Abstract
Inflammatory polyps are associated with significant aural or nasopharyngeal disease in cats. It has been proposed that chronic viral infection may induce the masses. Ventral bulla osteotomy (VBO) is usually recommended for definitive therapy but removal of masses from the nasopharynx or external ear canal by traction/avulsion is also used. A retrospective study of 28 cats with inflammatory polyps was conducted to correlate recurrence with mode of therapy. Tissues from 41 polyps were assayed for feline calicivirus and feline herpesvirus-1 by RT-PCR and PCR, respectively. Of the 14 cats initially treated by traction/avulsion, recurrence was detected in five of nine cats with radiographic evidence of bulla disease but none of the cats with normal bullae. Traction/avulsion is a reasonable treatment for inflammatory polyps if the bullae are radiographically normal. Failure to detect feline calicivirus and feline herpesvirus-1 suggests that tissue persistence of these viruses is not associated with the development of inflammatory polyps.
Inflammatory polyps are the most common nasopharyngeal disease of younger cats (Allen et al 1999) and the most common non-neoplastic mass of the feline ear (Rogers, 1988). Histopathologically, these masses consist of a core of loosely arranged fibrovascular tissue covered by a stratified squamous to ciliated columnar epithelial layer that is commonly ulcerated. Variable lymphoplasmacytic, lymphoid aggregates/follicles, and pleocellular inflammation are often present throughout the stromal tissues (Lane et al 1981). The causes of inflammatory polyps in cats are unknown and similar lesions are rarely detected in other animals. It has been hypothesised that feline nasopharyngeal polyps are congenital defects arising from remnants of branchial arches (Baker 1982) or are a response to irritation from chronic viral infections (Rogers 1988). Cats commonly have chronic respiratory tract inflammation induced by feline herpesvirus 1 (FHV-1) or feline calicivirus (FCV) and so it is possible that these two viruses are associated with feline inflammatory polyps. Green sea turtle fibropapillomatosis has similar histopathologic lesions and has recently been associated with an increased incidence of antibodies to a herpes virus (Herbst et al 1998).
Inflammatory polyps are thought to originate in the middle ear or eustachian tube (Faulkner & Budsberg, 1990). Polyps that expand into the nasopharynx result primarily in sonorous breathing noises and gagging. Polyps in the middle ear can induce Horner's syndrome or facial nerve paralysis and the mass can extend through the tympanum into the external ear canal. Ventral bulla osteotomy (VBO) is commonly recommended for cats with aural manifestations of disease (Anderson et al 2000) or all cats (Kapatkin et al 1990) and is clearly indicated for cats that present with clinical manifestations of middle ear involvement such as facial nerve paralysis or Horner's syndrome.
Optimal treatment of cats with only nasopharyngeal clinical signs, regardless of radiographic evidence of bullae disease, is currently unclear due to the small number of cases reported in the literature. Removal of inflammatory polyps through the mouth or external ear canal by use of traction/avulsion is usually possible and generally results in resolution of clinical signs. This is a less invasive and less expensive option for owners when compared to VBO. Additionally, VBO has a higher rate of morbidity than simple traction-avulsion, with a transient Horner's syndrome being the most significant complication, occurring in up to 80% of cats post-operatively (Kapatkin et al 1990). Recurrence of inflammatory polyps treated with traction/avulsion without VBO has been reported to be as high as 33% (Harvey & Goldschmidt 1978). The objectives of this study were to report clinical findings and management outcomes for a group of cats with nasopharyngeal polyps and to assess fresh or formalin-fixed tissues for the presence of FHV-1 DNA by use of polymerase chain reaction (FHV-1 PCR) or FCV RNA by use of reverse transcriptase PCR (FCV RT-PCR).
Materials and methods
Clinical cases
Medical records of the Veterinary Teaching Hospital, at Colorado State University (CSU-VTH) between March 1987 and July 2001 were reviewed. Key words searched included nasopharyngeal polyp, inflammatory polyp, middle ear polyp, and nasal polyp. Cases were included for retrospective clinical analysis if formalin-fixed tissue was available for FHV-1 PCR and FCV RT-PCR and the histological findings were consistent with those of inflammatory polyps. Clinical information and fresh tissues were available from four cats for assessment prospectively. Additional formalin-fixed tissues from cats with a histopathologic diagnosis of inflammatory polyps from one author (DG) employed by a commercial laboratory (IDEXX Laboratories, Broomfield, CO 80020, USA) were assessed by FHV-1 PCR and FCV RT-PCR but were not included in the clinical analysis.
Clinical data
Information obtained from the medical records included signalment, vaccination and FeLV/FIV status if available, age at onset of clinical signs, major presenting complaint (otic or respiratory), history of upper respiratory infection (URTI) within the last 3 months, presence or absence of bulla involvement by computed tomography (CT) or radiography, method of polyp removal (VBO or other), post-treatment complications and duration of follow-up. Owners were contacted to determine final rate of recurrence. Ventral bulla osteotomies were performed in a routine manner. Traction/avulsion removals were performed after visualising the polyp directly or with endoscopy, followed by grasping the polyp as close to the base as possible and removing with gentle traction.
Tissue analyses
Fresh tissues were immediately placed in sterile saline at the time of surgery and stored at −70°C until processed for FHV-1 PCR and FCV RT-PCR. For all tissues assessed, 5 μm thick sections were made by use of a microtome. Thirteen formalin fixed polyps and all four fresh frozen polyps had enough tissue available for an additional 60 μm thick cut to be assayed in an attempt to increase sensitivity. Formalin-fixed tissues were deparaffinised and DNA was extracted from all tissues via the methods described in a commercially available DNA isolation kit (DNeasy™ Tissue Kit, Qiagen Inc, Valencia, CA 91355, USA). Extracted DNA was immediately assessed by FHV-1 PCR at the Veterinary Diagnostic Laboratory at Colorado State University (Fort Collins, CO 80523, USA). After shipping to University of California (School of Veterinary Medicine, University of California, Davis, CA 95616, USA), additional tissue was deparaffinised as previously described and RNA was extracted using the Trizol reagent (Gibco BRL, Carlsbad, CA 92008, USA), according to manufacturer's recommendations. For reverse-transcription, 5 μl of sample RNA was denatured with 5-pmol oligo dT primer (Invitrogen, Carlsbad, CA 92008, USA) in a thin-walled PCR tube at 95°C for 5 min, and then quickly cooled in wet ice. Four microlitres of reverse-transcription mix was added, to a final volume of 50 mM tris-HCl, 75 mM KCl, 3 mM MgCl2, 10 mM dithiothreitol, 1 mM each of four deoxynucleotide triphosphates (dNTPs), 10 RNAsin (Promega, Madison, WI 53711, USA), and 50 U Moloney Murine Leukaemia Virus reverse transcriptase (Promega, Madison, WI 53711, USA). The reaction mix was incubated at 37°C for 1 h. Nested PCR was performed in a thermal cycler (MJ Research, Waltham, MA 02451, USA), using published primers and a protocol with modifications (Radford et al 1997). The primers used were primers 1 and 2 for the outer round and 4 and 5 for the inner round, Taq DNA polymerase (Promega, Madison, WI 53711, USA) was substituted for Pfu. Products of RT-PCR were electrophoresed on a 1% agarose gel containing ethidium bromide, and visualised by transillumination. Products were considered positive if there was a 265 bp band. For all PCR and RT-PCR assays, appropriate steps were taken to avoid DNA or RNA contamination and positive and negative controls were included.
Results
Clinical findings
Clinical information was included for 28 cats. For two cats, a second polyp developed on the contralateral side. Of the cats, 11 were male castrated, 14 were female spayed, one was male intact, and two were female intact. Breeds represented included 17 Domestic Shorthairs, five Siamese, four Domestic Longhairs, one Abyssinian, and one Maine Coon. All cats were vaccinated and no cats were known to be positive for FeLV/FIV. Median age at onset of clinical signs was 5.1 years, with average age at time of presentation being 6.1 years (range 0.25–18 years). On initial presentation, clinical signs were otic alone for 14 of 28 (50%) cats, respiratory alone for six of 28 (21%) cats, and a combination of otic and respiratory for eight of 28 (29%) cats. In two cats with polyps that spontaneously developed on the contralateral side (40 and 193 days after first polyp removal), the clinical manifestations were both respiratory and respiratory and otic on the first and second admissions, respectively. Only two of 28 (7%) cats had a history consistent with upper respiratory tract infection within the three months prior to the diagnosis of the inflammatory polyp.
Evaluation of radiographs (19 cats) and CT scan (eight cats) documented bulla involvement in 15 of 27 cats. One cat did not have imaging, but a pinch biopsy alone. As evidenced by CT or radiographs, neither of the two cats who developed disease in the contralateral ear had evidence of disease in that ear at the time of initial diagnosis.
Of the 28 cats, 14 were initially treated with traction/avulsion, 13 were treated with VBO, and one cat was biopsied but not treated. Minimum duration of follow-up was 2 months, with a median of 10 months and a range of 2 months to 6.5 years. Of the 13 cats treated with VBO alone, six had bulla involvement radiographically and seven were normal. One of the six cats with bulla involvement had a recurrence of the disease; the VBO was repeated in this cat and there was no subsequent recurrence. Of the 14 cats initially treated with traction/avulsion alone, three were removed from the external ear canal, eight were removed from the mouth, and site of removal could not be determined by medical record review for three. Of the 14 cats initially treated with traction/avulsion alone, five had recurrent polyps (three removed via the mouth and two removed via the external ear canal) necessitating VBO after which recurrences were not detected. Each of the five recurrent cases had radiographic evidence of bulla disease. Relapses in clinical signs necessitating a second procedure occurred between 19 and 129 days (median 53 days) after initial treatment. The remaining nine cats treated with traction/avulsion alone had no clinical evidence of recurrence; four of these had radiographic evidence of bulla disease.
FHV-1 PCR and FCV
Two polyps were available as both fresh and formalin-fixed tissue, 39 polyps were available as formalin-fixed tissue embedded in paraffin blocks, and two polyps were available as fresh tissue alone. All samples were negative for both FHV-1 and FCV regardless of the amount of tissue digested for analysis or whether or not fixation was used.
Discussion
Results of the retrospective aspect of this study are similar to those found in previous studies, in which no obvious breed or sex predilection was found (Allen et al 1999, Anderson et al 2000). However, our results were not compared to the overall patient population evaluated during the time frame of the study. An apparent older mean and median age at onset of clinical signs was found compared to previous studies (literature median range 0.9–5.0 years, literature mean range 0.4–6.1 years (Allen et al 1999, Anderson et al 2000, Kapatkin et al 1990)). Although this study was performed using the records of a secondary care facility, age at onset of clinical signs as opposed to age at primary admission was used to try to reduce artefact of continuous care inflating age at presentation. It has traditionally been taught that nasopharyngeal polyps primarily occur in younger animals; results of this study suggest that the disease should be on the differential list for middle-aged to older cats with otic and nasal signs as well.
While the majority of the clinical cases (14 of 28) presented for otic signs, four of five recurrent cases presented for both otic and respiratory signs, with the remaining recurrence initially presenting for otic signs. This is in contrast to a previous report (Anderson et al 2000) in which the authors concluded that a higher rate of recurrence was found in those animals presenting with otic signs. While there was a higher rate of recurrence in those cats treated with traction/avulsion when compared to those cats initially treated with VBO, evidence of bulla involvement was a reliable indicator of recurrence. Of the cats treated by traction/avulsion alone, recurrence was detected in five of nine cats with radiographic evidence of bullae disease but none of the five cats with normal bullae. The results of the present study suggest that traction/avulsion is a reasonable first line therapy for inflammatory polyps, however, it should be reserved for those cats without evidence of bulla involvement. The polyps in cats without evidence of bulla involvement may be arising from the eustachian tube but not middle ear, allowing more complete removal of affected tissue via traction/avulsion, as opposed to those with bullae involvement, in which complete removal of affected tissue cannot be achieved without VBO.
A history of a recent URTI was only reported for two cats. This finding suggests that URTI are not associated with development of inflammatory polyps or that the clinical manifestations of URTI occurred more than 3 months earlier. Failure to detect FHV-1 or FCV in the tissues from the cats described here could be due to several reasons. It is possible that these two viruses are not associated with inflammatory polyps. However, it is possible that FHV-1 or FCV infection is responsible for initiation of syndrome, but is cleared as inflammation develops and that viral persistence is not required for perpetuation of inflammation. This hypothesis is supported by findings with green sea turtle fibropapillomatosis. In this disease, herpesvirus antigen is detected by immunohistochemistry more commonly in the most immature, experimentally induced tumours (47%) than the older, spontaneously occurring tumours (7.5%) (Herbst et al 1999). It is also possible that FHV-1 or FCV were present in the tissues but the assays utilised were not sensitive enough. However, both assays have been shown to be as sensitive as immunohistochemistry or culture when used with fresh samples from other locations in the body for calicivirus (Sykes et al 1998) or herpesvirus-1 (Burgesser et al 1999). PCR and RT-PCR sensitivity is lower in formalin-fixed, paraffin embedded tissue, which could have affected our results. However, the four fresh samples assayed were also negative for FHV-1 and FCV. Lastly, it is possible that a virus other than FHV-1 or FCV or an entirely different disease process causes the disease.
Footnotes
Acknowledgements
The authors gratefully acknowledge the technical assistance of Melissa Brewer and Courtney Rand for performing the viral assays. Funding at the University of California was provided by the Center for Companion Animal Health. A portion of these results was previously presented as an abstract at the American College of Veterinary Internal Medicine Annual Meeting, Denver, 2001.