A case of feline aspergillosis with central nervous system involvement: clinical utility of cerebrospinal fluid PCR testing

Abstract

A 5-year-old cat presented with exophthalmos and neurological signs. Aspergillus udagawae was identified by histopathological examination and polymerase chain reaction (PCR) analysis of an orbital mass. Although magnetic resonance imaging suggested intracranial infection, cerebrospinal fluid (CSF) analysis and fungal culture were not clear. Serial CSF PCR detected Aspergillus sp., leading to diagnosis of central nervous system (CNS) infection and monitoring of treatment response. Following surgical debulking and administration of antifungal agents including voriconazole, the clinical signs resolved and CSF PCR reverted to negative. This case suggests that CSF PCR is a sensitive adjunctive tool for diagnosing and monitoring CNS aspergillosis in cats when conventional methods are inconclusive.

Feline aspergillosis is categorized into sino-nasal aspergillosis and sino-orbital aspergillosis. Sino-nasal aspergillosis is typically non-invasive, with lesions confined to the nasal cavity and paranasal sinuses. Sino-orbital aspergillosis is invasive, with lesions extending from the nasal cavity to the orbit and surrounding tissues and, in rare cases, infiltrating the central nervous system (CNS) and resulting in neurologic signs [7, 13]. Diagnosis is difficult when infection involves CNS. Cormack et al. reported that CNS infection was identified at necropsy despite the absence of overt neurologic signs in a cat [10]. Furthermore, imaging findings on computed tomography (CT) or magnetic resonance imaging (MRI) are non-specific, and imaging alone may not reliably distinguish aspergillosis from other mycoses or neoplastic disease [5, 6]. Recurrence and persistent infection after clinical remission have also been reported in sino-nasal aspergillosis and sino-orbital aspergillosis, often necessitating long-term antifungal therapy [6, 11]. It is important to have diagnostic methods that can accurately assess CNS involvement and measure treatment response. This report describes the clinical course and diagnostic approach in a cat with suspected CNS involvement in which cerebrospinal fluid (CSF) PCR testing enabled diagnosis of intracranial aspergillosis and continuing assessment of treatment response.

A 5-year-old spayed female British Shorthair was referred to Azabu University Veterinary Teaching Hospital (Day 1). The cat had a 2-month history of reduced appetite and food spillage during eating. The cat presented with exophthalmos of the left eye 10 days prior to presentation, and left circling began one day before presentation. Neurological examination revealed obtundation, bilateral absence of the menace response, anisocoria (L >R), and bilateral absence of physiologic nystagmus. No abnormalities were observed in trigeminal nerve function, facial nerve function, or tongue movement. There were no clinical abnormalities according to thoracic and abdominal radiography or routine hematologic testing. Based on the neurological findings, intracranial lesion was suspected and MRI and CT were performed.

Anesthesia was induced by propofol at a dose of 8.0 mg/kg and was maintained by isoflurane. MRI was performed using a 3.0 T unit (Vantage Titan 3T; Canon Medical Systems, Otawara, Japan). An orbital mass was identified by MRI, demonstrating hypointense to hyperintense signals on T2 weighted images (T2WI; TR, 4,300 msec; TE, 87 msec), and isointense to hyperintense signals on T1 weighted images (T1WI; TR, 600 msec; TE, 15 msec), with peripheral contrast enhancement (gadoteridol, Bracco, Tokyo, Japan; 0.2 mL/kg intravenous administration). In addition, poorly bounded intracranial lesions were observed in the left temporal and parietal lobes, characterized by T2WI hyperintensity and T1WI hypointense to isointense signals with faint meningeal contrast enhancement. These findings suggest intracranial inflammation or neoplasia with associated edema (Fig. 1). The CT scan (Aquilion Prime SP, Canon Medical Systems) was performed with scan thickness of 2.00 mm, 250 mA, and 120 kV. For contrast-enhanced imaging, the cat was administered with 2.0 mL/kg of non-ionic contrast medium (Iohexol, Fuji Pharma, Tokyo, Japan). Contrast-enhanced CT images were acquired during the arterial, portal venous, and delayed phases. An orbital mass with contrast enhancement and heterogeneous attenuation was observed on CT. Osteolysis of the ventral orbital bone was also observed, with extension of the mass into the oral cavity, together with mild enlargement of the left medial retropharyngeal lymph node (Fig. 1). Because MRI showed indistinct sulci suggesting increased intracranial pressure, CSF collection was not performed. Cytology and bacterial/fungal culture were performed using samples from the oral mass. Cytological examination revealed neutrophils, macrophages, eosinophils, and plasma cells, but no causative organism was identified. An incisional biopsy of the orbital mass was performed for histopathological evaluation. Histopathological examination revealed granulomatous necrotizing inflammation with fungal hyphae. Fungal hyphae were septate, with parallel walls and acute-angle branching, consistent with Aspergillus species, and were highlighted by Grocott’s methenamine silver staining (Fig. 2). PCR was performed using five fungal primer sets as follows: 5′-GTTGATAGAAYAATGTAGATAAGG-3′ (forward) and 5′-GACAGTAAGATTCCCCTTG-3′ (reverse) as primer set 1, fungal primer targeting the extended 28S rDNA region [12]; 5′-GCATCGATGAAGAACGCAGC-3′ (forward) and 5′-TCCTCCGCTTATTGATATGC-3′ (reverse) as primer set 2, ITS2 [15]; 5′-GTACCCGCTGAACTTAAGC-3′ (forward) and 5′-ATCCTGAGGGAAACTTC-3′ (reverse) as primer set 3, LSU 28S (LROR/LR5) [18]; 5′-TTAGCATGGAATAATRRAATAGGA-3′(forward) and 5′-TCTGGACCTGGTGAGTTTCC-3′ (reverse) as primer set 4 nu-SSU (0817/1196) [9]; and 5′-AATTGGTGCCGCTTTCTGG-3′ (forward) and 5′-AGTTGTCGGGACGGAATAG-3′ (reverse) as primer set 5, β-tubulin [1]. Using the broad-range fungal primer sets 1–4, PCR of DNA extracted from histopathological sections yielded amplicons with primer sets 1, 3, and 4 (Fig. 2); direct sequencing of these products identified the fungus as Aspergillus sp. PCR with primer set 5 yielded an amplicon with direct sequence showing 100% sequence identity with Aspergillus udagawae in an NCBI BLASTn search.

Fig. 1.

Transverse MRI at the level of the thalamus (A, B, C), and CT images at the level of the orbit (D, E, F) on Day 1. An orbital mass (white arrows) demonstrates hypointense to hyperintense signals on T2-weighted images (T2WI) (A), and isointense to hyperintense signals on T1-weighted images (T1WI) (B), with peripheral contrast enhancement (C). Poorly marginated lesions (arrowheads) are visible in the left temporal and parietal lobes, characterized by T2WI hyperintensity and T1WI hypointense to isointense signals with faint meningeal contrast enhancement, suggesting intracranial inflammation and associated edema. CT images are presented as bone window (D), soft tissue window (E), and post-contrast soft tissue window (F). These images show an orbital mass with heterogeneous attenuation and contrast enhancement (yellow arrows). Osteolysis of the ventral orbital bone and extension of the mass into the oral cavity are observed, along with slight enlargement of the left medial retropharyngeal lymph node.

Fig. 2.

Histopathological examination and PCR testing of the orbital mass. Histopathological examination of the orbital mass biopsy revealed septate fungal hyphae with parallel walls and acute-angle branching (Grocott’s methenamine silver staining, A). PCR analysis of DNA extracted from the orbital mass yielded amplicons with primer sets 1, 3, and 4 (arrowheads, B).

Antifungal treatment with itraconazole (7.3 mg/kg, once daily) was begun on Day 9. Oral prednisolone (0.5 mg/kg, once daily) was also prescribed for ocular swelling and cerebral edema. By Day 30, neurological signs had improved, and follow-up MRI showed improvement of the left temporal and parietal lesions and the associated edema. To further evaluate the intracranial lesions, CSF was collected from the cisterna magna. The orbital mass and surrounding inflammation showed no marked improvement. CSF analysis revealed no abnormalities (white blood cell (WBC), 0 cell/μL; total protein (TP), 6.7 mg/dL; glucose, 149 mg/dL), and fungal culture was negative. CSF PCR using the primer sets 1, 3 and 4 that had yielded positive results in the orbital mass demonstrated a positive result for Aspergillus. Intracranial Aspergillus infection was diagnosed, and antifungal therapy was changed to fluconazole (10 mg/kg, once daily) because of its favorable CNS penetration. Exophthalmos worsened on Day 39. Surgical debulking of the orbital mass was performed on Day 51, and the resulting surgical cavity was packed with a topical antifungal preparation (Mometaotic, MSD Animal Health, Tokyo). Follow-up CT on Day 87 found no evidence of regrowth at the debulked orbital site, and the clinical course was considered favorable. Fluconazole was continued at the same dosage postoperatively.

On Day 120, no neurological abnormalities were detected on examination. MRI revealed a new irregular contrast-enhancing lesion at the base of the brain, characterized by hypo- to isointense signals on T2W images and isointense signals on T1W images. CSF analysis remained unremarkable (WBC, 1 cell/µL; TP, 10.3 mg/dL; glucose, 86 mg/dL), and fungal culture was negative. CSF PCR was positive for Aspergillus. Based on the newly identified lesion on MRI (Fig. 3), antifungal therapy was changed to voriconazole (2.5 mg/kg, every 3 days). On Day 184, neurological examination was normal, and MRI confirmed resolution of the contrast-enhancing lesion at the base of the brain. CSF findings were unremarkable (WBC, 0 cells/µL; TP, 11.2 mg/dL; glucose, 78 mg/dL) and fungal culture was negative. Antifungal treatment was continued because CSF PCR remained positive.

Fig. 3.

Transverse MRI T2WI at the level of the thalamus, performed on Days 30, 120, 184, and 259. On Day 30, relative to Day 1, improvement of the lesions in the left temporal and parietal lobes and associated cerebral edema is observed (A). On Day 120, a new lesion is identified at the base of the brain (arrowhead, B). On Day 184, resolution of the new lesion at the base of the brain is confirmed (C). On Day 259, no abnormal findings are observed in the brain (D).

On Day 259, neurological examination was normal and MRI showed no abnormalities (Fig. 3). CSF analysis was unremarkable (WBC, 2.5 cells/µL; TP, 7.0 mg/dL; glucose, 80 mg/dL), with negative fungal culture and PCR results for Aspergillus. Although no fungal organisms were detected in the CSF, antifungal therapy was continued because of the possibility of fungal recurrence. On physical examination, new ulcerative lesions were identified in the oral cavity. An excisional biopsy was performed on Day 282. Histopathological examination revealed granulation tissue in the ulcerative lesion, but histopathology revealed no obvious pathogens, including fungi. The lesions were suspected to be granulation tissue unrelated to fungal infection. Oral antifungal medication was temporarily discontinued on Day 356 because anorexia developed.

On Day 435, neurological examination remained normal and MRI found no abnormalities. CSF analysis was unremarkable (WBC, 0.5 cells/µL; TP, 9.5 mg/dL; glucose, 88 mg/dL) with negative fungal culture and CSF PCR for Aspergillus. Remission was considered to be maintained despite discontinuation of oral antifungal therapy on Day 591.

In this case, Aspergillus infection was diagnosed by histopathological examination of an orbital mass, where routine CSF analysis and fungal culture failed to detect intracranial infection. CSF PCR enabled diagnosis of intracranial infection, and serial Aspergillus PCR testing allowed accurate monitoring and facilitated treatment planning. Diagnosis of feline upper respiratory tract aspergillosis typically involves visualization of fungal plaques via rhinoscopy, and confirmation of fungal hyphae on cytology and/or histopathology of biopsy specimens [5]. Additional supportive diagnostic tests include CT or MRI, fungal culture, fungal PCR, antigen testing, antibody testing, and Aspergillus species–specific IgG ELISA. Antigen testing of galactomannan has a sensitivity of 23% and a specificity of 78%, and antibody testing has a sensitivity of 43% and a specificity of 100% [4, 5, 21]. The Aspergillus species–specific IgG ELISA demonstrates excellent diagnostic performance (sensitivity 95%, specificity 93%), but is used primarily for research and is not currently available in diagnostic laboratories in Japan [4, 21]. Images including CT and MRI are useful for assessing lesion extent, osteolysis, and intracranial extension, although differentiation from other mycoses, encephalitis, or neoplastic lesions can be challenging [2, 5, 8]. Fungal culture is typically performed using tissue obtained from the lesion and has high sensitivity, with positive results in 22 of 23 cases in a previous report [3], although results may be negative when the fungal burden is low. PCR testing may be positive even when lesion-tissue culture is negative [7, 16]. CSF PCR has demonstrated high diagnostic performance (sensitivity 75%, specificity 98.3%) in human CNS aspergillosis [14]. This case suggests that PCR testing is useful for detecting CNS aspergillus infection in cats.

Repeated PCR testing of CSF enabled assessment of antifungal treatment response and infection status, allowing appropriate management of Aspergillus. In invasive aspergillosis, persistent infection has been reported even when clinical signs and imaging findings appear to indicate remission, making decisions regarding treatment discontinuation difficult [11]. Furthermore, there has been a case in which clinical signs and imaging suggested mainly upper respiratory tract, orbital, or pulmonary disease for a prolonged period after diagnosis, with no imaging evidence of CNS involvement, yet neurological signs developed in the terminal stage and necropsy findings suggested CNS involvement. These fact imply that clinical signs and imaging alone may underestimate CNS infiltration and/or lesion severity [10]. In the present case, clinical findings, imaging features, and fungal culture results did not necessarily correlate with intracranial infection status, whereas CSF PCR remained consistently positive during periods in which CNS infection was considered as continuing, supporting the utility of CSF PCR for monitoring CNS infection.

MRI findings of the orbital lesion in the present case were heterogeneous on both T1WI and T2WI and showed contrast enhancement, consistent with previous reports [5]. In contrast, there are no comprehensive reports describing the MRI findings of feline CNS aspergillus infection. Although several studies have reported MRI findings in dogs with CNS aspergillus infection, the imaging features are highly variable. In one study, 3 of 7 dogs had contrast-enhancing mass lesions, one had a hemorrhagic infarction, and the remaining 3 had no detectable abnormalities on MRI [20]. In another report, 2 dogs exhibited meningeal contrast enhancement with surrounding edema, whereas one dog had a contrast-enhancing intracranial mass [17]. In the present case, the predominant MRI findings were meningeal contrast enhancement and surrounding edema. These findings suggest that imaging findings alone are insufficient for the diagnosis of CNS aspergillus infection.

For the orbital lesion, surgical management and local therapy were undertaken. Because systemic antifungal therapy alone did not lead to significant clinical improvement of the orbital lesion, surgical debulking combined with a topical antifungal preparation was considered beneficial. In addition, antifungal agent was selected based on serial CSF PCR results for intracranial lesions, and voriconazole proved effective in the present case. Voriconazole is a potent azole antifungal agent with strong activity against Aspergillus spp.; however, neurological adverse effects and hepatotoxicity have been reported in cats, and caution is warranted when administering this drug [6, 19]. In this case, dosing adjustments were made taking into account a previous report [6]. Unfortunately anorexia, suspected to be an adverse effect of voriconazole, was observed on Day 340, and the drug was discontinued on Day 356.

This case report suggests that CSF PCR can act as a novel adjunctive test for diagnosing feline aspergillosis, and for evaluating treatment response. In conditions such as sino-orbital aspergillosis, which are highly invasive and raise concerns regarding CNS involvement, CSF PCR appears to be a potentially sensitive method for supporting the detection and monitoring of CNS aspergillosis in cats.

Conflicts of Interest

The authors declare that they have no conflict of interest.