorry I didn't get to get to all of these wonderful questions But uh, thank you very much for your attention. I think I'll wrap things up now. So Perfect. Thank you so much dr. Little for your time today We appreciate you being a part of this new conference and thanks to all of you for participating in this new conference Again, this will be made available on demand at MRI online calm in addition to all previous noon conferences Please follow us on social media at the MRI online for updates and reminders next week schedule Will be posted on the website within the next day or so So please register for those and we look forward to seeing you again. Thank you and have a great day