Oral malignant melanoma and OSCC are the most common oral tumors of dogs, and both have a high rate of metastasis to the regional lymph nodes.1,2 Evaluation of regional lymph nodes is critical for staging of oral cancers in dogs because results can provide prognostic information. In a study2 of dogs with OMM, those with metastatic disease at the time of diagnosis (n = 9, of which 8 had metastasis to lymph nodes only) had significantly shorter survival times, compared with dogs that did not have metastatic disease at the time of diagnosis. In human patients with intermediate-thickness melanomas and nodal metastasis, lymphadenectomy provides a therapeutic benefit by decreasing the risk of nodal recurrence, distant metastasis, and death.3
Multiple lymphocentra exist within the canine head, including the mandibular, retropharyngeal (medial and, less commonly, lateral), and parotid lymphocentra.4,5 Buccal lymph nodes have also been reported and were identified during anatomic dissection in 13 of 150 (9%) dogs in 1 study.6 The MLNs are the most accessible lymph nodes of the canine head and are commonly considered the first lymph nodes to which tumors of the head are likely to metastasize. However, some dogs develop metastases in the MRLNs or parotid lymph nodes without metastases to the MLNs.4,5 Dogs typically have 2 to 3 MLNs but can have up to 5 MLNs on each side, whereas most have only 1 MRLN.5,7 The presence of multiple MLNs makes cancer staging challenging, particularly when not all MLNs are evaluated.
Aspiration of MLNs for cytologic evaluation is routinely performed in patients with tumors of the head. In dogs, although aspiration of the MRLNs is possible, it typically requires imaging guidance. Palpation of parotid lymph nodes can be challenging, and these nodes are not routinely aspirated in clinical practice.5 Cytologic evaluation of lymph nodes can be unreliable for detection of metastatic disease, especially for melanocytic neoplasms owing to the lack of defined malignancy criteria.8,9 Moreover, cytologic evaluation is generally limited to a small portion of the lymph node, and metastatic cells can be missed, particularly in nodes with micrometastases.8,9 In dogs with OMM, pigmentary incontinence leading to the presence of melanophages can further complicate cytologic evaluation of lymph nodes for metastasis.8 Histologic evaluation of lymph nodes allows for assessment of more cells than cytologic evaluation and, when used in conjunction with immunohistochemical stains, can improve diagnostic sensitivity for detection of metastatic disease.
A combined surgical approach to the ipsilateral mandibular, MRLN, and parotid lymph nodes has been described for dogs,5 as has a surgical approach for bilateral mandibular and medial retropharyngeal lymphadenectomy.10 Many surgeons routinely remove the MRLNs from dogs in which the MLNs are being extirpated because the surgical approach is straightforward and associated with minimal morbidity.4,10 However, information regarding the frequency of metastasis to the MRLNs in the absence of metastasis to the MLNs is somewhat contradictory, with the reported prevalence of dogs with metastasis to the MRLNs or parotid lymph nodes without concurrent metastasis to the MLNs ranging from 0% to 46%.4,11
The true prevalence of metastasis to the MRLNs of dogs with OMM or OSCC is unknown, and cytologic or histologic assessment of MLNs alone might not definitively rule out metastasis to the MRLNs in such dogs. The objective of the study reported here was to retrospectively assess the benefits of evaluating both the MLNs and MRLNs for the presence of metastatic disease during tumor staging for dogs with OMM and OSCC.
Materials and Methods
Animals
Study dogs were recruited from VSSO member institutions. All VSSO members were informed of the study by an email and were provided access to a standardized data collection form. Ten VSSO members from 8 institutions chose to participate in the study, and the medical record databases of each of those institutions were searched to identify dogs that underwent unilateral or bilateral extirpation of MLNs and MRLNs during the same surgical procedure between January 2004 and April 2016. Only dogs with OMM or OSCC were considered for study inclusion. Dogs were excluded from the study if only 1 lymphocentrum (MLN or MRLN) was removed and evaluated or if histologic results for the primary tumor were unavailable. Dogs from which the MLNs and MRLNs were not extirpated at the same time were also excluded because the intervening period could have affected the histologic findings for those nodes.
Data collection
For each dog evaluated in the study, information extracted from the medical record included signalment, primary mass location and size (ie, largest dimension), diagnostic imaging results at the time of initial examination, histologic results for the primary tumor and all lymph nodes evaluated, and whether distant metastasis developed. The dates of initial examination for the primary tumor, lymphadenectomy, and surgical removal of the primary tumor were also recorded. Reasons for delays between initial examination and surgical removal of the primary tumor or lymphadenectomy were not recorded. The histologic reports for the primary tumor and all extirpated lymph nodes generated by the attending pathologists were used for analysis. The specific criteria used for diagnosis of metastatic disease and the use of immunohistochemical methods varied among those pathologists; however, all were board-certified veterinary pathologists. All diagnostic images were evaluated by board-certified veterinary radiologists.
Statistical analysis
Descriptive statistics were generated. The data distribution for each continuous variable assessed was evaluated for normality by means of a Shapiro-Wilk test. Results were reported as the mean (SD) for normally distributed continuous variables and the median (range) for continuous variables that were not normally distributed. Comparisons of the overall rates of lymph node metastasis and distant metastasis between dogs with OMM and dogs with OSCC were performed with χ2 or Fisher exact tests. All analyses were performed with commercially available software,a,b and values of P < 0.05 were considered significant.
Results
Study population
The study population included 27 dogs with OMM and 21 dogs with OSCC. The median number of dogs recruited from each of the 8 participating institutions was 5 (range, 1 to 15).
Dogs with OMM
The 27 dogs with OMM included 10 castrated males, 7 sexually intact males, 9 spayed females, and 1 sexually intact female and had a mean (SD) age of 10.9 (2.8) years. There were 4 mixed-breed dogs, 3 Cocker Spaniels, 3 Labrador Retrievers, 3 Yorkshire Terriers, 2 Golden Retrievers, and 1 each of 12 other breeds.
The primary tumor was located on the maxilla (n = 8), lip (8), mandible (7), hard palate (2), or tongue (2). Tumor size was reported for 21 dogs, and the median tumor size was 3 cm (range, 0.8 to 6.0 cm). Three dogs had a scar observed during initial examination resulting from a previous incomplete excision of the primary tumor.
Twenty-six dogs underwent thoracic imaging as part of the initial diagnostic workup for OMM, and 1 dog had evidence of metastatic disease in the lungs. Fourteen dogs underwent diagnostic imaging (ultrasonography [n = 8] or CT [6]) of the abdomen, and none had evidence of metastatic disease.
All 27 dogs with OMM underwent surgery to remove the primary tumor. Twenty-three dogs underwent lymphadenectomy in conjunction with primary tumor removal, whereas the remaining 4 dogs underwent lymphadenectomy at 148, 150, 264, and 401 days after primary tumor removal. The median duration between initial examination and lymphadenectomy was 9 days (range, 0 to 401 days).
The MLN and MRLN ipsilateral to the primary tumor were extirpated in 18 dogs, and the MLN and MRLN contralateral to the primary tumor were extirpated in 1 dog owing to a palpably enlarged contralateral MLN (Table 1). Eight dogs underwent bilateral lymphadenectomy of the MLNs and MRLNs. Ten of 27 (37%) dogs had metastatic disease in at least 1 lymph node. The dog that had distant metastasis to the lungs at the time of initial examination also had metastasis to the ipsilateral MLN and MRLN. Most dogs had a single MLN and MRLN extirpated regardless of lymph node laterality in relation to the primary tumor (Table 2). One dog with lymph node metastasis had > 1 ipsilateral MLN removed; however, only 1 of those extirpated lymph nodes underwent histologic evaluation to confirm the presence of metastatic disease.
Histologic results for MLNs and MRLNs extirpated from dogs with OMM or OSCC that underwent unilateral (OMM, n = 19; OSCC, 14) or bilateral (OMM, 8; OSCC 7) lymphadenectomy during the same surgical procedure.
| Â | Unilateral lymphadenectomy | Bilateral lymphadenectomy | ||
|---|---|---|---|---|
| Lymph nodes with metastasis | OMM | OSCC | OMM | OSCC |
| None | 12 | 10 | 5 | 5 |
| Ipsilateral MLN and MRLN | 4 | 1 | 1 | 1 |
| Ipsilateral MLN | 1 | 1 | 0 | 0 |
| Ipsilateral MRLN | 1 | 1 | 0 | 1 |
| Contralateral MLN and MRLN | 1* | 1†| 0 | 0 |
| Contralateral MLN | 0 | 0 | 1 | 0 |
| Ipsilateral and contralateral MLNs and MRLNs | 0 | 0 | 1 | 0 |
Values represent the number of dogs with the given findings. Ipsilateral and contralateral are used to refer to lymph node location relative to the location of the primary tumor. All dogs were evaluated as part of a retrospective multi-institutional (n = 8) study and were initially examined for the primary tumor between January 2004 and April 2016.
Lymphadenectomy of the contralateral lymph nodes was performed because of a palpably enlarged contralateral MLN.
Lymphadenectomy of the contralateral lymph nodes was performed because of diagnosis of metastatic disease in an ipsilateral MLN.
The median follow-up time for dogs with OMM was 243 days (range, 41 to 888 days). Eleven of the 27 (41%) dogs developed distant metastasis to the lungs (n = 9), spinal column (1), skin (1), and cervical region (1), with 1 of those dogs having distant metastasis to both the lungs and skin. Five of those dogs had no evidence of metastatic disease in extirpated lymph nodes, whereas the remaining 6 dogs had metastatic disease in at least 1 extirpated lymph node.
Dogs with OSCC
The 21 dogs with OSCC included 6 castrated males, 5 sexually intact males, 9 spayed females, and 1 sexually intact female and had a mean (SD) age of 9.2 (3.2) years. There were 4 mixed-breed dogs, 3 Yorkshire Terriers, 2 German Shepherd Dogs, 2 Golden Retrievers, 2 Labrador Retrievers, and 1 each of 8 other breeds.
The primary tumor was located on the maxilla (n = 9), mandible (5), tonsil (4), and lip (3). Tumor size was reported for all 21 dogs, and the median tumor size was 3.3 cm (range, 0.8 to 6 cm). Two dogs had a scar resulting from a previous incomplete excision of the primary tumor. All 21 dogs underwent thoracic imaging as part of the initial diagnostic workup for OSCC, and none had evidence of metastatic disease. Six dogs also underwent diagnostic imaging (ultrasonography [n = 5) or CT [1]) of the abdomen, and one of the dogs that underwent abdominal ultrasonography had evidence of suspected metastasis to the liver, which was neither cytologically nor histologically confirmed.
Of the 21 dogs with OSCC, 20 underwent surgery to remove the primary tumor; the owner of the remaining dog declined surgical removal of the primary tumor but allowed sampling of the lymph nodes. Nineteen dogs underwent lymphadenectomy in conjunction with primary tumor removal, whereas 1 dog underwent lymphadenectomy 2 days after removal of the primary tumor. The median duration between initial examination and lymphadenectomy was 9 days (range, 0 to 213 days).
The ipsilateral MLN and MRLN were extirpated from 13 dogs, and the contralateral MLN and MRLN were extirpated from 1 dog (Table 1). The dog that underwent lymphadenectomy of the contralateral lymph nodes had metastatic disease previously identified in an ipsilateral MLN. Seven dogs underwent bilateral lymphadenectomy of the MLNs and MRLNs. Six of the 21 (29%) dogs had metastatic disease in at least 1 lymph node. Among the 4 dogs with tonsillar OSCC, 2 had metastasis to at least 1 lymph node. The dog with suspected metastasis to the liver had no evidence of metastatic disease in the ipsilateral MLN or MRLN. Most dogs had a single MLN and MRLN extirpated (Table 2). None of the dogs with lymph node metastasis had > 1 lymph node within a lymphocentrum extirpated and examined.
Number of lymph nodes extirpated from each lymphocentrum for the dogs of Table 1.
| Lymphocentrum | No. of lymph nodes extirpated | OMM | OSCC |
|---|---|---|---|
| Ipsilateral MLN | 1 | 20 (77) | 13 (65) |
| Â | 2 | 6 (23) | 7 (35) |
| Ipsilateral MRLN | 1 | 25 (96) | 19 (95) |
| Â | 2 | 1 (4) | 1 (5) |
| Contralateral MLN | 1 | 6 (67) | 6 (75) |
| Â | 2 | 2 (22) | 2 (25) |
| Â | 3 | 1 (11) | 0 (0) |
| Contralateral MRLN | 1 | 9 (100) | 8 (100) |
Values represent the number (%) of dogs unless otherwise indicated.
See Table 1 for remainder of key.
The median follow-up time for dogs with OSCC was 149 days (range, 1 to 3,703 days). Distant metastasis was not identified in any of the dogs; however, 1 dog developed a new OSCC on the contralateral mandible.
All dogs
Of the 48 study dogs, 3 (6%) developed complications (lymphedema [n = 2] or an abscess in the cervical region [1]) following lymphadenectomy. The proportion of dogs with OMM that developed distant metastasis was significantly (P < 0.01) greater than that for dogs with OSCC. The frequency of metastasis to regional lymph nodes did not differ significantly (P = 0.54) between dogs with OMM and those with OSCC. Sixteen (33%) dogs had metastasis to at least 1 lymph node. For the 33 dogs that underwent unilateral extirpation of the MLN and MRLN, the histologic results for the MLN were consistent with those for the MRLN in 29 dogs (ie, 22 dogs had no evidence of metastasis in either lymph node, and 7 dogs had evidence of metastasis in both lymph nodes; Table 1). One dog with OMM of the right buccal mucosa and 1 dog with OSCC of the right tonsil had metastasis to the MLN without metastasis to the MRLN. One dog with OMM of the right buccal mucosa and 1 dog with OSCC of the right buccal mucosa had metastasis to the MRLN without metastasis to the MLN. For the 15 dogs that underwent bilateral extirpation of the MLNs and MRLNs, the histologic results for the 4 nodes were in agreement for 11 dogs (ie, 10 dogs had no evidence of metastasis to the lymph nodes, whereas 1 dog with OMM of the central tongue base had evidence of metastasis in all 4 nodes). One dog with OSCC of the caudal aspect of the right maxilla and 1 dog with OMM of the rostral aspect of the left mandible had metastasis to the ipsilateral MLN and MRLN only. One dog with OSCC of the rostral aspect of the left mandible had metastasis to the ipsilateral MRLN only, and 1 dog with OMM of the caudal aspect of the right hard palate had metastasis to the contralateral MLN only.
The contralateral MLN and MRLN were evaluated in 17 dogs, of which 15 underwent bilateral lymphadenectomy of the MLNs and MRLNs; the remaining 2 dogs underwent unilateral lymphadenectomy of the contralateral MLN and MRLN. Among those 17 dogs, 4 had metastasis to at least 1 contralateral lymph node.
Among all 48 study dogs, 3 (6%) had metastasis to an MRLN without concurrent metastasis to an MLN and 10 (21%) had metastasis to both an MRLN and MLN. Among the 16 dogs with metastasis to regional lymph nodes, 13 would have had residual disease if the MRLN had not been removed concurrently with the MLN.
Discussion
In dogs, tumors of the oral cavity (particularly OMM and OSCC) are common, and the complex lymphatic pathways of the head can make tumor staging challenging. Although the MLNs are peripherally located and at least 1 MLN is generally easy to access for palpation or aspiration, other lymph nodes may receive drainage from primary tumors. Among the 48 dogs with OMM or OSCC evaluated in the present study, 16 (33%) had metastasis to at least 1 regional lymph node. Three of those dogs had metastasis to an MRLN without concurrent metastasis to an MLN, and treatment for those 3 dogs may have differed if the MRLN had not been removed and histologically evaluated in conjunction with the MLN because it would have been assumed that metastatic disease was not present. That finding may be clinically relevant because chemotherapy is recommended for dogs with metastasis to regional lymph nodes.12 In fact, the MRLN was affected in 13 of the 16 (81%) dogs with regional lymph node metastasis, and all 13 of those dogs would have had residual disease had the MRLN not been extirpated in conjunction with the MLN. That finding was consistent with results of another study11 in which the MRLN was affected in 11 of 14 (79%) dogs with malignancies of the head and regional lymph node metastasis.
Many dogs have > 1 MLN on each side of the head, with some dogs having up to 5 lymph nodes in that region, whereas most dogs typically have only 1 MRLN on either side.5,7,13 Typically, only 1 MLN was extirpated from the dogs of the present study, although 2 to 3 MLNs were removed from some dogs. Metastatic disease may have been present in MLNs that were not removed; thus, it was possible the prevalence of metastasis to the MLN was underestimated for the dogs of the present study. Only 1 dog with lymph node metastasis had > 1 MLN removed from the same lymphocentrum; however, only 1 of the excised lymph nodes underwent histologic evaluation to confirm metastatic disease. Further studies are warranted to evaluate the frequency of metastasis to multiple lymph nodes within a lymphocentrum. Although the MRLN is located more deeply than the MLNs, it is larger and more anatomically consistent, which makes lymphadenectomy of that node straightforward.5 For the dogs of the present study, 1 MRLN was most commonly removed from 1 side; however, 2 MRLNs were extirpated from the same side in some dogs, albeit rarely. Ideally, all lymph nodes in a lymphocentrum should be removed during lymphadenectomy so they can be histologically evaluated for metastases, to allow for complete tumor staging, and to minimize the risk for residual disease. Results of the present study suggested that lymphadenectomy of both the MLN and MRLN lymphocentra should be considered in dogs with suspected metastatic OMM or OSCC. Computed tomography can be used to help determine whether multiple lymph nodes are present in a lymphocentrum, and an attempt should be made to remove all lymph nodes identified within a lymphocentrum that is receiving drainage from a primary tumor.13 On the basis of results of the present study and a previous study,11 bilateral lymphadenectomy should be considered in dogs with lateralized tumors of the head. In the present study, metastatic disease was detected in the contralateral lymph nodes for 4 of the 17 (24%) dogs that underwent lymphadenectomy of the contralateral lymph nodes. In the other study,11 metastatic disease was detected in the contralateral lymph nodes for 8 of 13 (62%) dogs with malignancies of the head. Collectively, the results of these 2 studies indicated that metastasis to contralateral lymph nodes is common in dogs with oral malignancies, and lymph nodes contralateral to the primary tumor should at a minimum be diagnostically evaluated for evidence of metastases (eg, aspirated for cytologic evaluation) if not removed. Potential complications associated with lymphadenectomy of the MLNs and MRLNs include seroma development and transient lymphedema; however, the incidence of those complications is low, and the lymphadenectomy procedure is straightforward regardless of the approach.5,10,11,14 Only 3 of the 48 dogs evaluated in the present study developed lymphadenectomy-associated complications, but given the retrospective nature of the study, it is possible that postoperative complications were underreported. Also, it can be difficult to differentiate whether lymphedema was the result of lymphadenectomy or primary tumor removal. Nevertheless, the benefits associated with the information gained from lymphadenectomy frequently outweigh the potential risks, and minimal additional dissection is required to remove the MRLNs in dogs that are undergoing lymphadenectomy of the MLNs.
Theoretically, biopsy-based management of sentinel lymph nodes should allow for more accurate tumor staging and lymphadenectomy in patients with OMM or OSCC. That procedure could facilitate identification of the first lymph node to which a tumor drains, and if the tumor has not metastasized to that lymph node, extirpation of additional lymph nodes could be avoided. However, if the tumor has metastasized to the sentinel lymph node, all other lymph nodes in that lymphocentrum should be extirpated. Use of this technique to manage patients with oral tumors would allow for evaluation of the parotid lymphocentrum, which is not often evaluated owing to the difficulty in identifying lymph nodes in that area.13,14 Although biopsy-based management of sentinel lymph nodes might improve the accuracy of tumor and lymph node staging, it is not routinely used in clinical practice. Therefore, on the basis of the results of the present study and a previous study,11 bilateral lymphadenectomy of the MLN and MRLN lymphocentra should be considered in dogs with OMM or OSCC and suspected lymph node metastasis.
In the present study, the prevalence of metastasis to regional lymph nodes did not differ significantly between dogs with OMM and those with OSCC. However, the prevalence of distant metastasis was significantly greater for dogs with OMM (11/27 [41%]) than for dogs with OSCC (0/21 [0%]). That finding was consistent with results of other studies in which the prevalence of distant metastasis ranged from 24% to 27% for dogs with OMM2,15 and was 13% for dogs with OSCC.1 Failure to detect a significant difference in the prevalence of regional lymph node metastasis between dogs with OMM and dogs with OSCC in the present study may have been caused by a lack of power owing to the small number of dogs evaluated in each group. It is also possible that OMM does not metastasize to regional lymph nodes prior to metastasizing to distant sites as commonly as does OSCC. In a study2 of 70 dogs with OMM, 6 dogs had regional lymph node metastasis, 14 dogs had distant metastasis to the lungs, and only 4 dogs had metastasis to both the regional lymph nodes and lungs. Unfortunately, it was unclear which lymph nodes were evaluated for the dogs of that study,2 and it is possible that lymph node metastasis was missed in some of those dogs, particularly if the MRLNs were not evaluated.
For the dogs of the present study, there was no obvious association between the presence of lymph node metastasis and distant metastasis. Although the median follow-up time was approximately 8 months for dogs with OMM and 5 months for dogs with OSCC, the evaluation procedure for metastatic disease was not standardized, and some dogs may not have undergone comprehensive evaluation for metastatic disease unless clinical signs of disease were present. Moreover, for veterinary patients, generally only a small portion of an extirpated lymph node is histologically evaluated for metastases, and metastases might have been missed if tumor cells were clustered within a node or only micrometastases were present. It is also possible that the presence or absence of lymph node metastasis may be a poor predictor of distant metastasis because metastatic patterns are subject to inter-individual variation and are seldom straightforward. Further research is necessary to determine whether there is an association between lymph node metastasis and distant metastasis in dogs with OMM and OSCC.
Another limitation of the present study was that the extirpated lymph nodes were not histologically evaluated for metastasis by the same pathologist. Because the lymph nodes underwent histologic evaluation by multiple pathologists at multiple institutions, it is likely that the preparation method and number of slides prepared for each lymph node varied among dogs as did the thoroughness with which each slide was evaluated and the interpretation of the histologic findings. It is also unlikely that the entire MLN lymphocentrum was extirpated for any dog, which limits the ability to assess the importance of removing all MLNs in any 1 patient. Furthermore, the study population was fairly small, which limited the overall power of the study.4,5,11
In the present study, the MRLN was affected in 13 of 16 dogs with regional lymph node metastasis, and all 13 of those dogs would have had residual disease if the MRLN had not been extirpated in conjunction with the MLN. Also, metastatic disease was identified in the contralateral lymph nodes for 4 of the 17 dogs that underwent lymphadenectomy of the contralateral lymph nodes, and all 4 of those dogs would have had residual disease if only the unilateral lymph nodes had been removed. On the basis of these findings, bilateral lymphadenectomy of the MLN and MRLN lymphocentra by means of a combined surgical approach as previously described5,10 is recommended for dogs with OMM or OSCC and suspected lymph node metastasis. Evaluation of only a single MLN (either ipsilaterally or bilaterally) is insufficient to definitively rule out regional lymph node metastasis in dogs with OMM or OSCC.
ABBREVIATIONS
| MLN | Mandibular lymph node |
| MRLN | Medial retropharyngeal lymph node |
| OMM | Oral malignant melanoma |
| OSCC | Oral squamous cell carcinoma |
| VSSO | Veterinary Society of Surgical Oncology |
Footnotes
JMP Pro, version 13.0.0, SAS Institute Inc, Cary, NC.
Excel 2016, Microsoft Corp, Redmond, Wash.
References
1. Fulton AJ, Nemec A, Murphy BG, et al. Risk factors associated with survival in dogs with nontonsillar oral squamous cell carcinoma 31 cases (1990–2010). J Am Vet Med Assoc 2013;243:696–702.
2. Tuohy JL, Selmic LE, Worley DR, et al. Outcome following curative-intent surgery for oral melanoma in dogs: 70 cases (1998–2011). J Am Vet Med Assoc 2014;245:1266–1273.
3. Morton DL, Thompson JF, Cochran AJ, et al. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N Engl J Med 2014;370:599–609.
4. Herring ES, Smith MM, Robertson JL. Lymph node staging of oral and maxillofacial neoplasms in 31 dogs and cats. J Vet Dent 2002;19:122–126.
5. Smith MM. Surgical approach for lymph node staging of oral and maxillofacial neoplasms in dogs. J Am Anim Hosp Assoc 1995;31:514–518.
6. Casteleyn CR, van der Steen M, Declercq J, et al. The buccal lymph node (lymphonodus buccalis) in dogs: occurrence, anatomical location, histological characteristics and clinical implications. Vet J 2008;175:379–383.
7. Bezuidenhout AJ. The lymphatic system. In: Evans HE, Miller ME, eds. Miller's anatomy of the dog. 4th ed. St Louis: Elsevier, 2013;535–562.
8. Grimes JA, Matz BM, Christopherson PW, et al. Agreement between cytology and histopathology for regional lymph node metastasis in dogs with melanocytic neoplasms. Vet Pathol 2017;54:579–587.
9. Ku CK, Kass PH, Christopher MM. Cytologic-histologic concordance in the diagnosis of neoplasia in canine and feline lymph nodes: a retrospective study of 367 cases. Vet Comp Oncol 2017;15:1206–1217.
10. Green K, Boston SE. Bilateral removal of the mandibular and medial retropharyngeal lymph nodes through a single ventral midline incision for staging of head and neck cancers in dogs: a description of surgical technique. Vet Comp Oncol 2017;15:208–214.
11. Skinner OT, Boston SE, Souza CH. Patterns of lymph node metastasis identified following bilateral mandibular and medial retropharyngeal lymphadenectomy in 31 dogs with malignancies of the head. Vet Comp Oncol 2017;15:881–889.
12. Liptak JM, Withrow SJ. Cancer of the gastrointestinal tract, section A: oral tumors. In: Withrow SJ, Vail DM, Page RL, eds. Withrow & MacEwan's small animal clinical oncology. 5th ed. St Louis: Elsevier Saunders, 2013;381–398.
13. Kneissl S, Probst A. Comparison of computed tomographic images of normal cranial and upper cervical lymph nodes with corresponding E12 plastinated-embedded sections in the dog. Vet J 2007;174:435–438.
14. Worley DR. Incorporation of sentinel lymph node mapping in dogs with mast cell tumours: 20 consecutive procedures. Vet Comp Oncol 2014;12:215–226.
15. Boston SE, Lu X, Culp WT, et al. Efficacy of systemic adjuvant therapies administered to dogs after excision of oral malignant melanomas: 151 cases (2001–2012). J Am Vet Med Assoc 2014;245:401–407.
