Radiographic Characterization of Enlarged Sternal Lymph Nodes in 71 Dogs and 13 Cats
In this retrospective study, radiographically enlarged sternal lymph nodes (LNs) were evaluated in 71 dogs and 13 cats for average size, location, and most representative radiographic view. Concurrent clinical diagnoses were also noted and grouped into one of three following categories: neoplastic, inflammatory, or hematologic. There were no statistically significant differences in LN size between lateral views within each species. Enlarged sternal LNs were more cranially positioned in dogs than cats. No statistical difference was noted between right and left laterals, as to on which projection the enlarged sterna lymph nodes was seen best. Neoplastic disease (78.9%) was the most prevalent condition seen in association with LN enlargement in dogs, followed by primary infectious or inflammatory diseases (14.1%) and various hematologic conditions (7.0%). In cats, neoplasia was also most common (69.2%), followed by inflammatory diseases (30.8%). No hematologic conditions were noted in cats. The most common etiologic agent seen concurrently with enlarged sternal LNs in both dogs (33.8%) and cats (38.5%) was malignant lymphoma. The results of this study provide a clinically useful representation of the average size and location of radiographically enlarged sternal LNs for dogs and cats. The diseases represented demonstrate the wide spectrum of potential causes of sternal lymphadenopathy.
Introduction
The sternal lymph nodes (LNs) are part of the parietal group of thoracic LNs. They are located immediately dorsal to the cranial sternum and are closely associated with the internal thoracic artery and vein on each side. In dogs there is typically a single node on each side of the sternum. Less commonly, the nodes may be absent, there may be a single node on either one side or the other, or two nodes may be present on a single side.1 The sternal LNs receive drainage from the pars sternalis, pars costalis, and pars lumbalis and are responsible for draining approximately four-fifths of all material removed from the peritoneal cavity.2 Afferent lymphatics arise in the cranial abdominal wall, with tributaries also originating from the ribs, sternum, serous membranes, thymus, and adjacent musculature. The sternal LNs reportedly drain the cranial and caudal thoracic mammary glands in cases of neoplasia; however, it is uncertain whether this occurs in normal mammary glands.3 Normal sternal LNs are not apparent on radiographic images; however, they can be visualized when enlarged sufficiently to distort the contour of the mediastinum in response to disease.1,4 Sternal lymphadenopathy in dogs appears as a radiopaque mass dorsal to the second sternebra and is most accurately visualized on lateral radiographs, but may also be seen on ventrodorsal (VD) projections if severely enlarged.5
Reports classifying the appearance of radiographically visible sternal LNs are limited. A single report studying the effect of position on the radiographic appearance of various thoracic structures in dogs described the sternal LNs as being soft tissue opacities seen over the caudal second sternebra. In that report, the average sternal LN was 30 mm in length and was evaluated preferably on the right lateral view. The authors of that study postulated that fat may have contributed to the overall size of the sternal LNs radiographically; however, when assessing the entire population it was noted that in each obesity-grading group the LN was only visible 25% of the time.6
To the authors’ knowledge, there are no case studies with large study populations that have investigated the appearance of radiographic sternal lymphadenopathy in dogs and cats. The purpose of this retrospective study was to characterize radiographic sternal lymphadenopathy in the dog and cat by evaluating average LN size, location, and most reliable radiographic view for each species. The diseases seen in association with each case of sternal lymphadenopathy were also noted. This information may provide a basis for radiographic interpretation and subsequently assist with the clinical diagnosis of dogs and cats with sternal lymphadenopathy.
Materials and Methods
Digital radiographs from the University of Florida Veterinary Medical Center and Kansas State University Veterinary Medical Teaching Hospital from 2000 to 2007 were reviewed by the authors for cases of radiographically apparent (enlarged) sternal LNs in cats and dogs. Cases were included in the study if two-view thoracic radiographs, radiographic evidence of sternal lymphadenopathy, and a clinical diagnosis were available. In almost all cases the clinical diagnosis was confirmed by histology. If no histologic diagnosis was available, the medical diagnosis was used. The diseases were then grouped into three main categories: neoplastic, inflammatory, and hematologic. Cases were excluded if pleural fluid, obesity, or other lesions obscured the boundaries of the LNs.
The medical records of the selected cases were reviewed, and the patient's species, breed, and age were recorded. Lateral and VD thoracic radiographs were retrospectively evaluated by one author (R.O.) for evidence of sternal lymphadenopathy, LN size (height, length, and width), and LN location. Each lateral view was also subjectively examined to determine which projection provided the best LN margin delineation. The height and length of the LNs were measured (in mm) on the left and right lateral views (Figure 1), and the width was measured on the VD view (Figure 2). The height was measured from the dorsal aspect of the closest underlying sternebral body to the maximum dorsal point of the LN. The location of the sternal LN was measured in all available lateral projections. The method developed by the authors for measuring sternal LN location for the purposes of this study uses the maximum dorsoventral (DV) height of the LN. The point of maximum height corresponded to a location on the underlying sternebrae or intersternebral space. There are eight sternebrae and associated intersternebral spaces in the dog and cat numbered one through eight. For this study, the eight sternebrae and corresponding intersternebral spaces were numbered sequentially from zero to seven, designating the manubrium as zero and the xiphoid as seven. Each individual sternebra and intersternebral space was then fractioned into tenths, with the cranial endplate at 0.0, the midsternebra at 0.4, the caudal endplate at 0.8, and the midintersternebral space at 0.9. Using this system, a sternal LN with its point of maximum DV height overlying the caudal endplate of the third sternebra was given a location designation of 2.8, interpreting the third sternebra as “2” and the caudal endplate as “0.8” (Figure 3).
Citation: Journal of the American Animal Hospital Association 48, 3; 10.5326/JAAHA-MS-5750
Citation: Journal of the American Animal Hospital Association 48, 3; 10.5326/JAAHA-MS-5750
Citation: Journal of the American Animal Hospital Association 48, 3; 10.5326/JAAHA-MS-5750
Statistical Analysis
A one-way, two-tailed analysis of variance was performed to compare length, width, and height LN measurements. Location measurements based on the lateral projections were compared using an unpaired t test with Welch's correction. A contingency table was performed using the Fisher exact test to determine if the left or right lateral view provided the best margin conspicuity of the enlarged sternal LN. All tests were evaluated using a level of significance of P≤0.05.
Results
The study population consisted of 84 patients: 71 dogs and 13 cats. The mean ages of the dogs and cats at the time of diagnosis were 7.6 yr (range, 1–15) and 8.3 yr (range, 1–18 yr), respectively. There were 23 different dog and six different cat breeds represented.
Sternal LNs were evaluated in three radiographic views: left lateral, right lateral, and VD. The left lateral view was not available for six dogs and one cat. Results of the statistical analysis are summarized in Table 1. The average size and location of the LN is reported for each species. No comparisons were made between LN size and breed or between LN size and disease process. There were no significant differences in size or location between views within each species. In dogs, the median location of sternal LNs was dorsal to the midbody of the second sternebra (position 1.50). In cats, the median location of the enlarged sternal LNs was almost one sternebra more caudal: dorsal to the cranial aspect of the third sternebra (position 2.25). There were slightly more dogs in which the enlarged sternal LN was better defined in the right (n=36) compared with left (n=29) lateral view, but this was not statistically significant. A similar trend was seen in cats. In individual animals, there was often a large subjective difference in margin definition of the enlarged sternal LN between the two lateral projections, but there was no consistency as to which view was most likely to be the best view for sternal LN discernment.
Units are the sternebrae, numbered 0–7.
A broad spectrum of neoplastic, inflammatory, and hematologic diseases was represented in the current study in association with sternal lymphadenopathy (Table 2). In 9/13 cats (69.2%) and 56/71 dogs (78.9%), neoplasia was the most common condition. Lymphoma (information on lymphoma type/location unavailable) was the most common neoplastic disease in cats (5/9) and dogs (24/56). In dogs, this was followed by splenic hemangiosarcoma (12/56). All cases of splenic hemangiosarcoma had concurrent hemoabdomen. One case of splenic hemangiosarcoma had complete resolution of the enlarged sternal LN on radiographs repeated 2 mo after the initial radiographs were taken. Round cell tumors represented the majority of neoplastic diseases in both the dogs (33/56) and cats (6/9). This was followed by sarcomas in dogs (14/56) and carcinomas in cats (3/9).
Concurrent hemoabdomen present.
Inflammatory processes were more diagnosed in 4/13 cats (30.8%), but only 10/71 dogs (14.1%). Pulmonary eosinophilic granulomatosis and blastomycosis represented 2.8% (2/71) of inflammatory conditions in dogs each, with single cases of pulmonary lymphoid granulomatosis, discospondylitis, bacterial pericarditis, bacterial hepatitis/peritonitis, histoplasmosis, and pythiosis. In cats, inflammatory lesions were limited to feline infectious peritonitis (FIP, 3/13) and osteomyelitis/sepsis (1/13).
Hematologic conditions were diagnosed infrequently in 5/71 dogs (7.0%) and were not seen in cats. Single cases of splenic hematoma, disseminated intravascular coagulopathy, splenic torsion, immune-mediated thrombocytopenia, and rodenticide poisoning constituted the diseases seen in dogs. Each of these conditions, with the exception of splenic hematoma, had concurrent hemoabdomen present.
Discussion
The statistical evaluation of radiographic sternal lymphadenopathy in the current study provides an assessment of average LN size and location for dogs and cats. The average length of the LN was greater than that noted in previous studies; however, the average location was similar. No previous studies of either size or location was available for cats.6 In the current study, there was no statistical evidence to support an argument for which lateral view was more reliable for radiographic sternal LN evaluation. The authors did observe, however, that there was frequently a large subjective difference in margin definition of the LN between the right and left lateral views, suggesting that both views may be required to adequately assess the sternal LNs radiographically. Further studies evaluating the size of the LNs based on breed and disease etiology are needed to determine if radiographic sternal LN appearance can be used clinically to assist in determining disease etiology.
The results of the current study indicate an important relationship between specific disease conditions and the observation of enlarged sternal LNs in dogs and cats. Lymphoma was the most common neoplastic disease associated with enlarged sternal LNs, which supports the findings from previous research. One study identified radiographically apparent sternal LNs in up to 60% of dogs with multicentric lymphoma and noted that that sternal lymphadenopathy was the most common thoracic lesion observed.7 Another study documented the presence of sternal lymphadenopathy in dogs with various grades (mild, moderate, severe) of lymphosarcoma. The results of that study indicated a significant negative correlation between sternal and cranial mediastinal lymphadenomegaly and survival.8 Whether the findings of the current study reflect a higher disease prevalence of lymphoma among dogs and cats compared with other types of neoplasms (8–10% of all tumors in dogs and 33% of all neoplasms in cats) or that sternal lymphadenopathy occurs more frequently in lymphoma than in other forms of neoplasia cannot be determined based on the results of the current study.8,9 What may be elucidated from these findings is that an important connection exists between radiographically apparent sternal LNs and lymphoma. Further studies are needed, however, to investigate the prevalence of sternal lymphadenopathy in correlation with the various lymphoma types (i.e., gastrointestinal, multicentric, etc.), as this information was not available in the current study.
Splenic hemangiosarcoma was the second most common neoplasm in dogs associated with enlarged sternal LNs. In all cases, concurrent hemoabdomen was present (confirmed either by an ultrasound exam or at surgery). The direct cause of the lymphadenopathy could not be determined based on the results of the current study, but several possible explanations exist, including metastatic hemangiosarcoma and/or hemoabdomen. A previous study also took note of the occurrence of sternal lymphadenopathy in relation to hemoabdomen, especially in cases of splenic hemangiosarcoma. Nonetheless, it could not be concluded whether the lymphadenopathy was due to metastasis or the draining of blood or blood products from the abdomen.10
Other types of cancer represented in the current study, including malignant histiocytosis and feline mammary carcinoma, have also been documented in the literature in association with sternal LN enlargement.11,12 Further research exploring the relationship between these neoplasms and sternal lymphadenopathy are needed to determine the existence of a statistically significant correlation.
Inflammatory conditions were less frequently identified in connection with enlarged sternal LNs in this study. In cats, FIP represented the majority of cases. In dogs, pulmonary eosinophilic granulomatosis and pulmonary blastomycosis were the most common diagnoses. To the authors’ knowledge, the observation of enlarged sternal LNs in association with FIP in cats and pulmonary eosinophilic granulomatosis in dogs has not been previously reported in the literature. Further research is therefore needed to substantiate these observations.
The finding of sternal lymphadenopathy in connection with blastomycosis concurs with the findings of one past study documenting this observation.13 In that study, <20% of dogs had sternal and/or tracheobronchial lymphadenopathy in association with blastomycosis.13 An additional case study reported an incidence of pulmonary thromboembolism associated with blastomycosis. The resulting sternal lymphadenopathy in that case was histologically determined to be reactive in nature.14 The cause of sternal lymphadenopathy in the current study is postulated to either be reactive in nature or due to spread of the organism by lymphohematogenous routes.
Multiple other inflammatory conditions were represented in the current study. Of these, the diseases previously reported in the literature in association with sternal lymphadenopathy include histoplasmosis and pulmonary lymphoid granulomatosis.15,16 Both of those studies reported sternal lymphadenopathy as an uncommon finding.
An unexpected finding in the current study was the observation of enlarged sternal LNs in association with hematologic conditions. Reports documenting this relationship are rare. A single case study documented the presence of sternal lymphadenopathy in association with a splenic abscess.17 The majority of the hematologic diseases in the current study were seen in association with hemoabdomen. It is unknown whether the cause of sternal lymphadenopathy in those cases is due to the primary disease processes, the concurrent hemoabdomen, or both. In a review of the literature concerning resorption of blood from the peritoneal cavity, it was established that the lymphatics draining the peritoneum via the diaphragm are capable of draining blood and blood products.18 This observation, when applied to the appearance of sternal lymphadenopathy in connection to hemoabdomen, indicates that the presence of blood in the abdomen may play a significant role in the cause of LN enlargement. This relationship was also noted in a previous study, where it was postulated that the sternal LNs may function in a similar manner to hemal nodes in certain clinical circumstances.10 The results of the current study support the notion that the sternal LNs, in cases of extensive hemoabdomen, may become radiographically enlarged. Further research is needed, however, to investigate the mechanism by which diseases with concurrent hemoabdomen cause sternal lymphadenopathy.
The results of this study indicate that a radiographically apparent sternal lymphadenopathy may occur with a wide variety of concurrent disease processes. This suggests that the observation of sternal lymphadenopathy clinically should provoke an investigation into a potential primary cause. Prospective studies are needed, however, to determine the role of sternal LN enlargement as an indicator of disease. The observations pertaining to concurrent diseases are limited by regional infectious disease prevalence. The statistical results evaluating the appearance of the sternal LNs radiographically do not support the argument that either the right or left lateral view is more accurate for observing the LNs. This study was limited by the lack of an objective method for determining margin conspicuity. Further studies are needed, perhaps using contrast computed tomography, to verify size independent of surrounding soft tissue structures and mediastinal fat.
Conclusion
Based on the evaluation of cases in this study, there are many diseases and affected organs of origin that result in sternal lymphadenopathy. Disseminated and intra-abdominal neoplastic, inflammatory, and hematologic conditions are commonly associated with enlarged sternal LNs. Sternal LNs may become enlarged when there is concurrent hemoabdomen. No significant advantage for evaluating the LNs based on lateral view was determined or any differences in size or location between lateral views within species noted. The sternal LNs in cats are more caudally positioned than in dogs.
Right lateral thoracic radiograph of a 12 yr old domestic shorthair with lymphoma. Note the enlarged sternal lymph node (white arrow) dorsal to the third sternebra.
Ventrodorsal radiograph of a 10 yr old spayed female cat with lymphoma. Note the enlarged sternal lymph node (black double arrow) measuring 15.32 mm.
Identical radiographic projection and cat presented in Figure 1. The lines indicate normal intrathoracic structures and the method for indicating location of sternal lymphadenopathy. In this example the center of the enlarged lymph node is dorsal to the caudal end-plate of the third sternebra and would be accorded position “2.8.”
Contributor Notes
K. Smith's present affiliation is University Place Veterinary Hospital, University Place, WA.
R. O'Brien's present affiliation is Department of Veterinary Clinical Medicine, University of Illinois at Urbana-Champaign, Urbana, IL.
