Lymphatic pathways of the head and neck were complex. In the scalp region, a rich avalvular lymph capillary plexus was located, originating from both the dermis and galea, which converged to form precollecting lymph vessels and then drained to the collecting lymph vessels that distributed densely in the subcutaneous tissue. The double structure of the lymph capillaries is a characteristic that may relate to the strong immunity present in the scalp. In the facial and cervical regions, lymph capillaries were found, arising from a single layer of the dermis, linked to precollecting lymph vessels and then drained to the collecting lymph vessels that distributed sparsely in the subcutaneous tissue. The lymphatic drainage patterns of the head and neck were different from person to person and even asymmetrical on each side of the same body (Fig. 4.1). Due to the unpredictable drainage pattern in individuals, it may confuse clinicians during the sentinel lymph node biopsy for treating head and neck cancer patients.

The term “sentinel node” was first described by Braithwaite (1923). Gould et al. (1960) suggested that during routine excision of the “sentinel node”, a frozen section assay should be completed before deciding whether or not to perform a radical neck dissection in parotid tumour patients. Morton et al. (1992) introduced an invasive method of intraoperative lymphatic mapping for treating early stage melanoma, which was the first time the concept to localize the sentinel nodes in skin cancer patients was employed. Alex and Kraig (1993) introduced a non-invasive technique by using a handheld gamma probe to localize radiolabelled sentinel lymph nodes for patients with melanoma in the head and neck. Currently techniques of lymphoscintigraphy and sentinel lymph node biopsy have become routine procedures to identify cancer spread and determine the need for lymph node ablation, especially with melanoma and breast cancer surgery.

After analysing nearly 1000 cases with melanomas in the head and neck, Uren et al. (2004) found that results of lymphoscintigraphy were often discordant with clinical predictions based on our “classic” knowledge of the lymphatic system. They found a false negative rate of sentinel node biopsy to be as high as 30%. This figure indicated that the surgeon would fail to remove nodes potentially containing metastatic disease in one of three patients.

Although this book demonstrates static anatomical images rather than in the dynamic physiological state, it updates the anatomical knowledge of the lymphatic system and may help explain the unexpected findings of lymphoscintigraphy (Pan et al. 2008a, b). The following points should always be of concern to surgeons when performing lymphoscintigraphy and were emphasized by Thompson et al. (1999, 2004) and Uren et al. (1999):

Besides a site where a lymphaticovenous shunt was found in the superficial tissue of the left head and neck in Chap. 2. The efferent lymph vessels of the superficial occipital lymph nodes formed a lymphatic network from which two small lymphatic vessels emerged to drain to a superficial occipital vein in the subcutaneous layer (Fig. 2.​122). This confirmed the clinical findings of Wallace et al. (1964). It should be noted that anastomosis may provide a systemic route for metastatic disease.

The incidence of postoperative prolonged oedema after rhytidectomy procedures is very low, but remains a frustrating complication for the surgeons and patients alike. After Baker et al. (1977) reviewed 1500 cases, they found five cases where the patient suffered from prolonged oedema after rhytidectomy. The correlation between lymphatic drainage and prolonged oedema was not mentioned. Guy et al. (1977) stated that the postoperative persistent oedema associated with rhytidectomy was rare and presumably related to lymph stasis. They did not provide any further discussion due to inadequate details of the lymphatic anatomy in this region. The anatomical details of the lymphatic drainage patterns in the superficial tissue of the head, face and neck have been provided in Chap. 5 of the book. Drainage patterns were different between individuals. Therefore, occasionally lymph vessels gathering in the preauricular area might be cut off by a crossing incision of rhytidectomy and result in the frustrating complication of oedema (Fig. 4.4). Besides, if over-dissecting were performed in the subcutaneous tissue of the facial region, extensive lymph vessels might be damaged and lead to oedema (Figs. 4.1, 4.4 and 4.5). This information may help surgeons avoid damaging lymphatic vessels in the region for preventing the lymphoedema during rhytidectomy.

The avalvular lymph capillary plexus arose in the upper and lower eyelids where lymph returned via the inner canthus, outer canthus and inferior eyelid lymph vessels (Fig. 4.6 left). If one of those vessels was destroyed or blocked, the others could still perform the function (Fig. 4.6 right). Eyelid oedema is one of the postoperative complications after blepharoplasty. It was believed that if over-dissecting were performed in the subcutaneous tissue of the outer canthal area during surgery, a majority of lymph vessels could be damaged which leads to eyelid oedema (Klapper and Patrinely 2007) (Fig. 4.7). Common oedema might be greatly subsided as the lymph might return via the inner canthus or inferior eyelid lymph vessels (Fig. 4.6 right).

Chronic lymphoedema of the eyelid is an uncommon condition presenting in many cases as a chronic form related to acne rosacea, irradiation and ocular surgery (Chalasani and McNab 2010). However acute eyelid lymphoedema is even rarer. Aveta et al. (2011) reported a case of acute eyelid lymphoedema after a major reconstruction of the medial canthus treating a recurrent basal cell carcinoma in the upper third of the nasal dorsum and the left medial canthus. Extended resection including a wide skin excision (the upper two-thirds of the nasal dorsum, medial third of both eyelids, the glabella and the head of the left eyebrow) was performed. Reconstruction was achieved by using a right forehead flap and a rotational cheek flap with lateral cantholysis and tarsoconjunctival sliding. After 10 days, a severe acute lymphoedema occurred in the left upper eyelid but not in the lower eyelid. It was likely, according to the description given by Pan et al. (2011a and b), that the medial lymphatic pathway was destroyed by the radical excision and the lateral path was interrupted as a consequence of the rotational cheek flap, while the inferior eyelid remained unaffected because of the inferior eyelid lymph vessel (Fig. 4.8).

There are several published articles regarding the lymphatic drainage of the ear, but most of them, except Sappey’s work (1874) (Fig. 4.9), provide simple sketches rather than actual lymphatic pathways (Figs. 4.10 and 4.11).

Results from Chaps. 2 and 3 of this book have revealed some differences, as shown through photographs and radiographs of actual lymphatic pathways (Figs. 2.​55, 2.​56, 3.​31, 3.​32, 3.​33, and 3.​34). Lymphatic vessels of the auricle were sparse when compared with the rich lymphatic vessels of the auricle seen in Sappey’s diagram (Fig. 4.9). This was probably due to (1) the different ages of the cadavers, (2) Sappey used mercury for the injection that could perfuse to the tiny lymphatic capillaries and (3) his drawings might be based on multiple cadaveric studies.

Regarding lymphatic pathways in different regions of the auricle, the conchal branches presented by Sappey correspond well to the anterior branch identified in this book, which both drained to preauricular lymph nodes. Lymphatic vessels of the lobule (inferior) identified by Sappey and in this book also concur well with both draining into infra-auricular lymph nodes (Sappey referred to these as parotid lymph nodes). However, differences were noted in the lymphatic vessels of the superior and middle regions (referred to as helical and antihelical branches by Sappey). In Chap. 4, these vessels are identified to drain to the infra-auricular lymph nodes (Figs. 3.​31, 3.​32, 3.​33, and 3.​34) that are consistent with the description by Haagensen et al. (1972) (Fig. 4.8), while Sappey found that these vessels drained to the retroauricular or mastoid lymph nodes (Fig. 4.9).

Based on Rouvière’s (1938) anatomical study, the lymphatic drainage area of the auricle can be divided into three principal territories (Fig. 4.10). He described lymphatic vessels in the anteromedial part of the anterior aspect of the auricle draining into the preauricular lymph node, the vessels of the lobule draining into the infra-auricular (parotid) lymph nodes. He stated that the lymph from the superiolateral part of the auricle could drain into either infra-auricular (parotid) or retroauricular lymph nodes. Haagensen et al. (1972) also divided the lymphatic drainage of the auricle into three regions based on their clinical investigation (Fig. 4.11), but they stated that lymph from the superiolateral part of the auricle would mostly drain into the infra-auricular lymph nodes. Based on lymphoscintigraphic studies in vivo, Aydin et al. (2005) divided the lymphatic drainage area of the auricle into two possible territories. The lobule portion was excluded due to a design blemish with the technique, where the target site (infra-auricular node) was always overlapped by the injection site in the lobule.

In Chap. 3 of this book, four lymphatic drainage territories are described based on four lymphatic vessels in the auricle (Figs. 3.​31, 3.​32, 3.​33, and 3.​34). It also shows that lymphatic vessels from the upper two-thirds of the anterolateral aspect and upper two-thirds of the posterior aspect drain to the infra-auricular lymph nodes. Sometimes these vessels diverge into two, one drains to the nearest node and one bypassing it to continue its course.

Primary malignant melanoma of the external ear is rare, but the prognosis is poor (Ward and Acquarelli 1967; Byers et al. 1980). Previous studies have reported a 5-year survival rate of >74% after the treatment, while a downtrend of the rate was noticed in patients after 5 years (Cole et al. 1992; Davidsson et al. 1993; Ravin et al. 2006). It was mentioned that the prognosis directly correlates to the thickness, stage and topography of melanoma (Mondin et al. 2005). While others hold a different view, it was discussed whether the location of melanoma on the auricle would significantly affect survival rates (Byers et al. 1980; Ravin et al. 2006). Variable relationship between the primary site of the auricular melanoma and the sentinel node has been revealed through lymphoscintigraphic studies that used mapping the sentinel lymph node in the treatment (Morton et al. 1992; Uren et al. 1999; Thompson et al. 2004). Thompson et al. (1999) reported a case that the lymphatic drainage from the tragus passed directly to a submental node but also to the ipsilateral upper cervical chain. Cole et al. (2003) stated the proportional location of auricular melanomas as the helix (47%), lobule (21%), ear-scalp junction (10.5%), posterior ear (10.5%), concha (5%) and tragus (5%). They found 24 sentinel lymph nodes in 19 patients. The sentinel lymph nodes were summarized as level II cervical lymph nodes (n = 10), mastoid (n = 1), parotid (n = 6), lower jugular chain (n = 2), supraclavicular (n = 1) and nonspecified neck (n = 4). It was concluded that the lymphatic drainage of the auricle was variable and unpredictable.

Shown in Fig. 4.12, the lymphatic vessels from the auricular apex drain to either the infra-auricular or sternocleidomastoid lymph nodes. While vessels from the lateral middle part of the auricle drain to infra-auricular lymph node or bypass the node and may drain into a distant node, which explains some anomalies seen in lymphoscintigraphy and update the knowledge of the lymphatic anatomy in the auricle for performing sentinel node biopsy in the treatment of melanoma.

Knowledge regarding the lymphatic anatomy of the nasal fossae, nasopharynx, soft palate and oropharynx from previous studies are still vague.

Sappey’s (1874) original diagrams have shown an abundant lymphatic network on the wall of the nasal fossae and the nasopharynx (Figs. 4.13 and 4.14). The radiographic and photographic images in previous chapters of this book have presented the lymphatic pathways of this region in detail, from the lymphatic capillaries to the precollecting vessels and from collecting vessels to lymph nodes (Figs. 2.​57, 2.​58, 2.​59, 2.​60, 2.​61, 2.​62, 2.​75, 2.​76, 2.​100, 3.​36, 3.​37, 3.​38, 3.​39, 3.​40, 3.​41, 3.​42, 3.​43, 3.​44, 3.​45, and 3.​46).

Based on Sappey’s work, Delamère and Cunéo (1913) also described the lymphatic drainage in this area. They reported that the collecting trunks network of the nasal fossae formed anterior and posterior groups. According to their description, it is assumed that their anterior trunks were equivalent to the precollecting lymph vessels arising from three turbinates that are presented in Figs. 2.​57, 2.​58, 2.​100, 3.​36, 3.​38, 3.​39, 3.​40, and 3.​41. The basis for this assumption is (1) they are situated in the mucous membrane; (2) there are no valves between these vessels and lymphatic capillaries where the injectant can flow back to; and (3) they are connected to both lymphatic capillaries in the mucosa and collecting lymph vessels in the paranasopharyngeal fat tissue. The posterior trunks described by Delamere and Cunéo were that collecting lymph vessels were actually situated in the paranasopharyngeal fat tissue.

The retropharyngeal lymph nodes were first discovered by Rouvière (1938), also named as Rouvière nodes, and discussed in previous studies (Delamère et al. 1913; Ballantyne et al. 1964; Watanabe et al. 1985; Batsakis et al. 1989; Hasegawa et al. 1994; Okumura et al. 1998; Saito et al. 2002; Földi et al. 2003; Standring 2005). However, in these reports the anatomical lymphatic pathways in this area were not seen. Images in Chap. 5 have shown the entire lymphatic pathway, from lymphatic capillaries in the nasal fossa to the retropharyngeal and lateral pharyngeal lymph nodes in the parapharyngeal space. There are two major groups of collecting lymph vessels. One is running along the outside of the lateral pharyngeal wall and the other one more posteriorly. The collecting lymph vessel on the lateral side of the wall divides several times and then crosses over the external carotid artery medially and the internal carotid artery laterally, before finally reaching multiple first-tier nodes: the lateral pharyngeal node, the subdigastric node and the third and fifth nodes of the retropharyngeal group (Figs. 3.​40 and 3.​41). Therefore, the third- and fifth-tier nodes of the retropharyngeal group could be multiple first-tier nodes for the lateral pharyngeal lymphatic vessels. These findings may guide clinical management of cancer treatment in this region.

Primary malignant melanoma in the mucosa of the nasal cavity is rare, but the prognosis is poor, despite the advances in radical neck surgery, postoperative radiotherapy and chemotherapy (Bhattacharyya 2002; Martin et al. 2004; Huang et al. 2007). Local recurrence may have an anatomical base. The abundance of avalvular lymph capillaries on the wall of the nasal fossa and the nasopharynx reaches multiple first-tier lymph nodes by passing three precollecting lymph vessels and two groups of collecting vessels. Meanwhile, there are numerous bypassing connections between nodes and pathways permitting cells to take alternative pathways particularly if the main path is blocked. In Fig. 4.15, if an injection is inserted in any of the three nasal turbinates, the lymphatic drainage could reach multiple lymph nodes in the parapharyngeal space.

According to neck dissection classification (Robbins et al. 2002), the lateral pharyngeal lymph nodes were included in the level II group, but the retropharyngeal nodes were excluded. This result suggests that for the treatment of patients with cancer in the nasal fossae or nasopharynx, multiple lymph nodes should be included.

The lymphatic anatomy of the tongue has been widely studied (Sappey 1874; Jamieson and Dobson 1920; Rouvière 1938; Haagensen et al. 1972; Moore 1992; Standring 2005). Most reports, except Sappey’s original drawings, which are based on results using a mercury injection, have only provided text and sketches rather than accurate radiographs and photographs.

The lymphatic distribution and pathways of the tongue have been presented as radiographs and photographs in Chaps. 2 and 3 of the book (Figs. 2.​20, 2.​74, 3.​47, 3.​48, 3.​49, 3.​50, 3.​51, and 3.​52). The difference between results of sparse lingual vessels in the book and drawings with numerous lymphatic vessels seen in Sappey’s work (Fig. 4.16) is probably due to (1) the different ages of cadavers, (2) mercury was used for injections which are able to fill more tiny lymph vessels and (3) Sappey’s drawings might be based on multiple results from different cadavers. In his book, a rich lymphatic network was shown on the dorsal tongue and the lymph-collecting vessels on the inferior aspect of the tongue mostly drained to the upper internal jugular lymph nodes. He did not include the submandibular lymph nodes in his diagrams.