Abstract
Spontaneous ejection of tissues from body orifices is rare in veterinary medicine. Here we underscore the diagnostic value of tissues spontaneously ejected from the nose or mouth of 21 dogs and submitted for histologic evaluation at 3 veterinary diagnostic institutions. Cases were retrospectively searched (2000–2024) from the Athens Veterinary Diagnostic Laboratory, Tifton Veterinary Diagnostic and Investigational Laboratory, and Antech Diagnostics web-based archive systems. Affected dogs were adults (x̄ age = 9.5 y) of several breeds. There were 13 male (8 castrated, 5 intact) and 8 spayed female dogs. Clinical signs consisted mainly of sneezing (19 of 21 cases) and epistaxis (11 cases), with spontaneous ejection of red-to-brown and fleshy-or-spongy nodules from the nose (19 cases) or mouth (2). Histologically, lesions consisted of neoplasms (19 cases) or clusters of fibrinous or suppurative exudate with hemorrhage (2). Epithelial neoplasms consisted of carcinomas and adenocarcinomas (3 cases each), and squamous cell carcinoma and a presumed adenoma (1 case each). Mesenchymal neoplasms consisted of spindle-cell sarcomas (4 cases), presumed osteosarcomas (2), and a chondrosarcoma and a chondrosarcoma/chondroblastic osteosarcoma (1 case each). Round-cell neoplasms included a B-cell lymphoma, a presumed lymphoma, and a plasmacytoma (1 case each). The presence of nasal mucosa and turbinates was supportive of nasal tumor in 4 cases. Although the anatomic origin of neoplasms cannot be determined in all cases, tissues ejected from the nose or mouth can be suitable for a histologic diagnosis.
The spontaneous ejection of reactive or neoplastic tissue from the nose, mouth, vagina, or anus has been reported rarely in human and veterinary medicine.1,4,6,7 In humans, maxillary sinus antrochoanal polyps may occasionally prolapse through the sinus ostium and into the nasopharynx. 4 These polyps are attached to the mucosa by a pedicle that may undergo torsion or constriction, with subsequent ischemia and autoamputation, leading to their expulsion through the nose or mouth.1,4 Similarly, vaginal expulsion of uterine leiomyoma and anal ejection of intestinal polyps and lipomas either spontaneously or during defecation have been reported.2,6,8,9 In veterinary medicine, a single case of anal expulsion of an osteosarcoma was reported in a cat. 7 Although the origin of the neoplasm was thought to be the lower gastrointestinal tract, this could not be confirmed because no autopsy was performed. 7 Although rare, veterinary clinicians and diagnostic pathologists need to be aware that nodular lesions spontaneously ejected from body orifices may be of diagnostic utility, and that affected patients can benefit from further diagnostic work-up to determine the origin of the lesions.
Here we report the clinical and diagnostic findings of tissues that were spontaneously ejected from the nose or mouth of 21 dogs and were submitted for histologic evaluation at 3 veterinary diagnostic institutions. Between January 2000 and December 2024, cases were retrospectively searched from the Athens Veterinary Diagnostic Laboratory and the Tifton Veterinary Diagnostic and Investigational Laboratory (College of Veterinary Medicine, University of Georgia, Tifton, GA, USA), and Antech Diagnostics (College Station, TX, USA) web-based archive systems using the keywords “expulsion”, “ejection”, “sneeze”, “sneezed”, “cough”, and “coughed”. Clinical forms and surgical biopsy reports from retrieved cases were reviewed for signalment, clinical signs, pathology findings, and diagnosis. Archived H&E–stained slides were reviewed, and lesions were classified as neoplastic or non-neoplastic, without knowledge by any of the authors of the original diagnosis. When deemed necessary, select cases underwent diagnostic immunohistochemistry (IHC) for more definitive diagnosis (Table 1).
Details of immunohistochemistry (IHC) performed on sections of neoplastic lesions expelled from the noses of dogs.
F8RA = factor 8–related antigen; NA = not applicable; RTU = ready-to-use; RT = room temperature; TTF1 = thyroid transcription factor 1. IHC was performed in an automated stainer (IntelliPATH; Biocare Medical) at the University of Georgia. Chromogen in all cases was diaminobenzidine. Antigen retrieval in all cases was 15 min at 110°C.
We included 21 cases in our study (Table 2). Affected dogs were 3–15-y-old (x̄ = 9.5 y, ±3.3 y). Patients included 13 male (8 castrated, 5 intact) and 8 spayed female dogs. Most individuals (5 cases) were mixed-breed dogs. Clinical signs consisted mainly of sneezing (19 of 21 cases), epistaxis (11), nasal discharge (2), cough (2), and nose swelling (1). Nodular tissues were spontaneously ejected from the nose (19 of 21 cases) or mouth (2 of 21 cases) after sneezing or coughing, respectively. A clinical or gross description of the expelled lesions was available in 4 cases, in which tissues were reported as red-to-brown and fleshy (cases 5, 9, 10) or red and spongy (case 17) nodules.
Signalment, clinical signs, and diagnostic findings of tissues spontaneously ejected via the oronasal route in 21 dogs.
CM = castrated male; M = male; MC = mitotic count in 2.37 mm2 (10 FN22/40× fields); SF = spayed female; NA = not applicable; NR = not reported.
Histologically, neoplasia was diagnosed in 19 of 21 cases and included 8 epithelial, 8 mesenchymal, and 3 round-cell neoplasms. Neoplastic lesions consisted of well-defined nodules with or without identifiable surrounding tissues (Figs. 1, 2). Carcinomas and adenocarcinomas (Figs. 3, 4) consisted of groups of neoplastic epithelial cells with polygonal cytoplasm that formed solid areas or tubules, respectively, and were separated by a fine stroma. Nuclei were round and had finely stippled chromatin and prominent nucleoli. One carcinoma (case 3) was originally diagnosed as a probable pulmonary carcinoma because of the clinical history of a pulmonary nodule based on thoracic radiographs. IHC for thyroid transcription factor 1 (TTF1) was negative, which did not support a possible pulmonary origin. Squamous differentiation, supportive of the diagnosis of squamous cell carcinoma, was present in case 8 (Fig. 5). One neoplasm (case 1) was composed of tall epithelial cells with low atypia and no mitotic activity, forming papillary projections supported by a fine stroma. Although morphologically consistent with an adenoma, the presence of tumor infiltration could not be assessed because of the lack of normal surrounding tissues (Fig. 6). For this reason, a well-differentiated carcinoma could not be ruled out.
Spontaneous oronasal ejection of neoplastic nodules by dogs. H&E.
Spontaneous oronasal ejection of neoplastic and non-neoplastic nodules by dogs. H&E.
The spindle-cell sarcomas were composed of interweaving bundles of elongate neoplastic cells with eosinophilic cytoplasm and oval-to-elongate nuclei with dense chromatin (Fig. 7). In cases 5 and 13, neoplastic cells were admixed with highly vascular areas containing hemorrhage. For that reason, tumors were subjected to factor 8–related antigen IHC to rule out a hemangiosarcoma; no neoplastic cell immunolabeling was observed. Presumed osteosarcomas consisted of bundles of elongate neoplastic cells with eosinophilic cytoplasm and oval nuclei with dense chromatin that surrounded irregular, scalloped, eosinophilic areas of bone matrix (Fig. 8). Neoplastic cells in the chondrosarcoma (case 9) were surrounded by clear chondroid lacunae and were arranged in nodules or sheets embedded in a faintly basophilic chondroid matrix (Fig. 9). Individual neoplastic cells had a moderate amount of round, eosinophilic cytoplasm and oval nuclei with dense chromatin. Case 16 was originally diagnosed as a possible cartilage-forming tumor but was reclassified as chondrosarcoma or chondroblastic osteosarcoma. Morphologic features were similar to case 9, but the tumor also had areas with eosinophilic matrix similar to osteoid. The advanced autolysis precluded histology and diagnostic confirmation.
Round-cell neoplasms in cases 11 and 20 were morphologically consistent with lymphoma and had sheets of round cells with scant eosinophilic cytoplasm and round-to-reniform nuclei with finely stippled chromatin (Fig. 10). Case 11 was subjected to CD3 and CD79a IHC; neoplastic cells had widespread membranous immunolabeling for CD79a, confirming a B-cell lymphoma. Blocks were unavailable for case 20; therefore, diagnostic IHC was not performed. One round-cell neoplasm was composed of sheets of round cells with eosinophilic cytoplasm with a perinuclear clear zone and round nuclei with coarse chromatin, consistent with a plasmacytoma (Fig. 11). Non-neoplastic lesions were diagnosed in 2 of 21 cases and consisted of clusters of fibrinous or suppurative exudate with hemorrhage and occasional clusters of epithelial cells with no organized tissue architecture (Figs. 12, 13).
Our findings support that tissues spontaneously expelled from the nasal or oral cavity are suitable for a routine pathology diagnosis, similar to the reports from human medicine.1,4 This unique phenomenon may offer the opportunity for a histologic diagnosis without surgical intervention. The greatest limitation of histologic assessment of ejected tissues is the lack of adjacent normal tissues and challenges in determining the anatomic site of origin of lesions. Respiratory clinical signs consisting mainly of sneezing and epistaxis were reported in 19 and 11 cases, respectively, suggesting that lesions likely originated from the upper respiratory tract or, less likely, the digestive tract. Although suggestive sites of origin should be relayed to clinicians, confirmation cannot be achieved based on histology alone, unless normal tissue is present adjacent to the neoplasm. In 5 of our cases (8–11, 15), adjacent nasal mucosa and/or turbinates were observed, and a diagnosis of a nasal lesion was possible.
A radio-dense pulmonary nodule was reported in case 3, which suggested a pulmonary carcinoma. However, neoplastic cells had no nuclear immunolabeling for TTF1, and a pulmonary origin could not be confirmed.
Tissue autolysis was present in cases 3 and 16, but did not substantially interfere with the routine diagnosis. However, it is still possible that some degree of autolysis could have precluded IHC results in case 3, in which autolysis was evident, and in other cases subjected to IHC.
Most of our cases (18 of 21 cases) were malignant neoplasms. The higher frequency of malignant sinonasal neoplasms over benign neoplasms is in accordance with the data in the veterinary literature.3,10 Although most of the malignant neoplasms in our study were diagnosed based on cell morphology and mitotic count, some samples lacked surrounding normal tissues to assess the degree of tumor infiltration, which could impact the degree of malignancy.3,5,10 For this reason, although the neoplasm in case 1 had morphologic features that were consistent with an adenoma, a well-differentiated carcinoma could not be ruled out definitively.
Although the anatomic origin of most neoplasms could not be determined based on histology, tissues ejected from the nose or mouth were suitable for a diagnosis in all 21 cases in our study. In the case of neoplasms, patients would benefit from imaging, such as MRI, computed tomography scan, or radiology, in an attempt to detect the primary site of origin of the ejected tumor.7 –9 Additional clinical information was not available for any of our cases, and our findings were not confirmed by autopsy follow-up.
Footnotes
Acknowledgements
We thank Dr. Moges Woldemeskel (Tifton Veterinary Diagnostic and Investigational Laboratory, Department of Pathology, College of Veterinary Medicine, University of Georgia) for contributing cases 18 and 19.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors declared that they received no financial support for their research and/or authorship of this article.
