Case Report: Spinal intramedullary fungal granuloma: clinical findings, surgery, and outcome in a dog

Abstract

A 7-year-old male neutered Miniature Schnauzer was assessed due to an acute and progressive history of pelvic limb weakness and incoordination. At the examination, the dog was non-ambulatory paraparetic, the postural reaction were abscent in both pelvic limbs, with intact withdrawal and patellar reflexes and increased pelvic limb muscle tone, which lead into paraplegia with intact nociception in the follow 24 h. A T3-L3 myelopathy was suspected, and an MRI of the thoracolumbar vertebral column was performed under general anaesthesia. An ill-defined, solitary, T2W hyperintense and strongly contrast enhancing intramedullary lesion was found, extending from the caudal endplate of T12 to the midbody of T13 and associated with severe spinal cord oedema and meningeal enhancement. A lumbar CSF analysis revealed a severe mononuclear pleocytosis (309 cells/μL) and markedly increased protein level; serology against Neospora caninum and Toxoplasma gondii was negative, and treatment with steroids for a suspected focal meningomyelitis of unknown origin was started. The patient deteriorated over the following days, becoming paraplegic with reduced nociception in the right pelvic limb. An exploratory surgery was carried out via T12-T13 right-sided hemilaminectomy, durotomy and myelotomy, and a well-defined, firm intramedullary mass was removed and sent for histopathology. The mass was consistent with an inflammatory pyogranuloma with intralesional fungal hyphae. Panfungal PCR was negative. Treatment with fluconazole was started, and a follow-up MRI was performed 6 months after surgery. At this time the dog was ambulatory paraparetic and MRI revealed complete resolution of the previous mass. Repeat CSF analysis revealed a mild mononuclear pleocytosis (6 cells/μL) and a mild increase in protein. Fluconazole treatment was continued for another 12 months; during all this time, the patient remained ambulatory with a mild paraparesis. The patient died for unrelated reasons 18 months after surgery when under general anaesthesia for routine removal of a subcutaneous lipoma.

1. Introduction

Fungal infections in the central nervous system (CNS) are uncommon, although this might vary across different geographical locations, and they are often associated with a poor prognosis (1, 2).

These cases can be challenging to identify, especially in non-endemic areas, as they can share a lot of clinical and imaging characteristics with other more common pathologies like neoplasia, sterile inflammatory (suspected immune-mediated) conditions and other infectious diseases (3, 4). Multiple fungal species have been reported as the causative agent of disease in dogs and cats (1), and the predisposition depends on the lifestyle of the animal (indoor vs. outdoor, city vs. countryside) (2), age, sex or even breed (4).

Fungal diseases involving the spinal cord or vertebral column are unusual, with fungal discospondylitis being the most widely reported in veterinary medicine (5, 6). In the case of solitary fungal granulomas involving the spinal cord, these are rarely reported in the literature (7–10), with a resultant lack of knowledge on diagnosis, treatment, and outcome,. In human medicine these cases are also considered rare, although diagnosed with increased frequency in immune compromised patients (11–13). In human medicine, surgical resection combined with prolonged antifungal medication is currently considered the gold standard of treatment. However, unlike in human medicine, the role and effectiveness of debulking surgery of intramedullary fungal granulomas remain poorly documented in veterinary medicine, with virtually no published evidence available.

The present case report describes the clinical, magnetic resonance imaging (MRI), and histopathological findings, and long-term outcome of a dog after surgical resection of a spinal intramedullary fungal granuloma.

2. Case presentation

A 7-year-old, male neutered Miniature Schnauzer was assessed due to an apparent acute onset of pelvic limb weakness that had progressed over the following 3 days before the referral consultation. At admission, the neurological examination revealed a severe ambulatory paraparesis with marked pelvic limb proprioceptive ataxia. The postural reactions were absent in both pelvic limbs, with an increase in pelvic limb muscle tone and intact pelvic limb spinal reflexes. The cutaneous trunci reflex was absent caudal to the level of L2, and there was marked apparent lumbar discomfort upon dorsal palpation. A T3-L3 myelopathy was suspected, and further investigations were scheduled for the following day. Considering the age, the breed, history and clinical sings as the more common differential diagnosis were intervertebral disc extrusion following by neoplasia, following for inflammatory immune-mediated and infectious diseases, although less likely.

Twenty-four hours later, the dog’s neurological status had deteriorated, becoming paraplegic with intact nociception in both pelvic limbs. Haematology and biochemistry were normal other than a mild uremia of 13.1 mmol/L (reference interval 2.5–9.6) and mild hypertriglyceridemia 1.2 mmol/L (reference interval 0.11–1.13).

The dog was anesthetised and magnetic resonance imaging (MRI) (Philips 1.5-T Gyroscan Intera) of the thoracolumbar vertebral column was performed. A well-defined, intramedullary, spherical lesion was identified at the level of the T12–T13 intervertebral disc space. It measured 1.3 cm in length, extending from the midbody of T12 to the cranial endplate of T13. The lesion was centrally located, although slightly left-lateralized, and infiltrated/compressed approximately 90% of the spinal cord diameter. It showed a heterogeneous signal in all sequences, being mainly hyperintense in T2w and isointense in T1w relative to the grey matter, with strong and homogeneous contrast enhancement. Cranial and caudal to the lesion, the spinal cord exhibited an ill-defined, intramedullary hyperintensity on T2w, extending from T1 to L4 and involving mainly the dorsal and ventral white matter, compatible with oedema/pre-syrinx formation (Figure 1).

Figure 1.

Sagittal T2-weighted (a), transverse T2-weighted (b), transverse T1-weighted (c), and post-contrast transverse T1-weighted (d) MRI images obtained at the time of diagnosis. The arrow in panel (a) indicates the intramedullary granuloma, which demonstrates marked contrast enhancement on the post-contrast sequences.

Following the MRI, a lumbar cerebrospinal fluid (CSF) sample was taken and revealed a severe mononuclear pleocytosis of 309 WBC/μL (reference interval 0–5), with 12 RBC/μL and hyperproteinorrachia 2000 mg/dL (reference <45). On cytological examination, the cells were often clumped together, and there was a pale eosinophilic fluid background containing rare erythrocytes. The nucleated cells population comprised predominantly mononuclear cells with a differential count being 68% lymphocytes, 27% large monocytoid cells, 3% neutrophils, and 2% eosinophils. Lymphocytes were mostly small mature lymphocytes. The monocytes present in the preparation were very active with lipid vacuolization of the cytoplasm (Figure 2).

Figure 2.

Cerebrospinal fluid (CSF) cytology stained with the Wright–Giemsa technique, showing predominantly mononuclear pleocytosis with two eosinophils identified within the cellular cluster.

Based on the MRI and CSF findings, an inflammatory granuloma, likely related to a focal granulomatous meningomyelitis, was suspected, although other differential diagnoses, such as a protozoal or fungal granuloma, or neoplasia such as histiocytic sarcoma, lymphoma or glioma, could not be ruled out.

Serology against Toxoplasma gondii and Neospora caninum was sent for analysis and while waiting for these results, the patient was discharged with prednisolone 1 mg/kg/24 h (Prednicare®, Ecuphar Ltd., Belgium) paracetamol 15 mg/kg/8 h (Paracetamol, Bova Specials Ltd., United Kingdom), and clindamycin 15 mg/kg/12 h (Zodon®, Ceva Animal Health Ltd., United Kingdom). The protozoal serology results subsequently returned as negative and the clindamycin was stopped.

One week after discharge, the patient was reassessed. He remained paraplegic, although deep pain perception now appeared diminished, especially in the right pelvic limb. After discussing the progression with the owner, options such as humane euthanasia or exploratory surgery followed by an excisional biopsy were considered, with the latter chosen.

The patient was booked for exploratory surgery the following day with the goal of acquiring samples for histopathology and achieving spinal cord decompression. As premedication, methadone 0.2 mg/kg IV (Comfortan®, Dechra Veterinary Products Ltd., United Kingdom) and dexmedetomidine 2 μg/kg IV (Dexdomitor®, Vetoquinol Ltd., United Kingdom) were administered, followed by induction with propofol (Propoflo®, Zoetis Ltd., United Kingdom) at an approximate dose of 2 mg/kg IV. For maintenance, inhalational anaesthesia with sevoflurane was used, followed by an ultrasound-guided erector spinae plane block with bupivacaine at the surgical site, and the patient was then administered neuromuscular blockade with atracurium during the entire surgical procedure.

A routine left dorsolateral approach to the thoracolumbar vertebral column was carried out for a T12-T13 left-sided hemilaminectomy. After creating the surgical defect using a high-speed bur, the spinal cord was exposed, and a small, rounded lesion was observed under the dura mater. The dura mater was incised immediately over this lesion in a straight fashion using a single-use corneal knife, revealing an intramedullary, red, well-demarcated, and firm lesion. Following this, a myelotomy parallel to the spinal cord was performed. Using blunt dissection and bipolar cautery, the mass was removed macroscopically, leaving surrounding visibly normal spinal cord tissue. Over the durotomy defect, a synthetic collagen-based dural graft (Duragel Plus®, Integra Ltd., United States) was placed, and routine closure of the muscle fascia, subcutaneous tissues and skin was performed in 3 layers using absorbable monofilament suture (PDS® and Monocryl®, Ethicon Ltd., United Kingdom) in a continuous simple pattern.

The patient recovered uneventfully from the surgery. The following day, he remained paraplegic, although with appropriate nociception in both pelvic limbs. He was discharged home after 2 days of hospitalisation with an indwelling urinary catheter in place and instructions for the management of this for his owners to assist with bladder management at home; this was removed 1 week after discharge following recovery of the motor function in the pelvic limbs.

The mass was fixed in a 10% formalin solution and sent to an external laboratory for analysis. This revealed a mass composed of multifocal coalescing inflammatory foci. Within the foci were large numbers of centrally placed epithelioid macrophages, which were surrounded by large numbers of lymphocytes, plasma cells, neutrophils, and macrophages. Some of these foci were also surrounded by a thin layer of fibrocytes (capsule). There were occasional septate hyphae 4 micrometres wide and up to 120 micrometres long (Periodic Acid-Schiff (PAS) and Grocott staining) (Figure 3). There were small-calibre blood vessels throughout the sections, many of which contained a perivascular lymphocytic and plasmacytic infiltrate. These results were consistent with a fungal granuloma and a panfungal PCR was run on the same sample. This was negative, potentially due to a delay in processing and low number of fungi.

Figure 3.

Histopathological examination of the excisional biopsy stained with hematoxylin and eosin (H&E) (a), periodic acid–Schiff (PAS) (b), and Grocott methenamine silver (GMS) (c). The images demonstrate an inflammatory granulomatous cell population, with arrows indicating fungal hyphae within the tissue sample.

Treatment with fluconazole 4.5 mg/kg twice a day (Diflucan® 50 mg capsules, Pfizer Limited, Kent, United Kingdom) was started following the histopathology results. The patient was reported to have a good response following surgery and anti-fungal treatment, showing progressive improvement in his mobility, and recovering independent ambulation on the pelvic limbs 2 months after the surgery.

After 6 months of antifungal medication, a follow-up MRI and lumbar CSF sample were performed. At this point in time, neurological examination revealed moderate ambulatory paraparesis and pelvic limb proprioceptive ataxia, with absent postural reactions in the left pelvic limb. The rest of the neurological examination was normal and there was no apparent discomfort on vertebral column palpation.

Magnetic resonance imaging showed a complete resolution of the previously described lesion, with a focal reduction of volume of the spinal cord and a small cavitary lesion at the surgical site (Figure 4). Analysis of CSF from the lumbar sample revealed a mild mixed pleocytosis of 6 WBC/μL (reference 0–5) and hyperproteinorrachia of 56.4 mg/dL (reference <45). Due to the possibility of this pleocytosis being related to a persistent or subclinical infection, the antifungal treatment was continued.

Figure 4.

Follow-up MRI performed 6 months after surgery, including sagittal T2-weighted (a) and transverse T2-weighted (b), transverse T1-weighted (c), and post-contrast transverse T1-weighted (d). No evidence of granuloma regrowth is observed. Pointed with an arrow. a focal spinal cord defect is visible at the previous site of the granuloma.

The patient remained ambulatory paraparetic but was reported to have an excellent quality of life over the following months after repeat MRI, without side effects related to the fluconazole. Eighteen months after the diagnosis, the owner reported that the dog died at their first opinion vets during a routine general anaesthesia for a subcutaneous lipoma removal.

3. Discussion

This case represents, to the authors’ knowledge, the first reported successful surgical resection of a spinal intramedullary fungal granuloma in a dog, followed by long-term antifungal therapy with a favourable outcome. There is one reported similar case of a cat with a fungal granuloma (Coccidioides spp.) (8), which also had a favourable outcome over a similar follow-up time.

Fungal infections of the CNS are rare in dogs, particularly in non-endemic regions, with multiple causative agents being reported causing granulomas in the CNS in veterinary and human medicine (4, 8, 14). A definitive diagnosis can be challenging to achieve in such cases, requiring a biopsy and histopathology to establish a final diagnosis.

The MRI characteristics in this case share multiple features of previously reported fungal granulomas, being heterogenous with predominantly hyperintensity in T2 and isointensity in T1-weighted sequences relative to the grey matter, and strong and homogeneous contrast enhancement (3, 8). However, these characteristics present a significant overlap with other pathologies, neoplasia in particular, meaning that a definitive diagnosis is not possible based on imaging alone. In some cases, the presence of extracranial involvement, such as sinusitis, uveitis or lung disease, may increase the index of suspicion for fungal involvement by species such as Cryptococcus and Blastomyces. However, these are more commonly associated with the formation of granulomas in the brain than the spinal cord (15–17). In the current literature, this correlation does not appear to exist for cases in which the lesion is restricted to the spinal cord, with the few reported cases as solitary granulomas with no other lesions reported (1, 8, 13).

Cerebrospinal fluid analysis may also help to narrow the list of differential diagnoses in dogs with focal spinal mass lesions. Although variable findings have been reported in the literature, a pleocytosis, often mixed or lymphocytic, has been commonly described in the presence of CNS fungal infections (1). However, this remains non-specific and may overlap with findings from other, more common, conditions such as meningoencephalomyelitis of unknown origin (MUO) or neoplastic processes like lymphoma. The presence of eosinophils is often reported in cases of fungal disease (1, 3, 14, 16), which may be less common for other conditions. However, CSF eosinophilia has also been reported for both MUO and neoplasia, such as histiocytic sarcoma, in dogs (18). In the present case, although eosinophils represented only 2% of the total cell population, this could have raised concern for an infectious origin, either fungal or protozoal. This prompted testing for toxoplasmosis and neosporosis in this case but given the rarity of fungal disease in the United Kingdom, this was not initially considered a likely diagnosis prior to surgery. Fungal organisms can be observed in the CSF, reported in as many as 81% of dogs and 54% of cats with Cryptococcus spp. (1, 8). However, the appearance of hyphae in the CSF is considered rare for other fungal species.

In the present case the final diagnosis was only achieved via excisional biopsy. The fungal species could not be identified, despite a panfungal PCR on the formalin fixed tissue. This PCR technique has been described to have moderate sensitivity of around 69.5% when performed in formalin fixed tissue (19), against a reported 93.6% in fresh tissue samples (20). This difference is likely due to the degradation of the DNA after contact with the formalin, particularly if testing is delayed as for this case. Fresh tissue samples were not collected at the time of surgery given the low index of suspicion for fungal disease at that time.

Other possible diagnostic techniques for fungal disease affecting the CNS in dogs include immune assays for specific fungal organisms in the blood. Serum antibodies against various fungal species can be detected, including Aspergillus spp., Histoplasma spp., Coccidioides spp., Blastomyces spp., and Cryptococcus spp. (1). The sensitivity of these tests varies depending on the disease presentation: for focal Aspergillus infections, the reported sensitivity is around 24%, whereas in cases of disseminated disease, sensitivity for all species ranges from 67 to 98%, with specificities between 82 and 100% (1). In our case, no further testing was performed because individual serological testing was not considered cost-effective due to its expected low sensitivity and the associated financial burden for the owner.

With regards to the treatment of fungal infections in the CNS, although the prognosis can depend on the species, in general it is considered poor, often associated with relapses and many months or lifelong treatment with antifungal medication (1, 3, 4, 16, 21). This is related to the fact that many antifungal medications have limited penetration into the CNS (1, 6). Fluconazole, however, is often considered a good option because it is able to cross the blood–brain/blood-spinal cord barrier (1). Furthermore, published evidence shows that debulking surgery in cases of fungal granulomas can significantly improve outcomes in both veterinary and human patients (3, 8, 14, 22, 23). This may be related to several factors; firstly, the debulking of a granuloma might reduce the mass effect over the brain or spinal cord. Secondly, it may also reduce the fungal load in the affected tissue and possibly help the penetration of the antifungal drug into the lesion. Moreover, it allows a histopathologic diagnosis and initiation of appropriate treatment, as in the present case. Regarding the length of antifungal therapy, there are currently no definitive guidelines, but it is frequently continued for many months. Due to concerns regarding relapse, it is currently suggested that discontinuation of the antifungal therapy is only attempted if there is a lack of macroscopic disease based on MRI and/or normal CSF findings (1, 14, 16). The unfortunate death of this dog 18 months after surgery during anaesthesia for an unrelated subcutaneous mass removal means that the longer term prognosis and whether relapse could have subsequently been seen is unknown.

Most of the cases of CNS fungal granulomas described in veterinary medicine that have undergone surgery are lesions involving the forebrain, with only one case of a spinal intramedullary lesion in cat (8). Surgical management of intramedullary lesions is rarely reported in veterinary medicine, and knowledge regarding complications and outcome remains limited. Transient postoperative neurological deterioration is, however, a recognized complication of this type of procedure. In human neurosurgery, between 9 and 34% of patients can present an acute neurological deterioration after intramedullary tumour resection which, in between 25 to 41% of cases, will resolve entirely over the course of approximately 6 months (22). In this case, no neurological deterioration was observed, with a recovery of ambulation over the following 2 months. However, the dog had residual paraparesis and pelvic limb ataxia during the following 18 months after surgery, which may have been related to the underlying pathology and/or the nature of the decompressive procedure.

4. Conclusion

In conclusion, this report describes the diagnosis, treatment, and long-term outcome of an intramedullary spinal fungal granuloma in a dog treated with debulking surgery and adjunctive antifungal therapy. This case highlights the importance of obtaining a biopsy in patients with granuloma-like lesions on MRI, especially when CSF analysis reveals marked inflammation with the presence of eosinophils. Surgical resection may offer a good functional outcome even for intramedullary lesions; however, prolonged antifungal treatment remains essential, with ongoing monitoring through follow-up imaging and CSF evaluation to assess therapeutic response.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. Linnaeus Veterinary Limited, which is part of Mars Veterinary Health, supported the cost of the Open Access Publication Charges. The author(s) declared that this work received funding from Linnaeus Veterinary Limited. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article, or the decision to submit it for publication.

Data availability statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Ethics statement

Written informed consent was obtained from the participants for the publication of this case report.

Author contributions

MB-B: Formal analysis, Validation, Writing – review & editing, Methodology, Supervision, Data curation, Project administration, Software, Conceptualization, Visualization, Investigation, Funding acquisition, Writing – original draft, Resources. TE-R: Conceptualization, Validation, Writing – review & editing, Visualization. MB: Writing – review & editing. FC-C: Formal analysis, Resources, Data curation, Writing – review & editing, Methodology. EI: Conceptualization, Investigation, Validation, Supervision, Writing – review & editing, Data curation, Methodology, Visualization.

Conflict of interest

MB-B, TE-R, MB and EI were affiliated to Anderson Moores Veterinary Specialists, Part of Linnaeus Veterinary Limited.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The handling editor KS declared a past co-authorship with author MB-B at the time of review.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.