Retrospective study of feline tracheal mass lesions

Abstract

Objectives

This multicentre, retrospective observational study aimed to describe the clinical presentation, diagnostic methods, treatment and outcomes of cats with tracheal masses.

Methods

Eighteen cats from five academic or secondary/tertiary animal hospitals were included.

Results

The median age at diagnosis was 10.7 years (mean 9.5; range 1–17). There were nine castrated males, seven spayed females, one intact male and one intact female. Fourteen (78%) were domestic shorthairs, one (6%) was an Abyssinian, one (6%) was an American Shorthair, one (6%) was a Bengal and one (6%) was a Scottish Fold. The most common presenting complaints included chronic respiratory distress or dyspnoea (n = 14), followed by wheezing/gagging (n = 12), coughing (n = 5) and voice changes (n = 5). There was cervical tracheal involvement in 16/18, and two showed involvement of the intrathoracic trachea. The following methods were used for diagnosis: ultrasound-guided fine-needle biopsy (UG-FNB) and cytology (n = 8), bronchoscopic forceps biopsy and histopathology (n = 5), surgical resection and histopathology (n = 3), forceps biopsy via an endotracheal tube (n = 1) and histology of tissue sputtered from a cough (n = 1). Lymphoma was most often diagnosed (n = 15), followed by adenocarcinoma (n = 2) and squamous cell carcinoma (n = 1). Most lymphoma cases received chemotherapy with or without radiation according to various protocols, and partial (n = 5) or complete responses (n = 8) were noted. Kaplan–Meier survival data for cats with lymphoma revealed a median survival time of 214 days (95% confidence interval >149 days), which was significantly longer than that of other types of tumours (21 days).

Conclusions and relevance

Lymphoma was the most prevalent diagnosis, and showed a good response to chemotherapy with or without radiation therapy. Various diagnostic procedures were performed, and UG-FNB and cytology are good diagnostic procedures for cervical tracheal lesions. Owing to the variety of treatment protocols at different centres, it was impossible to compare outcomes.

Introduction

Tracheobronchial mass lesions are occasionally encountered in cats and can cause airway obstruction and life-threatening respiratory distress. Although the true prevalence has not been determined, they have been reported sporadically and are diagnosed as lymphoma,^1–7 epithelial and non-epithelial tumors^1,8–16 and inflammatory lesions.^1,17 There are several diagnostic methods, including thoracic and cervical radiographs, CT,^{3,4,7,12,14–16} bronchoscopy with brush cytology ² or forceps biopsy,^11,12 bronchoalveolar lavage fluid analysis, ⁴ cytology of materials obtained by fine-needle biopsy ³ and surgical excision with subsequent histopathology.^{12,14,16–18}

Surgical debulking, ¹² excision,^7,14,16,19 stenting of the trachea, ¹⁵ chemotherapy^{2 –4} and radiation therapy^2,3,14 have been reported for feline tracheal mass lesions. The prognosis varies among cases, and different survival times have been reported for the same histopathological diagnosis and treatment.

Therefore, we designed a multicentre retrospective observational study to look at the clinical presentations, diagnostic methods, treatments and outcomes of cats with tracheal masses.

Materials and methods

The present study was a retrospective multicentre cohort study. The primary investigator (HK) contacted five academic or secondary/tertiary animal hospitals (Japan Small Animal Cancer Center [Saitama, Japan], Nippon Veterinary and Life Science University Veterinary Medical Teaching Hospital [Tokyo, Japan], Veterinary Medical Center of the University of Tokyo [Tokyo, Japan], Hokkaido University Veterinary Teaching Hospital [Hokkaido, Japan] and DVMs Animal Medical Center Yokohama [Kanagawa, Japan]) to participate in this study and electronic medical records were collected, along with accompanying data, such as diagnostic images, imaging and histopathological reports.

Cats with tracheal mass lesions and a final diagnosis based on cytological and/or histopathological examination between 2010 and the end of June 2022 were included in the study. Data retrieved from medical records included breed, sex, age, weight, history/clinical signs, final diagnosis, diagnostic/surgical procedures and outcomes. Cases in which only a presumptive diagnosis was made using cytology or diagnostic treatment were excluded.

Survival analysis was performed using Kaplan–Meier curves and log-rank tests using R software.^20,21 Statistical significance was set at P <0.05.

Results

Eighteen cats were included (Table 1): nine neutered males, seven neutered females, one intact male and one intact female. Fourteen (78%) were domestic shorthair, one (6%) was an Abyssinian, one (6%) was an American Shorthair, one (6%) was a Bengal and one (6%) was a Scottish Fold. The median age at diagnosis was 10.7 years (mean 9.5; range 1–17). Feline leukaemia virus (FeLV) antigen testing was positive in 1/11 cats tested, and feline immunodeficiency virus (FIV) antibody was positive in 1/11 cats.

Table 1.

Demographic data and summary of 18 cats with a tracheal mass included in the study

Cat	Age (months)	Sex	Breed	Body weight (kg)	Diagnosis	Biopsy procedure	Location		Duration of clinical signs (days)	Metastasis	Treatment	Response to treatment	Survival time (days)	Outcome	PARR	IHC
1	89	MN	SF	4.92	Lymphoma	Bronchoscopic forceps biopsy	C7	Ventral	30	None	Chemotherapy	CR	85	Survived	Negative	CD20
2	210	MN	DSH	5	Lymphoma	Surgical biopsy	C2	Ventral	96	None	Chemotherapy, surgery	CR	884	Died	Negative	CD20
3	151	FI	Bengal	4.6	Lymphoma	UG-FNB	C2–4	Ventral	90	None	Chemotherapy, radiation	CR	140	Died	Negative	ND
4	177	FN	DSH	6.16	Lymphoma	UG-FNB	C3–4	Dorsal	30	Lung, suspected	Chemotherapy	CR	784	Survived	IgH	ND
5	132	FN	DSH	4.4	Lymphoma	Surgical biopsy	T3	Filled	20	None	Surgery	ND	6	Died	ND	CD20
6	88	FN	DSH	4.1	Lymphoma	Bronchoscopic forceps biopsy	T1	Filled	65	Mediastinal lymph node	Chemotherapy, radiation	PR	153	Died	ND	ND
7	125	MN	Abyssinian	4.35	Lymphoma	Bronchoscopic forceps biopsy	C5–6	Dorsal	36	None	Chemotherapy	PR	149	Died	ND	ND
8	29	FN	DSH	4.54	Lymphoma	UG-FNB	C6–7	Filled	14	None	Chemotherapy	ND	6	Survived	ND	ND
9	134	FN	DSH	3.26	Lymphoma	UG-FNB	C2	Ventral	17	None	Chemotherapy	CR	785	Died	ND	ND
10	23	MI	DSH	5.5	Lymphoma	UG-FNB	C2	Filled	11	None	Chemotherapy	PR	0	Survived	IgH	ND
11	54	MN	DSH	4.7	Lymphoma	Bronchoscopic forceps biopsy	C5–6	Ventral	21	None	Chemotherapy, radiation	CR	168	Survived	ND	CD20
12	135	MN	DSH	5	Lymphoma	UG-FNB	C3–4	Ventral and dorsal	148	Lung, suspected	Chemotherapy	CR	136	Survived	ND	ND
13	132	MN	DSH	5.18	Lymphoma	UG-FNB	C3–4	Ventral	9	None	Chemotherapy, radiation	CR	394	Survived	ND	ND
14	106	FN	DSH	2.73	Lymphoma	UG-FNB	C2	Ventral	90	Larynx	Chemotherapy, radiation	PR	214	Died	ND	ND
15	55	MN	DSH	4	Lymphoma	Bronchoscopic forceps biopsy	C2–4	Ventral	25	None	Chemotherapy, radiation	PR	93	Died	IgH	ND
16	163	FN	DSH	3.55	Adenocarcinoma	Histopathology of the sample obtained by forceps biopsy via an endotracheal tube	C2–6	Filled	21	Lung, suspected	Surgery	ND	15	Died	NA	NA
17	102	MN	ASH	7	Adenocarcinoma	Surgical biopsy	C7	Dorsal	46	Lung, suspected	None	ND	81	Survived	NA	NA
18	154	MN	DSH	3.95	SCC	Surgical biopsy	C4–5	Dorsal	ND	Lung, suspected	None	ND	21	Died	NA	NA

PARR = PCR for antigen receptor rearrangement; IHC = immunohistochemistry; MN = male neutered; SF = Scottish Fold; CR = complete remission; DSH = domestic shorthair; FI = female intact; UG-FNB = ultrasound-guided fine-needle biopsy; ND = not determined; FN = female neutered; PR = partial response; MI = male intact; NA = not available; ASH = American Shorthair; SCC = squamous cell carcinoma

The most common presenting complaints were dyspnoea or respiratory distress (n = 14; 78%), followed by wheezing/gagging (n = 12; 67%), coughing (n = 5; 28%) and voice change (n = 2; 11%). The duration of these clinical signs was described in 17 cats as ranging from 9 to 148 days (median 30).

In all cases, the lesions were evaluated using cervical or thoracic radiography performed at the participating institutions or submitted by a referral veterinarian. Ultrasonography was performed on eight cats with cervical lesions, and CT was performed in 11 cases. Sixteen of 18 cats presented with involvement of the cervical trachea, and two with involvement of the intrathoracic trachea. Of the 16 cervical lesions, eight protruded from the ventral side of the trachea, four from the dorsal side and one from both sides. The lesion filled the lumen in the remaining three cats, and ventrodorsal localisation could not be determined. One thoracic lesion protruded from the ventral side and the other filled the lumen. In the craniocaudal direction, lesions were evenly distributed from the C2 to the T3 level (Figure 1). Ultrasonography of the cervical lesions showed a hypoechoic mass adjacent to the air in the trachea that narrowed the lumen and occasionally destroyed the wall of the trachea (Figure 2). CT confirmed the radiographic and ultrasonographic findings, and better visualised the extension of the lesion (Figure 3) or pulmonary metastasis in several cases.

Figure 1.

Representative lateral thoracic radiographic image in a lymphoma case. The trachea demonstrates narrowing by a dorsal soft tissue opacity at the level of the thoracic inlet (arrows)

Figure 2.

Transverse ultrasonographic image of a cervical lymphoma case showing a hypoechoic mass adjacent to the air in the lumen of the trachea (arrows). The line of the tracheal wall appears unclear on the right side due to the invasion of the lesion

Figure 3.

Lateral thoracic radiograph and transverse enhanced CT image of a thoracic lymphoma case. Radiography shows a ventral protrusion at the thoracic inlet (a, arrows), whereas CT imaging revealed the mass to be more extensively infiltrated, surrounding the trachea dorsally to the left and ventrally (b, arrows)

The following methods were used for diagnosis: ultrasound-guided fine-needle biopsy (UG-FNB) and cytology (n = 8), bronchoscopic forceps biopsy and histopathology (n = 5), surgical resection and histopathology (n = 3), forceps biopsy via an endotracheal tube (n = 1) and histology of tissue sputtered from a cough (n = 1). Lymphoma was diagnosed in 15 cases (eight by cytology and seven by histopathology). Other diagnoses included adenocarcinoma (n = 2) and squamous cell carcinoma (SCC; n = 1). PCR for antigen receptor rearrangements (PARR) analysis was performed on six lymphoma cases: three showed IgH clonal rearrangement, whereas three were negative for both T-cell receptor and IgH genes. Immunohistochemistry showed CD20 positivity in the four cats tested. In total, 7/15 lymphoma cases were determined to be of B-cell origin, and the T-cell/B-cell phenotype in the remaining cases was unknown.

Pulmonary involvement was suspected but not confirmed on either chest radiography or CT in two lymphoma cases. Sternal lymph node involvement was cytologically confirmed in another case of lymphoma. A laryngeal mass lesion was also noted in another lymphoma case but was not histologically or cytologically examined. Two carcinoma cases and one SCC case had single or multiple pulmonary nodules suggestive of metastasis.

A lymphoma patient underwent resection of the trachea and anastomosis but died 7 days after surgery. Although necropsy was not performed, tracheal dehiscence was suspected based on the findings of inflammatory pleural effusion. Resection of the tumour was performed in another cat with lymphoma after a partial response to two administrations of L-asparaginase followed by the finding of mediastinal emphysema, suggestive of tracheal perforation. The cat was discharged uneventfully, followed by multiagent chemotherapy. All other lymphoma cases received chemotherapy with (n = 6) or without (n = 7) radiation according to various protocols, including agents such as L-asparaginase, vincristine, cyclophosphamide, doxorubicin, cytarabine, nimustine, and bleomycin; partial (n = 5) or complete (n = 8) response was noted during the protocol in all 13 cases in which the outcome was followed.

In the case of adenocarcinoma, the cat became dyspnoeic upon awakening from anaesthesia for CT and underwent tracheal resection and anastomosis. The patient was discharged on postoperative day 10 but died at home on postoperative day 11. A patient with SCC was scheduled for radiotherapy but died of respiratory distress 21 days after the diagnosis.

Kaplan–Meier survival data for cats with lymphoma revealed a median survival time of 214 days (95% confidence interval >149, which was significantly longer than that of the other types of tumours (21 days; Figure 4).

Figure 4.

Kaplan–Meier survival curve showing lymphoma (solid line) vs other tumours (dotted line). X represents censored patients. The median survival time of 214 days (95% confidence interval >149) in lymphoma cases was significantly longer than that of other tumour types (21 days)

Discussion

In this retrospective study evaluating cases of feline tracheal masses, cats of various breeds had a chronic course of signs related to airway obstruction such as wheezing, gagging and dyspnoea. Most cats were relatively old, except for three with lymphoma, which were 1, 2 and 4 years old, respectively. The lesions were distributed from beneath the larynx to above the carina and were attached to either the ventral, dorsal or almost the entire lumen of the trachea. Therefore, tracheal masses should be included in the differential diagnosis of cats with chronic respiratory signs, and the neck and thorax should be carefully examined.

Most tracheal masses in the present study (n = 15/18) were lymphomas, which agrees with a previous report in which lymphoma was the most frequent diagnosis (n = 3/7). ¹ Lymphoma cases were more likely to be included than other diagnoses in the present study, because lymphoma can be diagnosed cytologically more frequently than other types of lesions. It is possible that, in the case of lymphoma, a response to corticosteroids administered prior to diagnosis allows time for the owner and primary care veterinarian to consider referral, and more cases can be seen at a secondary care facility compared with non-lymphoma cases. For non-lymphoma tracheal masses, owners often give up further testing at the point of no response to experimental treatment. Moreover, it is possible that the percentage of lymphomas increased in this study because steroid treatment improved the condition of lymphoma cases and reduced the risk, allowing more cases to undergo diagnostic testing. Corticosteroids were administered before referral in 9/15 lymphoma cases. Although the exact proportion of lymphoma in the tracheal mass could not be determined due to the retrospective nature of the present study, it seems plausible that lymphoma is most frequently diagnosed in feline tracheal masses.

Tracheal lymphoma in this study occurred in young to older cats, mostly without FIV/FeLV infection. No breed or sex predisposition was identified. The lesion was located in both the neck and thorax but was predominant in the cervical trachea, and clinical signs included wheezing, gagging, respiratory distress, dyspnoea and coughing. This clinical picture follows those of previously reported cases. Cervical and thoracic radiography was the primary diagnostic modality for tracheal lesions, and ultrasonography, CT and bronchoscopy were the subsequent advanced imaging procedures that revealed details of the location, distribution and presence of metastasis. Lymphoma cases in the cervical trachea were diagnosed most frequently by UG-FNB and cytology, followed by endoscopic tissue biopsy. In a previous report, the diagnostic potential of ultrasonography and UG-FNB was not described in detail, and the results of this study emphasise the usefulness of this procedure for tracheal lesions. The high prevalence of lymphoma further supports the usefulness of UG-FNB, because there is a good chance of diagnosis by cytological examination. PARR analysis may support the diagnosis of lymphoma if cytology does not provide sufficient findings. Bronchoscopic forceps biopsy was performed on four cats without complications, and the samples obtained were sufficient for definitive diagnosis. The same method has also been reported,^11,12 and other sampling methods using bronchoscopy include brush cytology^2,9 and bronchoalveolar fluid analysis. ⁴ However, the number of cells obtained using brush cytology is not always sufficient for diagnosis. ⁷ Although severe complications such as massive haemorrhage and air leakage should always be considered, forceps biopsy under bronchoscopy is a valuable method for histological diagnosis of tracheal masses. No complication was noted with UG-FNB and endoscopy. However, cats with intratracheal mass lesions are at greater risk from general anaesthesia and bronchoscopic procedures. UG-FNB is preferable if the cat is not uncooperative and cervical ultrasonography can be safely performed without anaesthesia.

In this study, only 1/11 cases examined was positive for serum FIV antibody and FeLV antigen. This result was in agreement with the previous studies: none of the 13 cases examined were positive for FIV/FeLV.^{1 –5,7,18} However, FeLV antigen was immunohistochemically detected in 54% of the cases in a recent study of feline upper respiratory tract (namely nasal and nasopharyngeal) lymphoma, suggesting the role of FeLV at the genomic level. ²² The immunohistochemical investigation of FeLV antigen was not possible in this study due to the retrospective nature of the study and the small number of cases with histological investigation. Considering the similarities between nasopharyngeal and tracheal lymphomas, such as B-cell origin and diffuse large B-cell lymphoma (DLBCL) predominance, further studies on the relevance of FeLV in feline tracheal lymphoma are highly desirable.

PARR analysis or immunohistochemistry was performed in seven cases; six were of B-cell origin, whereas the immunophenotype of the remaining case could not be determined. B-cell origin has also been reported in feline tracheal lymphoma, ⁷ but the immunophenotype was not clarified in most previous reports.^1,2,4,5,7 The results of the present study suggest that B-cell lymphoma is the predominant type of tracheal lymphoma in cats, in contrast to humans. In particular, a histological diagnosis of two cases specifically noted the histological type of DLBCL. This histological type is most frequently seen in feline lymphomas of other organs, such as lymphomas of the eye, ²³ retina, ²⁴ lung ²⁵ and nose,^22,26,27 except for gastrointestinal lymphomas, ²⁸ where T-cell lymphomas predominate. Although anatomically different, the biological similarity of the same histological type in other organ lymphomas may help provide insight into the treatment and prognosis of tracheal lymphoma. However, further studies are needed to evaluate the true prevalence of each histological classification.

All lymphoma patients with a confirmed response to chemotherapy showed at least a partial response at any time point in the protocol, and 62% (n = 8/13) showed a complete response. This good response to chemotherapy has not been well described in the previous literature; however, long-term survival has been reported.^1,2,4,7 Six cats received radiation and chemotherapy, whereas nine cats received only chemotherapy, and the survival time was not different between cats with and without radiation. However, because the protocols in each case were not standardised in this study, the appropriate chemotherapy with or without a radiation protocol was not determined.

Two adenocarcinomas and one SCC were included in the study. Although a small number of cases were included, signalment, clinical signs and imaging features were similar to those of lymphoma, and all cases had suspected pulmonary metastasis. Only one patient underwent aggressive surgical treatment, but the prognosis was poor, in contrast to the lymphoma cases. A high proportion of possible lung metastases differed from the previous study,^{6,8,9,12,14,19} which could be due to the difference in population.

Two patients that underwent resection and anastomosis died shortly after surgery. This procedure has been reported sporadically in the veterinary literature with occasional severe complications, such as cardiac arrest and laryngeal paralysis.^{7,12,14,16,18,19,29} Other reported surgical procedures include bronchoscopic debulking and stenting.^12,15 Because of the small number of patients that underwent surgery, the feasibility of surgical procedures for tracheal masses in cats could not be determined. However, appropriate non-invasive diagnostic procedures should be pursued before invasive surgery for feline tracheal masses, as lymphoma is the most likely diagnosis and there are other reasonable treatment options for lymphoma, such as chemotherapy and radiation.

The limitations of this retrospective study include the bias of the case selection and the variation of the diagnostic tests, case management and treatment regimens. In particular, the lack of a complete histological report limited comprehensive histopathological subtyping in lymphoma cases that might potentially lead to the establishment of an ideal treatment protocol based on subtypes. Further study is recommended for this disease entity including prospectively collected cases with standardised diagnostics/treatment.

Conclusions

This study described a relatively large number of cats with tracheal masses. Lymphoma was the most prevalent diagnosis, with a good response to chemotherapy with or without radiation therapy. UG-FNB and cytology are reasonable diagnostic procedures for feline cervical tracheal masses.