Pulmonary rhodococcosis in a cat

Abstract

Feline Rhodococcus equi infection is rare, despite the bacteria is widespread in the environment. R equi infection is typically observed in equine species but the infection has also been reported in dogs, cats and other domestic animals. There are a few reports regarding pulmonary R equi infection in cats and the disease appears to be limited to the skin and the subcutaneous tissue. This report describes the pathological, microbiological and the virulence features associated with an acute necrosuppurative pneumonia in a cat. To the best of our knowledge, this is the first report of feline pulmonary R equi infection in Italy.

A 6-year-old, neutered, regularly vaccinated, indoor male domestic shorthair cat, was brought in for post-mortem evaluation by a referring veterinarian. Anamnesis included initial gastroenteric signs, followed by severe respiratory failure and death in a period of 10 days. The cat was negative for feline immunodeficiency virus (FIV) and feline leukaemia virus (FeLV) by a commercially available enzyme-linked immunosorbent assay (Snap FIV/FeLV combo test, Idexx Laboratories). The animal was recently brought to a farm housing horses. A complete post-mortem examination was performed. At necropsy a moderate amount (∼20 ml) of serohaemorrhagic fluid was found to be present in the pleural spaces. Lungs were distended, not collapsed, and brightly pink in colour with mottled red areas. The pleural surface was shiny and multifocal scattered whitish foci were dispersed throughout the lobes (Fig 1). The lung parenchyma had a moderately increased consistency, but no truly consolidated areas were present. The cut surface was very moist, and abundant turbid frothy mucopurulent exudate oozed both from airways and parenchyma. Tracheobronchial lymph nodes were moderately enlarged and contained small whitish foci. Peripheral lymph nodes were smaller than normal. The wall of the small intestine was uniformly thickened and the mesenteric lymph nodes were enlarged. Both liver and kidneys were slightly enlarged and yellowish. Tissue samples were collected for histological, bacteriological and biomolecular analysis. Samples for histopathology were fixed with 10% buffered formalin, routinely processed, cut at 5–6 μm and stained with haematoxylin and eosin.

Fig 1.

Diffuse lung oedema and red mottled appearance with whitish pin-point foci.

Histologically, most of the lung parenchyma was effaced; alveolar spaces were filled with detached pneumocytes and a great number of activated macrophages, intermingled with scattered neutrophils, lymphocytes and very few plasma cells. Occasionally, large multinucleated giant cells were present in alveolar lumina. These cells, were two to four times larger than normal, vesiculated, peripherally located nuclei with prominent nucleoli and a large vesiculous lightly eosinophilic cytoplasm. Pleomorphic coccobacilli, 1–2 μm in length, were also found in the cytoplasm of some macrophages. In addition, multifocally distributed small necrotic foci with prevalent neutrophilic inflammation were also present.

The lung lesions were referable as diffuse severe subacute lymphohistiocytic and necrosuppurative pneumonia with multinucleated giant cells (Fig 2).

Fig 2.

Lung: multifocal scattered necrotic foci. Haematoxylin and eosin, magnification 4×.

Tracheobronchial lymph nodes contained small aggregates of activated macrophages in the medullary sinuses.

In the intestines, a monomorphous lymphocytic infiltration was observed in the mucosa and submucosa, also extending to muscular and serosal layer. The cells presented a hyperchromatic, round, small nucleus and very scant basophilic cytoplasm; neither cellular atypia nor mitotic figures were present. This lymphocytic infiltration was considered neoplastic rather than inflammatory due to the infiltration into the submucosa and the muscular layers. Lymphoid infiltration was not present in any other site. Neoplastic lymphocytes were immunoreactive to CD3 (Pan-T marker) and negative to CD79a (Pan-B marker), suggesting an intestinal T cell lymphoma. The liver and kidney had moderate degenerative changes, mostly in centrolobular and cortical areas, respectively. Other organs were unremarkable.

Bacteriological examination was performed on lung samples applied to blood agar with 5% defibrinated sheep blood and MacConkey agar. The plates were incubated at 37°C in aerofilia. After 48 h of incubation, evidence of bacterial growth was found in blood agar. Numerous, large, mucoid, non-haemolytic salmon-pink colonies formed after aerobic incubation of pulmonary fragments at 37°C for 48 h suggesting a R equi infection. The isolate tested positive by the CAMP test (Christie, Atkins, Munch, Petersen). This test can be used as quick presumptive test for R equi, interacting with the β-haemolysin of Staphylococcus aureus. A positive CAMP test is indicated by an arrow-head of complete haemolysis.

The isolate was also tested by the API Coryne test (BioMérieux) to evaluate several biochemical parameters and their compatibility with the hypothesised R equi. API Coryne and the other bacteriological identification tests used were all suggestive of R equi.

Antimicrobial susceptibility testing showed the isolate to be sensitive to clarithromycin, azithromycin, erythromycin, rifampin, imipenem, amoxicillin–clavulanate and lincomycin. The combination of rifampicin and erythromycin is considered to be the therapy of choice for rhodococcosis, especially in foals. These drugs are lipid soluble and penetrate caseous material and phagocytic cells. ³ Drugs such as lincomycin, amoxicillin and gentamicin have been recommended for treatment of rhodococcosis in cats, ^4,5 but β-lactam antibiotics such as amoxicillin/clavulanic acid have poor intracellular activity. ³

DNA was extracted from lung tissue and from a bacteriological isolate to perform a polymerase chain reaction (PCR). In order to confirm bacteriological findings and to establish the degree of virulence of the strain, the amplification of the R equi 16S ribosomal RNA (16S) gene and 85–90 kb plasmid (VapA) ⁸ was performed. The amplified fragments were sequenced and submitted to Genbank (accession number: HM025947 and HM025945). These fragments showed 100% homology with R equi 16S gene and VapA gene. Recent studies demonstrate that VapA-positive R equi have been isolated from dogs and cats. ¹¹

R equi is a telluric ubiquitous Gram-positive coccobacillus; it is an optional intracellular microorganism that is able to infect macrophages by interfering with the phagolysosomal fusion.

It is principally pathogenic in horses, in which it can produce pyogranulomatous bronchopneumonia and ulcerative enteritis with a high mortality rate in 1- to 3-month-old foals; it has also been reported to cause ulcerative lymphangitis, subcutaneous abscesses and arthritis. ^3,7 R equi has also been isolated from pigs, cattle, rabbits, dogs, cats and humans, in particular from immunocompromised patients. ¹⁰ Three virulence levels of R equi have been identified: virulent, intermediate and avirulent. ⁹ Virulent R equi is characterised by the presence of VapA and virulence plasmid DNA 80–90 kb long and has been found in foals (murine LD₅₀=10⁶). The intermediate virulence strains are characterised by the presence of VapB and virulence plasmid DNA 79–100 kb long; these strains have been isolated from the submaxillary lymph nodes of pigs (murine LD₅₀=10⁷). Avirulent R equi strains show no evidence of either virulence-associated antigens or plasmid DNA (murine LD₅₀>10⁸) and are widespread in soil.

The infection in pet animals is very rare, it usually incites soft tissue pyogranulomatous lesions, with primary involvement of the extremities. Rarely, vaginitis, hepatitis, osteomyelitis, myositis, and lymphadenitis, particularly to mediastinal and mesenteric lymph nodes are observed. ^1,6

It appears that the respiratory tract is not a primary target for R equi in pet animals, as opposed to that which is observed in horses. ^3,9 In cats, only a single suspected case of pulmonary involvement, based on clinical findings and cytological examination of tracheal exudate, was reported in the literature, but without necropsy confirmation. ²

In humans, rhodococcosis affects essentially immunocompromised patients. Most of the reported cases are in acquired immunodeficiency syndrome (AIDS) or cancer patients and in individuals treated with immunosuppressive drugs. ¹⁰

Unfortunately, in this case, no specific tests were performed to evaluate the immune status of the cat, nevertheless the cat had intestinal lymphoma, in itself enough to cause immunosuppression.

Previous studies indicate that the plasmid profiles of VapA-positive isolates from cats and dogs are closely associated with those isolated from horses. It has been suggested that given their high degree of similarity, it is possible that VapA-positive R equi isolates from cats and dogs could derive from horses or their environments. ⁶ In humans, the isolated strains of virulent R equi from subjects with and without AIDS were similar to the isolates from cats and dogs. ¹⁰

The cat in this case report spent a short period of time in a horse farm immediately prior to the onset of signs emphasising this hypothesis.

In horses, rhodococcosis is generally acquired by the aerogenous route. In cats, on the other hand, the most common route of infection is by transcutaneous infections following penetrating wounds. In this particular case, it is likely that the infection occurred via the aerogenous route.

The histopathology report suggested intestinal lymphoma, so the authors believe that the underlying neoplastic condition may have contributed to immunosuppression.

In humans, a defect in cell-mediated immunity, as seen in AIDS patients, is necessary in order to develop rhodococcosis. Similarly, in cats a previous viral infection by FeLV or FIV could represent the predisposing factor. Other possible causes of immunodepression in cats include corticosteroid therapy, nutritional deficiencies, age, genetic disease, stress and underlying infectious and infestive diseases. Although, in this case, not all of these factors were evaluated, the intestinal lymphoma was considered to be the main cause of immunodepression in this cat, even if this consideration is speculative.

In conclusion, to the best of the authors’ knowledge, this is the first report of R equi infection in a cat in Italy and one of the few described feline infections, of this aetiological agent with respiratory involvement.

Epidemiology and pathogenesis of rhodococcosis in cats still remain unclear and require further study in order to fully understand the risk factors involved. The role of the cat as a carrier of potentially severe opportunistic infections to immunodepressed humans should be considered.