Pitfalls: Question and Discussion (3)

Pitfalls in Scan Comparison for Cancer Patients

Rick, I really like the point that you made about the potential pitfall of not just comparing to the most recent prior scan in somebody with cancer or suspected cancer.

I think that is such a common problem among radiologists, and I have to, in almost every oncology case, remind the residents or fellows, you gotta go back and look further.

Anything else you might wanna say about that?

Of course, yes, of course.

It's not only seeing that something was smaller on the older CT, sometimes things are bigger and get smaller and bigger and smaller, and that indicates that it's a metastasis, usually.

Or a tumor that's being treated and untreated and treated and untreated.

So it isn't simply a progression you need to look for, but things changing in size from one scan to the next.

And it isn't even just cancer.

I have seen a couple of malpractice cases where people are following an aortic aneurysm and every dictation says no significant interval change compared to the scan from a short time ago.

Whereas if you look at something from several years ago, it's clearly gotten bigger.

And so I see that as a frequent pitfall.

Challenges in Accessing Historical Scans

Lemme ask, although the old scans are always harder to get?

Yeah. Is it delay in bringing 'em up or whatever?

Yeah. Or not at even at the same institution or things that are increasingly a problem.

Also, I know that situation occurs for in musculoskeletal with myositis specific hands, if you can determine when the patient's first injury took place, or find out if it's an avulsion.

'Cause sometimes that's the follow up film is actually over interpreted as something much more sinister than it is.

Mm-hmm.

Minimizing IV Contrast Artifacts

Todd, you showed some artifacts from dense IV contrast in veins.

I know we've tried to minimize that by using as contrast injectors, those that have the dual barrel, so you can do a saline chaser.

I'm wondering if that's in more widespread use or your own comments about that.

I think many of the academic institutions do use the dual injectors.

I don't know, I guess out in the community. It just depends on their resources.

But that does actually reduce the artifact from pooling contrast in the veins.

Adding Value to Radiology Reports for Orthopedists

Nancy, I have a tough question for you here, but I think an important one, and it has to do with the value added with our reports.

So when we're dealing with somebody like sophisticated orthopedists who are probably pretty good at looking at their own scans, what do we, as radiologists face as a challenge in really providing a report that adds value over what they can determine on their own?

I think that's actually a great question.

When we stand here in meetings like this and we try to distill for you the real practical and clinically important things to convey in the report, the orthopedic surgeons don't want a list of your findings.

They can do that themselves.

They want a summary of your findings.

They want, what do you think is really going on?

And if you can pair it down to your assessment, maybe help them with the mechanism of injury or point out some of the nuances of being definitive or more definitive.

One of the constant complaints that I hear is, I don't need a laundry list or differential.

I just stopped reading the reports because I can read them better myself.

And it's because we're not doing that.

We're not being more definitive.

We're not willing to say, this is what I think is going on.

And as you can draw these conclusions and you go to these lectures and you hear, people are saying, take that home and try to be more clinically relevant when you're interpreting the studies.

And of course, that applies throughout every area, not just musculoskeletal, maybe even more so.

I mean, there's no problem with a modern scanner of saying there's a recognizing that there's a hypervascular mass in the liver.

They can see that.

What they want to know is, in your judgment, can you make a specific diagnosis or at least point me towards one additional study that will allow me to make a confident diagnosis for management.

Management of Adenomas and Hyperplasia

Rick, I have a question for you about the adenomas hyperplasia.

I hope I'm not misquoting you, but I think you said the small lesions that have characteristics of adenoma to hyperplasia don't need to be followed.

But then I was struck by your slide that said, these are considered to be pre-malignant lesions.

So how do we resolve that?

Pre-malignant means not malignant.

Yeah.

And in studies that have looked at these lesions and followed them over time, they don't tend to change into a overt carcinoma.

And even if they do, they're still that relatively benign type that can go on for years.

But that's the current recommendation.

Yeah. That even though these things less than, or five millimeters or less can be cancers, there's no recommendation to follow them at present.

Yeah. That one I find a little troubling because

Well, it doesn't make a lot of sense to me either.

Yeah. But it, I think it's practical in the sense that these are very small, and if they grow, they're gonna grow very slowly.

And most patients we're talking about are probably 50 or 60 and would die with their lesion rather than of it.

Analogies to Colonic Polyps and Fleischner Society Recommendations

Probably an analogy in my world would be the small colonic polyp.

Right. So one of the criticisms is that on CT colonography, we're very good at finding the 6, 8, 10 millimeter polyps and above, and we're not good at the five millimeters.

And our comeback is, five millimeter lesions are almost never cancers.

And so our current recommendation is not to follow up on those or recommend colonoscopic removal.

When the original Fleischner Society recommendations for follow up of nodules came out, I thought it would be incredibly controversial.

Mm-hmm. That there would be letters to the editor flying back and forth, everybody rolled over because it was so much better, so much better way to follow people that had lung nodules.

Yeah. And I think this is probably what's gonna happen with the ground glass nodules too, that it seems a little aggressive at this point, but I think it's the right answer.

Multiple Adenocarcinomas in Situ and Treatment Approaches

While I've got you on the point here, I wanted to also follow up on the multiple adenocarcinomas in situ two, the hardest, previously known as BAC and TAF Quebec.

So we know that these patients have, I think you, and maybe even Todd, showed a case where somebody had gone to a unilateral pneumonectomy only to have additional lesions in the other lung.

Right. Other lung, one case went on to transplantation.

It reminded me of also in the world of abdominal imaging, how we're finding more and more lower grade cancers that are being treated with something other than removal of the entire organ.

And I think you briefly started to allude to that, but I wonder if you could comment on that.

Well, it's a distinction to be made between someone who has multiple lesions that represent invasive mucinous adenocarcinoma, which is the thing we've been calling diffuse BAC for years, and it's a very bad thing.

And you can't treat that with resection.

It's just going to get worse and worse and there's no hope, basically.

But there are patients with adenocarcinomas in situ two maybe mixed with invasive adenocarcinomas that present with multiple bilateral isolated lesions.

Mm-hmm. And those people, I think, will be treated with resection of the most aggressive looking tumors.

So thoracoscopic resection generally?

Yeah.

I would suspect so, but that they will be followed and lesions that look more aggressive because they develop soft tissue in the center of their homogeneously dense or whatever, that they will have these individual lesions resected.

And it's not that uncommon to see someone who's had a cancer resected before.

And you look in the opposite lung, and there are all these little ground glass things.

Excuse me. And again, drawing on analogies from the world of abdominal imaging, we're going more and more towards regional therapies, focal ablations and so forth.

Right. I know there's been some talk about it.

I maybe I just haven't kept up with the literature, but where do we stand with image guided ablations of focal lung lesions?

I'm not the person to ask.

Oh, okay. It's a good question.

But I think that certainly in this sort, in that sort of situation where you have multiple bilateral lesions that you can be reasonably confident or adenocarcinomas, that some sort of ablation or radiation gonna go that way.

Surveillance in Hepatocellular Carcinoma

Think I can tell you in for hepatocellular carcinoma, when I started my career, many moons ago, if we diagnosed HCC, you were dead in six months, no matter what you did.

Now, part of the difference is that we're doing a better job of surveillance.

So somebody that has chronic liver disease, we enter into surveillance program.

So we don't expect to find the 10 centimeter HCC, we expect to find the one or two centimeter HCC.

And we know that given the fact that they've got underlying bad liver, they're probably gonna develop more HCCs.

But in fact, we can keep 'em alive for a long time or even.

Yeah. I'm not sure it's clear that in someone who has these multiple adenocarcinomas situ to or whatever they need to do anything at all.

Yeah. And now that we're just recognizing them, I think we'll have to figure that out.

Audience Questions: Nodules in Known Cancer Patients

Questions, comments from the audience?

Yes. Question.

Dr. is the society working on recommendations for small pulmonary nodules in patients who have known cancer?

Small pulmonary nodules in people with known cancer have cancer?

Yeah. See, you gotta, you got renal cell carcinoma.

Right? And staging, you find these little not.

So the question is, how do you follow those, right?

So rather than being in a surveillance program or asymptomatic or whatever, you've now discovered a pulmonary nodule in a patient with a known cancer.

Does the Fleischner Society have recommendations for that?

Well, to answer your first question, no, to my knowledge, we're not working on anything specific for that.

But if you look at the original Fleischner Society thing from 2005, there is a mention in there that if the patient has a known carcinoma and you see one of these nodules, the follow up protocol should depend on what the tumor is.

And that's what we usually dictate.

If we're reading a CT and we see a small nodule or small nodules in someone with a known carcinoma, we do not recommend follow up based on the Fleischner Society criteria.

We recommend follow up as is clinically indicated.

The Fleischner Society criteria for solid nodules are primarily intended for people with no known carcinoma who are being screened for lung cancer, who are suspected of having lung cancer, or maybe have metastases from an unknown primary.

But if someone has a known primary, you do something different.

Did you understand that?

Well, I just was wondering what you do.

What I do, what I do is if the patient has a known tumor, and this comes up every day, 10 times, we say there's a two millimeter, three millimeter, four millimeter nodule that either we haven't seen before, is unchanged, or whatever follow up is recommended per clinical protocol or follow up significance of this nodule cannot be determined without appropriate follow up.

But we don't use that schedule that is part of the fleischner recommendations.

So they typically will then call and ask, well, I don't know when I should call.

This is the only finding that I would think if somebody, if an oncologist is taking care of someone with renal cell carcinoma and there's a potential metastasis, they probably should know at what interval that person should be followed.

They certainly would know it better than I would for individual tumors.

So you would just say followed based on,

I would say patient should be followed based on what is appropriate clinically no specific recommendations?

No.

Okay. We avoid specific recommendations when someone has a known tumor.

Bayesian Analysis in Reporting

I would like to interject a little different spin on that.

And I think you need to use a little, the formal term would be Bayesian analysis, but I do this a lot.

So you have somebody with colon cancer, not rectal, but colon cancer who has no liver lesions at all, and they've got a couple of little four millimeter lesions in the lung.

I say in the absence of liver metastases, these are very unlikely to represent metastases.

But we're not necessarily looking at a liver scan or an abdominal scan.

And,

Well, I know, but I see the lungs when I'm looking at that.

Yeah. Well, I don't see the, well, I ignore the liver.

Yeah. But I know it's there.

But I,

PET Challenges in Renal Cell Carcinoma

Todd, I was really glad that you included in your challenge cases, a patient who had the mixed cystic solid renal cell carcinoma.

That's totally, actually you showed two cases, one of a solid renal cell carcinoma, one of a cystic solid that were totally not apparent on the PET images.

And I think that is so important to emphasize.

I have personally seen about six cases missed like that.

If it's not FDG avid, it can't be important.

And man, that is so important not to blow by those.

Yeah.

I think it's particularly important given the changes in reimbursement that have occurred over the past basically six years.

Like I said, it used to be that the legacy indications, head and neck carcinoma, melanoma, lung cancer and so forth, all of them were FDG avid.

So if you didn't have uptake, it's a pretty good chance that it's not tumor, although you could always miss a secondary tumor.

However, now that all these other indications are covered, hepatocellular carcinoma, renal cells covered and so forth, I think it makes it much more challenging.

And certainly having a good contrast enhanced CT helps quite a bit in the interpretation.

And I think probably nationwide, only about 20% of centers use intravenous contrast.

But even that has increased significantly over the past three to four years.

Utility of FDG PET in Musculoskeletal Lesions

I know that, sorry to interrupt it.

The is her microphone on?

Yes. I'm just not talking into it.

Ah, there we go.

Just the utility of FDG PET in musculoskeletal where we deal with lesions and interos lesions.

So on, I'm happy to say I was so excited. I was high fiving Rick that I got the two bone cases correct.

Nice.

But it, you know, we would see a lesion and the as FDG PET was coming into vogue.

Mm-hmm. People were getting referred all the time so that if it were demonstrating activity, then it suddenly was thought to be sinister.

We learned a lot about benign lesions having uptake.

And I think that people are starting to appreciate that, maybe dial back a little bit more on referring everything because we were starting to treat a benign lesion.

Right. Or you're treating an x-ray rather than the patient and or the real findings.

So you can't understate to make sure that you're looking at the CT in with respect to the patient and patient care.

Yeah. Unfortunately, there are centers still to this day where they read PET CT and if this pet's negative, they don't look at the CT and they don't read the CT.

Yeah. I think that is a mortal sin.

I really do. And this has been discussed ag nauseum for since 2004.

We came together as a group of PET CT interpreting physicians and said, look, if you're doing PET CT and you're radiating the patient, someone has to be reading the CT part of the exam and dictating any relevant findings.

But still,

PET Pitfalls in Differential Diagnosis

I wanted to pick up again on something Nancy said.

'Cause I think it's a really important point.

A lot of times if your differential diagnosis includes some chronic fibroin inflammatory process as well as cancer in various forms, I see people making what I think is a mistake very frequently.

PET alone will be the deciding factor.

You showed an ICE case of a carcinoid tumor, but I have seen fibrosing mesenteries be FDG avid and have people say, that settles it. This has gotta be cancer.

You can't ignore the other volume of knowledge that we have based on other imaging criteria.

You showed a retroperitoneal fibrosis that was FDG avid.

Same thing.

So, and yesterday we talked about IgG four disease, the autoimmune pancreatitis and its first cousins.

I've seen many mistakes made where people do a PET scan and say, this lesion in the pancreas is FDG avid. That settles it. This is definitely cancer.

So you have to use other hints in that, the speakers have pointed out to you.

So it's PET is a wonderful advance.

I love PET. I can't imagine practicing oncology without it.

But the stuff that Todd has covered here today, and he, I'm sure he could give a week long course himself just on other potential pitfalls, exceedingly important.

Don't worry, we won't hold you to that.

Quiz on PET/CT Artifacts and Findings

Okay. First question.

Brown fat can be diagnosed on PET CT by measuring the attenuation in an area where there is increased FDG activity.

The Hounsfield units in the area of interest should measure approximately negative 800 to negative 60, negative 150 to negative 50.

50 to 150 or 600 to 800.

Good.

Yes. So the correct answer is two.

Which of the following lymph nodes can cause false positive FDG uptake due to an attenuation correction artifact on PET CT number one, necrotic nodes number two, normal nodes number three, calcified nodes, or number four malignant nodes.

Perfect.

Yes. So calcified nodes interestingly can give you, it's not common, but it can give you, depending upon the density of the calcification in a postmenopausal woman.

FDG activity seen within the uterine cavity on a PET scan is most likely due to one endometrial carcinoma, two ovarian carcinoma, three menstruation or four cervical carcinoma.

Good.

Yes. So endometrial carcinoma would be the primary consideration.

Yep.

All the following are likely to cause a CT based attenuation correction artifact except number one, IV contrast number two, barium based oral contrast number three, negative for water-based oral contrast.

Number four, high attenuation material chemotherapy port.

Perfect.

Yes. So, as we said, any of those high attenuation structures can potentially give you an attenuation correction artifact.

One way to avoid an attenuation correction artifact on PET CT using oral contrast is to one, double the amount of barium.

Two, use higher density barium.

Three, use water-based oral contrast such as Volumen four use barium followed by two cups of water.

Yes. That was sort of a gimme from the last one.