Evaluation of Indeterminate Renal Masses with CEUS: A Diagnostic Performance Study - HD

Introduction

Hi, I'm Richard Barr from South Woods Imaging.

Today I'd like to give you a talk on characterization of indeterminate renal masses using contrast enhanced ultrasound.

Today I'd like to review a study that we've done evaluating indeterminate renal masses with contrast enhanced ultrasound, a diagnostic performance study.

These are my disclosures.

The purpose of our study was to determine the utility of contrast enhanced ultrasound and characterization of indeterminate renal masses.

This study has been published in radiology in 2014. As listed on this slide.

Renal Masses and Bosniac Classification

In determining renal masses are a common clinical problem. It is estimated that more than half of patients over the age of 50 have at least one renal mass.

Many renal masses are found incidentally during an imaging examination.

The majority of all masses are benign simple cysts, which can be confidently diagnosed as such with cross-sectional imaging alone, excluding inflammatory masses, vascular abnormalities, and pseudo tumors.

Most solid enhancing renal masses are malignant complicated cystic lesions with enhancing soft tissue components, excluding the actual cyst wall are all malignant.

However, there is a wide range of complicated cysts between the obviously benign and the obviously malignant lesion.

These complicated cystic lesions are classified using the Bosniac classification. This was originally used to describe CT findings, but is now widely applied to other modalities in this classification.

There are four categories.

A bosniac one is a simple cyst,

A BOSNIAC two is minimally complicated cyst with a single thin less than one millimeter septation and or thin calcified wall. Well marginated non enhancing high attenuation regional lesions less than three centimeters are also included in this category.

There's a subcategory of bosniac two F, which have an increased number of septa, minimally thickened or enhanced SEPTA or walls or thickened calcifications. Hyperdense greater than three centimeters with no enhancement are included in this category.

A BOSNIAC three lesion is thick or multiple septations mural nodules and or hyperdense on CT and bosniac four lesions or solid vascular masses based on the literature lesions with the BOSNIAC one score have a 0% chance of malignancy and do not require additional follow up.

Bosniac two lesions have an essentially 0% chance of malignancy, and usually you're not recommended for additional follow up.

However, the subgroup of Bosniac two F lesions have approximately 5% chance of malignancy and serial monitoring is advised for these lesions.

BOSNIAC three lesions have been reported to have a 31% to 100% percent malignancy. And these lesions are usually either biopsied or go to surgical resection.

A bosniac four lesion has an approximately a hundred percent chance of malignancy and should be surgically resected.

Contrast-Enhanced Ultrasound

Contrast enhanced ultrasound uses a true intervascular agents. These agents are not nephrotoxic or hepatotoxic.

They have a very short half-life, usually approximately five minutes and allow for multiple injections at one setting.

The ability to subtract out background soft tissue, I mean that is a contrast only image is extremely helpful, especially when we're looking at small nodules and or septations. As we'll see later on in this talk.

The narrow ultrasound beam also allows for improved visualization of vascularity in small structures such as SEPTA or small neural nodules as compared to MR and CT E.

The vascular enhancement patterns of contrast enhanced ultrasound in the kidney are similar to those of CT and MRI in the arterial phase, but may be different in the delayed phase as CEUS agents do not extravasate.

Therefore, you'll not see any enhancement within the caly, fibia or renal pelvis, and this is sometimes very helpful that you do not have the overlapping attenuation from contrast within the collecting system.

Study Design and Methods

This was a retrospective performance study that was approved by I-R-R-I-R-B and was HIPAA compliant with waiver of informed consent.

Patients included 721 patients referred to contrast enhanced ultrasound with 1018 indeterminate renal masses from 1999 to 2010. These were identified initially on an imaging study, be it A-C-T-M-R or a non enhanced ultrasound.

The patients were 44% female and 56% male. Their ranges of age was 17 to 95 with a mean of approximately 70 years.

Lesion size varied from two millimeters to 161 millimeters with a mean of 27 millimeters.

Because this was done over a approximately 10 year period of time, the exams were performed on a state ofthe art equipment with contrast specific software that was appropriate for that time that the examination was performed.

We did examinations using either optisan or definit, and the doses range from three ccs in 1999 to 0.3 ccs more recently. And the dose changed depending on the equipment as it improved over time.

The doses were adjusted to be most appropriate for that level of contrast software.

We had our lesion classification based on contrast enhancement ultrasound enhancement patterns. We considered a lesion with no flow within the lesion as benign.

If there were occasional bubbles in the septi, usually seeing a single bubble going through the septation, we would consider that benign.

If there was enhancement equal to normal cortex in all enhancement phases and there was a presence of a central pyramid, we consider that benign as a pseudotumor echogenic mass with enhancement less than renal cortex, often with a peripheral distribution. We considered a benign angiomyolipoma, but always obtained a CT or MRI to confirm the presence of macroscopic fat in the lesion.

When we began the study, if there was a constant flow of bubbles in a fine septation without nodularity, we consider that indeterminate. And as you'll see as we go through this talk at the end, we now believe that that actually is a benign finding.

Echogenic mass with enhancement equal to or greater than normal renal cortex. With washout, we considered malignant in any other mass with blood flow solid or cystic was considered malignant.

Results and Examples

Here is an example of a simple cyst. We can see that there's no bubbles at all within the cyst.

These dotted arrows point to a medullary pyramid. The again, because we do not have excretion of the ultrasound contrast like in CTN mr, we can see the difference in blood flow between the medullary portion of the kidney and the cortex with the medullary portion receiving less blood flow.

And again, we can see the renal pelvis in fibula without any bubbles. Again, because there is no excretion of the contrast, this is an example of a bosniac two cys with a fine septation.

The image on the right is our reconstructed B mode image, and you can see here is that fine septation. And if you look on our contrast ultrasound study, we can see three individual bubbles transversing that septation during the course of the examination. And again, we consider this as a benign finding.

Here we have on CT a fat containing lesion, which was an angiomyolipoma. On standard ultrasound, we had increased echogenicity. And on the contrast and hand ultrasound, we often saw that these lesions did not have as much below as renal cortex. They often had a little bit of blood flow as you see here in the periphery, and we consider that for this study an angiomyolipoma.

In this patient, you can see there was a little contour abnormality on the ct, and here you can see in the arterial phase of the contrast ultrasound that the lesion is hypervascular to the renal cortex.

And on delayed images, you can see that we have washout of the lesion. And again, this was our classification of a malignant lesion.

We did a per patient analysis, so if a patient had one or more contrast in hands positive lesions, the patient was classified as ultrasound contrast positive. If the patient had a ultrasound contrast indeterminate lesion and did not have a positive lesion, the lesion was classified as indeterminate.

If a patient did not have a positive ultrasound enhancement or a contrast enhance indeterminate lesion, they were classified as a contrast enhanced ultrasound negative.

If there were multiple lesions that were within the same CEUS category, the most complex lesion based on the bosniac criteria was selected as the patient's lesion.

The contrast enhanced ultrasound enhancement patterns were used to characterize masses as benign or malignant lesions with a definitive diagnosis. By that, I mean they were biopsied or a definitive diagnosis was able to be made on MR and CT such as macroscopic fat in the lesion as an angiomyolipoma.

We had 306 out of the 1018 lesions, or 30%. These were correlated with their ultrasound enhanced findings.

We had another 167 lesions that were benign in that category, and we had 139 lesions that were malignant. Again, this is the category where we had a definitive diagnosis, made.

Lesions without a pathological diagnosis, or about 70% were followed for up to 10 years.

The diagnostic accuracy measurements were calculated using the pathological diagnosis as the reference standard, as well as stability of a lesion at three years and five years.

On a per patient analysis contrast enhanced ultrasound had a sensitivity of 100%, with a confidence interval of 97 to 100%, a specificity of 95% with confidence levels of 90% to 98%. The positive predictive value was 95%, and then negative predictive value was 100%.

We did have five false, five false positive masses, which included three oncocytoma and two bosniac classification, three cystic lesions.

Of the 290 lesions that had at least a 36 month follow up, none of the lesions changed to reclassify the lesion.

If we include these lesions, assuming lesions classified as malignant were malignant. And by that I mean we have some patients that had enhancing masses that were small and elderly patients that were elected to be followed and not biopsied.

Then we have a total of 596 lesions. The sensitivity was, again, 100%. The specificity was 97% a positive predictive value of 92% and a negative predictive value of 100%.

This is our flow chart of our patients. Again, we had 721 patients.

We had 265 that had confirmed diagnosis. 126 were malignant and 139 were benign.

You can see on the malignant lesions, all 126 had a positive ultrasound contrast, and zero had a negative ultrasound contrast. These were mostly renal cell carcinomas, although we had one case of transitional cell carcinoma and one case of lymphoma.

If you look at the benign lesions, five had the ultrasound positive, 132 ultrasound negative, and we considered two indeterminate. And these were lesions that had a constant blood flow through a syn sensation. And based on our results, we would now classify those as benign.

And again, you see, we had a wide range of benign pathology.

If you look at our 721 patients, 456 of these had follow up. 256 had less than 36 month follow up, 200 had to greater than 36 month follow up, and 137 had greater than six month follow up. And you can see the distribution of those.

On this slide, again, if we look at the confirmed diagnoses, all of the lesions that were malignant had a positive ultrasound. There were no malignant lesions that had a negative ultrasound. And again, a wide range of pathology all listed here.

And we had five false positives, three of which were oncocytoma, so they were solid masses with blood flow. And the other two were bosniac three lesions that we felt there was some blood flow within the lesion.

Let me show you some examples and compare them to ct. Again. We have a large amount of referrals for ultrasound contrast in kidney cases where the patient cannot receive CT or Mr. Contrast.

And in this unenhanced ct, you can see there are some septations that are pretty high attenuation that may represent calcification, but it's very difficult to determine if this mass is benign or malignant in the associated enhanced ultrasound.

With contrast, you can clearly see that there is a solid component along the wall that enhances, and this was a cystic renal cell carcinoma.

Here's a lesion on ct. There are some material within a lesion that has a little bit higher attenuation than simple fluid.

And if you look on the ultrasound, we can see that soft tissue material within the cyst. But if you look on the enhanced ultrasound, there's absolutely no enhancement of any of the components within the cyst In this patient that had a hematoma.

This was a patient that was unable to have ultrasound contrast. And you can see that this patient had a funny shaped kidney.

This inferior portion on this slice was a little bit more lower in attenuation, the remainder of the lesion.

You can see on the enhancement with ultrasound that that portion is markedly enhances and has minimal enhancement enhancement within the central portion representing necrosis in this necrotic renal cell carcinoma.

This is a case with mr. You can see that this is the pre contrast mr, and this is the post contrast, Mr.

Here's the portion of the normal kidney, and on the without contrast, we have a little bit of signal, which we felt was probably noise after we gave contrast.

We can see some addit enhancement or increased linear signals, but it's often very difficult to determine if this is secondary to patient motion or could be noise.

But on the ultrasound with contrast, we have a much higher resolution because we have much thinner slices, we can see that a marked network of irregular septations is present, all of which are enhancing. And again, in this cystic renal cell carcinoma.

This is a 65-year-old male with renal failure. He had a right renal mass noted on CT that has some rim calcification on B mode image.

The lesion looked like it had low level internal echoes and was solid. However, when we gave contrast, there was no blood flow within this lesion at all, and this was just a benign, complicated cyst.

This slide gives a distribution of our enhanced ultrasound findings per patient in patients with a confirmed diagnosis on the Bosniac classification of this study.

And again, you can see that our indeterminate lesions were classified as bosniac threes. And again, I think we would've now considered these benign and moved them up to a benign category.

On ultrasound, we've had again, if you look, the majority of our masses we're done on enhanced ct. A few cases on enhanced mr.

This is the lesion distribution based on enhanced ultrasound. You can see that the vast majority of our patients had one lesion, 107 had two lesions, and list that here all the way to one patient having nine lesions.

And you can see that the malignant appearance was seen in the majority of the one and twos with the one of the lesions in the patient with nine lesions was a malignant lesion.

Conclusion

So to conclude, ultrasound contrast has a high sensitivity, which is a hundred percent with confidence levels of 97 to a hundred percent. In our study, it had a very high specificity of 95% with confidence levels of 90 to 98% in our study, in characterizing indeterminate renal masses, visualization of occasional blood flow bubbles or constant flow bubbles in a septation without nodularity or within cystic masses of benign finding.

16 of those we had on all 16 were followed for greater than five years, and they did not change. So we now believe those are benign findings.

Angio myel lipomas in our hands enhance less than normal regional cortex on the arterial phase. They often have some enhancement in a peripheral distribution.

While echogenic renal cell carcinomas, which we had a small number, have a diffuse intense enhancement with washout contrast.

Enhance ultrasound has a high positive predictive value of 92% and a negative predictive value of 100% in characterizing indeterminate renal masses as benign or malignant.

I will note that based on our classification scheme, benign oncocytoma will be considered as malignant lesions. And in our institution, our urologists remove these. So actually our positive predictive value was actually higher based on the way our urologists treat oncocytoma.

The application of contrast enhanced ultrasound findings to the bosniac classification can provide a valuable contribution to the accurate classification of cystic lesion.

Contrast enhanced ultrasound evaluation of renal masses could decrease the need for CT with the associated radiation, or MRI with its high expense.

CEUS can play an important role in renal mass characterization in patients with contra indications to both CT or MR contrast.

Thank you for your attention.