Percutaneous image-guided biopsy has largely replaced open surgical biopsies for many head and neck (H&N) lesions, being very safe and minimally invasive. Although the radiologist plays the primary role in these cases, it requires a multidisciplinary approach. Depending upon numerous factors, these biopsies can be either fine-needle aspiration or core needle biopsy, using ultrasound for superficial lesions and computed tomography for deep neck lesions. The most crucial part of H&N biopsies is planning a trajectory to avoid injury to critical anatomic structures. This article outlines the standard biopsy approaches and key anatomical considerations for H&N procedures.
Key points
- •
Excellent anatomical knowledge is required for head and neck biopsies given the critical structures in the region, including vital nerves and vessels.
- •
Core needle biopsies provide larger tissue samples with higher diagnostic accuracy as compared with fine-needle aspirations.
- •
Ultrasound should be the preferred modality for most superficial lesions, whereas computed tomography is generally the choice for deep neck lesions.
- •
Meticulous planning, and good pre- and post-procedural care are essential parts of a successful procedure.
- •
Although minor complications are occasionally seen, major complications with image-guided head and neck biopsies are extremely rare.
Introduction
Image-guided biopsies have largely replaced open surgical techniques as the preferred method for establishing a histologic diagnosis for most of head and neck (H&N) pathologies and add tremendous value in terms of accuracy, safety, efficacy, lower cost, and overall better patient care. Dedicated H&N radiologists or neuroradiologists with expertise in H&N disease, now form an essential part of the multidisciplinary ear nose throat (ENT) team. They can best provide high-quality care, both in the diagnostic and interventional realm, given the critical anatomy in the region of the biopsies, including major nerves and vessels. Over the past decade, there has been increasing subspecialization in the field of neuroradiology with many large centers having dedicated H&N radiologists or a small set of neuroradiologists with subspecialty expertise. , As the number of requisitions for percutaneous image-guided biopsies has increased tremendously, so has interest among radiologists. H&N intervention is still an evolving field, and this service line is managed differently at different centers by interventional radiologists, abdominal radiologists, musculoskeletal radiologists, neurointerventional surgeons, and neuroradiologists. We believe that diagnostic H&N neuroradiologists who regularly interpret these scans, participate in ENT tumor boards, and gradually develop a high level of anatomical knowledge provide optimal care.
Image-guided biopsies can be either fine-needle aspiration (FNA) or core needle biopsy (CNB) and can be performed under computed tomography (CT) or ultrasound (US) guidance. In general, if a lesion is palpable or superficial in location, it should be amenable to US guidance. On the contrary, CT is the modality of choice for deep neck lesions. MRI-guided H&N interventions have failed to garner much support and widespread acceptability, despite the advancement in open high-field magnets and are generally felt to be unnecessary given the wider availability and ease of CT-guided procedures. Thyroid and parotid biopsies are usually FNAs, almost always done under US, and a detailed discussion of these is beyond the scope of this article. , In this review article, we focus primarily on the anatomical approaches and biopsy techniques for deep-seated neck lesions with an emphasis on CT-guided procedures. Suprahyoid neck biopsies, including skull-base biopsies, are more challenging than infrahyoid neck lesions, given the complex bony structures and narrow windows. Although SCCa forms the most common indication for image-guided biopsies, the range of lesions is diverse, including other benign and malignant neoplasms, infectious conditions like skull-base osteomyelitis, and inflammatory conditions such as immunoglobulin G4 (IgG4) disease. As with any other interventions, a comprehensive knowledge of anatomy and proper planning are critical. The anatomic approach is one of the most important factors in H&N biopsies, given the vital anatomic arrangement in this region. Despite the complexity, major complications with H&N biopsies are rare, with few minor complications reported in the literature, like small hematomas and minor infections. Surgical biopsies are usually reserved for mucosal lesions easily accessible on endoscopic exams or for cases where image-guided biopsy is inconclusive.
Protocol, review, and scheduling
The importance of thorough pre-biopsy planning cannot be overstated, and often requires more time and effort than performing the actual procedure. Fig. 1 outlines the algorithm used at our institution before scheduling the patient for image-guided biopsy. This includes review of the case at the ENT tumor board, obtaining all current and prior imaging, and ordering additional imaging if needed. Input from the radiation oncologist is crucial for both previously treated cases and for post-biopsy management. A significant number of biopsy requests that arise from outside the multidisciplinary team are either felt to be unnecessary or can be delayed in lieu of serial imaging. The list of indications for H&N biopsies is extensive, with a primary diagnosis or follow-up of SCCa (primary lesion or nodal) being the most common. However, there is a list of “do not touch” lesions, which includes anatomic variants, “pseudo-masses,” venolymphatic malformations ( Fig. 2 ), and benign neurogenic tumors, which should be identified and completely avoided, especially if near the skull base or critical anatomic structures. Apart from imaging, the team needs to review serological markers, including the levels of circulating tumor DNA (liquid biopsy), which has gained prominence recently for the surveillance of cancer recurrence in human papillomavirus (HPV)-associated squamous cell carcinoma (SCCa) ( Fig. 3 ). Each H&N biopsy is protocolled at our institution by a member of the H&N biopsy group and scheduled on a day when a member of the H&N biopsy team is scheduled for procedures. Any need for the patient to be nothing per oral (NPO) should also be outlined in the protocol and communicated with the patient before the procedure.
Pre-procedural workup
Pre-procedural workup for H&N biopsies is like that of any interventional case. This includes withholding any anticoagulants such as aspirin, Plavix, warfarin, or medication that can alter platelets, prothrombin time (PT), and international normalized ratio (INR). Although lab evaluation for most H&N biopsies are not as stringent as spinal or vascular procedures, we do routinely check the INR (<1.5) for patients with a history of warfarin use. The patient usually checks in 2 h before the procedure in the “Prep and Recovery” room followed by routine clinical evaluation by the nursing staff. Informed consent is usually standard, with a brief discussion about the procedure, indication, alternatives, risks, benefits, and common side effects. For parotid and deep H&N biopsies, we inform the patient about the risk of transient facial weakness and numbness. Apart from general allergies documented in the electronic medical record, it is prudent to rule out allergies to lidocaine and latex, as these frequently go undocumented. The decision regarding no sedation, or mild or moderate sedation is made on a case-to-case basis with general anesthesia rarely necessary. We perform most of our deep H&N biopsies under mild sedation (Fentanyl: 25 to 50 μg) with continuous nursing support and vital monitoring during the procedure. Moderate sedation is preferred for skull base and osseous lesions, the latter requiring bone-cutting needles, such as osteo-site. , No sedation or nursing support is needed for most parotid and superficial biopsies. However, informing the patients regarding the high sensitivity of facial structures helps in set realistic expectations, as despite sedation, H&N biopsies are usually associated with some element of pain and discomfort.
Modality selection: ultrasound versus computed tomography
The decision regarding US versus CT guidance for H&N biopsies is usually straightforward. Superficial (within 3 cm of the skin surface) and palpable lesions and generally easily accessible by US. Thyroid and superficial parotid lesions are almost always done under US. US is cost-effective and safe with the benefit of real-time visualization and needle localization along with instant maneuvering of the needle trajectory for accurate tissue sampling. US requires minimal procedural time and is easier to schedule in comparison to CT; the latter being used by other interventional teams with limited time slots. If there is ambiguity regarding the modality, we usually double book under CT and US and make every attempt to use US guidance ( Fig. 4 ). Optimizing US parameters for proper lesion delineation is very important. This includes adjusting the depth and focus to the region of interest with the added advantage of a Doppler to evaluate the vascularity of a lesion and avoid any vessels along the tract. Linear high-frequency transducers are best suited for large skin surfaces, whereas compact high-frequency transducers (17 to 5 MHz; “hockey stick”) are better for narrow windows like the parotid angle. The curved-array small footprint transducers (8 to 5 MHz) are ideal for lesions in a narrow window with a steep angle ( Fig. 5 ). Although more straightforward than CT-guided procedures, US is not without its limitations. First, US procedures require more technical experience and are highly operator dependent. This is further accentuated by the fact that most neuroradiologists do not read US scans, leading to abdominal radiologists performing these at most centers. Second, lesions deep to osseous structures, calcifications, or metallic hardware are obscured on US. , Furthermore, US becomes even more challenging in the post-treated neck secondary to many factors, including lymphedema, chemoradiation-related fibrosis, and metallic hardware. There have been many instances where a large superficial lesion on CT was obscured on US due to dermal thickening and scarring associated with acoustic impedance. Apart from deep lesions, CT is usually the modality of choice for lesions at or near the skull base. CT provides excellent soft-tissue and osseous details and does not have the drawbacks of US. Near real-time CT fluoroscopy systems are available, with newer systems that feature a unique tablet-based mobile workflow for intuitive scanner operation. These systems also support users with innovative planning software. These newer scanners offer reduced radiation dose to protect the patient and the interventionist, whereas the tablet interface displays a dose thermometer that can monitor radiation levels in real-time. These also feature artifact reduction for the needle tip and intuitive touchscreen functions that help in finding the proper position for the needle while measuring relevant distances.
Fine-Needle Aspiration versus Core
FNAs and CNBs are the two most common tissue sampling techniques, each with pros and cons ( Table 1 ). The preferred method of tissue sampling should be decided upon before the procedure; however, the flexibility to switch from FNA to CNB always exists. By convention, FNA is defined as a gauge (G) of ≤ 22 and CNB is performed with thicker needles ranging from 14 to 20 G. In general, FNA is less traumatic but provides smaller tissue samples than core biopsies. FNA is usually sufficient for most parotid and thyroid lesions with few exceptions (like lymphoma) and offers a high diagnostic yield. FNA is also preferred for most nodal biopsies and lesions near neurovascular structures. For nodules with an initial nondiagnostic FNA, repeat FNA is futile and nondiagnostic in 28% to 53% of cases, whereas CNB provides a diagnostic sample in more than 95% of cases. In many cases, it is appropriate to perform FNA first with a cytopathologist present and then to proceed with core needle sampling if the cytologic samples are inadequate. The complication rates of CNB, such as infection, bleeding, and nerve injury, are not significantly higher than FNA if carried out under US guidance. The concern for seeding along the biopsy track is unfounded as the incidence is very low and usually related to the diameter of the needle, rather than the technique. , After inserting the needle into the lesion, applying suction to the syringe, and moving the needle back and forth (around 5 to 10 times) results in a better and more cellular sample ( Fig. 6 ). Suction should be released when withdrawing the needle and should also be discontinued immediately if the blood becomes visible in the needle hub. If infection is on the differential and a discrete pocket of fluid or abscess is seen, it is better to use suction to aspirate. Frequent and rapid changes in the direction of the needle is painful for the patient and with poor results. The first pass usually yields the most diagnostic material; however, the diagnostic yield can be improved by using up to three passes. More than three passes are usually unnecessary and do not increase the yield. Suction should be avoided for thyroid and vascular lesions to minimize bleeding. For cystic neoplasms, FNA is frequently superior to CNB as the latter does not allow for content aspiration. ,
| FNA | CNB |
|---|---|
| Rapid, less traumatic, lower complication rate | Higher sensitivity and diagnostic accuracy |
| Less expensive, rapid pathology turnaround | Provides larger tissue sample, larger intact tissue with preserved architecture |
| Preferred collection method for flow cytometry of lymphoid lesions | Preferred collection method for immunohistochemistry and most molecular markers |
| Standard for superficial parotid and thyroid lesions | Preferred for most deep neck and skull-base lesions |
| Limited tissue architecture characterization | More painful for the patient with need for sedation |
| Poor yield for fibrotic and cohesive lesions | Longer tissue-fixation and processing time |
| Poses limitation for immunohistochemistry and molecular testing | Higher complication rate including bleeding, tumor seeding etc. |
Instrumentation
Lidocaine (1%) is generally used for local anesthesia, even in sedated patients, with a 25-G needle used to create a good skin wheal. A deeper neck lesion requires deep anesthetic injection, using a variable length (3.5 to 7 inches) 22-G spinal needle. If the lesion is transosseous, a spinal needle can be used to inject anesthetic beneath the periosteum. For superficial FNAs, including parotid and thyroid lesions, a simple hypodermic needle (22 to 25 G) is usually sufficient with no need for coaxial technique. Approximately 3 to 5 passes are obtained depending upon sample adequacy as assessed by the cytopathologist. Echogenic needles can be used for US-guided procedures, providing a clearer, more defined image from any angle with reduced acoustic shadowing. Core biopsies (and occasional deep neck FNAs) are performed with a co-axial technique which involves the initial placement of an introducer thin-wall guide needle close to the target lesion followed by advancement of the biopsy needle through this needle to obtain multiple tissue samples with a single needle pass. , The coaxial needle size (length and thickness) is usually decided on a case-by-case basis. We usually use a 17 to 19 G introducer coaxial system with an 18 to 20 G biopsy needle, which has been shown to be safe with no adverse effects ( Fig. 7 A ). Although thicker bore needles provide more tissue, there is increased trauma to tissue and risk of hemorrhage. Moreover, needle gauge has been shown to have little effect on tissue diagnosis when at least an 18-gauge needle is used. Small caliber (as thin as 25 G) core biopsy needles are now available, which can consistently provide high-quality specimens with very low complication rates. Most biopsy needles have a “throw” length of 1 or 1 and 2 cm ( Fig. 7 B), with ultra-short throw needles available for smaller lesions. A 2-cm throw will provide the best tissue sample but with more tissue damage, and is rarely needed, with a 1 cm throw being sufficient for most H&N lesions. Blunt needle ( Fig. 7 C) systems, such as Hawkins-Akins, permit an atraumatic approach to the target structure and allow for the deflection of vascular structures, which can be used for lesions very close to major vessels. Alternatively, the sharp diamond tip stylet can be replaced by a blunt tip needle after crossing a particular landmark.
