Pitfalls in Image Guided Tissue Sampling in the Head and Neck




Image-guided tissue sampling is becoming increasingly important for management of head and neck cancers. Ultrasound-guided fine-needle aspiration (UG-FNA) is safe, effective, and has many advantages compared with palpation-guided FNA and computed tomography-guided FNA. The technique of UG-FNA is highly operator and experience dependent; however, understanding the complex anatomy, disease processes, and patterns of nodal spread in the head and neck make this technique ideal for the neuroradiologist. Proper technique and recognition of pitfalls are critical to successful UG-FNA. Computed tomography–guided FNA is valuable for tissue sampling from deep lesions and for those without a sonographic window for UG-FNA.


Key points








  • The ability to easily and accurately establish a histologic diagnosis from the soft tissues of the neck using image guidance is becoming increasingly important.



  • Definite histologic characterization of deep and nonpalpable targets often requires image-guided tissue sampling.



  • Patient management frequently depends on accurate characterization of suspicious or indeterminate findings in the neck to determine initial stage malignancy or the presence or absence of residual, progressive, or recurrent malignancy.



  • This article reviews the technique and relative advantages for tissue sampling guided by ultrasonography and computed tomography.






Introduction


The ability to easily and accurately establish a histologic diagnosis in the soft tissues of the neck using image guidance is becoming increasingly important. Improved sensitivity afforded by advances in computed tomography (CT), magnetic resonance (MR) imaging, ultrasonography (US), and CT-positron emission tomography (PET) for diagnosis, staging, and surveillance now depict subclinical, small, and deep lesions that are inaccessible by palpation-guided (PG) histologic sampling. For example, CT-PET is increasingly being used for staging and surveillance of head and neck squamous cell carcinoma (HNSCC), sometimes resulting in indeterminate standardized uptake values for fluorodeoxyglucose (FDG) uptake in lymph nodes. Postoperative infection or posttreatment inflammatory changes can result in false-positive FDG uptake in the setting of recent surgery or radiation therapy, and small or necrotic lymph nodes can lead to false-negative FDG uptake. US is commonly used in initial staging and for routine surveillance of patients with papillary thyroid carcinoma (PTC), resulting in detection of new, enlarging, or mildly hypervascular lymph nodes, which may be suspicious for but not diagnostic of PTC.


Although mucosal lesions in the head and neck are readily biopsied under direct clinical inspection or endoscopically, the definite histologic characterization of submucosal or deep primary sites and nonpalpable lymph nodes often requires image-guided tissue sampling. Patient management frequently depends on accurate characterization of suspicious or indeterminate findings in the neck to determine initial stage or presence or absence of residual, progressive, or recurrent malignancy. These factors have made the ability to perform image-guided tissue sampling in the head and neck a critical skill.


Options for Tissue Sampling in the Head and Neck


There are several options for tissue sampling of nonmucosal lesions in the head and neck, including PG fine-needle aspiration (PG-FNA), US-guided FNA (UG-FNA), core biopsy, CT-guided FNA (CTG-FNA) or core biopsy and open surgical biopsy. Frequently, masses found on imaging studies are deep and nonpalpable, precluding the simplest and most inexpensive option, PG-FNA. UG-FNA has many desirable features for tissue sampling in the head and neck, making it ideal for evaluating nodal disease. CTG-FNA is extremely useful for lesions within the deep face and skull base that are too deep or do not have an adequate sonographic window for UG-FNA. Surgical excision should be reserved for cases in which image-guided biopsy is inconclusive or in which performing image-guided biopsy would not affect the decision to perform surgery.


Observation with repeat US, CT, MR imaging, or CT-PET is also a reasonable alternative to tissue sampling in certain circumstances, but despite considerably increased cost, short-interval follow-up examinations may not be definitive to exclude malignancy. Before initiating chemotherapy or radiation therapy regimens or performing potentially risky surgical procedures, referring services may desire a tissue diagnosis of malignancy regardless of the follow-up examination results, mandating image-guided tissue sampling. The small percentage of nondiagnostic image-guided FNAs/biopsies can be repeated at a fraction of the cost of follow-up imaging, if needed.




Introduction


The ability to easily and accurately establish a histologic diagnosis in the soft tissues of the neck using image guidance is becoming increasingly important. Improved sensitivity afforded by advances in computed tomography (CT), magnetic resonance (MR) imaging, ultrasonography (US), and CT-positron emission tomography (PET) for diagnosis, staging, and surveillance now depict subclinical, small, and deep lesions that are inaccessible by palpation-guided (PG) histologic sampling. For example, CT-PET is increasingly being used for staging and surveillance of head and neck squamous cell carcinoma (HNSCC), sometimes resulting in indeterminate standardized uptake values for fluorodeoxyglucose (FDG) uptake in lymph nodes. Postoperative infection or posttreatment inflammatory changes can result in false-positive FDG uptake in the setting of recent surgery or radiation therapy, and small or necrotic lymph nodes can lead to false-negative FDG uptake. US is commonly used in initial staging and for routine surveillance of patients with papillary thyroid carcinoma (PTC), resulting in detection of new, enlarging, or mildly hypervascular lymph nodes, which may be suspicious for but not diagnostic of PTC.


Although mucosal lesions in the head and neck are readily biopsied under direct clinical inspection or endoscopically, the definite histologic characterization of submucosal or deep primary sites and nonpalpable lymph nodes often requires image-guided tissue sampling. Patient management frequently depends on accurate characterization of suspicious or indeterminate findings in the neck to determine initial stage or presence or absence of residual, progressive, or recurrent malignancy. These factors have made the ability to perform image-guided tissue sampling in the head and neck a critical skill.


Options for Tissue Sampling in the Head and Neck


There are several options for tissue sampling of nonmucosal lesions in the head and neck, including PG fine-needle aspiration (PG-FNA), US-guided FNA (UG-FNA), core biopsy, CT-guided FNA (CTG-FNA) or core biopsy and open surgical biopsy. Frequently, masses found on imaging studies are deep and nonpalpable, precluding the simplest and most inexpensive option, PG-FNA. UG-FNA has many desirable features for tissue sampling in the head and neck, making it ideal for evaluating nodal disease. CTG-FNA is extremely useful for lesions within the deep face and skull base that are too deep or do not have an adequate sonographic window for UG-FNA. Surgical excision should be reserved for cases in which image-guided biopsy is inconclusive or in which performing image-guided biopsy would not affect the decision to perform surgery.


Observation with repeat US, CT, MR imaging, or CT-PET is also a reasonable alternative to tissue sampling in certain circumstances, but despite considerably increased cost, short-interval follow-up examinations may not be definitive to exclude malignancy. Before initiating chemotherapy or radiation therapy regimens or performing potentially risky surgical procedures, referring services may desire a tissue diagnosis of malignancy regardless of the follow-up examination results, mandating image-guided tissue sampling. The small percentage of nondiagnostic image-guided FNAs/biopsies can be repeated at a fraction of the cost of follow-up imaging, if needed.




UG-FNA: technique and pitfalls


Rationale for and Advantages of UG-FNA


There are multiple advantages of UG-FNA in the neck over PG-FNA and CTG-FNA. In experienced hands, UG-FNA is safe and simple to perform. Although a lesion may be palpable, its depth relative to the neck musculature and relationship to surrounding vascular structures may be unknown, precluding safe PG tissue sampling if previous imaging is not available documenting the location of the lesion. In the neck, lymph nodes are typically in close relation to the carotid arterial system or internal jugular veins. UG-FNA allows for precise targeting of small nodes with a larger safety margin, because the needle tip can be imaged in real time, and adjacent vascular structures can be clearly delineated using color Doppler techniques, allowing safe sampling of even small targets ( Fig. 1 ). Targets for aspiration may be heterogeneous in composition and contain cystic or necrotic spaces. Under US guidance, the needle can be precisely positioned to a specific abnormal area of a mass such as solid or cystic portions to maximize yield depending on the suspected diagnosis ( Fig. 2 ), and areas of hypervascularity including hilar vascularity ( Fig. 3 ) can be avoided to decrease the likelihood of aspirating purely bloody or blood-diluted samples. UG-FNA has shown improved sensitivity and specificity over CT and MR imaging for staging the N0 neck in HNSCC, and for detecting neck recurrence after previous treatment.




Fig. 1


Safe and precise tissue sampling adjacent to vascular structures using UG-FNA. Transverse US image showing successful needle targeting of a 3-mm nodule ( arrow ) adjacent to the internal jugular vein (IJV) and common carotid artery (CCA).



Fig. 2


UG-FNA of heterogeneous lesions. Transverse US images through the right parotid gland. ( A ) A heterogeneous centrally cystic and peripherally solid mass. ( B ) The needle is targeting the solid wall of the lesion.



Fig. 3


Avoidance of the hilar vascularity on UG-FNA. Longitudinal US images through the left neck. ( A ) Note a hypoechoic but normal reniform appearance to a lymph node with echogenic hilum ( arrow ). ( B ) Doppler US image shows normal flow in the hilum ( arrow ), which should be avoided when performing UG-FNA.


The UG-FNA procedure itself can be performed quickly and with minimal setup time. With an experienced technologist, much of the preparation, including equipment setup, target measurement, and initial optimization of the US acquisition parameters, can be standardized and completed before the radiologist’s involvement in the procedure. Compared with CTG-FNA, there is less time spent bringing the patient in and out of the CT gantry and repositioning the needle. Particularly, if the first pass is nondiagnostic, the setup time for a second pass is typically shorter for UG-FNA than CTG-FNA, unless a larger introducer needle has been used during CT guidance.


With CT guidance, a portion of the procedure is performed blind, with safeguards in place for the length of needle throw during positioning before aspiration. Unless iodinated intravenous (IV) contrast material is used for the CTG-FNA, a complex solid and cystic mass may have fairly uniform attenuation under CT, so solid elements may not always be visible and adjacent vasculature may have similar attenuation. In contrast to CTG-FNA, there is no ionizing radiation used in UG-FNA, which is particularly problematic when repeated thin-collimation CT imaging is used in the same location. Although radiation is not a critical issue for patients with known malignancy and who have undergone or will undergo radiation therapy, it is generally favorable to try to limit exposure whenever possible, particularly in the pediatric population or in younger patients, who are less likely to have a malignant diagnosis. The lack of need for IV placement, renal function testing, and injection of iodinated contrast material if vascular visualization is required is also an advantage of UG-FNA. In general, we reserve CTG-FNA for deep targets with no sonographic window for UG-FNA (masticator space, parapharyngeal space, retropharyngeal space, pterygopalatine fossa, and deep lobe of the parotid gland).


Advantages of a Neuroradiologist-Run UG-FNA Service


There is great variability in who performs UG-FNA in the neck, ranging from the neuroradiologist/head and neck radiologist, body radiologist, head and neck surgeon, and cytopathologist. This situation is often dictated by local referral patterns and personal expertise in the technique. However, there are many potential advantages of having a neuroradiologist perform this procedure. Thorough knowledge of the neck imaging anatomy, normal and abnormal appearance of the postoperative/postradiotherapy neck, and specific patterns of nodal spread for various head and neck malignancies is a major advantage over a body radiologist.


The neuroradiologist often interprets the imaging that may recommend FNA, leading to familiarity with the case and deliberation on the advantages and disadvantages of UG-FNA, CTG-FNA, and follow-up. At institutions with multidisciplinary tumor boards, cases requiring FNA for pathologic diagnosis often have the imaging presented by the neuroradiologist/head and neck radiologist at tumor board, so there is familiarity with the specific question to be answered by UG-FNA, which is important in complex cases with more than 1 potential target. There is also an automatic trigger for radiologic-pathologic discordance, so that mismatches can be discussed and further passes/alternative targets selected during the UG-FNA. Compared with surgical and cytopathologic colleagues, we are familiar with the imaging anatomy on multiple modalities and may already have skill sets from other image-guided procedures, including CTG-FNA.


Important to the success of a UG-FNA service is the presence of a dedicated US technologist with familiarity with setup of UG procedures. Also critical is the presence of an on-site cytopathologist. At our institution, all image-guided FNAs are performed with a cytopathologist present so that the appropriate number of needle passes are performed to obtain an adequate tissue sample for diagnosis. This process leads to immediate feedback on whether or not the procedure is being performed properly, and nondiagnostic passes can lead to adjustments in technique and sampling of different areas to achieve the highest possible yield, without performing additional unnecessary passes. For example, Fig. 4 shows a case of metastatic lymphadenopathy from melanoma that resulted in several nondiagnostic bloody passes through an obviously abnormal hypervascular lymph node. Attention was paid to an adjacent node with less vascularity, yielding melanoma on the first pass. In practices in which an on-site cytopathologist cannot be present, learning to properly smear and stain slides from a specimen is a desirable skill for the neuroradiologist performing UG-FNA. Maintaining a record of diagnostic yield and following up final pathologic diagnosis on each case, and troubleshooting nondiagnostic UG-FNAs, are critical to ongoing improvement with the technique.




Fig. 4


Selection of optimal target on UG-FNA when multiple available. Longitudinal US image through the right neck, showing 3 hypoechoic lymph nodes. The lymph node on the right ( large arrow ) shows markedly increased Doppler flow and yielded only blood during multiple passes. The lymph node on the left ( small arrow ) was then sampled, yielding metastatic melanoma.


Whether or Not to Perform UG-FNA


Knowledge of the disease processes involving the head and neck, as well as the advantages and disadvantages of UG-FNA versus CTG-FNA, is a major benefit of a neuroradiologist determining how a lesion should be sampled. The choice of US or CT guidance depends on a variety of factors. In general, the size, depth, and location of a lesion influence the choice of technique. Most lesions less than 3 cm from the skin surface on cross-sectional imaging are accessible with UG-FNA and a standard 3.8-cm (1.5-in) 23-gauge needle, because after compression of the skin by the transducer, this results in a depth less than 2 cm and an oblique needle trajectory less than 4 cm, leaving sufficient needle length for sampling. Fig. 5 depicts the general area accessible by UG-FNA with a standard 3.8-cm (1.5-in) needle. Deeper areas can be reached with a longer needle if required, although the resolution of the lower frequency of US needed to penetrate deeper may prohibit adequate needle and target visualization. On occasion, cases are referred for findings reported as indeterminate on CT or MR imaging, which the authors on review believe are probably benign. These cases are scheduled for an UG-FNA but may be converted to a diagnostic neck US examination without FNA if the neuroradiologist feels it has a benign sonographic appearance.


Mar 18, 2017 | Posted by in NEUROLOGICAL IMAGING | Comments Off on Pitfalls in Image Guided Tissue Sampling in the Head and Neck
Premium Wordpress Themes by UFO Themes