Guidelines for FNA: Update 2013 - HD

Introduction

Hi, my name's Michael Beland.

I'm an associate professor of diagnostic imaging at Brown University in Providence, Rhode Island.

Today I'm gonna be giving you a talk on guidelines for thyroid FNA, an update in 2013.

I have no disclosures.

Overview of the Talk

First I'll start by briefly reviewing the SRU criteria, published in the mid two thousands, and then go over some of the criteria that have been published in the interval up to the current time, and try to put it all together to give current, contemporary recommendations on what to do with thyroid nodules, and then discuss some of the issues that I think are unresolved that need to be dealt with in determining when a thyroid nodule FNA needs to be performed.

SRU Consensus Statement (2004)

In 2004, a multidisciplinary panel met, to create the SRU consensus statement, this was to establish nodule, guideline biopsies for, the thyroid gland.

From that article, we have this table which summarizes the recommendations of that consensus conference.

And based on features and size, we have some guidelines on when to do a biopsy.

If we have a solitary nodule, we're recommended to do a biopsy at the one centimeter threshold if microcalcifications are present, but can increase the threshold to one and a half centimeters.

If the nodule is solid or containing course calcifications.

And as nodules begin to demonstrate cystic components, we can increase the size threshold further, we say that generally and ultrasound guided FNA is probably not necessary, nodules which are, entirely cystic or nearly entirely cystic or have demonstrated long-term stability.

Endocrinologist Response to SRU

The endocrinologist had a response to the SRU, consensus statement, and I think it's best summarized here when they stated We cannot support the exclusive and limited approach published by the SRU for the management of thyroid nodules detected at ultrasound, one of their primary concerns was the use of size threshold, that was provided in the consensus statement feeling that by using the current size threshold, we may miss smaller thyroid cancers.

So they employed their own, multiple half day meeting sessions to come up with some biopsy recommendations.

And their goal was to create evidence-based recommendations to assist in the management of patients with thyroid nodules and thyroid cancer.

And they were hoping to come up with guidelines that would provide optimal care for patients with these disorders.

The guidelines published included multiple, recommendations for the treatment of cancer patients, but a significant portion of that, article gave us new recommendations for what to do with thyroid nodules.

This table from that article summarizes their new recommendations.

And the important difference from the SRU criteria are that we now have clinical risk stratification in patients with a high risk history generally characterizes those with a first degree relative who had thyroid cancer, a history of thyroid cancer in themselves, or a history of prior radiation or potentially familial syndromes would undergo FNA of thyroid nodules at a smaller size using a five millimeter threshold.

And the recommendations for larger nodules in patients without a high risk history were fairly similar, and we can summarize them here.

American Thyroid Association Recommendations for High-Risk Patients

So the revised American Thyroid Association recommends for high risk patients that a nodule greater than five millimeters undergo biopsy, particularly when suspicious ultrasound features are present.

But even without suspicious ultrasound features, we should consider biopsy of nodules in high risk patients, although they note that there's no real evidence or consensus to support this suspicious features, according to American Thyroid Association included microcalcifications a hypo coic morphology increased internal vascularity infiltrated margins, and a taller than wide morphology.

American Thyroid Association Recommendations for Low-Risk Patients

The American Thyroid Association had recommendations as well for low risk patients.

So for the remainder of patients that we do ultrasound of the thyroid on and recommended biopsy for one centimeter nodules with microcalcifications or hypo coic solid nodules, there's a gray zone for the one to one and a half centimeter nodules that are iso or hyper coic and then mix cystic and solid nodules, slightly larger, biopsy threshold.

But they do recommend that we pay attention to suspicious ultrasound features within the solid components, which may trigger a biopsy at a smaller size of 1.5 centimeters.

They mention spongiform nodules and that we should wait till two centimeters to perform a biopsy, and if abnormal cervical lymph nodes are present that we should biopsy the cervical lymph nodes and they can be biopsied in lieu of the thyroid nodule and do recommend that we don't biopsy purely cystic nodules.

Post-ATA Guidelines and Studies

So since the American Thyroid Association recommendations were published, multiple articles have come out applying the various size criteria thresholds to determine if we can improve our diagnostic yield with thyroid biopsy.

In addition, at least two more sets of guidelines have come out trying to indicate when a thyroid biopsy may be indicated.

One in 2010 was a joint effort, headed by the American Association of Clinical Endocrinologists as well as two European societies.

And in 2011, a task force on thyroid nodules by the Korean Society of Thyroid Radiology both came up with publications that gave us new recommendations.

AACE Guidelines

The clinical endocrinologists recommend now that we biopsy any size thyroid nodule with a high risk patient history and that we biopsy all hypo coic nodules, which are over a centimeter.

If they have the suspicious features of an irregular margin, hyper vascularity, a tall configuration or microcalcifications, the Korean task force recommends that if we see suspicious features and nodules, that we basically biopsy all nodules, including nodules less than five millimeters.

If the clinician is comfortable, with doing that biopsy and that if a nodule is suspicious in ultrasound and the FNA is benign, that we should actually repeat that biopsy.

They do give us some guidelines on probably benign nodules and recommend that if they're less than a centimeter, we probably don't need to follow them at all nodules greater than a centimeter.

We do, have follow up suggested, although the interval has been extended to two years and then potentially after that every three to five years.

And if a nodule is probably benign in more than two centimeters, we can consider the use of biopsy.

Summary of New Criteria Since SRU

So these new criteria since the SRU can be summarized, as follows, for solid nodules, all three of the societies recommend that we biopsy any thyroid nodule greater than a centimeter.

If it has suspicious features.

The clinical endocrinologist and Korean society recommend in addition that we perform FNA on nodules less than a centimeter if suspicious features are present.

And the American Thyroid Association states that we should do FNAs on these smaller nodules.

If a high risk patient history is present applying these criteria to cystic nodules, we see that the American Thyroid Association recommends FNA of all complex cysts over one and a half to two centimeters.

The clinical endocrinologists don't offer as specific guidelines on the size, but recommend that we do FNA of all complex cysts because there is a risk of cystic papillary thyroid cancer.

And the Korean recommendations state that we should do an FNA of a complex cystic nodule if we see suspicious features in the solid component.

There are times when we're not gonna perform a biopsy based on these new criteria, but they are quite limited.

The American Thyroid Association recommends that we wouldn't biopsy nodules smaller than five millimeters in any patient solid nodules less than a centimeter in low risk patients or patient or nodules without microcalcifications in purely cystic nodules or spongiform nodules less than two centimeters.

And in smaller complex cystic nodules, the clinical endocrinologists generally recommend that we're not gonna perform a biopsy in a nodule less than a centimeter if there are no high risk factors or suspicious ultrasound features.

And they also note that if you have a hyper functioning nodule on scintigraphy, that biopsy is likely not indicated.

The Korean recommendations state that we're not gonna biopsy nodules less than a centimeter if there are no suspicious features present.

And in spongiform nodules and simple, smaller, simple cysts,

TI-RADS (Thyroid Imaging Reporting and Data System)

I'd like to touch base on a concept called TI-RADS.

You can see here it stands for Thyroid Imaging Reporting and Data System.

And the idea is to create a categorical system that will state how likely it is that a nodule is malignant and therefore guide management.

It was first described by Dr. Horvath in 2009 and further detail in 2011 in a article in radiology where they looked at nearly 1700 nodules and applied the criteria, to determine risk of malignancy.

See, In summary, TI-RADS uses ultrasound features that are usually associated with thyroid cancer and decides if multiple features combined together will increase the probability of cancer.

And they have shown that if you increase certain combinations of features, you do increase the probability of cancer.

And in summary, the risk of malignancy with one suspicious feature is quite low at 3.3%, but if you add up five suspicious ultrasound features, you can reach an almost 88% likelihood of malignancy.

Further work in this area level likely occur in the future, although it's not widely used at this time.

Challenges: Too Many Biopsies and Benign Features

I'd like to shift gears at this point though and say that even with these recommendations, we are still doing too many thyroid nodule FNAs, far too many benign thyroid nodule FNA results are being obtained at this time, and we need to improve our selection of nodules for fine needle aspiration to decrease the number of benign biopsies.

We of course, don't want to miss clinically relevant thyroid cancer, but the guidelines that have been published since the SRU criteria with continually decreasing size thresholds will likely address the second issue, but are definitely going to increase the number of biopsies and therefore increase the number of benign biopsies as well.

So how can we do fewer biopsies?

Well, we can identify features or scenarios that are highly likely to be benign to decrease the number of FNAs potentially offering follow-up ultrasound rather than the more aggressive sampling.

We also need to deal with the management of the non-diagnostic FNA.

And one question that comes up frequently is, how long do we follow these nodules?

Classic Benign Patterns

So one way to find the benign features is to look for classic patterns, and this was described initially by Dr. Reading in 2005 and from his paper he noted for what he would call leave alone patterns.

And they're listed here being a colloid cyst, so a cystic lesion with a large echogenic focus and ring down artifact, a spongiform nodule.

So a cystic nodule with small intervening septae, a predominantly cystic nodule.

Now these will occasionally be malignant, but we know that the likelihood of malignancy is very low in this pattern of a Hashimoto's gland, generally a hypo coic gland with innumerable small nodules sometimes separated by fibrous septae.

Further along this, Dr. Bonavita looked at 500 nodules and presented his data in 2009 in a JR and they found four specific benign patterns where in their series of 500 nodules were 100% specific for benignity recognition of these benign patterns.

They found to significantly reduce the biopsy rate up to 61% fewer biopsies in their series.

And their benign patterns included spongiform nodules, which we saw in the prior slide cyst with a colloid clot or basically cystic nodule with this echogenic focus with ring down.

And then two additional findings of what they call giraffes and white knights.

The giraffes being in the bottom left of your screen, echogenic nodules or cluster essentially of echogenic nodules in a Hashimoto's gland in the white knights, which is a very hyper coic nodule in a Hashimoto's gland.

Now depending on your experience, you may occasionally come across nodules that look like this in Hashimoto's glands that are malignant anecdotally, but in their series of 500, they found that this was highly likely to be a benign finding at our institution.

We published a study in 2011 where we tried to look at the echogenic foci and thyroid nodules to see if that could help us determine more benign patterns.

And this schematic represents the four patterns of echogenic foci that we identified.

Pattern one being that typical colloid pattern two being wide echogenic linear foci without any shadowing.

Pattern three are smaller rounded foci that don't quite meet the criteria for psammomatous calcifications, and most people would consider these to be indeterminate foci.

And then pattern four is a typical micro calcification or psammomatous pattern where we saw 31%, cancer incidence, which is consistent with the current literature.

So maybe these wide echogenic foci without shadowing, particularly if they have ring down artifact, are benign features and can sway us to avoid biopsy in most patients.

It's currently a joint study performed between our institution and USC headed by Dr. Grant, where we looked at 701 nodules with echogenic foci categorize things slightly differently and you can see the categorizations here.

But the big, picture is that we were measuring these comet tail lengths and we had two categories, one with small comet tails being less than a millimeter in length and larger comet tails being greater than a millimeter in length.

And we looked at these, we found that the larger comet tail artifacts are not very common, but when they are present, they're strongly associated with benign frequently colloid nodules, 98% in this series.

However, the small comet tail artifacts are quite common being present in 29% in this series, and malignancy was frequent in these nodules.

So, this highlights the fact that you have to be very careful about your determination of colloid and comet tail artifact and that you wanna be strict and make sure that when you're calling comet tail artifact that this is in a large comet tail artifact with a prominent echogenic focus.

Management of Non-Diagnostic FNA

What do we do with a non-diagnostic FNA?

Because this is a source of difficulty clinically when you have a non-diagnostic FNA generally is going to result in repeated FNAs and potentially surgery.

The American Thyroid Association has, published guidelines on this and they recommend that all non-diagnostic FNAs be repeated, ideally with ultrasound guidance and onsite cytology.

But they also note in their publication that up to 7% of nodules will continue to be non-diagnostic despite multiple repeat FNAs.

It may turn out to be malignant at time of surgery.

So they recommend quite strongly that if you have a nodule that's undergone a repeated non-diagnostic FNA, that if a nodule is solid, surgery should be performed.

And if it's not solid, you should consider surgery or at least very close observation.

However, there was a study more recently in the American Journal of Surgery where they looked at, a relatively small number of non-diagnostic FNAs 51 in total, which were done with both ultrasound and without ultrasound guidance.

So their non-diagnostic rate was slightly higher than we see typically with ultrasound guided FNAs at 21%, they found cancer in seven of these 51 non-diagnostic FNAs for a rate of 14%, and that is reasonably high.

But even with that, they said that most patients with a non-diagnostic FNA will have benign disease, which is true, and that low risk patients with a repeat non-diagnostic FNA probably warrant more, conservative follow up, meaning follow up thyroid ultrasounds rather than continued repeat FNAs or thyroidectomy.

So we wanted to look at this data at our institution as well, and we looked at, over 5,300 FNAs performed at Rhode Island Hospital, and we found 405 non-diagnostic nodules, which had adequate clinical surgical or FNA follow-up in these 405 nodules.

18 FNAs that were repeated were felt to be positive or suspicious.

However, at surgery, only two of the 18 nodules were actually cancer giving a rate of 0.5% in ultrasound guided non-diagnostic FNAs, which is very, very low.

So we felt that clinical and ultrasound follow up may be more appropriate than repeat FNA in the vast majority of patients.

And when we looked at the patterns of nodules that were non-diagnostic, we found that if a nodule was cystic or felt to be spongiform, in retrospect, had eggshell calcifications or these larger indeterminate echogenic foci that those, we did not see any cancers in those types of nodules on repeat, non-diagnostic FNAs.

And in younger patients, particularly females less than 45, we did not have any incidence of cancer in these.

So in these specific indications, it seems that repeat FNA would really not be necessary in that ultrasound followup would be most appropriate.

There has been some recent literature on the role of core needle biopsy after non-diagnostic FNA and three studies that are quoted here from 2012 and 2013 show fairly similar results that if you have a non-diagnostic ultrasound guided FNA that repeating the biopsy using a core needle will give you a higher diagnostic rate.

At best, we saw non-diagnostic biopsies, on the range of 1.6% with a core needle biopsy versus 28% with a repeat FNA.

So in some institutions, they are performing core needle biopsies after a non-diagnostic thyroid FNA.

Although based on our earlier literature, we have not really been doing this because we feel like the non-diagnostic FNA is essentially a diagnostic result, meaning that the likelihood of malignancy is very, very low in most patients.

Follow-Up of Thyroid Nodules

How long do we follow thyroid nodules?

Well, the SRU guidelines basically stated that there was no consensus, so they had no specific recommendations, which is true the American Thyroid Association has recommendations on follow-up, but with the caveat being that they're recommending follow-up, specifically nodules that have been previously biopsied and found to be benign, they're quite aggressive with their follow-up initially in that they recommend an ultrasound after a benign FNA at six as early as six months, as late as 18 months.

And they state that if the nodule size is stable on that follow-up, then you can begin to space out your follow-up intervals perhaps every three to five years.

But that if a nodule grows by palpation or ultrasound, then the FNA should be repeated.

It's noted that they do have specific criteria for a change in nodule size in that, change is felt to be significant if there's more than a 50% increase in volume or greater than a 20% increase in at least two nodule dimensions of at least two millimeters in solid nodules or in the solid portion of the mixed cystic and solid nodule.

We all know from doing ultrasound that a two millimeter change is within the limits of, variability in the way you measure or perhaps how you placed your calipers or even just the track ball.

It's hard to control it in these smaller sub two mil or two millimeter range.

So these criteria, are, difficult to apply clinically and usually I tend to err more on the side of this 50% change in volume from one study to the next.

Recommendations at Rhode Island Hospital

So what do we do at Rhode Island Hospital?

And we try to combine the indications between the SRU and the American Thyroid Association.

We generally, recommend biopsy of nodules that have microcalcifications or that are solid and specifically markedly hypo coic, meaning at least as hypo coic as the strap muscles, if not even, less echogenic than the strap muscles at one centimeter.

For all the remaining solid nodules, we have a one and a half centimeter threshold.

If we have a mixed cystic and solid nodule with truly suspicious features in the solid portion, we'll recommend FNA at one and a half centimeters.

The remaining cystic and solid nodules, we have a two centimeter threshold as well as for spongiform nodules, and we do not biopsy purely cystic nodules.

Of course, these recommendations can be modified based on patient risk factors, so it's important to go in the room and talk to the patient when you scan them, as well as talk to your referring physician or endocrinologist to determine the risk history of the particular patient you're seeing.

Conclusion

I'd like to thank you for your attention. I.