SE

M I N A R S I N

DI

A G N O S T I C

P

A T H O L O G Y

32 (2015) 284–295

Available online at www.sciencedirect.com

www.elsevier.com/locate/semdp

Advances in head and neck fine-needle aspiration and ultrasound technique for the pathologist Joseph D. Jakowski, MDa,n, Laurence J. DiNardo, MDb a

Division of Anatomic Pathology, Department of Pathology, Virginia Commonwealth University Health Systems, Gateway Building, 6th Floor, Room 6-203, 1200 E Marshall St, Richmond, Virginia 23298 b Department of Otolaryngology/Head and Neck Surgery, Virginia Commonwealth University Health Systems, Richmond, Virginia

article info

abstract

Keywords:

The success of fine-needle aspiration (FNA) biopsy in the evaluation of head and neck

Ultrasound

(H&N) masses has already been established. Herein we outline the most recent advance-

Fine needle

ment for the pathologist who performs traditional palpation-guided FNA (PGFNA) in the

Aspiration biopsy

H&N while also incorporating ultrasound-guided FNA (UGFNA) into their practice. We

Head and neck

provide an overview of the educational and training opportunities in H&N ultrasound and UGFNA with emphasis on the advantages and limitations for the pathologist. Throughout are useful clinical and technical pearls, many of which may also interest those who practice PGFNA, including local anesthesia use and FNA procedures in pediatric patients. & 2015 Elsevier Inc. All rights reserved.

Introduction Fine-needle aspiration (FNA) biopsy continues to be a safe, rapid, cost-effective, and accurate procedure that has been accepted as the initial biopsy procedure for the evaluation of superficial head and neck (H&N) masses. It is especially useful for diagnosis whenever clinical and radiographic findings are equivocal, to help in initial tumor staging and treatment planning, and to aid in the surveillance of postoperative neck lymph nodes. The American College of Radiology has further stated that ultrasound (US) has increasingly demonstrated its usefulness in adults and children in differentiating solid and cystic neoplasms, in assessing vascular lesions, and in facilitating biopsies.1 It is therefore within reason that a pathologist who currently practices palpationguided FNA (PGFNA) may, with proper training, expand their scope of practice to include the use of US for ultrasoundguided fine-needle aspiration (UGFNA). A number of n

Corresponding author. E-mail address: [email protected] (J.D. Jakowski).

http://dx.doi.org/10.1053/j.semdp.2014.12.010 0740-2570/& 2015 Elsevier Inc. All rights reserved.

interventional cytopathologists in both private and academic practices have previously made the case for pathologistdirected UGFNA, with the advantages summarized in Table 1.2–5 The College of American Pathologists (CAP) has also endorsed pathologist-directed UGFNA through the creation of their own Ultrasound-Guided Fine-Needle Aspiration Advanced Practical Pathology Program (UGFNA AP3).6 Furthermore, several other national medical organizations are now offering training and continuing medical education for pathologists in US medicine and UGFNA, which we will later detail.

General concepts Current UGFNA training courses and previously published pathology practice models have helped define the goal of US use by the pathologist to include evaluation and confirmation of a target lesion for FNA and for real-time image guidance of

S

E M I N A R S I N

D

I A G N O S T I C

P

Table 1 – The potential patient care and practice advantages of incorporating ultrasound into the fine-needle aspiration (FNA) procedure. Patient care advantages Increase patient satisfaction by applying the latest technology to their care Enhance the physical examination Confirm a suitable target lesion is present for FNA Help identify “pseudotumors” or normal anatomy that do not warrant FNA Practice advantages Ability to target non-palpable superficial lesions Not left out due to advancing technology Address clinician demand for faster patient turnover through greater accessibility to UGFNA Expand your technical skill as wells those of other staff such as cytotechnologists who can assisting in the ultrasound-guided FNA procedures Potential to increase practice marketability, the number of FNAs performed, and increase revenues You are in control of the UGFNA procedure and not limited to only performing an adequacy assessment while someone else performs the biopsy. This has the following additional benefits: Proper sample triaging (e.g., PTH or thyroglobulin washout levels and cultures) Obtaining sufficient material for ancillary studies (e.g., flow cytometry) Increasing contact and direct communication with the patient and referring physician

the needle during the FNA procedure. Although providing diagnostic medical US is not the end goal for the pathologist who incorporates US into their FNA practice, there is certainly a need to become as skilled as possible with its use to ensure the success of the UGFNA. The cost and benefits of UGFNA must be considered as it entails considerably more monetary and training investment compared to PGFNA. For example, initial UGFNA training cost is 4$3000, and a suitable US machine can range from $25,000 to more than $40,000. As with all laboratory equipment, a routine preventive maintenance program will be needed to ensure the safety and quality of the US machine, and there should be quality improvement procedures in place to monitor the appropriateness, technical accuracy, and accuracy of the interpretations of the UGFNA performed. A process of retention, secure storage, and transfer of US images is also required. Detailed standards of practice for medical US have been published by the American College of Radiology (http:// www.acr.org/Quality-Safety/Standards-Guidelines) and the American Institute of Ultrasound in Medicine (http://www. aium.org/resources/guidelines.aspx), and it would be in the best interest of the pathologist to adhere to all the guidelines that are applicable to their practice. If considering hiring an ultrasonographer to enhance your UGFNA service, their average yearly salary may be $60,000 or more depending on experience. Hospital-employed pathologists who are considering incorporating UGFNA into their already existing PGFNA service must also apply for this additional privilege to confirm their competency in UGFNA and to define the scope of their UGFNA practice. We have specifically defined our scope

A T H O L O G Y

32 (2015) 284–295

285

of UGFNA practice to include superficial anatomical sites that are usually above the fascia and can be reached by a needle of 2.5–3 cm in length. Excluded are any targets within a body cavity such as the abdomen or thorax. Excellent in-depth articles on starting an UGFNA practice in an academic or private practice setting have already been published.7,8 Although we are not aware of any consistent reimbursement issues concerning pathologist-performed UGFNA, a growing number of insurance companies are requiring US practice accreditation for non-radiologists (e.g., endocrinologists) who perform diagnostic US before reimbursement is approved. We are, however, aware that some interventional pathologists in private practice are seeking this additional US accreditation as a precaution against any such future requirement for reimbursement for UGFNA procedures.

UGFNA training and education The ideal way to train and educate a pathologist in UGFNA would be throughout their pathology residency or cytopathology fellowship much like traditional PGFNA has been taught for decades. Integrated UGFNA training is being performed in our pathology department, not only for our own pathology residents and cytopathology fellows, but also for endocrinology fellows as well. We realize, however, that we are an exception as there are currently very few pathologists in academic teaching hospitals in the U.S. performing UGFNA. The current state of UGFNA training and education for the pathologist is therefore one of a post-cytopathology fellowship endeavor. It has been our impression, however, and that of many other interventional cytopathologists, that initially acquiring proficiency in PGFNA certainly makes the transition to UGFNA an easier task. Regardless of the type of FNA procedure performed (PGFNA or UGFNA), the pathologist needs to be very skilled in making good-quality aspirate smears as well as being able to triage the specimen for special studies through an immediate preliminary examination. Even the best technically performed UGFNA may prove clinically or diagnostically useless if poor-quality smears are made or special studies were not sent to aid in a definitive diagnosis. Currently, several national organizations offer US and UGFNA training for a pathologist that, upon completion, will issue either a certificate of recognition/participation or certification. Evaluation of competency occurs only with certification and this will entail passing a comprehensive examination in addition to submitting US and UGFNA cases for formal review and validation to ensure that a standard of practice is achieved. The most logical place for a pathologist to acquire their initial UGFNA training is through the CAP’s UGFNA AP3 course (http://www.cap.org/apps/cap.portal). This course offers 16 hours of continuing medical education (CME) and a certificate of recognition in UGFNA that is valid for a 5-year period.6 The course covers topics not only in the H&N (thyroid, lymph nodes, and salivary gland) but also breast and soft tissue. It entails approximately 6 hours of pre-study followed by attending a 2-day workshop with combined lecture presentations, small group activities, and hands-on

286

S

E M I N A R S I N

D

I A G N O S T I C

P

training in smear making and UGFNA. Other noted highlights include passing a slide preparation and US needle placement assessment and a multiple-choice test of 60–70 questions. The maintenance and renewal requirement for the UGFNA AP3 certificate consists of paying a renewal fee (currently $200) and submission of 5 original UGFNA pathology reports that are evaluated for completeness. The American Association of Clinical Endocrinologists (AACE; https://www.aace.com/) probably offers the most comprehensive US and UGFNA training and education for a pathologist in the H&N with a focus on thyroid, parathyroid, and cervical neck nodes. Their Diagnostic Endocrine Neck Ultrasound and UGFNA Course (https://www.aace.com/meet ings/symposia) is a 2-day workshop consisting of didactic lectures and hands-on US and UGFNA training in which the attendee will receive a letter of participation and 15 CMEs upon completion. ACCE additionally offers an Endocrine Certification in Neck Ultrasound (ECNU) Program (http:// www.aace.com/ecnu) in which there is a pathway for American Board of Pathology-certified cytopathologists to obtain professional certification in neck ultrasound and UGFNA for thyroid and parathyroid. This certification is valid for 10 years. Another major benefit of completing the ECNU program is that the American Institute of Ultrasound in Medicine (AIUM; http://www.aium.org) now allows those with ECNU certification to apply for AIUM practice accreditation. Accreditation not only helps in standardization of US and UGFNA practice, but it may also be important in the future for reimbursement as previously discussed. AIUM practice accreditation is good for 3 years and is renewable. The American College of Surgeons (ACS; http://www.facs. org) and the American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS; http://www.entnet.org) also offers a 1-day H&N US and UGFNA course in which non-surgeons may also attend. The ASC and AAO-HNS courses cover thyroid, parathyroid, and cervical nodes and are held during their respective annual meetings in which you receive CME that is also accepted for application for ECNU certification. For both courses, there is a prerequisite CD-ROM module entitled “Ultrasound for Surgeons: The Basic Course” that must be completed prior to attending these programs and can be purchased separately online (http://www.acs-resource. org). An additional excellent resource to augment any formal UGFNA training would be to spend time visiting a pathologist who is currently practicing UGFNA or attending the many US courses sponsored by any one of the US manufacturers. There are also a variety of textbooks and peer-reviewed articles on US and US-guided procedures by surgeons,9 interventional radiologists,10 endocrinologists,11 and anesthesiologists12–15 that will prove useful to the pathologist. Although there are no current required standards for the number of CME requirements in US medicine for the pathologist, a possible benchmark to consider is 15 hours of CME every 3 years, as required by both the AACE and AIUM for their members. All of the courses above will provide the pathologist with an understanding of the basics of US instrumentation, UGFNA, and the US criteria one can use in daily practice to help determine the most appropriate target for FNA.

A T H O L O G Y

32 (2015) 284–295

However, one must also invest additional time to practice operation of the US machine and needle placement under US-guidance on homemade16 or commercially available phantoms until these skills are automatic before making the transition to UGFNA on patients. The point at which one feels confident in their skill set to transition from phantoms to patients will vary from pathologist to pathologist. One of the authors (J.D.J.) engaged in practicing US-guided needle placement in phantoms for 10 hours over 3 months while others have spent 8 hours over a 1-month before moving onto patients.3 An interesting comparison concerning interventional US training may be found in the anesthesiology literature where they demonstrated that approximately 28 supervised trials were needed for anesthesiologist trainees to acquire competency in US needle visualization using cadaver models.17 Another practical point is that the novice pathologist should first become proficient in performing UGFNA on palpable masses before moving to non-palpable ones, as the latter can be technically more difficult. For example, one of the authors (J.D.J.) performed approximately 25 UGFNAs on palpable masses of various sizes before attempting a truly non-palpable target in a patient and Dr. Lieu previously reported performing 100 UGFNAs on palpable masses before progressing to non-palpable ones.3

UGFNA procedure The H&N can be one of the more technically challenging areas to perform UGFNA given that there is limited room to work between the chin and clavicles. There are also many anatomical structures that need to be avoided during the biopsy, including the major neck vessels (jugular and carotid) and the lung apex in supraclavicular sites. The angle of the mandible, clavicle, or other anatomical protrusions will also potentially interfere with flush seating of the transducer on the skin. This is especially problematic in very thin patients and those status post-neck dissection. One may even consider purchasing a smaller transducer that could fit better in these smaller anatomical areas. To open up workspace in the neck and make access to the FNA target easier, the patients should be placed supine with the neck hyperextended with a pillow under the shoulders and lower neck for support. Turning their head to the contralateral side of the target or performing the biopsy from the head of the table instead of from the side of the table may also help. A pictorial essay on how to perform an UGFNA has already been published for pathologists and we have included a summary of this method in Fig. 1.18 As previously noted, those already skilled in PGFNA will find that the basic aspiration steps are transferable to UGFNA. There are some differences, however, that must be considered when transitioning to UGFNA. These include the loss of tactile sense in targeting non-palpable masses with US, the finer hand–eye and motor coordination needed for US guidance, and the additional awareness of the spatial relationships involved in orientation of the transducer with the US display screen, the patient, and the biopsy needle. One of the biggest challenges for the pathologist undertaking UGFNA is achieving an acceptable level of competency

S

E M I N A R S I N

D

I A G N O S T I C

P

A T H O L O G Y

32 (2015) 284–295

287

Step 1. Locate the correct or most suspicious target for FNA. 28 year-old woman with one palpable left thyroid nodule (N1) and one incidental nodule (N2) in the right lobe. N1 is benign appearing and colloid rich with comet tail artifact.

N2 has suspicious features including hypoechogenicity and chaotic internal vascular flow.

Step 2. Determine the best needle approach to the target. Path 1 is slightly longer than Path 2 however, Path 1 avoids going through the SCM as this may cause undue discomfort and/or plug the needle up with skeletal muscle. Path 1 is the best approach through the strap muscles.

Step 3. Place the needle under ultrasound guidance into the target and perform the FNA. The hyperechoic biopsy needle (arrows) is being advanced in the parallel plane of the transducer and on the correct path into the target. UGFNA confirmed papillary thyroid carcinoma in N2. Subsequent total thyroidectomy confirmed the FNA diagnosis and N1 was proven to be a benign colloid nodule.

Fig. 1 – Basic steps to performing an ultrasound-guided FNA. SCM, sternocleidomastoid muscle; sk, skin; isth, thyroid isthmus; trach, trachea; CC, common carotid artery.

in US examination that includes obtaining the best-quality US images and identifying specific features of the target and its anatomical surroundings as listed in Table 2. Ways to

improve one’s US skills include frequently scanning all of your FNA patient’s lumps and bumps, seeking out the help of a certified ultrasonographer, routinely comparing your own

288

S

E M I N A R S I N

D

I A G N O S T I C

P

Table 2 – Specific ultrasound features of the target to evaluate that may assist in performing the FNA. Size Margins (well circumscribed versus irregular/infiltrative) Depth from skin Solid versus cystic versus necrotic Vascularity Relationship to other important anatomical structures (e.g., major vessels or apex of lung) Identifying the “higher yield or most suspicious” target if many are present Determining the ideal part of the target for sampling (e.g., mural nodule in a cyst) Determining the best needle path

US findings with those performed by a radiologist, practicing US scanning and identifying normal anatomy on family and friends, or even US examination of gross specimens at the surgical pathology bench. A basic knowledge of US physics and terminology is also key, and it may be acquired through the references provided.19–23 The pathologist should find that their US skills quickly develop using a combination of these educational practices as well as after each UGFNA performed. For example, non-radiologists have been able to achieve acceptable levels of US competency, defined as 490% concordance with a radiologist’s interpretation of the same examination, after 25–50 focused US examinations.9 Patients referred for UGFNA should be required to bring any previous radiology reports, and ideally, the pathologist should have access to view the actual imaging studies. Real-time correlation of one’s own limited US findings with previous imaging studies should be routinely practiced to confirm the correct FNA target. If any doubt exists, which may occur when multiple pathologic processes or suspicious targets are present, reviewing the patients imaging studies with the referring clinician can avoid confusion. If by reasonable means, one fails to identify a target by US or physical exam, you will want to explain the situation to the patient, defer the biopsy, and contact the referring physician to determine the next best course of action.24 As with all biopsy procedures, one will need to develop practices that will minimize false negatives. In UGFNA, these include prioritizing targets to biopsy according to their most suspicious US features, sampling several areas of the target, sampling any solid component of a cystic lesion, and repeating a biopsy on any previously diagnosed “benign” target that develops concerning clinical or radiographic features on follow-up. Lastly, additional studies employed on FNA material may be crucial for arriving at a clinically useful diagnosis. These additional studies include immunohistochemical staining of cell blocks, microbial testing and culture, molecular testing, and flow cytometry. Parathyroid hormone (PTH) washout on FNA material can help to rule in or out a parathyroid lesion, and calcitonin washout is excellent for recurrent or suspected medullary thyroid carcinoma. Thyroglobulin washout levels from lymph nodes in patients with suspected metastatic papillary thyroid carcinoma (PTC) are very desirable, especially in the absence of diagnostic cells in the aspirate material. Comparison with the patient’s serum levels will be needed for proper interpretation of these

A T H O L O G Y

32 (2015) 284–295

washout levels. Immunohistochemistry for surrogate highrisk HPV marker p16 performed on metastatic squamous cell carcinoma is useful not only to suggest a metastasis from Waldeyer’s ring but also for prognostic purposes.

Pediatric UGFNA US is one of the preferred imaging modalities for the initial evaluation of lumps and bumps in a child and to aid in FNA biopsy. As with adults, US can be a helpful distraction during the procedure, as the technology and display screen can be used to redirect the child’s attention. The child will also find the initial US examination for target location a painless process, and this will build some initial trust for the rest of the FNA procedure. Other useful distractions to use during the FNA include having the child read books, watch videos, view photographs, or listening to music on their parent’s cell phone. However, the needle biopsy itself should be performed rather quickly as pediatric patients may give you only a small window of opportunity and goodwill. Having everything set up in advance for the procedure and filling out any forms as much as possible, including the consent and answering any questions from the parents, is highly recommended. In addition, using the Zajdela (needle-only) technique for the FNA has the added advantage of allowing the practitioner to “palm” the FNA needle and thus keep it out of sight. In our practice, we find that being honest with the child that the FNA biopsy may pinch or sting, but that it will not hurt for long, is the best approach. Supportive parents are often essential to the success of the biopsy and telling scary stories, applying threats, or trying to scold the child into getting the biopsy is counterproductive and should be discouraged. Willing parents may also help immobilize their child by gently holding their head in position or keeping their arms crisscrossed on their chest to prevent them from reaching for the biopsy needle. Occasionally, the biopsy may be performed with the child sitting in the parents lap or by getting the parent to lie down on the exam table with the child to provide additional comfort. For infants, wrapping them in a blanket (swaddle) with their arms straight at their sides may prove useful. If you find ahead of the child’s FNA appointment that the parents are very concerned that their child may have a hard time coming to see you for the procedure, you can suggest that doing medical play with the child at home may be beneficial and you can explain what to do to prepare them for the visit. Experiences from nurses who routinely perform percutaneous injections and vaccinations on pediatric patients have additional tips at decreasing pain at the needle stick site that are also applicable to FNA. These include stroking the skin or applying pressure close to the needle insertion site, flattening and pulling the skin taut at the insertion site (this can be performed with the help of an assistant and the transducer), inserting the needle through the skin in one quick motion, and keeping the bevel of the needle facing up during skin insertion. In addition, when disinfecting the skin with alcohol, let it completely dry or wipe off any excess before you stick the skin with the needle.

S

E M I N A R S I N

D

I A G N O S T I C

P

Anesthesia There is no universally accepted recommendation for the use of local anesthesia in FNA procedures, as some practitioners may never offer it, while others will offer it to all their patients. If local anesthesia is used, an effective technique for administration should be employed, and this has been previously described in detail.25 Typically, 0.5–1 mL of a 1% or 2% lidocaine solution is sufficient to provide anesthesia. If a patient is allergic to lidocaine, bupivacaine may be used instead, as long as the patient is not also allergic to bupivacaine. A history of an allergy to Novocain is not a contraindication to the use of lidocaine because they are chemically different and cross-reaction is rare. In our FNA practice, we prefer to use lidocaine without epinephrine so that we do not have to be concerned about other rarely associated epinephrine reactions, such as tissue necrosis (in some specific anatomic sites) and any additional allergic or cardiovascular effects. Using an ultra-thin 30-gauge needle and a slow injection rate makes for a relatively pain-free administration, and we never make a skin wheal during the injection as this can cause undue discomfort and is not necessary to provide adequate anesthesia. We often prefer to inject the anesthesia under US guidance as we feel we can better direct it into the tissues along the proposed biopsy path to the target. For example, if performing FNA on a lymph node or thyroid gland, we will infiltrate lidocaine into the subcutaneous tissues along the planned path of the biopsy needle and around the lymph node or thyroid capsule. Using US for the anesthetic injection is also a way to “practice” for the actual biopsy as you can ascertain the correct needle angle to be on track to hit the target for the FNA. Beware that using too much anesthetic may obscure anatomic detail and may make it difficult or even obscure visualization of the target on US. Once injected, anesthesia generally develops in 2 minutes and may last for 1–2 hours, which is long enough that readministration is rarely, if ever, needed. Alternatives to injectable lidocaine include placing an ice pack before the procedure, using topical lidocaine spray, lidocaine gel, or applying skin refrigerant (e.g., ethyl chloride). Beware that ethyl chloride has been reported to erode the plastic on some transducers,4 and damage in this manner may void the manufacturer’s warranty. It is therefore imperative that the US user determine, before purchasing of the US equipment, whether the transducer is susceptible to certain chemical damage that it may have contact with during use, including disinfectant products. In children, we rarely find that they need to be sedated for the FNA procedure if the previously discussed techniques are employed. EMLA (lidocaine and prilocaine) cream applied to the skin 30–60 minutes before the procedure is usually very effective by itself at nearly eliminating pain from the FNA. Local injectable lidocaine anesthetic may also be used alone as described above; however, if using in conjunction with EMLA cream, cumulative doses must be considered. Note that EMLA cream dosing, application area, and application times are based on the child’s age and weight. After application, the cream should be covered with a bandage while letting it take action and careful observation by an adult will be needed in order to prevent any accidental ingestion or eye contact by

A T H O L O G Y

32 (2015) 284–295

289

the child. Among the rare side effects of any -caine anesthetic, methemoglobinemia needs special mention, as it is a rare and potentially fatal side effect. Therefore, -caine anesthetic use is contraindicated in patients with this condition or any condition or drugs that may be associated with it. Signs and symptoms include cyanosis and/or decreased oxygenation (dizziness or excessively sleepy). If this occurs after applying topical anesthetic like EMLA, it should be removed immediately. Early recognition of methemoglobinemia is essential and supportive care with supplemental oxygen and treatment with methylene blue has been employed with great success.

US technology in medicine As with all imaging modalities, one must be aware of not only their usefulness but also their limitations. For example, although there are reported US features that may help define benign thyroid nodules and lymph nodes, well-differentiated thyroid malignancies (e.g., minimally invasive follicular carcinoma) and small metastatic nodal disease may demonstrate otherwise benign-appearing US features. The reverse is also true, in that, the US features of many reactive and benign conditions may overlap with those of a malignant process. Therefore, it is recommended that FNA evaluation should be performed whenever the clinical and imaging findings are equivocal. Also consider that just because a target of interest was visualized on another imaging modality such as MRI, CT, or PET scan, this does not necessarily mean that identification will easily occur by US imaging, as low-contrast targets or ill-defined lesions may be virtually invisible on US. This is especially problematic in patients with obese necks, patient’s who have undergone radiation, and/or who have significant scar tissue or edema. Furthermore, US is not well suited for evaluation of certain H&N regions and structures including bone, the base of the skull, parapharyngeal space, some deep lobe parotid lesions, deep-lying lymph nodes, and substernal extension of disease. A good estimate is that any target lying within 4 cm depth of the neck skin will have the best chance of being visualized and biopsied by the pathologist using a high-frequency “small-parts” linear array transducer. Despite the above limitations, recent advances in US engineering and its application to medicine have challenged the boundaries of what was previously possible for its use as an imaging modality and in interventional procedures. What follows are a few of the new developments in US medicine that are being investigated in clinical practice, as well as some new types of US technology that may be available to the pathologist now or in the future as pricing and availability improves.

Needle visualization technology A number of US manufacturers have recently developed US technology with the specific focus on increasing biopsy needle reflectance relative to the surrounding tissues. Such enhancement is especially useful during steep-angled UGFNA procedures that may occur in scenarios such as performing a biopsy in an obese patient’s neck or in sampling deep cervical nodes or posterior thyroid nodules.

290

S

E M I N A R S I N

D

I A G N O S T I C

P

US fusion US fusion allows one to incorporate previously obtained CT or MRI scans into the US-guided biopsy procedure. Real-time navigation and “marriage” of the cuts of the CT or MRI scans to the real-time US imaging enhances US target identification, especially for low-contrast lesions that normally would not be easily visualized using US alone. This includes the possible expansion of US biopsy for intraosseous lesions. An absolute contraindication is its use on patients with a cardiac pacemaker, as these systems use an electromagnetic navigation system. A major disadvantage of US fusion machines is their current high cost at 4$100,000.

US elastography US elastography, or stiffness imaging, is a dynamic technique that assesses mechanical stiffness. It involves measuring target tissue distortion in response to external compression by the operator using the transducer. US elastography may help in characterization of thyroid nodules, detection of lymph nodes containing metastatic disease, and in the assessment of soft tissue masses in the H&N, as malignancies will usually have a higher elastographic (stiffness) measurement than benign processes. Several US manufacturers have been offering US units with this technology for some time, and it has the added advantage of largely being a software-based technology that can be updated as improvements occur. One ongoing problem is that elasticity image quality appears to be dependent on the operator, in addition technical factors that may adversely affect diagnostic accuracy and reproducibility. This also adds difficulty in comparing results from different studies or even from different US systems. Furthermore, cystic and heavily calcified targets and coalescent nodules, for example, in a multinodular goiter, are not suitable for elastographic analysis. Additional studies are continuing in hopes of improving the diagnostic accuracy of this technology.

US contrast media Intravenous contrast agents, most of them consisting of gas microbubbles, have been recently introduced for use in US imaging and UGFNA. Contrast enhancement may provide some additional information in the evaluation of H&N pathology, such as in malignant cervical lymph nodes, and in defining the size and limits of necrotic zones in targets. Its routine clinical practice is limited, particularly because it is time consuming to administer, expensive, and usually offers only a modest improvement over the information obtainable with traditional US Doppler examinations.

US examples in clinical practice What follows is a brief review of normal sonographic H&N anatomy that will be needed for evaluation of the patient and their lumps and bumps for UGFNA. Additional practice points for US use and UGFNA in the H&N are given wherever possible. We also have incorporated images from cases we

A T H O L O G Y

32 (2015) 284–295

have encountered in our clinical practice to illustrate these and additional interesting findings.

Lymph nodes An understanding of the level system used by surgeons and radiologists for describing lymph node locations in the H&N, along with recognizing the anatomical structures on US that make up the boundaries of these levels, will be needed in order to practice UGFNA. Illustrations of the anatomical landmarks and a simplification of the lymph node level system are provided in Fig. 2. Familiarity with the predictable routes of metastatic cervical nodal spread by primary and non-primary H&N tumors will also prove beneficial. On US, normal lymph nodes in the H&N are often difficult to detect because of their small size, and their internal US appearance may be similar to that of surrounding tissues. Level I nodes are those found in the submental and submandibular space. The US landmarks of the submental space are the triangular boundary of the anterior belly of the digastric muscles and the hyoid bone inferiorly. The submandibular space nodes are all extrinsic to the gland and the posterior border of the submandibular gland is a means of demarcating the posterior aspect of Level I from Level II. Level II nodes (upper jugular) are identified by scanning transversely from the tail of parotid and then descending down along the internal jugular vein (IJV) and carotid artery to the hyoid bone. By continuing to descend along these vessels, Level III (middle jugular) nodes are found between the hyoid bone and the cricoid cartilage and Level IV nodes (lower jugular) between the cricoid cartilage and clavicle. The left supraclavicular lymph node known as Virchow’s node is in Level IV. Level V nodes (including supraclavicular and posterior triangle group) are found in the boundary delimited by the posterior border of the sternocleidomastoid muscle (SCM) and the anterior border of the trapezius muscle, the mastoid process superiorly and clavicle inferiorly. The lateral extent of Level V is the acromion process. In everyday practice, the distinction between Levels II, IV, and V may be somewhat subjective and may be altered by neck positioning. Level VI nodes (anterior cervical neck or central compartment) are located between the hyoid bone and suprasternal notch, and the common carotid arteries (CC) define the lateral boundaries. Evaluating the size, shape, and internal architecture of a lymph node on US can help classify the node as likely being normal versus abnormal, which can help determine the most suspicious node for FNA (Fig. 3A–D).26 Cervical nodes larger than 8–10 mm in the shortest axis, those with a longitudinal to transverse (L/T) ratio o2 (rounded), loss of an echogenic hilum, presence of calcifications, central necrosis, or increased vascularity on Doppler US are all considered abnormal and concerning for possible malignant involvement.27 However, US criteria alone cannot fully distinguish malignant nodes from reactive hyperplasia and inflammatory lymphadenitis as these benign conditions can have overlapping US features. Nodes harboring only micrometastatic disease and partial nodal involvement by lymphoma may also have benign-appearing US features and lead to a falsenegative US impression. Other exceptions to these criteria

S

E M I N A R S I N

D

I A G N O S T I C

P

A T H O L O G Y

32 (2015) 284–295

291

Fig. 2 – (A) Anatomical landmarks in the neck. (B) Simplified lymph node level system. Level I, submental and submandibular; Level II, upper jugular; Level III, middle jugular; Level IV, lower jugular; Level V, supraclavicular and posterior triangle; and Level VI, anterior cervical. Hy, hyoid bone; Cric, cricoid cartilage; Thy, thyroid gland; SCMc, clavicular head of the sternocleidomastoid muscle; SCMs, sternal head of the sternocleidomastoid muscle; par, parotid gland; sub, submandibular gland. (White arrows, thyroid cartilage.)

exist. For example, normal submental nodes tend to be rounded (L/T o2), and small quiescent nodes may lack an obvious echogenic hilum. Importantly, node size is most useful when monitored on serial examinations, and a node that continues to enlarge over time, even in the absence of additional abnormal US features, should warrant a biopsy. This is especially important for patients with a previously known malignancy.

Thyroid High-resolution US examination and FNA biopsy results guide the current management of thyroid nodules. Guidelines set forth by the AACE/AME/ETA on the US criteria for FNA biopsy of thyroid nodules are an essential resource for anyone performing thyroid UGFNA, and we only provide an introductory overview.28,29 On US, the average normal adult thyroid lobe dimensions are 4–6 cm in height (longitudinal), 2–3 cm in width (transverse), and 2 cm in thickness (anterior to posterior). The thyroid lobes are often slightly asymmetric (R 4 L) and the isthmus varies from 1 to 2 cm longitudinally and typically 0.5 cm in anterior to posterior dimension. Normal thyroid parenchyma is characteristically homogeneous and groundglass in appearance (Fig. 3E). The various anatomic structures and tissues adjacent to the thyroid are easily identifiable on US, and many of these must be avoided during UGFNA (Fig. 3E–H). The muscles, including the SCM, longus colli, anterior scalene, and strap muscles, are darker (hypoechoic) relative to the normal thyroid parenchyma and have

internal structure consisting of brighter (hyperechoic) lines representing perimysial connective tissue. The CC artery is difficult to compress with the transducer and appears as a rounded and black (anechoic) structure with a bright (echogenic) wall. The IJV is easily compressible and is therefore not always visible, but it can assume an anechoic ovoid or slit-like shape and will dilate upon Valsalva maneuver. The esophagus may be seen at the posteromedial border of the left thyroid as a semicircular bull’s eye and should not be mistaken as a posterior thyroid nodule or other mass. Having the patient swallow and watching this movement in real-time US can confirm its location. Posterior to the thyroid and longus colli muscle will be the anterior surface of the cervical vertebrae, which appears as a hyperechoic line with marked posterior shadowing. The US findings in thyroid nodules considered suspicious enough to recommend FNA include hypoechogenicity (Fig. 3G), irregular or microlobulated margins, microcalcifications, taller-than-wide appearance on transverse view, intranodular vascularity, and significant growth (420%). Nodules o1 cm are rarely biopsied unless 2 or more suspicious US criteria are found, or if there is a concerning clinical history, including previous neck irradiation, previously diagnosed thyroid cancer, MEN 2 in first-degree relative, or an increased calcitonin level. Patients referred for thyroid FNA should have TSH results available, and if a radioisotope scan has been performed, FNA of “hot” nodules should not be performed. In a multinodular goiter, it is rarely necessary to biopsy more than 2 nodules when proper US criteria are applied; however, some

292

S

E M I N A R S I N

D

I A G N O S T I C

P

A T H O L O G Y

32 (2015) 284–295

Fig. 3 – Normal sonographic anatomy of the lymph node, thyroid, and salivary glands and various ultrasound images of head and neck conditions evaluated for ultrasound-guided FNA. (A) Benign Level I lymph node extrinsic to the submandibular gland (L/T 4 2 and echogenic hilum). (B) Hodgkin lymphoma involving a Level III node (hypoechoic with loss of hilum). (C) Metastatic squamous cell carcinoma to a Level I node (submandibular). (D) Irradiated neck with metastatic squamous cell carcinoma to a Level IV node with extracapsular extension (very ill defined margins on US). (E) Normal thyroid transverse view. (F) Benign spongy appearing thyroid nodule incidentally dsicovered on CT scan. (G) Anaplastic thyroid carcinoma (large hypoechoic mass). (H) Metastatic mucinous lung cancer to the thyroid and Level IV node. (I) Normal tail of parotid gland. Notice the deeper gland visualization is lost behind the mandible. (J) Salivary duct carcinoma of the parotid (arrows define this hazy hypoechoic lesion). (K) Normal submandibular gland. (L) Sialolithiasis (arrow) with sialadenitis (hypoechoic area) of the submandibular gland. (M) Lipoma of anterior cervical neck presenting as a thyroid mass. (N) Foreign body (arrow at bullet fragment) of the face, status post gun shot wound. (O) Calcified facial artery with posterior shadowing, status post radiation for carcinoma (inset is Doppler showing vascular flow). (P) C2 transverse process presenting as a neck mass, status post neck dissection and radiation (arrow at boney protuberance). Thy, thyroid gland; SCM, sternocleidomastoid muscle; par, parotid gland; sub, submandibular gland; sk, skin; sq, subcutaneous tissue; AoM, angle of mandible; mylh, mylohyoid muscle; LCM, longus colli muscle; trach, trachea; CC, common carotid artery; IJV, internal jugular vein; ASM, anterior scalene muscle; VB, vertebral body; srm, strap muscles; cor, cortex; h, hilum; aLN, abnormal lymph node.

practitioners may biopsy up to a total of four during one visit. Most thyroid nodules are within 2 cm depth of the skin and can be reached with a 1–1.5-in. FNA needle. Rarely, as in

posterior thyroid nodules or in a morbidly obese patient’s neck, does one have to employ a 3.5-inch spinal needle to reach the nodule. Most solid thyroid nodules yield

S

E M I N A R S I N

D

I A G N O S T I C

P

satisfactory material with proper FNA technique; however, those with cystic change, especially 475%, may have a higher nondiagnostic rate. Using US for the FNA in these cystic nodules does allow direct sampling of any solid component and/or areas of suspicious calcifications. Other nodules that can cause trouble in acquiring diagnostic material include those with sclerosis, a thick or calcified capsule, hypervascularity, or necrosis. If a suspicious or malignant thyroid nodule is confirmed on preliminary FNA interpretation and there is known suspicious cervical lymphadenopathy, FNA biopsy of the abnormal lymph node is also warranted. An increasingly common occurrence in our practice is the PET-positive thyroid “incidentaloma,” which is typically found in patients being evaluated for a known nonthyroidal malignancy. PET-positive nodules should also undergo US evaluation and UGFNA biopsy because of the higher risk of malignancy. Particularly challenging are thyroid bed lesions that develop in post-thyroidectomy patients, especially those with a history of thyroid cancer. Scar tissue and distortion of normal anatomy is the main problem encountered in US examination and biopsy of these patients. Furthermore, the distinction between recurrent thyroid cancer and benign lesions in the thyroid bed on US cannot be made with any accuracy and UGFNA is usually required. Numerous thyroid bed lesions, besides recurrent carcinoma, have been described that are potential pitfalls and include residual benign thyroid, parathyroid tissue, reactive lymph nodes, foreign body giant cell reaction or suture granuloma, strap muscle, cysts, tracheal cartilage, and fat necrosis.30

Parathyroid Normal parathyroid glands are rarely visualized on US because of their compressed and minute size (3–6 mm). They also have internal echogenicity similar to that of the thyroid or other neck tissues. Parathyroid adenomas and hyperplasia will have similar US features. They present as well-defined nodules that are typically homogeneous, hypoechoic, and located near the posterior margins of the thyroid. Their shape may be oval or conform to the pressure of their surrounding structures; however, elongated or “cigar-like,” discoidal, or lobulated forms have been described.31 On Doppler US examination, a vascular pole, peripheral arc of flow, and/or prominent internal vascularity may be demonstrated. Enlarged or cystic parathyroids can share US features that are similar to exophytic posterior thyroid nodules or cervical neck nodes and vice versa. Especially problematic is the US recognition of an intrathyroidal parathyroid. Collection of aspiration washings for PTH from suspected or questionable lesions can prove diagnostic and is recommended. In patients with hyperparathyroidism, difficulties in US location of the parathyroid may occur with small adenomas and from obscuring factors such as an enlarged thyroid goiter or in locations close to air-containing structures (trachea, pharynx, and esophagus) and substernally. Lieu has published a helpful review of pathologist-directed UGFNA of parathyroid lesions.32

A T H O L O G Y

32 (2015) 284–295

293

Salivary glands The superficial location of the parotid and submandibular gland makes them ideal targets for FNA and US examination. US is particularly useful to help distinguish extrinsic from intrinsic salivary gland targets. Tumors of the sublingual gland are rare, as they comprise only 1% of salivary gland neoplasms. Our discussion will focus on parotid and submandibular gland only. The US appearance of all the salivary glands is similar to that of the thyroid and typically shows a homogenous ground-glass appearance that is brighter than skeletal muscle (Fig. 3I and K). However, their US appearance may vary depending on the proportion of intraglandular adipose tissue, and significant fatty replacement may make US evaluation of small and deep-seated targets difficult. The parotid gland occupies the area just below the zygomatic arch and lies between the anterior border of the ramus of the mandible and the mastoid process. A portion of the parotid gland “tail” extends into Level II. The facial nerve cannot be visualized on US, but the retromandibular vein lying in close proximity is readily identified and can be used as a surrogate marker to divide the parotid into superficial and deep lobes. The superficial parotid gland is the most accessible to US and UGFNA as the deep lobe is incompletely visualized as it extends behind the mandible and into the parapharyngeal space.33 CT-guided FNA may be needed to reach these deepseated parotid areas. Intra-parotid lymph nodes are not easily identified, except when enlarged, and are most often found in the preauricular and the tail of the gland. Stensen’s duct is not typically visualized unless dilated from obstruction. The submandibular gland is pinecone-shaped and lies between the mandible and mylohyoid muscle34 (Fig. 3K). The entire gland is easily examined on US, and there are normal extrinsic, but no intraglandular lymph nodes. Wharton’s duct may be identified even if not distended and the facial vein and artery may be identified as they cross the gland and over the mandibular bone. US is often not useful in distinguishing between benign (Fig. 3L) and malignant salivary gland tumors, although the latter may be suspected if there are findings such as irregular margins and posterior shadowing (Fig. 3J) or they have a heterogeneous internal echotexture. We have additionally discovered that tumors of the same histologic type, for example, Warthin tumor and pleomorphic adenoma, can have varying US appearances. Another difficulty we have encountered is the US identification of non-palpable targets in irradiated salivary glands (and in thyroid as well), as these glands can become hypoechoic and inhomogeneous after this treatment. Furthermore, radiation-induced fibrosis and atrophy make acquiring diagnostic material difficult. Lastly, we have had particular success in applying US in our H&N service for re-biopsy of salivary gland lesions in which a previous outside PGFNA result yielded limited or unsatisfactory cells for interpretation. With US, we can specifically identify the longest axis of the target and make longer needle excursions through this portion, which we believe helps increase our diagnostic yield.

294

S

E M I N A R S I N

D

I A G N O S T I C

P

Soft tissue and miscellaneous lesions There are numerous other H&N tumors and tumor-like conditions that arise in the soft tissue outside of the glands already discussed for which US evaluation and UGFNA can prove useful. Particularly, thyroglossal duct cysts, ranulas, epidermal inclusion cysts, and branchial cleft cysts are lesions for which US can be used to help determine the anatomical location for classification and aid in biopsy if needed. Likewise, we find that CT scan is occasionally unreliable in distinguishing solid from cystic neck lesions and that dynamic US can be more telling. The classic US features of a cyst include compressibility with the transducer, well-defined margins, posterior acoustic enhancement, no evidence of internal blood flow on Doppler imaging, and realtime movement of internal cyst contents. Moreover, US enables one to sample the walls of the cysts for biopsy. Preand post-drainage interrogation helps evaluate for any solid component that may additionally be biopsied. Lipomas are often easily suggested by US exam alone, and this can be reassuring to the pathologist when only benign fat is aspirated, as occasionally these tumors are clinically misdiagnosed as an enlarged lymph node or thyroid mass if on the anterior neck (Fig. 3M). Neural tumors may be suggested on US by connection to a nerve. Pre-biopsy US exam may be advantageous in suggesting a carotid body paraganglioma as there is a perceived risk for biopsy-related bleeding even though many uncomplicated biopsies have been reported for these tumors. On US, carotid body paragangliomas show characteristic splaying of the internal and external carotid arteries and are usually intensely vascular on Doppler imaging. We have also encountered a number of examples of nonneoplastic and normal soft tissue structures in our practice, which have included foreign bodies (Fig. 3N), prominent or calcified vascular structures (Fig. 3O), and the transverse process of cervical vertebrae (Fig. 3P), that were initially referred for FNA as they simulated a clinically concerning lesion. Ultimately, these were recognized on US exam and spared the patient an unneeded and likely nondiagnostic FNA procedure.

Summary In conclusion, formal training programs in H&N US and UGFNA are now well established for the pathologist. We hopefully provided some direction and insight for the pathologist who is considering incorporating US and UGFNA into their FNA practice, and more importantly, demonstrated some of the benefits to patient care and to the practice of pathology. Finally, a close working relationship between the pathologist and the treating otolaryngologist/head and neck surgeon is necessary to ensure optimal patient outcome.

r e f e r e n c e s

1. American College of Radiology. Appropriateness criterias neck mass/adenopathy. Available at: 〈https://acsearch.acr.org/ docs/69504/Narrative/〉.

A T H O L O G Y

32 (2015) 284–295

2. Abele JS. The case for pathologist ultrasound-guided fineneedle aspiration biopsy. Cancer. 2008;114(6):463–468. 3. Lieu D. Cytopathologist-performed ultrasound-guided fineneedle aspiration and core-needle biopsy: a prospective study of 500 consecutive cases. Diagn Cytopathol. 2008;36(5):317–324. 4. Rollins SD. Decade of wisdom on ultrasound-guided FNAs. CAP Today. 2008;8(8):51. 5. Powers CN, Jakowski JD. Ultrasound-guided fine-needle aspiration biopsy: the cytopathologist’s next frontier. Pathol Case Rev. 2013;18(1):1–2. 6. College of American Pathologists. Available at: 〈http://www.cap. org/apps/docs/education/UGFNA_Manual.pdf〉. 7. Abele JS. Initiating private practice ultrasound-guided fineneedle aspiration: beginner’s road map and survival guide. Pathol Case Rev. 2013;18(1):53–57. 8. Ljung BM, Pitman MB. Establishing ultrasound-guided fineneedle aspiration in an academic setting. Pathol Case Rev. 2013;18(1):48–52. 9. Coltrera MD. Clinician-performed thyroid ultrasound. Otolaryngol Clin N Am. 2014;47(4):491–507. 10. Iro H, Alessandro Bozzato, Johannes Zenk. Atlas of Head and Neck Ultrasound. Stuttgart: Thieme; 2013. 11. Baskin HJ, Duick DS, Levine RA. Thyroid Ultrasound and UltrasoundGuided FNA, 3rd ed. New York: Springer Publishing; 2013. 12. Sites BD, Spence BC, Gallagher JD, Wiley CW, Bertrand ML, Blike GT. Characterizing novice behavior associated with learning ultrasound-guided peripheral regional anesthesia. Reg Anesth Pain Med. 2007;32(2):107–115. 13. Chin KJ, Perlas A, Chan VW, Brull R. Needle visualization in ultrasound-guided regional anesthesia: challenges and solutions. Reg Anesth Pain Med. 2008;33(6):532–544. 14. Sites BD, Brull R, Chan VW, et al. Artifacts and pitfall errors associated with ultrasound-guided regional anesthesia. Part I: understanding the basic principles of ultrasound physics and machine operations. Reg Anesth Pain Med. 2007;32(5):412–418. 15. Sites BD, Brull R, Chan VW, et al. Artifacts and pitfall errors associated with ultrasound-guided regional anesthesia. Part II: a pictorial approach to understanding and avoidance. Reg Anesth Pain Med. 2007;32(5):419–433. 16. Harvey JA, Moran RE, Hamer MM, DeAngelis GA, Omary RA. Evaluation of a turkey-breast phantom for teaching freehand, US-guided breast biopsy. Acad Radiol. 1997;4(8):565–569. 17. Barrington MJ, Wong DM, Slater B, Ivanusic JJ, Ovens M. Ultrasound-guided regional anesthesia: how much practice do novices require before achieving competency in ultrasound needle visualization using a cadaver model. Reg Anesth Pain Med. 2012;37(3):334–339. 18. Jakowski JD. Pictorial essay on how to perform an ultrasound guided fine needle aspiration biopsy. Pathol Case Rev. 2013;18 (1):18–23. 19. Jakowski JD. The basics of sonography and ultrasound terminology. Pathol Case Rev. 2013;18(1):12–17. 20. Lieu D. Ultrasound physics and instrumentation for pathologists. Arch Pathol Lab Med. 2010;134(10):1541–1556. 21. Ihnatsenka B, Boezaart AP. Ultrasound: basic understanding and learning the language. Int J Shoulder Surg. 2010;4(3):55–62. 22. Smith RS, Fry WR. Ultrasound instrumentation. Surg Clin North Am. 2004;84:953–971. 23. Feldman MK, Katyal S, Blackwood MS. US artifacts. Radiographics. 2009;29(4):1179–1189. 24. Jakowski JD, Powers CN. Ultrasound guided fine needle aspiration practice survey. Pathol Case Rev. 2013;18(1):58–63. 25. Pitman MB, Abele J, Ali SZ, et al. Techniques for thyroid FNA: a synopsis of the National Cancer Institute thyroid fineneedle aspiration state of the science conference. Diagn Cytopathol. 2008;36(6):407–424. 26. Jakowski JD, Kraft A, Powers CN. Lymph node ultrasound guided fine needle aspiration: two case reports addressing the

S

E M I N A R S I N

D

I A G N O S T I C

P

approach to non-palpable targets for biopsy and factors that affect sonographic image quality. Pathol Case Rev. 2013;18(1): 31–34. 27. Ying M, Bhatia KS, Lee YP. Review of ultrasonography of malignant neck nodes: grey scale, Doppler, contrast enhancement and elastography. Cancer Imaging. 2014;13(4):658–669. 28. Gharib H, Papini E, Paschke R, et al. American Association of Clinical Endocrinologists, Associazione Medici Endocrinologi, and European Thyroid Association medical guidelines for clinical practice for the diagnosis and management of thyroid nodules: executive summary of recommendations. Endocr Pract. 2010;16(3):468–475. 29. Gharib H, Papini E, Valcavi R, et al. American Association of Clinical Endocrinologists and Associazione Medici Endocrinologi medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract. 2006;12(1): 63–102.

A T H O L O G Y

32 (2015) 284–295

295

30. Shin JH, Han BK, Ko EY, Kang SS. Sonographic findings in the surgical bed after thyroidectomy: comparison of recurrent tumors and nonrecurrent lesions. J Ultrasound Med. 2007; 26(10):1359–1366. 31. Frasoldati A, Valcavi R. Challenges in neck ultrasonography: lymphadenopathy and parathyroid glands. Endocr Pract. 2004;10(3):261–268. 32. Lieu D. Cytopathologist-performed ultrasound-guided fineneedle aspiration of parathyroid lesions. Diagn Cytopathol. 2010;38(5):327–332. 33. Bialek EJ, Jakubowski W, Zajkowski P, Szopinski KT, Osmolski A. US of the major salivary glands: anatomy and spatial relationships, pathologic conditions, and pitfalls. Radiographics. 2006;26(3):745–763. 34. Ching AS, Ahuja AT. High-resolution sonography of the submandibular space: anatomy and abnormalities. Am J Roentgenol. 2002;179(3):703–708.