I, first of all, want to thank APPLIED RADIOLOGY and the sponsors for inviting me. So I'm going to be talking about the rest of the chest not, the heart but radiation considerations in the chest. And let me be specific. The three topics I'm going to cover are pulmonary embolism, assessment of the lungs, lung parenchyma and aortic imaging. And to give you a sense of the equipment we're using, we have a Phillips 64, which we use for a lot of our general lung assessment. And then the more challenging kinds of cases, we now also have a Phillips 256 ICT that we use for some of the more involved studies that we do.
So I'm going to start with pulmonary embolism. And, of course, this remains a huge, huge problem in a lot of different ways, with a large number of cases, 50,000 deaths a year, and an 8% case fatality rate. And a number of autopsy studies, going back 50 years now, about 70% of PE is overlooked, based on autopsy data. The other piece of it ... of course, we think of venothromboembolic disease as being a systemic kind of thing, but the reality is somewhere around 25% of patients ... only 25% ... have symptomatic DVT. So it's a real challenge, separating the two for one thing, and making a diagnosis.
There have been, over the years, a number of attempts used to try to parse this, to try to get us a better sense, clinically, and these include things like the Wells criteria, which are clinical criteria, and D-dimer, which is more of a blood test. But even these have their limitations. So a lot of this has really come on the shoulders of imaging. And we have a whole series of options, again dating back many years ... and I think everybody here is familiar with what these are ... and obviously the first two of these have major limitations; one in the sense that chest radiograph is not a particularly accurate technique. Pulmonary angiography is invasive. For the purpose of PE study, V/Q scanning certainly has value, even today, but there are many indeterminates, and that can be a potential limitation. And the leg studies is really, although in certain circumstances worthwhile, it's an indirect measure.
So de facto, in many cases and in many practices, including our own, CT pulmonary angiography (CTPA) is the gold standard. How good is CTPA? It really depends on what study you look at. This is probably the largest study in recent years, and this is the PIOPED II study, using actually what was a composite scan. And that's one of the limitations, meaning the standard reference that they used was not any one thing; it could be a pulmonary angiography, it could be, I think, a high-probability a V/Q scan. So there are some challenges with that.
But you can see that even using that, the sensitivity and specificity were quite high. And, in particular, the negative predictive value, meaning the likelihood of coming back with pulmonary embolism after a negative study, is really outstanding. So then there are other things which I will ... the other point of it is that, unlike, let's say, V/Q scanning, where if it is negative you don't know what the patient has an alternative. In the context of CTPA there are many other things that can be diagnosed, and account for the patient's chest pain.
That's the god news. Of course, the bad news, as I think many of us know, is that there has been an incredible increase in usage. This is from our institution. And you can see ... this is plotted since 1999 ... and you can see there's been an order of magnitude, almost, increase in the use of PE, which is a huge number, if you think about it. And there's, not surprisingly, been a concomitant decrease in the percent-positivity for PE, which is problematic; down from about 15% to on the order of 7% or8%. Some of the literature suggests that it's down around 5%, in some centers. So we've increased the volume enormously, and decreased the positive rate in the process.
And that's only, of course, a piece of it. The other piece of it is that it's part of the use of CT overall, and this is certainly one of the components of that increased use of CT, and the attendant radiation concern that exists with that. And, of course, this is the article from The New England Journal of Medicine that highlighted a lot of this and, I think, threw a scare into a lot of people; perhaps, as we talked about earlier, overblown, but that kind of number certainly is an attention-grabbing number.
So just to give you a sense of the ... since this is a seminar on radiation, to give you a sense of some of the different doses that are applied for different types of scanning related to pulmonary embolism work-up. And these, again, are in millisieverts (mSv). And, as we've said earlier in the seminar, there are limitations with looking at this. But it does give you something of an apples-to-apples comparison, if nothing else. And you can see that, of the technique, CTPA is somewhat or pretty close to equivalent to what's now called invasive pulmonary angiography. And it's, of course, far greater than chest radiography, and somewhat greater, at least, than a full V/Q scan. And, in fact, it's somewhat less than a gated cardiac CT, but it is actually higher in a prospectively ... by which I meant retrospectively gated ... but it is actually higher than many of the prospectively gated CTs. So if you do a gated CTPA vs. a gated cardiac CT, you're probably giving ... depending on, again, some of the factors we're talking about ... you're giving at least an equivalent dose, if not a higher dose, for CTPA.
And then, in addition to that ... that's sort of for the general population ... there's a group of patients who are at even more increased risk, and I think we're going to hear more of that from Don Frush, in a little while. But certainly children and adolescents are at increased risk of cancer. It's been estimated that it's 10 times the risk. Young females and girls, as well. And pregnant females, from the perspective of fetal susceptibility. So let me talk about some of the strategies in line with some of those high-risk groups, and in general.
So the general strategies, some of which are controllable by the technologist and the physicians, and some which are not ... dose modulation is certainly one of the huge innovations for general body work, including CTPA, that we can use. And there are basically the 3 types: Related to patient size; related to the Z axis; and related to angular rotation. And then I've listed some other things, as well, including decreasing the kVp or the mA. And this is true with pretty much anybody, where these are strategies you can adopt. I'm going to talk about them, to a large degree, in the chest here. But the other thing you can do is reduce the Z-axis coverage, the longitudinal coverage. And some of the other things that have been talked about, I think, are less useful in this setting is increase the pitch and/or the sections thickness.
So let me at least discuss dose modulation, at least just a tad. And in the chest, it's actually quite advantageous, because you can see that there's a relatively low dose that you can get away with, because the chest contains mostly air that there's an opportunity to modulate in a way that you may not have in, let's say, the abdomen or shoulder girdle. And so this is, again, the same thing depicted in a different way, showing how the dose can actually be ramped down as you get through the chest.
Another approach, which has been looked at in some detail, is decreasing the kVP or the mAs. And so there have been a couple of studies, now, looking at that ... perhaps more ... which have shown that you can substantially decrease the dose of the mA with maintenance of the imaging parameter.
In other words, the image quality can be kept constant with a considerable reduction of dose. This is just one paper that I cite here. And, similarly, although it's much more difficult ... and less predictable ... you can reduce the kVp, the kilovoltage potential, and the additional advantage of doing that, if you can get away with it ... which you often can in thinner patients ... is that actually with a lower kVp, the lower kVp puts you at a level closer to the K-edge of iodine. So for a CPTA, that actually potentially will increase the contrast that you see in the pulmonary arteries. And that's something that actually has been shown in a couple of papers, but this is the one I'm citing here. So I think mA is easier to fiddle with, and kVp is a little less flexible. But kVp has the additional advantage of perhaps increasing your intravenous contrast.
The other thing that we've tried to do, a little bit, is cone down the coverage. So your normal chest CT that you'd expect, well, let's go from the apices down into the upper abdomen, and make sure we get the tips of the lungs at the very bottom ... but that's really not necessary with a CTPA. Because, really, you're dealing with the tiniest sort of branches when you're at the apex and at the very lowest part of the base. So why not, instead, do this? And, again, because of the Z-axis coverage, you'll get some dose savings with that.
One of the big discussions, of course, is girls and young women. Again, the estimates of what the organ-specific dose is to breast tissue are variable. But you can see that they range from sort of10 mGy to 17 mGy, but not that the number ... I put that in context, in some sort of reference ... it exceeds, by far, what is recommended for mammography. So even the lower end is three-fold times what is the limits for the ACR for mammography. Perfusion imaging actually gives a much lower breast dose, which is one of the advantages. And then the further point is that dose of 10 mGy may again, according to some of the extrapolated data, lead to an increased risk of breast cancer.
So one of the approaches to addressing this that hasn't actually caught on, for a number of reasons, is to use a bismuth shield to ... an in-plane shield%u2026 to try to reduce the dose to the breast. And there have been a couple of studies looking at this, and this is one I've cited here, where they found a 41% dose reduction in women, and in the phantom it was actually less, without a change in image quality. Others have been less ... you can also use different configurations, two-ply, four-ply, and what-not. These are now commercially available. Others have had problems with near-field artifacts, and that's why. And it just has not caught on perhaps as much as it should have. But it is certainly, in young women and girls, something that ought to be taken into consideration. And here is from their paper, just showing actually the shield, the way it looks. And also their detector, and the fact that the image quality actually is pretty good.
All right. The pregnant patient is similar in one sense, but there's additional considerations. One is that the D-dimer may be falsely elevated. And since that is one of the big clinical uses or clinical ways to triage patients, the fact that it's elevated means the patient is ... or is often elevated means that the patient is more likely to undergo CTPA than perhaps a non-pregnant patient. Another important consideration is not only do we have to worry about the patient, but we also have to worry about the fetus. And that's something that obviously isn't an issue in other groups of patients.
The other thing ... there's some early data that suggests that CTPA in pregnant patients may be less reliable. Whether that is because of flow artifacts, or valsalva maneuvers, it's not entirely clear, but there are a couple of studies that are coming out, one of which should be coming out shortly, which suggests that the non-diagnostic rate of CTPA in pregnant patients can be quite high. And for these reasons, this is one of the situations where venous leg ultrasound may be ... one of the Fleischner papers, which is the one I've cited here ... it may be worthwhile trying to go to the legs first, to look for pulmonary embolism in that group of patients, because a positive tells you, you have venothromboembolic disease, and you're going to have to anticoagulate the patient. So that has been one of the situations where you may go to an alternative study.
Another option, because of the high false-positive ... the poor quality of CTPA in pregnant patients, is to potentially start with a V/Q scan in patients where the chest radiography is unremarkable. So there are some real considerations.
Now, I mentioned, in addition to the patient, there's the fetus. And there have been a number of papers, some contradictory, about what the actual dose to the fetus is. It's clearly higher, now that we're using multi-detector CT, because, with single-slice, the dose to the fetus was generally viewed as being lower with CT than with V/Q scan. But now with ... and this is done with 16 detector CT from Duke University, actually ... and you can see, at or around conception, the dose was kind of comparable to the fetus, CTPA vs. V/Q. But, at three months, the dose was actually excessive, or exceeded ... not excessive, but exceeded the dose for V/Q scan by almost double. So that might influence which way you decide to go, in terms of CTPA vs. V/Q scan. Now, there are some caveats with this study. One is that the CTPA dose in the study was not optimized. At least not ... there were factors that probably could be reduced in the pregnant patient that were not. So some of the assumptions that were made may be not as conservative as they could have been. But it's also true that the V/Q scan dose can be decreased. And certainly one of the things that we've done in pregnant patients is to avoid re-breathing. You can decrease the amount of perfusion agent, and you can eliminate the ventilation part, and still get, in many cases, an acceptable V/Q scan. So this is an interesting topic to which, I think, the resolution is still awaited. There are a lot of different opinions on this.
Having said that, you certainly can do a CTPA in pregnancy. A couple of examples that turned out to be negative, which is unfortunate in the sense that we perhaps gave unnecessary radiation. But if you have to do it, then, as long as you know the full set of facts, then that's a reasonable thing. Using a bunch of different factors, this is a recent article that came out in Radiographics. You can potentially ... so here there's modification of a variety of the different factors, and you can perhaps reduce the dose of a CTPA by about four-fold, which is pretty dramatic, if you choose the right factors. The right parameters.
So in terms of pulmonary embolism ... and then I'll move on to some of the other things, briefly ... there are multiple strategies to reduce the dose. In young female patients, certainly breast shielding would be one thing. In pregnant patients, you've got the option of trying to start with the legs. And if the chest radiograph is negative, then maybe V/Q scan is a reasonable place to start.
I'm going to talk about lung parenchyma, then. And one of the things that makes this a difficult topic is that there's a huge variety of parameters that are used in clinical practice. This is from a series of papers, and you can see that the kVp and mA ... especially the mA ... is all over the place. But there is at least potential to reduce dose with lung nodules, following up malignancy; following up pneumonia, which is in fact we do reduce the dose for that; and interstitial lung disease. And just again to give you an idea of the different ... there have been about 10 different papers with lung nodules. And you can see that they've played around with different parameters. But in a nutshell, what they've found is ... the papers have determined that, for the most part, you can reduce the mAs, if that's what you choose to do, to about 50, at least ... maybe a little further ... without compromising imaging quality. If you go below that, it's mixed. In the National Lung Screening Trial, which is of course the lung cancer low-dose screening trial, they use a 30 mAs. Also, obviously, you can reduce the kVp.
So here's a couple of examples, just showing the difference. I mean, clearly, the nodule that you see at the right posterior costophrenic angle is visible with 200 mAs, as well as 60 mAs, but you can see the image quality is a little greater. It's still very diagnosable, however. High-resolution CT. We use a lower dose on the axial. In fact, with this one, there's a bit of an apples- and-oranges comparison here. But you can see that the axial imaging is at least as good if not better than the helical. And there are several reasons for that.
So, in the last couple of minutes, let me just turn to the aorta. And obviously CT of the aorta is widely used. Dissection, aneurysm, in this case, due to trauma, are a couple of the reasons. But there are some huge challenges, including the large area to be covered; potentially the entire chest, abdomen, and pelvis. If you are doing a dissection protocol, the pulsation artifact at the root can be a real problem at times. Usually, it's not, in experienced hands, but there are times where even those of us who do this all the time have a problem. And, therefore, ECG gating can be used to resolve some of the ... especially the root issues, and perhaps bring some clarity to that. However, there's a potential increased radiation dose associated with that.
So here's an example of a non-gated study, non-diagnostic ... or at least was felt to be. And with the gated study, it's clear there's no aortic dissection. So that's useful. The other thing with retrospective gated ECG is that you can actually get a little more data on the flap. For instance, now we see the entry site, which we couldn't see on the still. You can see that the flow in this lumen, the true lumen, is faster. It's a little bit more dense. We can see what's called the beak sign. And also were there to be ... this is not in your branch vessel ... but we'd be able to see dynamic obstruction. So there are some major advantages, potentially, to doing gating, but at a radiation cost.
The other thing is the fact that we can see coronary arteries very nicely. So, in the same exam, a nicely gated aorta with a clear intimal flap, and we can see the coronaries here. This is the false lumen; this is the true lumen; they're both coming off the true lumen. And we can go beyond that, and look at the flap vs., let's say in this case, the take-off of the left coronary artery, a little closer. And then we can also assess the coronary arteries. Let's say the patient needs to be taken to surgery. In this case, we're going to say, well, maybe you want to look carefully at the LAD and see is this something that you might want to consider bypassing. Or the converse, which is the coronaries are fine, go ahead and do it.
I think we've talked a lot about gating in the previous two talks. But one of the solutions to this aortic problem is similar. That is we can go to prospective gating and use this on the aorta. And we can get some very nice images. And so here's the entire aorta. You can see that this ... you can see the coronary here. You can see the aortic root. It's very clear. And this was done prospectively.
And with prospective gating, I think you can appreciate that we're down to ... so if we look just to look at the 30-centimeter level, that length, we can see the non-gated aorta, you're up two and a half or so ... and certainly one-and-a-half, even with retrospectively gated with dose modulation. With prospective, you're back down close ... at least depending on what assumptions you make and the factors you use ... you're getting back close to a non-gated CTA. So perhaps we can do prospectively gated aortas with little if any dose penalty.
And then, of course, in the pediatric population, with prospective gating ... and I've listed some factors here ... you can also get some nice images. And potentially, if you move the factors in an optimal setting, or an optimal way, that you can get down even to the 1 mSv range, which is pretty impressive for the image quality.
Essentially, the considerations are vital and I think we have many, many tools that we can use, and should use, to try to reduce the dose for the chest indications. Thank you very much.
