7 Modified Radical Neck Dissection Abstract A modified radical neck dissection is defined as functional resection of the lymphatic levels I to V in the neck. This procedure was pioneered by Suarez and Bocca, who were the first to understand the concept of nodal groups defined by fascial places during the second half of the last century. This operation revolutionized the practice of head and neck oncology, and laid the foundation for the development of ever less invasive approaches, even decades after its inception. In this chapter, critical aspects of the surgical anatomy of the neck are reviewed, and the reader is presented with strategies to avoid common pitfalls in the execution of the procedure. Keywords: neck dissection, modified, functional, technique, complications. The concept of neck dissection refers to systematic resection of lymph nodes from well-defined fascial compartments in the neck. These will be, the vast majority of times, performed in the context of mucosal aerodigestive tract, salivary, cutaneous, or endocrine malignancies. The oncologic rationale for the procedure involves resecting lymph nodes that show clinical or radiological evidence of involvement (clinically positive necks or cN +) or are at risk of harboring microscopic disease (clinically negative necks or cN0). As such, depending on the clinical neck status, neck dissections are classified as therapeutic for patients with cN + and as elective or prophylactic in patients with cN0. Throughout its more than 100 years of history, neck dissection has continuously evolved from a mutilating radical surgery to an elegant, anatomically bound ablation with minimal functional impact. Similarly, as it relates to the nomenclature and indications, the last decades have seen a significant degree of consolidation that has laid the foundation for a common understanding of its oncological and technical aspects. The current nomenclature for the boundaries and contents of the nodal groups in the neck is summarized in Table 7.1 and illustrated in Fig. 7.1. A modified radical neck dissection, which is the most comprehensive form of functional neck dissection, entails the resection of the nodal groups I through V, and is still considered the standard of care for management of the cN + neck. The width and scope of this surgery mandate a polished surgical technique and thorough knowledge of the anatomy. While this surgery can be analyzed from multiple perspectives, this chapter will primarily focus on its anatomical and technical aspects, which are essential to perform this surgery safely, effectively, and efficiently. As such, it is strongly recommended that the reader reviews Chapter 3 to get intimately familiar with the surgical anatomy of the neck.
A modified radical neck dissection is defined as functional resection of the lymphatic levels I to V in the neck. This procedure was pioneered by Suarez and Bocca, who were the first to understand the concept of nodal groups defined by fascial places during the second half of the last century. This operation revolutionized the practice of head and neck oncology, and laid the foundation for the development of ever less invasive approaches, even decades after its inception. In this chapter, critical aspects of the surgical anatomy of the neck are reviewed, and the reader is presented with strategies to avoid common pitfalls in the execution of the procedure.
Keywords: neck dissection, modified, functional, technique, complications.
The concept of neck dissection refers to systematic resection of lymph nodes from well-defined fascial compartments in the neck. These will be, the vast majority of times, performed in the context of mucosal aerodigestive tract, salivary, cutaneous, or endocrine malignancies. The oncologic rationale for the procedure involves resecting lymph nodes that show clinical or radiological evidence of involvement (clinically positive necks or cN +) or are at risk of harboring microscopic disease (clinically negative necks or cN0). As such, depending on the clinical neck status, neck dissections are classified as therapeutic for patients with cN + and as elective or prophylactic in patients with cN0. Throughout its more than 100 years of history, neck dissection has continuously evolved from a mutilating radical surgery to an elegant, anatomically bound ablation with minimal functional impact. Similarly, as it relates to the nomenclature and indications, the last decades have seen a significant degree of consolidation that has laid the foundation for a common understanding of its oncological and technical aspects.
The current nomenclature for the boundaries and contents of the nodal groups in the neck is summarized in Table 7.1 and illustrated in Fig. 7.1. A modified radical neck dissection, which is the most comprehensive form of functional neck dissection, entails the resection of the nodal groups I through V, and is still considered the standard of care for management of the cN + neck. The width and scope of this surgery mandate a polished surgical technique and thorough knowledge of the anatomy. While this surgery can be analyzed from multiple perspectives, this chapter will primarily focus on its anatomical and technical aspects, which are essential to perform this surgery safely, effectively, and efficiently. As such, it is strongly recommended that the reader reviews Chapter 3 to get intimately familiar with the surgical anatomy of the neck.
Content and boundaries
Level I (submental and submandibular)
The lymph nodes between the mandible and hyoid bone. The posterior (lateral) boundary is the vertical plane defined by the posterior edge of the submandibular gland. This level is divided into two sublevels.
Sublevel IA (submental)
The lymph nodes within the triangular boundary of the anterior belly of the digastric muscles and the hyoid bone.
Sublevel IB (submandibular)
The lymph nodes within the boundaries of the anterior belly of the digastric muscle, the stylohyoid muscle, and the body of the mandible. The submandibular gland is usually included within the specimen when the lymph nodes of this triangle are removed.
Level II (upper jugular)
The lymph nodes located around the upper third of the internal jugular vein and spinal accessory nerve, extending from skull base to the level of the inferior border of the hyoid bone. The anterior (medial) boundary is the vertical plane defined by the posterior edge of the submandibular gland and the posterior (lateral) boundary is the posterior border of the sternocleidomastoid muscle.
The lymph nodes located anterior (medial) to the spinal accessory nerve.
The lymph nodes located posterior (lateral) to the spinal accessory nerve.
Level III (middle jugular)
The lymph nodes located around the middle third of the internal jugular vein extending from the inferior border of the hyoid bone (above) to the inferior border of the cricoid cartilage (below). The anterior (medial) boundary is the lateral border of the sternohyoid muscle, and the posterior (lateral) boundary is the posterior border of the sternocleidomastoid muscle.
Level IV (lower jugular)
The lymph nodes located around the lower third of the internal jugular vein extending form the inferior border of the cricoid cartilage (above) to the clavicle below. The anterior (medial) boundary is the lateral border of the sternohyoid muscle and the posterior (lateral) boundary is the posterior border of the sternocleidomastoid muscle.
The lymph nodes located along the lower half of the spinal accessory nerve and the transverse cervical artery. The supraclavicular nodes are also included in the posterior triangle group. The superior boundary is formed by the sternocleidomastoid and trapezius muscles, the inferior boundary is the clavicle, the anterior (medial) boundary is the posterior border of the sternocleidomastoid muscle, and the posterior (lateral) boundary is the anterior border of the trapezius muscle. This level is also divided at the level of anterior cricoid arch into sublevels VA and VB.
Contents of level V above the level of anterior cricoid arch.
Contents of level V below the level of anterior cricoid arch.
Level VI (anterior compartment group or central group)
The pretracheal and paratracheal nodes, precricoid (Delphian) node, and the perithyroidal nodes, including the lymph nodes along the recurrent laryngeal nerves. The superior boundary is the hyoid bone, the inferior boundary is the suprasternal notch, and the lateral boundaries are the common carotid arteries.
Level VII (upper mediastinal group)
Contains the paratracheal lymph nodes and fibrofatty tissue located between the suprasternal notch and the innominate artery.
7.1.1 History and Classification
Historically, the first accounts of nodal resection for treatment of cancers date back to the 19th century, with Kocher and Packard describing the removal of cervical lymph nodes as treatment of oral cavity cancers. Franciszek Jawdyński is credited with the first description of a radical neck dissection in the Polish literature in 1888. However, George Crile was the surgeon who popularized the technique and established its oncologic rationale and indications.1 In his description of radical neck dissection, Crile advocated for en bloc resection of cervical nodes and the internal jugular vein (IJV), sternocleidomastoid muscle (SCM), and spinal accessory nerve (SAN). The procedure was widely adopted and endorsed by prominent head and neck surgeons such as Vilray Blair and Hayes Martin and became the standard of care for neck management through the first half of the 20th century. In the second half of the 20th century, most of the modifications to the procedure were aimed at limiting morbidity, ushered by Ward, who described a SAN-sparing approach to neck dissection in 1951. This trend toward less radical procedures was fueled by further understanding of the fascial boundaries of the neck compartments, which led to the description of a comprehensive neck dissection sparing the SAN, IJV, and SCM by Suarez in 1963, which he named functional neck dissection.2 It was Bocca and Pignataro,3 however, who was largely responsible for disseminating and promoting the technique in the United States and Europe. The next incremental steps in the evolution of this procedure became possible thorough a more thorough understanding of the patterns for nodal spread depending on the location of the primary tumor.4 This ushered in the concept of selective neck dissections, which only address the nodal basins at risk and spare other lymphatic groups, and largely reducing perioperative morbidity. Although these were initially implemented in the context of the cN0 neck, they are more commonly being used for treatment of cN +. As it relates to the classification of the procedure, the current nomenclature has not changed since 2002 ( Table 7.2) and modified radical neck dissection formally entails the resection of all lymphatic groups in the neck (I–V) with preservation of the IJV, SAN, and SCM. This procedure is still the standard of care in the management of the cN + neck, although there is growing evidence supporting the role of selective approaches in this setting as well. Given the low rate of involvement of level V, this level is commonly spared, yielding to the common practice among head and neck surgeons of colloquially calling a selective neck dissection of levels I to IV, a modified radical neck dissection. The history and the evolution of the procedure are reviewed in detail in Chapter 2.
The following section describes a step-by-step approach for the dissection of neck levels I to V in a sequential fashion. Decision points and critical aspects of the surgical anatomy will be addressed as the process is described, with the intention of contextualizing the information within the flow of the operation. For the same reason, intraoperative images were preferred over diagrams whenever possible. From the perspective of a practicing head and neck surgeon, the importance of having a systematic approach to the neck cannot be overemphasized, and this chapter has been written with that goal in mind. In this author’s experience, this consistency will reward practitioners with a level of confidence, and familiarity with the anatomy, that will prove essential when clinical judgment calls for an alternative surgical approach.
Radical neck dissection
Radical neck dissection
Classification and terminology of neck dissection has not changed.
• Boundaries between levels I and II, and levels III/IV and VI
• Terminology of the superior mediastinal nodes (level VII)
• The method of submitting surgical specimens for pathologic analysis
Modified radical neck dissection
Modified radical neck dissection
Selective neck dissection
Selective neck dissection: each variation is depicted by “SND” and the use of parentheses to denote the levels or sublevels removed
Extended neck dissection
Extended neck dissection
(Committee for Head and Neck Surgery and Oncology of the American Academy of Otolaryngology–Head and Neck Surgery)
(American Head and Neck Society and Committee for Head and Neck Surgery and Oncology, American Academy of Otolaryngology–Head and Neck Surgery)
(American Head and Neck Society and Committee for Head and Neck Surgery and Oncology, American Academy of Otolaryngology–Head and Neck Surgery)
As it relates to the decision-making process before and during the operation, there are countless valid variations in planning and execution that truly epitomize the art of surgery. A such, the information presented by no means represents an exclusive view of this operation, but rather a compendium of collective knowledge and personal experience, validated over years of academic practice.
7.2.1 Patient Positioning
Once the airway is secured, the bed is routinely turned 90 to 180 degrees. Before placing a shoulder roll, the neck is inspected in the supine position under gently anterior flexion to identify and mark the location of skin creases. A shoulder roll is then placed at the level of the scapulas in order to gently lift the superior chest and neck. Care must be taken to ensure the head is not hanging, especially in older patients. The head must be supported with a soft device that holds it in position, such as a donut-shaped gel cushion. While the procedure can be performed in straight supine position, the best exposure is achieved in beach chair position, elevating the back of the table about 30 degrees, and gently lowering the legs. The head is extended and rotated to the opposite side; this maneuver not only provides the best exposure, but also brings the relevant anatomical structures to a more horizontal plane, significantly facilitating the dissection ( Fig. 7.2).
At this point, inspection and palpation are used to identify landmarks in surface anatomy such as the angle of the mandible, suprasternal notch, and mastoid tip. The neck can be gently flexed to identify natural skin creases, which should also be marked.
The placement of the surgical incisions for a neck dissection can have a profound impact on access, perioperative morbidity, and reconstructive considerations. During the evolution of neck dissection, many surgical approaches have been described ( Fig. 7.3) and are associated with the pioneers of the procedure. Nowadays, most of these approaches are seldom performed. As the collective knowledge has matured, a more utilitarian approach is overwhelmingly favored by most surgeons; as such, modern-day incisions are usually apron-type or transversal incisions at a skin crease.
The factors that must be considered for surgical planning are the following:
• Skin flap viability.
• Appropriate access to nodal groups of interest.
• Cosmetic outcome.
• Reconstructive needs (need for skin resection/rotational flaps).
• Reoperative exposure (risk of ipsilateral or contralateral neck recurrence).
Of these, skin flap viability is perhaps the most important consideration, and is affected by many factors, most of them outside the control of the surgeon (previous neck radiation or surgery, malignant cutaneous extension, vasculopathy). Regardless of the chosen approach, some common considerations are helpful in minimizing the risk of flap failure:
• Skin flaps must be broad based to maximize blood supply.
• Flap elevation should be in the subplatysmal plane and limited to the areas of oncologic interest.
• Trifurcate incisions should be avoided whenever possible.
• Placement of secondary limb incisions should consider the risk of carotid exposure.
• Preoperative planning is essential when performing cervicalbased rotational flaps for reconstruction (i.e., platysma or submental flaps).
In terms of flap viability, the platysma plays an important role in neck dissections. The muscle covers the vast majority of the anterior neck, and it has a rich blood supply that originates primarily from the submental artery, with tributaries from the superior thyroid, occipital, and posterior auricular arteries.5 For this reason, incorporating this muscle in the flaps maximizes blood supply to the overlying skin, thus significantly decreasing the risk of flap loss. This muscle is also important as it defines the appropriate fascial layers of the procedure. The platysma is enveloped by the superficial cervical fascia; elevating the skin flaps in the plane immediately deep to it will preserve the integrity of the superficial layer of the deep cervical fascia (SLDCF; considered the superficial boundary of the dissection).
Fig. 7.3 Neck incisions for modified radical neck dissection. (a) Hockey stick or half-apron. (b) Boomerang. (c) MacFee. (d) Modified Schobinger. (e) Apron or bilateral hockey stick.
There are some situations that mandate flap elevation in a supraplatysmal plane. These include cases where there is concern for involvement of the muscle by the disease process, such as extracapsular nodal extension, or when a myocutaneous platysma flap is considered for reconstructive purposes. While it has been largely replaced by other reconstructive options, the myocutaneous platysma flap should be part of the armamentarium of every head and neck surgeon, as it provides an elegant alternative for closure of small to medium sized intraoral defects, especially in patients with significant comorbidities. In a recent literature review, Eckard reported a flap success rate ranging from 71 to 100%6 and concluded that most failures were related to history of previous radiation or ligation of the facial artery. From the technical standpoint, patient selection and surgical planning are of outmost importance if this flap is to be attempted. The flap can be harvested as a superiorly or posteriorly based island flap ( Fig. 7.4). Superiorly based flaps have the advantages of an increased arc of rotation, robust arterial supply, and capability of sparing the cervical branch of the facial nerve, while posteriorly based flaps have better venous drainage. Regardless of the chosen approach, these flaps must be marked preoperatively, and they dictate the location of the rest of the incisions for the neck dissection. In a similar fashion, submental rotational flap and cervicofacial flaps require thoughtful preoperative consideration when planning for incision placement.
Fig. 7.5 Initial elevation of the subplatysmal flap. EJV, external jugular vein; GAN, greater auricular nerve.
7.2.3 Exposure: Skin Flap Elevation
Anatomically, a modified radical neck dissection requires wide exposure of the anterior and posterior neck triangles, so the boundaries for skin flap elevation are the mandibular margin and tail of parotid superiorly, the clavicle inferiorly, the strap muscles medially, and the anterior border of the trapezius muscle posteriorly.
Once the incision placement has been decided, the skin is incised with Bovie electrocautery and the incision is taken through the subcutaneous tissue and platysma, carefully avoiding violating the fascial layer below. Alternatively, the skin incisions may be performed with scalpel, prior injection with diluted lidocaine and epinephrine. At this point, skin flaps are elevated in the subplatysmal plane wide enough to expose all the anatomical landmarks and provide access to the nodal groups of interest. This is a relatively avascular plane that is only intermittently pierced by cutaneous perforators, so flaps can be raised with sharp instrumentation or with Bovie electrocautery ( Fig. 7.5). For this reason, the rate of bleeding can be used as a surrogate for appropriateness of the surgical plane: if bleeding is significant, the skin flaps are likely being raised in the wrong plane (usually too deep).
In the midline, the platysma has a variable degree of dehiscence, and fibers that decussate in the submental area. For this reason, it is recommended that for patients undergoing bilateral neck dissection, lateral flaps are elevated first and subsequently addressing the anterior neck, once the correct plane has been established. If an apron skin flap is used for access, the superior elevation should not go above the level of the mandible, as there is risk to damage the lower branches of the facial nerve. Inferiorly, the skin flaps should be elevated to the level of the clavicles, although this plane can be extended into the anterior chest for reconstructive purposes, such as a cervicothoracic flap. This maneuver completes the skin flap elevation over the anterior triangle ( Fig. 7.6).
Fig. 7.6 Completed elevation of the skin flaps over the anterior triangle. EJV, elevated jugular vein; GAN, greater auricular nerve; SCM, sternocleidomastoid muscle; SMG, submandibular gland.
Posteriorly, the platysma extends roughly to the level of the external jugular vein or slightly beyond it, so elevating the skin flaps beyond this boundary is significantly more challenging. The external jugular vein can also be used as a landmark; since the SLDCF invests the SCM and the vein, the plane of dissection is immediately superficial to the vessel. The same is true for other veins of the superficial plexus, such as the anterior jugular veins. On a technical point, if the external jugular vein is injured to the point that requires ligation, the stumps must be maintained in the same anatomical plane (i.e., attached to the SCM) and not elevated with the skin flaps, a common mistake among neophyte surgeons.
Similarly, posterior to the platysma, cranial sensory nerves (greater auricular, lesser occipital, and transverse cervical) traverse over the SCM and are invested in the SLDCF, so they are reliable points of reference as the surgical plane is immediately superficial to them. This same surgical plane must be maintained as flap elevation proceeds over the posterior triangle. This is significantly more challenging in this area as there no consistent points of reference, so it is recommended to frequently compare the skin flap thickness to the anterior flaps. In terms of elevation of skin flaps, in the posterior triangle there is also risk of injury of the SAN. The nerve is exposed to injury along its entire course in this area as it has a very superficial location and it is only protected by a thin fascial layer. Flap elevation with electrocautery may lead to abrupt nerve stimulation (especially in the coagulation setting), causing a violent trapezius contraction that brings the tissue toward the instrument, posing an exceptional risk. Furthermore, this may happen even if the patient is fully paralyzed. As such, it is recommended that this portion is performed with scalpel or with the electrocautery in a low intensity, pure cut setting.
To fully expose the contents of level V, it is often necessary to drop an accessory limb from the main incision; ideally, this limb must merge with the main incision at a 90-degree angle to minimize the risk of tip necrosis. When planning a trifurcation incision, acute angles (< 45 degrees) should always be avoided. Most importantly, the point of trifurcation must be placed over the SCM when possible. This is the most likely area for wound complications, so placing the trifurcation point over a critical structure, such as the carotid or the brachial plexus, can lead to significant morbidity in this event. If there is significant risk of carotid exposure (such as history of radiation or radical neck dissection), consider an elective interposition of myofascial pectoralis flap to protect these critical structures. This posterior limb can usually be placed within a dominant skin crease and may be extended to the posterior midline (or beyond) when performing a posterolateral neck dissection, which also encompasses the suboccipital and postauricular lymphatic groups.
Taking these points into consideration, the posterior elevation of skin flaps proceeds until the anterior border of the trapezius muscle is widely exposed ( Fig. 7.7).
7.2.4 Preservation of the Marginal Mandibular Branch of the Facial Nerve
A thorough understanding of the anatomy of the facial nerve and its relationships with neck structures is critical to preserve the function of its marginal mandibular branch (marginal nerve) and consequently perioral musculature. The marginal nerve is the fourth most inferior branch of the facial nerve, and after it has originated from its lower division, it exits the anteroinferior portion of the parotid near the angle of the mandible and remains deep to the investing cervical fascia. Near the mandibular mid-body, the nerve swings upward and perforates the DCF near the mandibular border to continue within the fibroareolar tissue between the deep fascia and platysma. It continues anteriorly at this plane to innervate the depressor labii inferioris, depressor anguli oris, and mentalis muscles ( Fig. 7.8). The effects of each of these muscles are as follows:
• Depressor labii inferioris draws the lower lip downward (depresses) and laterally (irony).
• Depressor anguli oris depresses the oral commissure (frown).
• Mentalis raises and protrudes lower lip and wrinkles skin of chin (pout).
On the other hand, the cervical branch of the facial nerve further divides into an upper and a lower branch ( Fig. 7.9); these will innervate the platysma muscle and may have an effect on preserving lip symmetry. Distally, the marginal nerve can anastomose with other branches of the facial nerve (most commonly buccal) to create a plexus that will innervate the perioral musculature. The location of the nerve in relationship to the lower border of the mandible is variable; furthermore, this relationship may change depending on the segment of the nerve analyzed (anterior vs. posterior to the facial artery), as shown in Table 7.3. This anatomical variability was also demonstrated by Al-Qahtani et al,7 who showed that the nerve has a sinuous pattern in three-fourths of the cases, and in 15% it loops posteriorly beyond its point of origin ( Fig. 7.10). The nerve almost invariably (90%) courses as a single branch as it exits the parotid and throughout its course. However, distally, as it approaches the effector musculature, it expresses a branching pattern in more than 80% of the cases.8 From the surgical standpoint, there are two ways to spare the nerve function:
Fig. 7.7 Completed neck elevation over the posterior neck triangle. DG, digastric muscle; GAN, greater auricular nerve; SCM, sternocleidomastoid muscle; TRP, trapezius muscle; XII, hypoglossal nerve.
Fig. 7.9 Illustration showing the anatomical relationship of the marginal and cervical branches of the facial nerve.
• The Hayes Martin maneuver, which entails the ligation of the facial vein two fingerbreadths below the lower border of the mandible, and subsequent cephalad retraction of the superior stump ( Fig. 7.11). Since the anatomy is constant in the sense that the marginal nerve is always superficial to the facial vessels at the mandibular edge, retracting the stump assures that the nerve will be contained within these fasciae. This is a time-tested approach that works well for most cases, but it has some limitations such as in cases where there is direct disease extension to the submental triangle, or significant nodal disease level Ib or in facial lymph nodes.
• Exploration of the marginal nerve. In this approach, the fascia overlying the submandibular gland is incised horizontally, about 2 cm below the midpoint of the mandibular body, which closely correlates with the location of the facial artery notch. Meticulous subfascial dissection is performed in a caudal-to-cephalad direction while maintaining a bloodless surgical field and minimizing the use of electrocautery. The facial vessels—and more specifically the facial artery—may be used as landmarks to identify the nerve, which will always be in a plane immediately superficial to the vessels and appear as a structure parallel to the mandible ( Fig. 7.12). Once the nerve is identified, antegrade and retrograde dissections are performed until it is released from underlying attachments, and it can safely be reflected cephalad. Posteriorly, this dissection may course through the tail of the parotid gland, which in this case is reflected inferiorly and kept in continuity with the contents of level II. The level of the dissection must be continued until the posterosuperior boundaries of level IB (the lower border of the mandible and the stylohyoid muscles) are exposed. This approach is particularly useful if the anatomy is distorted or if there are facial lymphadenopathies.