Working Up Thyroid Nodules: A Logical Approach - SD

Introduction

Hello, my name is Dr. Mary Freddy's. I'm an associate professor of radiology at Harvard Medical School in Boston, and the assistant director of ultrasound at Brigham and Women's Hospital in Boston.

This morning I'm going to be speaking about working up thyroid nodules, a practical approach.

Today's topic is working up thyroid nodules, a logical approach.

Background on Thyroid Nodules

Just as some background, we all know that approximately 5% of adults have palpable thyroid nodules.

In addition, somewhere between 10 and 41% of adults will have nodules identified at ultrasound.

However, 50% of adults will have nodules identified at autopsy.

We also know that the frequency of thyroid nodules increases as the patient age increases more background.

We know that thyroid nodules are more common in women than men, and approximately 50% of thyroid nodules are solitary nodules.

Thyroid Cancer Overview

A few words about thyroid cancer.

Thyroid cancer is of sufficiently high incidents, 24,000 new diagnoses each year to make it an important diagnosis for us to be able to identify.

In addition, it's the eighth most common malignancy in women.

It's also increasing in frequency across the world, but nodular disease of the thyroid is much more common than thyroid cancer.

The overall incidence of cancer in thyroid nodules is somewhere between 9.2 to 13% with both higher and lower rates reported in the literature, mostly based on selection criteria used in each project.

Cancer rate per patient is the same for patients with a solitary nodule or multiple nodules.

The cancer rate in nodules is the same, whether the nodule is palpable or non palpable.

Again, a little background on the epidemiology of thyroid cancer.

A T one cancer is a cancer smaller than two centimeters confined to the thyroid.

These cancers have extremely low morbidity and mortality, and both morbidity and mortality increase with patient age.

The 10 year recurrence risk for differentiated thyroid cancer is 10% for stage one, 20% for stage two and 30% For stage three,

the yield of cancer when we look at thyroid nodules, is higher in patients who are high risk.

In addition, clearly patients with clinical disease, either a palpable mass in the neck or regional metastatic disease benefit from early diagnosis and treatment.

However, the patient with an occult or an incidental cancer may not show that same clear benefit from an early diagnosis, except there is a low but persistent rate of distant metastatic disease.

Even with the small cancers PEG GRE at all identified in J-J-C-E-M in 2006, when they looked at nodules under 15 millimeters, that approximately one in four of the patients in their series had recurrent or persistent disease again, in their small thyroid cancers.

Chow in cancer in 2003, who looked at cancers under 10 millimeters identified a 5% incidence of lymph node recurrence and a 2.5% incidence of distant metastatic disease.

So the more we image the neck with sonography, the more nodules we'll identify.

The more nodules we find means the more procedures we perform and the more cancer we'll find.

But it also increases the number of complications that patients will suffer back to the same number we started with.

10. 10 to 14% of nodules are malignant.

The question remains which ones.

The challenge in thyroid ultrasound is to reassure the ma majority of patients who have benign nodular disease and diagnose the malignant minority.

Risk Factors for Malignancy

Clinical Characteristics

The clinical characteristics that increase the risk of thyroid malignancy include the following, number one, probably the history of radiation therapy to the head or neck.

In addition, a family history of thyroid cancer and age under 20 or over 60 years.

Patients who have undergone surgical menopause patients with breast cancer, those with an increased BMI all increase the risk of malignancy.

Interestingly, lower risk for thyroid cancer is found in patients who smoke and patients who exercise, I recommend exercise.

I do not recommend smoking.

Clinical findings that increase the suspicion of thyroid malignancy include rapid growth of a mass in the neck, a firm mass at palpation, masses that are fixed to surrounding structures or that have caused vocal cord paralysis and local adenopathy.

Patients presenting in this group will benefit from rapid diagnosis and treatment.

Sonographic Characteristics

In terms of sonographic characteristics of malignancy, most predictive of malignancy are the presence of microcalcifications solid nodules, nodules with, extensive internal vascularity and nodules that present with a associated cervical adenopathy.

Somewhat less predictive are hypoechoic echo texture of the nodule and irregular margins of the nodule.

Not predictive is size.

Now, if you could tell which nodule was cancer just by looking at it with an ultrasound, no one would ever need to have a TAL aspiration.

But the data show that no sonographic appearance is predictive enough for cancer that we can avoid the FNA and go straight to thyroidectomy.

Data from Brigham and Women's Hospital Thyroid Nodule Clinic

So let's see some data in the Brigham and Women Hospital thyroid nodule clinic that, this is data that began in 1995 and went through 2003.

We saw 1,985 patients and evaluated 3,483 nodules in this patient population.

The prevalence of thyroid cancer was approximately 15%.

We had 295 patients over that eight year, period with thyroid cancer.

There was a statistically significant difference in the prevalence of cancer in men with a rate of 20.2% versus women with a rate of 14.1%.

We also identified that a solitary nodule had a higher likelihood of malignancy than a non solitary nodule, 14.8% in the solitary nodules, 8% in the non solitary nodules.

For the 120 patients in our series who had multiple thyroid nodules over the size of 10 millimeters, the cancer was multifocal in 46% of those patients, 72% of the cancers occurred in the largest nodule.

And as the number of nodules in the thyroid increased the frequency of cancer in the largest nodule decreased In the 120 patients with multiple nodules in thyroid cancer.

If we had done an FNA on the largest nodule only, we would've found 86% of the cancers in patients who had only two nodules.

We would've found 50% of the cancers in patients who had three or more nodules.

So the more nodules the patient has, the less likely it is that the cancer is in the largest nodule In this patient series.

If we had sampled the four largest nodules, we would've found 100% of the cancers.

If we stopped at the three largest nodules, we would've found all but one of the cancers.

However, if we chose the nodule that was dominant by size only 69% of the cancers would've been found in this group of patients and nodules that I just presented, we looked at a subset of, patients with 885 nodules, and we looked at these, the sonographic criteria of all of these nodules in the 729 patients in this subset of patients, the malignancy rate overall was 10.8%.

In this group, when we separate the nodules out in the basis of size and we look, at the right hand column, we see that there was no statistically significant difference in malignancy rate, whether the nodule was 11 millimeters or over 30 millimeters.

This is an example of what we're talking about in this particular patient.

We have a very large nodule on the right side, solid 4.5 centimeters in transverse diameter in contrast to the left lobe where we have a very small nodule, eight millimeters transversely biopsied because it was 14 millimeters in the sagittal plane nodule on the right, benign nodule on the left malignant.

Nodule Composition and Malignancy Rate

Now, when we look at nodule composition and rate of malignancy, we can see that with a completely solid nodule, there's a malignancy rate of 14.3%, and as the nodule becomes progressively more cystic, the malignancy rate drops to zero.

And this is a statistically significant, change.

Here's a solid nodule, sage and transverse.

This is a malignant lesion in contrast to this nodule, mostly cystic over 75% cystic, a little bit of solid tissue in the back corner.

And this of course is a benign lesion.

Nodule Echogenicity

What about nodule echogenicity?

Well, we looked at those nodules that were over 50% solid, and in those nodules, those that were determined to be hypoechoic had a 15% risk of malignancy.

And the hypoechoic or iso coic rate was 9.2%.

Also statistically significant.

Here's a solid nodule, 14 millimeters.

This is a follicular cancer.

So not every cancer is hypoechoic.

Follicular cancer is a little bit more common to be echogenic.

Calcifications

What about calcifications?

Well, we separated calcifications in this series to punctate coarse or rim calcifications were grouped together and nodules were described as having no calcifications.

And this was a statistically significant relationship with punctate calcifications having the highest risk of malignancy course or rim calcifications having a higher than baseline risk and nodules with no calcifications having the lowest risk.

Here's a typical hypoechoic nodule with punctate calcifications, and this is a papillary thyroid cancer.

Another nodule with, again, multiple punctate lesions as well as some coarse calcifications, also a papillary cancer.

Logistic Regression Analysis

So we took the statist, the most statistically significant, identifying characteristics, number one, male versus female.

Number two, solitary versus multiple.

Number three, presence of calcification and number four, nodule composition and performed multiple logistic regression of these sonographic characteristics to create this table looking at likelihood of malignancy in a solitary nodule.

So this table that we're looking at again is for a solitary nodule.

This is for a female patient.

If you look at nodule composition in the left hand column, moving from completely solid to completely cystic and then looking at presence of punctate calcifications course or rim calcifications versus no calcifications, we can see that the risk of malignancy in a solid nodule solitary with punctate calcifications in a woman is approximately 32.7%.

Contrast that, for example, to a mixed cystic and solid lesion with no calcification, the risk of malignancy is down to 6%.

Same statistics for male patients with a solitary nodule.

Again, here's the number of particular interest.

A solid nodule with punctate calcifications nearly a 50% risk of malignancy in this group compared to a mix cystic or solid, only a 10% risk.

Here's the same table with a non solitary nodule in a female patient risk is down to 18.4% in that same category.

And you can see that that, even the mostly solid or the mixed cystic and solid lesions have a lower rate of malignancy than the background rate in our population.

If no calcifications are present, here's the same table for male patients, multiple nodules close to a 30% risk for that solid calcified nodule and down in the four to 5% risk in these lower categories.

So these tables could help you in your practice stratify or give a risk assessment to each individual nodule that you identify at thyroid sonography.

Follow-up and Growth of Nodules

What about nodules following them over time?

Well, we know that benign nodules when we follow them for five years, have an average growth of 15% over five years.

So because we can follow benign nodules and see that this is the natural history of thyroid nodules, we've defined that too much growth of a nodule is more than the 15% that we've identified as baseline in the interval of the five years.

Therefore, when we see a nodule that grows more than 15% in an interval follow-up exam, we will perform a repeat fine needle aspiration.

Consensus Guidelines

Society of Radiologists in Ultrasound (SRU) Consensus Conference 2004

Another way of thinking about thyroid nodules is to use the, information developed at the Society of Radiologists and ultrasound consensus Conference held in fall in the fall of 2004.

This was a statement developed to assist physicians in deciding when fine needle aspiration is indicated in a thyroid nodule that is identified at ultrasound.

The recommendations of this consensus conference allow flexibility for the practitioner in determining which nodules require can, intervention.

The questions that this consensus conference identified that does not, that do not have answers is, number one, does early diagnosis of thyroid cancer help?

And number two, what's the cost benefit ratio of diagnosing a cancer early versus sending multiple benign nodules to surgery?

So for the purposes of this discussion, a nodule was defined as any discreet lesion measuring at or greater than one centimeter identified at thyroid ultrasound.

And this is for two reasons.

There's some uncertainty whether earlier diagnosis of smaller lesions is valuable.

Do these sub centimeter cancers sometimes termed micro carcinomas need to be diagnosed at an early rate.

In addition, there was some concern of the panel for a possibility of having a higher number of non-diagnostic fine needle aspirations if a lower threshold for intervention was identified.

Things to think about when you look at the consensus conference recommendations is to realize that the criteria for fine needle aspiration change as the size of the nodule changes nodules that are smaller with high risk criteria.

FNA is recommended at the smaller size with nodules that have low risk criteria.

The the consensus conference felt that you could wait for some growth of the size of that nodule.

So the general recommendations of the consensus conference for solitary nodules over one centimeter in maximum diameter is to strongly consider doing a fine needle aspiration for any nodule over one centimeter.

If microcalcifications are present, we strongly suggest that you consider a fine needle aspiration for any nodule measuring over 15 millimeters.

If it's a solid nodule or almost entirely solid, or of course, calcifications are seen within that nodule, the panel recommends that you consider fine needle aspiration for nodules over 20 millimeters in largest diameter if the nodule is mis mixed, cystic and solid, or if this is an entirely, almost entirely cystic nodule with a solid mural component.

In addition, the panel suggests that you consider fine needle aspiration for any nodule that has grown substantially since the prior ultrasound.

Although the term substantial was not defined in the guidelines For patients with multiple nodules over measuring over one centimeter and maximum diameter, the recommendation is to consider aspiration of one or more of the nodules, prioritize the selection of which nodule to biopsy based on the previously stated criteria.

In other words, a nodule with microcalcifications should have aspiration performed preferentially over a mixed cystic and solid nodule without calcifications present.

In addition, fine needle aspiration is likely unnecessary in diffusely enlarged glands with multiple confluent nodules of similar sonographic appearance.

Importantly, the presence of abnormal lymph nodes overrides sonographic features and requires biopsy of the lymph node and or if necessary, the ipsilateral thyroid nodule.

American Thyroid Association (ATA) Guidelines 2009

Now the American Thyroid Association released guidelines in 2009 for the A TA guidelines in a high risk patient, a nodule with suspicious sonographic features should be, should undergo fine needle aspiration.

If the nodule is greater than five millimeters in the group of nodules without suspicious features that measured between five and 10 millimeters, the a TA felt there was insufficient evidence to recommend that these nodules be sampled.

Similar to the SRU abnormal nodes in any patient will prompt FNA of any size I ipsilateral nodule.

In the non-high risk patient, the presence of microcalcifications should prompt FNA at a size of 10 millimeters or greater.

A solid and hypoechoic nodule should also be FNA at 10 millimeters or greater.

And a solid and either iso or hyper or coic nodule could be FNA anywhere between a 10 or 15 millimeter lower threshold.

Mixed echogenicity nodules with suspicious features should be FNA beginning with a size of 15 millimeters.

And mixed echogenicity nodules without suspicious features could wait until they reached a diameter of 20 millimeters to undergo fine needle aspiration.

The A TA identified that in sponge offor nodules, in other words, nodules with multiple microcystic components.

FNA was likely not indicated.

As the data does suggest that this appearance is benign.

A purely cystic nodule does not need fine needle aspiration.

The A TA also identifies that if there's substantial growth in the nodule, which is defined as a 50% change in volume or a 20% increase in at least two dimensions at the 16 to 18 month follow up, that nodule should have an FNA repeated of note for the thi For the a TA uh, guidelines, ultrasound guided finding aspiration is recommended for nodules that are non palpable, predominantly cystic or located in the posterior aspect of the thyroid.

Otherwise, fine needle aspiration is done with palpation alone.

Practice at Brigham and Women's Hospital

Now what do we do at Brigham and Women's Hospital?

Well, at Brigham and Women's, we perform ultrasound guided fine needle aspiration of all nodules that measure over 10 millimeters in size.

We do not, stratify our nodules based on a risk criteria of, Identified sono graphically abnormal characteristics.

Basically what we do is start with the two largest nodules or if we, if we feel that a particular nodule is sono graphically suspicious, and we will perform fine needle aspiration of two, two nodules in each patient, and then we'll discuss having the patient return for additional fine needle aspirations at another visit.

And it may take us several visits to clear all of the nodules, we typically will do as many as three nodules in each side of the gland before we stop and watch.

Exceptions to this practice would be the very old, older patient and those patients with shortened life expectancies.

In other words, with a malignancy or a disease that, would preclude chasing down all of the nodules in the gland.

Fine Needle Aspiration Procedure

Now, ultrasound guided fine needle, a fine needle aspiration has been proven to be safe, accurate, and inexpensive.

The complications of fine needle aspiration are rare and usually minor and typically include pain or hematoma or both.

Ultrasound guidance assures that the sample comes from the nodule in question and can, in addition, can direct the nodule into the solid or vascular areas of each individual nodule.

Here's an example of a right-sided thyroid biopsy.

You can see that the needle is coming in from the medial aspect of the neck, from the region of the patient's trachea, carotid and jugular would be over here.

Here's the thyroid nodule and the needle is moving back and forth in the solid portion of the nodule of avoiding the cystic portion of the nodule for a left-sided thyroid biopsy.

In this instance, the trachea is here, the needle is coming in from the just, lateral to the trachea.

Here's the carotid artery, and you could see the tip of the needle moving back and forth inside this hypoechoic small thyroid nodule.

Another left-sided biopsy, uh, jugular vein, carotid artery trachea is here we come straight in just lateral to the trachea, directly into the hypoechoic thyroid nodule.

Management of FNA Results

Looking At, the results from the Brigham and Women's Thyroid Nodule Clinic.

When we receive a suspicious diagnosis or a diagnostic uh, result from of papillary thyroid cancer, these patients go directly to the operating room.

Patients with atypical results on their fine needle aspiration will have at least two res aspiration attempts until we can identify either that the nodule is benign or the nodule is suspicious for malignancy.

Any patient with an insufficient diagnosis will addition ha be re aspirated at least twice?

No, uh, nodules will have a thyroid, uh, stimulating hormone or A TSH checked to identify that they are functioning as functioning.

Thyroid nodules are highly unlikely to be malignant and do not need to have surgery.

And then those patients either atypical or insufficient that cannot be cleared will go to the operating room.

All benign fine needle aspiration nodules are reevaluated in nine to 12 months to identify for growth.

Outcomes and Recommendations

Looking at the data from our thyroid nodule clinic through 2004, we have seen approximately 3,600 patients.

2,587 have undergone fine needle aspiration of at least one thyroid nodule.

We've aspirated approximately 3,500 nodules and identified 373 cancers.

Now, if you look at the traditional cancer rated surgery, when patients with palpable nodules are sent directly to surgery, the rate of cancer in those patients with palpable nodules and nothing else is approximately 10%.

The overall rate for patients sent to the operating room after having a positive fine needle aspiration turns out to only be approximately 33%.

At Brigham and Women's Hospital right now, we've been able to increase our positive rate at surgery to over 56% with this approach.

So what do you do? Well, your options are to follow the Society of Radiologists and ultrasound consensus guidelines to look at the a TA guidelines and follow those.

Or you can use the Brigham and Women's system of finding little aspiration of each nodule over 10 millimeters in size with repeats for nodules that grow over 15% per year.

Unfortunately, there's an in, there's insufficient data right now to use a PA pattern approach at present, but it is important to learn the signs that you can use to recognize cancers or worrisome nodules.

We are hoping for a multi-institutional study that would allow us to direct, uh, finding needle aspiration to a smaller number of thyroid nodules.

In addition, hoping for some, mostly for some don't touch characteristics, for some nodules or for some patients that would allow you to say that nodule does not need to be stuck.

Future Directions

Now, as we're still hoping for a better selection process on the horizon or in the future may be elastography at the present time.

There isn't any data that show us that elastography can convincing convincingly keep us from doing fine needle aspiration for a definitive diagnosis.

But let's look at some biochemical markers that may be helpful in the future.

One of the things that we're attempting to address is, as the SRU brought up at the very beginning, does early diagnosis help anyone?

And again, let's look at that cost benefit ratio of identifying cancers early versus having patients undergo thyroidectomy unnecessarily.

We know that between 15 and 30% of fine needle aspirations are indeterminate.

This is across the board.

And when you take those indeterminate nodules to surgery and then give the nodule to the pathologist, the majority of them end up being benign and the patient has undergone what some would say is non indicated or unnecessary surgery.

So what is currently in process is a multi-institutional study, which is testing the ability of a novel molecular classifier to accurately identify benign thyroid nodules.

This classifier uses signals from approximately 200 genes to identify benign nodules and thus allowing the patient to avoid surgery.

This classifier is applied to a single fine needle aspiration sample, and the preliminary results show a 96% negative predictive value from this single sample.

The data is available only in abstract form right now, but the full data is soon to be released.

Interestingly, Brigham and Women's Hospital participated in this study sending in one sample on each nodule that we aspirated.

Our data or the samples that we sent from Brigham and Women's were 100% acceptable.

There was enough tissue in 100% of our samples to run this test.

However, samples from most of the other institutions did turn out to be unsatisfactory.

So the goal here would be that a single needle pass would give you the results.

If multiple needle passes are necessary, that's not as useful as of a test.

In addition, there is a mutation, uh, current, uh, identified as the BRAF F mutation, which is present in 30 to 50% of papillary cancers.

And interestingly, it is identified in the more aggressive papillary cancers and is not present.

The BAF negative cancers tend to be the more indolent or less aggressive type cancers.

So this might be a very useful way of determining which cancers require, uh, surgical intervention and which cancers could potentially be watched for growth over time.

So that is a extremely valuable piece of information to know about a particular thyroid cancer.

However, it at the present time is extremely expensive, running approximately $600 per test.

So at the present time, we are using this at Brigham and Women's Hospital to help identify surgical planning.

Does a full lymph node dissection need to be done at the initial time of diagnosis?

Probably in BRAF positive case patients, that is the case.

B BRAF negative patients maybe can have a, a smaller surgery.

What about galatin three?

Well, this is a protein that assists in identifying the presence of cancers in those indeterminate cytology cases or atypical cytology cases.

What happens is that the cytology slide is stained for the presence of this protein.

Now, the results on GALATIN three have mostly come into the literature from Europe and it's not being universally used in the United States.

And one issue here is that at least at our institution, we do not create a slide stain that could be stained for galatin three.

Our results are sent in liquid form, uh, with a CytoLight rinse to our cytopathologists, and we don't know that Galatin three staining will be as useful in the liquid cytology samples.

Again, our goal is to increase the percentage of surgical cases to 100% cancers so that no patient with a benign nodule undergoes thyroidectomy and to identify those aggressive cancers to help the patients who really are going to have trouble with their thyroid cancer.

However, to add currently to add all possible tests on to a, to work up a thyroid nodule adds a bill of approximately $2,500 per nodule.

So at the present time, we're waiting for more results.

Conclusion

I hope this has been a useful way for you to consider how you can address the presence of thyroid nodules in the patients that you see in your practice.

Thank you very much.