Pitfalls: Chest, Part 2

Introduction to the Presentation

In this presentation, I am going to deal principally with pulmonary embolism diagnosis. Probably two thirds or so of the talk will be in regard to pulmonary embolism and then the remainder, some miscellaneous interesting cases that I have missed in the past. This, the space shuttle flying over the Golden Gate Bridge on the way to Los Angeles, not too long ago.

Examples of Pulmonary Embolism

Okay, this is a nice example of a patient with pulmonary embolism. We see a number of clots filling a number of different vessels, and here we see a clot that's totally impacted within the pulmonary artery with a little bit of arterial dilatation, a very common finding that's present. Here's something we would call the donut sign where the acute embolus is centered in the vessel. And then contrast is around its outside a very typical appearance, I guess you could say. This is the railroad track sign where you see a clot along its length within an artery with contrast on either side, but again, very typical appearances and findings for a patient with acute pulmonary embolism.

Pitfalls in Pulmonary Embolism Diagnosis

Now, there are a lot of pitfalls in pulmonary embolism diagnosis. First of all, it's a very common study and the more common a study is, the more likely you are going to screw it up somehow. The clinical diagnosis of pulmonary embolism, at least in my experience, is quite poor. They get these studies in patients who really have a very low likelihood of pulmonary embolism. So there's a good chance you may make a misdiagnosis of PE when it's not present. In my experience, the dimers don't help very much in screening. One of the best negative predictors for the diagnosis of pulmonary embolism is a positive D dimer. CTO is overly utilized quite clearly. Most patients don't have PE who have PE studies, and usually PE is considered to be an indication for treatment. If you see a pe, most algorithms indicate that the patient should be treated, but I'm not sure that's the case. Anticoagulation certainly has risks. The diagnosis is based on a very simple finding, the diagnosis of pulmonary embolism, you're looking for a clot in a vessel, but there are a number of different things that may mimic that appearance. Technical factors can do it. Vascular abnormalities other than pulmonary embolism, interpretive abnormalities, PE mimics, there are lots of different things that can lead you to make a misdiagnosis. And I'm going to talk about some of those and I'm going to finish my discussion of pulmonary embolism diagnosis with the question, is treatment always necessary? And this is something that's not entirely clear, but I think the answer is no, it is not always needed.

Window Settings and Visibility of Pulmonary Embolism

Okay, let's talk first about window settings and visibility of pulmonary embolism. Usually if we're going to be reading a CT for a diagnosis of a vascular abnormality, we're gonna pick a window that's maybe a mean of 40 and a width of four 50, something like that, a so-called mediastinal window. But that window setting is really not optimal for the detection of pulmonary embolism. And I usually either raise the window mean here, 204 50, or make the window wider, 700 ho field units here. And that can allow pulmonary emboli to be seen when otherwise they would be invisible. And I'm not exactly sure how I do this when I'm doing it, but I take the mouse and I wiggle it around until I start to see pixels within the contrast column. And when I start to see pixels, I know it's a good window for pulmonary embolism diagnosis, whether it's a higher mean or wider window, I sort of move the mouse in a circle and at some point in time it gives me a good window.

Necrotic Lymph Node Mimic

Okay, so here is a patient with renal cell carcinoma, a rule out pulmonary embolism. This is a scan with a rather contrasty window. It actually appears contrast here on the screen that it is in real life, but it looks as if we're seeing a donut sign there, a clot in a vessel with contrast around it. But by changing the window and stretching it out, you can see that this actually represents a necrotic lymph node in the hilum, something that's not uncommon in renal cell carcinoma.

Edge Enhancing Algorithm Artifact

And here is a patient with a scan that seems to show lots of examples of what you'd call the donut sign or the railroad track sign where there is a low attenuation thrombus in the middle of the lumen and then contrast on the outside. But this in fact is not even an enhanced CT scan. And this is an example of a CT that's reconstructed with a very edge enhancing algorithm. So first of all, here it seems as if we're seeing filling defects within the vessels, but if you look at edges here between lung and soft tissue structures, you can see this dense white line and that's overshoot at the interface between two very differing attenuations. It's something you get with an edg hanen edge enhancing algorithm. And when you do PE CT scans, it's really pretty important to use a relatively smooth algorithm or an algorithm that does not have this problem with edge enhancement. And this is the same scan reconstructed with a better algorithm without so much edge enhancement. And you can see that those are normal looking vessels.

Missed Subsegmental PE

Now, here is a scan that was read by me as normal, and I probably would read this 10 times in a row as normal, but my fellow was reviewing scans from the night before and he was looking at this and he noticed that here we see this nice vessel and then at a slightly lower level, the vessel looks smaller and pacified. And then he pulled out here a MIP reconstruction and you can see that there is a small vessel there with a clot in it. A case that I called normal and clearly shows a subsegmental pe. However, it's not going to change the way I read scans. I'm still gonna miss this case every time I read it. I read PE studies by looking at the axial images. I don't look at reconstructions. It takes too much time and there are too many artifacts that can be introduced by the different reconstructions. And as you will see as I proceed through this presentation, missing a clot like this is not necessarily a bad thing. But in any case, in some patients you can see clots with reconstructions that are harder to detect on the trans axial images.

Studies on Reconstructions

Now, here are a couple of studies that have looked at this question and quite clearly MultiPlan or reformation, sometimes helpful in questionable cases, the Fleischner society is stepping into the fray again, and they have decided that post-processing generally is unnecessary in clinical practice when you're looking at PE studies, and I would certainly agree with that here.

Opacification Issues and Technical Pitfalls

Here's a study where we see an odd combination of excellent opacification of the superior vena cava, excellent opacification of pulmonary veins, but poor opacification of the pulmonary artery. So we see structures on the right side that are opacified and then structures on the left side of the circulation that are opacified, but the pulmonary artery in the middle is relatively pacified. Now, what causes this, it seems hard to explain that, but it's not something that is uncommon. It tends to occur when patients take a deep inspiration prior to the scan and or do a Valsalva maneuver during their breath hold for the scan. And if you see this combination of aorta, opacified and vena cava opacified but not pulmonary artery, then you can have them practice not doing a val salva not doing a big deep inspiration prior to the scan and repeat the scan and you'll get good vascular opacification. So this a technical pitfall that can be avoided, although it often will require a second injection.

Timing of Contrast and Distinguishing Arteries from Veins

Now, when you obtain a pulmonary embolism study, you can time the contrast or time the scanning to opacification of the pulmonary artery or to the aorta or left atrium at one hospital where I read cases, it's time to the pulmonary artery at another hospital, it's time to the left atrium. I prefer timing to the left atrium because it avoids a number of problems and can you can allow you to diagnose things you would otherwise miss. So here's a patient with a pe PE study time to opacification of the pulmonary artery, and it's certainly very nicely opacified, but the big question when you read the scan, this scan at the lung basis, which of these is an pacified artery because of clot? And what is an pacified vein just because of timing and it can be difficult to make that determination.

Now, I have a couple of hints for you. If you're trying to distinguish arteries from veins. Here's a patient who had a PE study and scans near the lung bases. There is a vessel, and I guess you could wonder if that's a pulmonary embolus in that vessel, but there are few ways that you can figure out actually that that's a vein and not an artery in something you don't have to care about. PE versus pacified vein number one in the lower lobe, in the lower lobes, veins tend to lie in the plaintiff scan and you see them along their length while arteries are seen in cross-section as just basic anatomy, this vessel is seen along its length. Number two, arteries are associated with bronchi of similar size and shape. Arteries and bronchi travel together and veins are alone. This vessel is seen by itself. There's no bronchus nearby, so it's likely a vein. And on the other hand, you can see these three vessels are oval and shape and adjacent to them are bronchi of similar size and shape. Those arrows didn't come up very well, but they're bronchi adjacent to those. And number three, veins may have thin perpendicular branches while arteries branch in a y shape pattern. And if you look at that vessel with the question mark, you can see these small perpendicular branches coming off it. Those are vein branches that go into interlobular, septa and marginate pulmonary lobules. And that's a way that veins branch in arteries don't branch that way. So there's several ways that you can determine that this pacified vessel is a vein and of no concern, but you can avoid the problem altogether by timing to the left atrium.

Paradoxical Embolization Case

Now this is a patient I saw at San Francisco General, a patient who came into the ER with syncope and rule out pulmonary embolism. And you can see quite clearly there are large pulmonary emboli in both inter lobar pulmonary arteries. And there's right ventricular dilatation with displacement of the ventricular septum towards the lv, which indicates high right-sided pressure. So this is looking really good for pulmonary embolism right now. These are scans higher up that were done as part of the pulmonary embolism study. And in this patient you can see the timing was to the left side of the circulation. Aorta is nicely opacified and what we can see here is clot. That looks like a typical pulmonary embolism in the aorta and up into the left carotid artery. So this is a patient who has pulmonary embolism with paradoxical embolization to the left carotid artery. And that is what was explaining his symptoms. And further, if we look at scans through the heart in this patient, you can see this little sort of collection of contrast in the region of the inter atrial septum in between the RA and the la and that is the appearance of a patent for Raymond oval. If you reconstruct that and the sensitivity and specificity of this finding, of seeing this little collection of contrast about 70%. But if you reconstruct this here in some sort of a oblique sagittal plane, you can see that little collection of contrast and in relation to it, this little flap of tissue and that represents a little flap that should be closing off the foramen, which is not in this patient. And here you can see the clot up in the aorta again. So here a patient with pulmonary embolism, high right sided pressures, patent for Raymond oval and paradoxical embolization that explains his, I can't remember what his symptoms were, d dizziness or passing out or something of that sort. So if you have a ification of left-sided structures, you can sometimes make other diagnoses that are explaining the patient's problem.

Flow Artifacts Mimicking Pulmonary Embolism

Okay, here is a patient with a pulmonary embolism study and you can see that there's a relatively a relative decrease in attenuation within the intra lobar pulmonary arteries that mimics the appearance of pulmonary embolism. Although I have to say this again, is more contrasty on the image I'm looking at than it was on the original scan. This simply represents laminar flow. This is something mimicking pulmonary embolism due to laminar flow. And you can see that in this patient, the right atrium and pulmonary artery or starting to wash out, and we are seeing pacified or relatively pacified blood in the center of the lumen of those vessels mimicking the appearance of a filling defect. And here at the level of the right pulmonary artery, the same sort of appearance. And if you see something like this and you're thinking, you're looking at a flow artifact, you can have the workstation measure attenuation in the region that you're concerned about. And this is too high to be clot. Clot usually in the range of 35 to 85 on the very high end of things. And this is much higher attenuation. And as a rule of thumb, if something I am thinking is a clot is higher in attenuation than muscle, it's a flow art effect, a true clot should be lower in attenuation or about the attenuation of muscle. And if something looks higher than this and you can just eyeball this, that looks higher than muscle here, you can be pretty confident you're looking at a flow artifact.

Recent Case with Localized Flow Artifact

This is a patient who we saw overnight just about three weeks ago and this person had hemoptysis and had a PE study that was read overnight by the resident and then first thing in the morning by my fellow, and there's a filling defect here in the left pulmonary artery, something here in the inter lobar pulmonary artery and something down there as well, and no other flow artifacts visible anywhere else in this patient. And these were called pulmonary emboli by the resident and then by the fellow. And then these came to me for interpretation and I was stroking my chin and being professorial. But these, the lung windows in this patient, you can see there's a lot of bronchiectasis. The l left lung is relatively lucid, bronchiectasis, again, seeing a lot of left upper lobe collapse. And I would point out here that the left pulmonary vein is not yet densely opacified, even though the right one is. Now, here again are those same images that were read as pulmonary embolism, putting a cursor over that it measures about two 50. And again, just simply compare the attenuation of what looks to be a filling defect with chest wall musculature. And this is brighter than musculature suggesting the presence of a flow artifact. That's what I said it was. And this patient actually had bronchiectasis with a lot of decreased vascularity here and slow flow within the left lung that was resulting in a localized flow artifact within the left pulmonary artery, not visible anywhere else. Evidence for that is the slow venous opacification left pulmonary vein. Partially opacified right is densely pacified. And this we confirmed with a repeat study six hours later, there's no PE present. So these flow artifacts can occur in funny ways, in ways you can't always explain. But if you can see that what looks like a clot is denser than musculature, you're halfway in making the correct diagnosis.

Extreme Flow Artifact in Congenital Anomaly

Now this is a an extreme example of flow artifacts mimicking pulmonary embolism. This is a patient who has a single ventricle pulmonary atresia and the SVC and IVC are directly aas the most to the pulmonary artery somewhere in here. And you can get this very profound streaming of contrast where it looks as if a pulmonary embolism is present just because of the mixture of pacified and opacified blood.

Recent Example of One-Slice Artifact

Now this is another case I just saw recently and it was just a good example. So I put it into this talk. We can see three scans, three adjacent scans here from a pulmonary embolism diagnosis case. It was another overnight case, and it looks like on this one scan as if there's a filling defect in that artery. Now it's a little bit small to compare to musculature. I certainly can't put a cursor over there and measure it's attenuation. But notice that we don't see anything at the scan above or the scan below. Is there such a thing as a one slice pe? There isn't. If you're looking at a vessel and cross section, and it looks as if you're seeing a PE only on one slice, it's some sort of artifact. Don't worry about what kind of artifact it is. It is an artifact. Pulmonary emboli in a vessel that you're looking at in cross-section are going to be long, they're going to be worm-like, and you'll see them on multiple adjacent scans. So some sort of artifact, no such thing as a one slice pe. Okay?

Motion and Volume Averaging Artifacts

This is a scan that was done with thicker collation that we would usually use for a PE study. But I would have to say that about at least half of the PEs we diagnose are on routine cts done for other reasons. So it's not uncommon to see a pulmonary embolism on a scan with relatively thick slices, half a centimeter or something of that sort. So what do we have? We have a pacified vessel. This is the lung window image at the same level opacified vessel lung window. At the same level on this scan, it looks as if there's a filling defect in the lateral aspect of that pulmonary artery branch. But I would point out on the lung window images that we are seeing two images of the pulmonary artery at that same level. There is the pulmonary artery at one point as the scanner is looking at it as the patient is breathing presumably, and here is the same artery at another moment in time. As the scanner is looking at it motion during the scan where the two little ovals overlap, it looks as if the artery is opacified and where they don't overlap here, out here laterally, it looks like there's a filling defect. So anytime you have respiratory motion during the scan or cardiac motion with pulsation, you can get these ghosting artifacts. And looking at the lung windows can be very helpful in making that identification.

Another relatively thick slice PE diagnosis here, a patient with a an apparent worm-like filling defect in the left upper lobe pulmonary artery. This is the lung window scan. At that level we don't see much, but the next scan down, you can see the upper lobe bronchus or this branch of the upper lobe bronchus having exactly the same appearance as the apparent filling defect on the mediastinal window slice. And this is simply volume averaging between the pulmonary artery and its adjacent bronchus. Again, arteries and bronchi travel together, and if you're looking at a thicker slice, you can have this sort of artifact, not uncommon. Okay, there we don't see anything in there. We did see the bronchus.

Unenhanced CT Detection of PE

When we do a aorta protocol, and I'm sure you do it the same way, if in someone with an acute aortic syndrome or rule out dissection or something of that sort, we first do an unenhanced series through the aorta to look for a high attenuation intramural hematoma. And this is a patient seen in our ER for rule out dissection. And these are the unenhanced, or this is an unenhanced CT scan from that study, notice that we're seeing something high attenuation In the right pulmonary artery, at least overlying right pulmonary artery and extending over towards the left. That is a thrombus within the right pulmonary artery being diagnosed without contrast injection simply because it's higher attenuation than blood. Just like intramural hematoma can be higher attenuation than blood. So very occasionally you may be able to diagnose a PE simply on the basis of an unenhanced scan. So keep your eye out for this sort of thing.

Atrial Fibrillation and Associated Findings

A cause of An association with pulmonary embolism is atrial fibrillation. One of the causes of AFib is pulmonary embolism. So if someone comes in with new atrial fibrillation, often pulmonary embolism will be questioned. And this is a pulmonary embolism study in a patient with atrial fibrillation. And we don't see a pe but what we do see is filling the effect here in the left atrial appendage and down here as well. And people with atrial fibrillation have a tendency to develop left atrial thrombus, atrial appendage thrombus. And it's an important thing to look at when you're doing a PE study. Now, notice that this thrombus is very sharply marginated. That is the way that thrombus should look real thrombus, very sharply marginated. It's being outlined by contrast on both sides. And if you compare its attenuation to that of muscle, if anything, the muscle looks brighter than the thrombus. And that's the way it should be. If this was, if what appeared to be thrombus was brighter than muscle, we would think that it's a flow artifact act.

Whoops, Here we go. Okay, this is another case. Atrial fibrillation, rule out pe. So this is being done for PE diagnosis in a patient with atrial fibrillation. Well, there's certainly something an apparent filling defect within the left atrial appendage. It's rather sharply marginated, again, very sharply marginated contrast here that looks clearly less opaque. If you compare that attenuation to musculature. Is it brighter or darker than muscles? It's brighter and measuring its attenuation, it was about 200 hansfield units. And this is simply a flow artifact or an artifact with inadequate contrast mixing in the left atrial pennage. It's something we see all the time, and usually it's not quite as sharply marginated as a real clot and its attenuation can be measurably higher or just eyeball higher than musculature.

Pulmonary Vein Abnormalities

Now, along with the left atrium, you need to look at the pulmonary veins when you're doing a pulmonary embolism study because you can see in some patients thrombus within pulmonary veins for the same reason. You can develop left atrial thrombus, you can develop vein thrombosis and they both need to be treated, treated the same way with anticoagulation, but you can see other vein abnormalities as well. Here's a patient with a PE study for pulmonary embolism diagnosis, and there are certainly rps and various branches of the pulmonary artery. And then sort of in a saddle configuration going from the left to the right pulmonary artery, but also notice a filling defect here in the left superior pulmonary vein, very sharply marginated. At the other levels you can see the filling defect. And then here something that's rather large and mass like, and this is a lung cancer with pulmonary vein invasion. Lung cancers like to invade the pulmonary veins and they can grow right down into the left atrium. And of course, anybody with cancer has a higher likelihood of developing a pulmonary embolism. So on occasion you'll see these two in association.

Now, just like you can see contrast streaming or contrast flow artifacts in pulmonary arteries, you can see them in pulmonary veins. And if you ask me to explain the hemodynamics here, I would not be able to do it. But these are vein branches that look like they have filling defects in them that represent some sort of a flow artifact, in complete mixing of contrast and pacified blood. So it's something you can see in pulmonary veins as well as arteries.

Neoplasm Mimicking Pulmonary Embolism

Okay, we're almost to our first question. Here's a 20-year-old with dyspnea. What we can see here is a large filling defect in the right pulmonary artery going into the inter lobar pulmonary artery and then into a couple branches, and then this a lung window image through the lower lobe. In the same patient with these very funny looking pulmonary arteries, the most likely diagnosis is flow artifact pulmonary embolism neoplasm. Well, we have Okay, good. We have a lot to discuss. Okay. The diagnosis in fact is neoplasm. This appearance a clear cut filling defect with this being, if anything darker than musculature is a true filling defect and could be a big pulmonary embolism. It's not a flow artifact, it's too well defined and we're seeing it too extensively and no other reason for abnormal flow. This appearance in the lung periphery is not something I have ever seen with pulmonary embolism. These nobby looking dilated vessels. On the other hand, this appearance is what you see with intravascular tumor embolization where you have tumor that is growing in pulmonary arteries and gives them this very nod, nobby sort of vascular appearance. And that is what's going on here. This is tumor that has embolized to the right pulmonary artery and is growing in vascular branches. This patient has an osteogenic sarcoma, and you can see this with sarcomas, you can see with lots of different tumors, but these really large intravascular filling defects, often with sarcomas or tumors that have a large intravascular component like hepatoma or renal cell carcinoma. And these are more images here, this large definite filling defect in the pulmonary artery intravascular mets. These are images through the lung and you can see these funny dilated nobby looking vascular branches that have tumor within them. A very typical appearance for intravascular term tumor embolization. And this, you can, this was followed and you can see that these mets in the vascular branches are getting larger and these eventually ossified. So intravascular tumor embolization is something that occurs not infrequently, little unusual to see such large macroscopic filling defects with this abnormality, but it may be seen.

This is another example. I just pulled off the my workstation a couple weeks ago as I was putting this together. But a patient with intravascular mets from transitional cell carcinoma, these vascular branches in the lung periphery that are grossly enlarged and have this nobby sort of branching appearance very typical of intravascular tumor embolization.

This is a patient we saw for pulmonary metastases or lung nodules that were diagnosed on chest film. And this nodule we had done a percutaneous biopsy on and had diagnosed as a sarcoma. And on the scan that was done actually for staging purposes. It was noted that there were a number of filling defects within the left pulmonary artery, and these were diagnosed as pulmonary emboli occurring in a patient with cancer. But in fact this is the primary tumor. This is an angiosarcoma of the pulmonary artery and the nodules we were seeing in the lung peripheral to it were metastasis. And very occasionally you will see a primary sarcoma of a pulmonary artery mimicking pulmonary embolism that the met and that the primary tumor, and he was treated with a pneumonectomy and did okay for a while, but then had a recurrence. And you can see this filling defect within the main pulmonary artery representing the angiosarcoma.

Evolution of CT in Pulmonary Embolism Diagnosis

Okay, let's talk about pulmonary embolism diagnosis and CT and how it all fits together because there's a few interesting things that are have come out just recently. Now, first of all, this is from studies published in 2000 in the internal medicine literature. They reviewed, or the authors reviewed a whole bunch of studies that were performed using single detector CT with three to five millimeter collation for PE diagnosis. And they found that the sensitivity of CT range from 64 to 93%, that CT had low sensitivity for small or subsegmental clots, and that the utility of CT as a primary imaging modality was questionable. So they were not all that hot for CT for making this diagnosis. They were still thinking that angio or VQ VQ scan should be the primary modality.

Now, not too long after that, a number of studies were reviewed here, 23 studies were reviewed more than 4,000 patients in which there was clinical follow up of at least three months after a negative ct. That that is a CT that was read as no pulmonary embolism, mostly was single detector scanners. Again, sort of the same group that they was reviewed that were reviewed in the pulmonary or in the internal medicine literature. And it was found that there was a negative predictive value of venous thromboembolism after a negative CT of basically 99% negative predictive value of mortality, 99.5%. And that was as good as any other exam that was available. So this just a few years after the other study, and this meta-analysis indicates that CT is just fine for pulmonary embolism diagnosis, and if the CT is negative, the patient does not need to be treated. And then the RSNA, of course published this issue, negative CT results can safely rule out pulmonary embolism. And if they say it, it's gotta be true.

Subsegmental Pulmonary Embolism and Overdiagnosis

Now, here is a patient seen in our emergency room with dyspnea. CT was done for diagnosis. This obviously multi detector CT with 1.25 millimeter slices and there is clearly a subsegmental pulmonary embolism in the right lower lobe, no doubt about it. That is a PE filling defect here. It's branching into two smaller branches. No other PEs were seen. Is this person sick because he has a pulmonary embolus or is he sick because of this large pleural effusion? How would you dictate this pulmonary embolism artifact? Subsegmental PE subs segmental pe, which may be incidental. That's good. Yeah, you have to qualify this. That is a little teeny weeny pulmonary embolism that cannot possibly be causing this person's problems. The people we have working in our ER are not good enough to diagnose or detect that pulmonary embolism clinically. If this person is dys, it's because of something other than that little teeny pe and you have to indicate that it is a minimal abnormality. In fact, it may not even be considered abnormal. And it may be entirely incidental to his situation. I think it's very important that we now begin to qualify these interpretations when we're seeing one or two subsegmental clots.

This is, this was published in the New York Times, what, a couple months ago? July 3rd, 2013. CT scans may be too good at finding lung problems. Study finds. Now the lung problems they're talking about are pulmonary embolism. This is the article that they're referring to. This was in the British Medical Journal analysis. Too much medicine is the heading when a test is too good? How CT pulmonary angiograms find pulmonary emboli that do not need to be found. Now, it's interesting, when we started off the internal medicine, people were saying CT is not sufficiently sensitive to diagnose pulmonary embolism. The pendulum has now gone to the point where people are saying it is too sensitive in diagnosing pulmonary embolism.

Here a study that was published a while ago. It's clearly multi detector CT clearly has increased sensitivity for diagnosing subsegmental emboli. 15% of positive cts show only subsegmental clots. Is MD CT two sensitive? Is overdiagnosis occurring? Are segmental clots important? Do subsegmental clots need to be treated? You might be surprised that there's very little literature that deals with this question. This is one paper that's available. A review of 192 patients with isolated subsegmental clots. 65 did not receive anticoagulation for one reason or another. There was no recurrent PE or death within three months in those 65 that were not treated in the 127 anticoagulated patients. Only one had a recurrent embolism. And the historical recurrence rate for PE after treatment is 6%. So here we have patients with a pulmonary embolism not being treated and no recurrence when the historical recurrence rate is considerably higher. And the recurrence rate in treated patients is also higher. Other than this study, however, there's very little to suggest or very little data that suggests that these clots need to be treated or not treated. There's just nothing there in the literature. But there is some evidence that these clots are being overdiagnosed with multi detector ct.

Here, 1,417 patients with intermediate or high clinical probability of PE randomized to VQ scan or ct PE was seen more frequently with ct, no difference in mortality over three months. Even though we were seeing more more clots with with CT and those patients were being treated and the patients that were not being seen with VQ were not being treated. No difference. So some evidence here that we're seeing clots that don't need to be treated or at least don't need to be seen. And in this bri British Medical Journal paper, some current guidelines recommend anticoagulation for all pulmonary embolized. Some of the major societies have come out with that recommendation. Others suggest that anticoagulation may not be warranted in all cases because of uncertainty about the balance of benefits and harms in treating isolated subsegmental. Pulmonary emli treated patients with subsegmental emboli have a 5% chance of major bleeding. And I was contacted by a lawyer not too long ago, maybe a month ago, to be an expert witness in a case. And I don't do that anymore because I find it too irritating. But the issue was a patient who had a sub segmental clot, diagnosed on CT was anticoagulated and had a big retroperitoneal bleed. So this is going to be an issue, I think. And I think the balance of information at this point in time suggests that it's not necessary to anticoagulate all sub segmental clots. And we need to indicate to our referring physicians when we see a subsegmental clot that it may not be all that important. And I'm not exactly sure how to do that yet, but I think we need to do that anyway.

Incidental Pulmonary Embolism

Incidental pe, we see it a lot. We see PEs all the time when we're not expecting it to be there. Higher incidents in patients in the hospital and patients that are being ventilated and patients that are old and trauma patients, all sorts of things like that. Incidental pe, very common. We're probably getting them right now. You're sitting here in this room listening to lectures all day. One or two of you probably has a subsegmental pe

This a case I just saw a week or so before I came. A lung cancer patient being followed for possible metastases or whatever. There's a definite subsegmental PE in the right lower lobe. They're branching into two little sub-branches. That's a 100% accuracy diagnosis of pulmonary embolism. There's no doubt about it. Now I'm not going to ignore that. I'm going to dictate this as showing a pulmonary embolism, but I am going to dictate it as a pulmonary embolism that is very likely incidental. And that's what I did in my report. So I think we're gonna have to start doing that. And physicians on the ward or in the in the clinics are going to have to start dealing with or finding PEs that we're suggesting might be incidental.

References and Recommendations

Now there are a couple articles or reviews I'm going to refer you to and you can look at them if you wish, and I'll quit after I review those. First of all, panel discussion recently published in the A JR towards the end of last year. The panel of radiologists concludes that CT is now capable of over-diagnosis. Not all small isolated clots need to be treated. Treatment criteria for small emboli are not well established. That is absolutely true, true. This, the Fleischner Society recommendations for diagnosis of pulmonary embolism and what studies to do in different clinical situations, pregnant patients and things like that. You can look this up if you wish from 2007 and this another one that was published in a JR in 2011, reviewing various things in regard to pulmonary embolism diagnosis and when different sorts of scans or modalities should be used to make this diagnosis.

Alright, thank you very much.