Hi, this is Elliot Fishman, and welcome to our latest series of talks, and this is going to be on small bowel obstruction. Now, we all know that small bowel obstruction is a very important application for CT scanning, and one where CT makes a very big impact on patient management. What I'll try to do in this talk, or in this series of talks, is look at some of the key factors, some of the real important diagnostic points that we need to look at, and how we can avoid some of the challenges and potential pitfalls and errors in the evaluation of suspected small bowel obstruction.

Now, at the end of the day, the main complication of small bowel obstruction, or the one we're most concerned about, is intestinal ischemia. With intestinal ischemia and bowel infarction, there's high morbidity and high mortality. What you want to be able to do with small bowel obstruction is define what patients can be managed conservatively and will do fine, and which patients need to go to surgery.

The patients that need to go to surgery need to go to surgery as soon as possible. One of the things we've seen in the published literature, as the delay in taking patients to surgery by as little as 24 hours will significantly increase the number of patients with high morbidity and high mortality rates, just simply by delaying what needs to be immediate surgery.

Now, when you think about small bowel obstruction, what are the key CT findings? Well, it's dilatation of bowel, and then we follow the dilated bowel and look for a transition point. One of the key things about CT is the ability to find the transition point. By looking for the transition point, we can be very specific, if we find it, usually, as to the cause of obstruction. Is it adhesions? Is it a mass? Is it a hernia? Is it a volvulus? So, it's very important, and the accuracy in determining cause is up to 90%, but we all know from reading scans, it's never quite that easy when you're doing things in clinical practice.

We know, and we'll speak about protocol, that we scroll through the images and scroll up and down looking for transitions. We do find that using coronal imaging is particularly helpful. Sagittal can be helpful sometimes. Where sagittal is most helpful is really letting you look at the patients' SMA and celiac axis, and look for any changes in the vessel, including vessel thrombosis or occlusion. But the sagittal view is not as helpful, perhaps, usually, in looking for transition of the bowel as the coronal is. So that becomes very, very important.

In terms of specific protocols, we like to give 1000 cc of water. We used to talk about positive contrast, now neutral agents are ideal. We then give IV contrast, injecting 4 to 5 cc a second. When we're looking for small bowel ischemia, we are going to do dual-phase acquisition. A lot of times people will get a single phase, but if you really want to look at the vessels on both the arterial and venous side, and look for changes in enhancement, dual-phase is necessary.

Now, sometimes people will say, "Gee, do I really need to give IV contrast?" For some reason, people always don't want to give IV contrast. Well, we know from many articles published over the last couple of years that the risk of renal injury and the risk of hypersensitivity reactions was always, in a sense, blown. When you give IV contrast and the patient is hydrated, the chance of renal injury is small. Also, very few patients have reactions to contrast. If you need to, you can premedicate. But, again, IV contrast is critical because, in this article by Shaishe in JAMA Surgery, made the point that non-contrast CT was 30% less accurate than contrast-enhanced CT for evaluating abdominal pain. 30%, that's massive, and this is from a surgical perspective. It's not a radiologist saying, "Well, I could see things better with IV than without IV." It's not those silly articles in emergency radiology... about emergency radiology in non-radiology journals that talk about, "Oh, radiologists want IV contrast because they like it and they're used to it, there really is no need." This article made the point: 30% less accurate when you do not use IV contrast material.

In this article by Yeh, "Adding IV contrast material provides the benefit of increased diagnostic accuracy at the cost of potential adverse reactions like contrast-induced nephropathy, extravasation, and/or allergic reaction." Again, these are small numbers, and when you look at the importance of accuracy, you need to give IV contrast material. "When we choose a noncontrast approach, we should be intentionally accepting diagnostic uncertainty, although this is already well accepted for certain diseases." Yeh goes on to say... Again, just read the article, I think it's worthwhile to read it. "In the low-risk patient, complications of IV contrast is often negligible, and the benefits of routine IV contrast far outweigh the risks." Bottom line, give IV contrast unless you have a really good reason not to.

I mentioned non-contrast scans and delayed scans we do not do, and we focus on dual-phase, at about 35 seconds arterial, 30 to 35 seconds to be exact, and about 65 to 70 seconds for venous phase imaging. We always use thin-section CT, 0.75 by 0.5. The better scanner you have, the faster the study, but 64-slice or better, which all of us have, is really ideal for these patients, so not a problem.

And in terms of analysis, axial and multiplanar are critical. Volume rendering and cinematic rendering can be helpful looking at transition points. Also, 3D mapping of the vessels can be very helpful as you're looking for branches of the SMA. For looking just at the SMA or celiac proximally, obviously just the sagittal view will work very nicely, but when you try to look for patency or occlusion of distal vessels, you really need to be looking at the volume display, or the MIP display, or cinematic rendering.

Now, one of the things of course with IV contrast and dual-phase imaging is the ability to see things and recognize specifically what they are. So, in this case of a patient with abdominal pain, stomach is well distended, there's a mass which looks like maybe it's coming from the pancreas or it's coming from the duodenum. There is no pancreatic duct dilatation and no common duct dilatation, and if this was truly in the head of the pancreas, you would expect to see duct dilatation. When you see this enhancing lesion, you recognize that what you're dealing with is a duodenal lesion. Now, it's interesting when you look at the coronals, it's off the medial wall, and still you begin to be concerned: am I certain it's going to be bowel, am I certain that I'm dealing with something perhaps in the duodenum or between the duodenum and the pancreas, what exactly am I dealing with? And this is a nice example also showing you the MIP imaging that we not only can recognize the presence of masses, in this case pushing on bowel but not obstructing bowel, but because of its enhancement pattern, we know we're dealing with one of two things here: we're dealing with a carcinoid tumor or a GIST tumor. Both of them are very vascular. GIST tumors tend to be exophytic and carcinoids tend to be within the lumen. So this case was a GIST tumor, again, very nicely shown as you look at the cinematic rendering. Again, a really good look at that. Here it is in the cinematic rendering showing you the GDA and the feeding vessels. And here, again, really, really nice look at the duodenal folds, the haustral folds, and then the mass being present, which is a gastric GIST tumor.

Now, it's a good example of the patient having symptoms but the bowel not being obstructed, but our ability to recognize the presence of mass. And although GIST tumors can occur in the esophagus, the stomach (majority in the stomach), small bowel, and large bowel, the ones that are smaller tend to be more vascular, and the duodenum, particularly those in the second portion of the duodenum, is a very common location. I will say, as I mentioned, the challenge to distinguish a GIST from a carcinoid, or in this location to make certain you are not dealing with a pancreatic mass like a neuroendocrine tumor. I showed you some cinematic rendering, and in this article we made the point, cinematic rendering helps demonstrate the extraluminal growth and relationship of GIST to underlying mucosa, overlying serosa, and vessels, which has the potential to further surgical or endoscopic treatment planning. So again, cinematic and the role of 3D is something we need to remind you about.

And cinematic rendering in bowel in general is excellent, but in this case like with the GIST tumor, the ability to recognize the relationship of the tumor to bowel, to the pancreas, and to vessels becomes critical, and for surgical planning, indeed, it becomes very, very important. Another case, very similar in location. Here it looks like again, could this be off the pancreas or near the pancreas? Is it coming from the stomach or the bowel? Again, a very vascular lesion. The bowel is not obstructed, though the patient had abdominal pain suggesting bowel obstruction. And when you look at it on the image on the right, you could say, "Well, am I sure this is not a neuroendocrine tumor of the pancreatic head or uncinate?" I don't see pancreatic duct dilatation, which I should see, particularly with a mass this large, and I don't see common duct dilatation either. When you look through these images, it's very similar to the last case, but not exactly the same location. You can see the lesion washes out. It's kind of more exophytic than the last case was in terms of position and orientation. Here it is on the coronal volume-rendered 3D views. Again, a vascular mass simulating pancreas, but it's coming from the duodenum and that was a GIST tumor as well.

When we look at small bowel, we look at transition points. We also look at the mesentery carefully. In the last two cases, I spoke about how we looked at that mass to determine whether it was bowel or pancreas. Here's an example of abdominal pain in a patient with Crohn's disease, where we see fibrofatty proliferation of the mesentery, right? You see the mesentery, prominent vessels, okay, and this mass effect by the bowel. You then see the terminal ileum markedly narrowed, with stricturing here, the inflammation. There's also a secondary of stricturing here, and a third area here. With Crohn's disease, although we think about bowel being a focal area where we're looking for obstruction, you need to keep looking carefully, because although you may see one area of involvement, like the terminal ileum in this case, it was actually the more proximal bowel that was problematic and giving the patient symptoms. So that indeed becomes very, very important. We talk about the comb sign, the prominent vasa recta extending to the inflamed bowel, very nicely seen here on these MIP images and on the volume-rendered images.

And so when you think about looking at bowel and we look for small bowel obstruction, Crohn's is one of the things we'll think about, and with Crohn's disease, it's usually long segments of bowel, fibrofatty proliferation of the mesentery, and prominent vessels, the so-called comb sign, particularly well seen by the distal ileum. And again, really nicely seen here. Also, we should note that nodes are not uncommon. You can get nodes from malignancy like carcinoid tumors or even adenocarcinoma, obviously, but even with Crohn's disease, particularly with long-standing Crohn's disease, multiple nodes, typically small nodes, are something we will commonly see. Here's a nice example with the cinematic showing you the increased edema in the bowel in a patient with inflammatory bowel disease, and we wrote this article a couple of years back talking about how cinematic is helpful in detecting the presence of inflammatory bowel disease as well as helping us determine the extent. The article makes the point: "It remains important to utilize multiple presets and to examine the axial reconstructions and multiplanar reformations in order to ensure that important pathology is not obscured." And what that means is 3D is an additional part of our evaluation. You still need to look at the axial, the multiplanar, and the 3D. It's our feeling if you look at everything, your accuracy is surely going to increase.

While there are suggestions that cinematic rendering may improve the speed of understanding complex anatomy and could better display the range of tumors, the data remain sparse. One of the things that is frustrating to all of us these days, our new techniques like cinematic rendering are not really used as much in practice as they should be, and that's because they're financially not reimbursable at this point in most places, which limits use. Money drives everything, it seems, in medicine these days. Also, the fact is everyone is so busy, everyone is so short-staffed, you're trying to do less rather than do more. And, of course, that can be challenging, but we really try to use cinematic rendering and 3D processing in our practice because it is indeed very helpful and the clinicians love it.

Now in Crohn's disease, just to make the point: submucosal enhancement, wall thickening, narrowing. You talk about those ring signs, right? Very nicely, multiple long loops of bowel being involved. You can see the extent of Crohn's disease, the areas for strictures, the fibrofatty proliferation of the mesentery, and again, the prominent vessels. Those vasa recta when they look like that, to me the vessels are patent, but it's this inflammatory process. And Crohn's disease to me gives the best look at prominent vasa recta, and you could see on the volume rendering also some areas of relative narrowing.

Now, in terms of numbers, this article by Paulson: "Small-bowel obstruction continues to be a substantial cause of morbidity and mortality, accounting for 12% to 16% of hospital admissions..." So, indeed, it's very important. "The mortality of small-bowel obstruction ranges from 2% to 8% but can increase as high as 25% if ischemia develops." And this is why the point about doing the study correctly, making the diagnosis early, not simply recommending a... recommend a follow-up study in a day or two, okay? CT is the study of choice, sensitivity and specificity in the range of 95%, also very good at picking up complications.

Now, if you look at small bowel obstruction and you look at a big picture, what is the causes? Adhesions is by far number one, then hernias, then neoplasm. That's what we're going to be thinking, and that's how we're going to look at the images. However, when you look at the causes of bowel obstruction, it's not three lines: extrinsic, adhesions, hernias, tumors, inflammatory disease, vascular disease, radiation, gallstone ileus, foreign bodies—there's a range of things. But, regardless of how long that list is, and I can give you a list three times longer, it's how we approach the problem. And I think let's stop here and come back and think about how we approach the problem of suspected small bowel obstruction when that's the requisition we get from our referring clinician. See you in a few minutes.