Central nervous system feline infectious peritonitis with a mass lesion in the fourth ventricle revealed by magnetic resonance imaging Authors: Yoshimichi GODA, Yoshiyuki INOUE, Kohei KURODA, Yuki HIDAKA, Hiroyuki SATOH Published: 2026 ABSTRACT An 11-month-old neutered male mixed-breed cat was presented to the referring hospital with the chief complaint of progressive ataxia. Magnetic resonance imaging (MRI) revealed a well-defined, ovoid, solitary mass on the left side of the fourth ventricle. Immunostaining of surgically excised tissue led to a diagnosis of feline infectious peritonitis (FIP). To our knowledge, this is the first report of a central nervous system (CNS)-FIP presenting as a solitary fourth ventricular mass identified by MRI. CNS-FIP should be considered as a differential diagnosis when encountering cases of young cats with a solitary mass lesion in the fourth ventricle. Feline infectious peritonitis (FIP) commonly occurs in cats less than two years of age [15, 17, 20, 26], and is reported to occur more often in purebred [15, 17]. It also occurs at a higher rate in male cats than in female cats [15, 17, 20, 26]. The causative agent is FIP virus, which arises from a mutation in feline enteric coronavirus in cats [18, 25]. Central nervous system (CNS)-FIP, which is present in approximately 35% of affected cats [7], is fatal [4]. A previous study of 24 cats with CNS-FIP reported a median survival time of 14 days (range: 2–115 days) from the onset of clinical signs [4]. The definitive diagnosis of FIP is made by histopathology and immunostaining [16, 23]; however, ante-mortem diagnosis using these methods is generally difficult [5, 22]. Clinically, measurement of feline coronavirus (FCoV) antibody titers in serum and cerebrospinal fluid (CSF) is used to aid in diagnosis [7, 24], although low FCoV antibody titers have been reported in some CNS-FIP cases [2, 7]. In recent years, magnetic resonance imaging (MRI), which enables noninvasive intracranial evaluation, has been used to aid ante-mortem diagnosis of FIP [4, 7, 9, 14]. Previous MRI studies of cats with CNS-FIP have reported ventriculomegaly, periventricular contrast enhancement, meningeal contrast enhancement, choroid plexus contrast enhancement, ependymal contrast enhancement, foramen magnum herniation, syringomyelia, subarachnoid diverticulum, and suppressed CSF signal reduction on fluid-attenuated inversion recovery (FLAIR) [4]. However, to the best of our knowledge, there have been no prior reports of intracranial masses in CNS-FIP detected by MRI. Herein, we report a case of a cat diagnosed postmortem with FIP, in which MRI revealed a mass in the fourth ventricle. An 11-month-old neutered male mixed-breed cat weighing 2.36 kg was presented to Miyazaki University Veterinary Teaching Hospital with a chief complaint of progressive ataxia persisting for 2 months. At presentation (day 1), the cat was recumbent and had showed gait abnormalities. Blood tests revealed lymphopenia, hyperproteinemia, mild hyperglycemia, and slightly increased levels of alanine aminotransferase, aspartate aminotransferase, and creatine kinase (Table 1Table 1.Blood examination results on day 1ParameterResultReference intervalWhite blood cells (/µL)12,9005,500–19,500Neutrophils (/µL)11,3522,500–12,500Lymphocytes (/µL)1,2901,500–7,000Monocytes (/µL)2580–850Eosinophils (/µL)00–1,500Red blood cells (×104/µL)862550–1,000Haemoglobin (g/dL)13.88.0–14.0Haematocrit (%)37.824.0–45.0Mean cell volume (fL)43.940.0–55.0Mean cell haemoglobin (pg)16.013.0–17.0Mean cell haemoglobin concentration (g/dL)36.530.0–36.0Platelets (×104/L)24.330.0–80.0Blood urea nitrogen (mg/dL)19.917.6–32.8Creatinine (mg/dL)1.080.90–2.10Total protein (g/L)7.35.7–7.8Ammonia (µg/dL)2623–78Alanine aminotransferase (U/L)20022–84Alkaline phosphatase (U/L)210–58Aspartate aminotransferase (U/L)6918–51Creatine kinase (U/L)31387–309Glucose (mmol/L)28471–148Total cholesterol (mg/dL)11795–259Calcium (mg/dL)10.28.8–11.9Phosphorus (mg/dL)3.52.6–6.0Magnesium (mg/dL)2.31.6–2.8Sodium (mmol/L)152147–156Potassium (mmol/L)3.93.4–4.6Chloride (mmol/L)112107–120Serum amyloid A (µg/mL) <3.750–5.49). Neurological examination showed that the cat was conscious and had ataxia, tetraparesis, and muscle atrophy of the extremities. The placing reaction was normal. Cranial nerve examination revealed facial asymmetry of the temporalis muscles and decreased jaw tone. Table 1.Blood examination results on day 1ParameterResultReference intervalWhite blood cells (/µL)12,9005,500–19,500Neutrophils (/µL)11,3522,500–12,500Lymphocytes (/µL)1,2901,500–7,000Monocytes (/µL)2580–850Eosinophils (/µL)00–1,500Red blood cells (×104/µL)862550–1,000Haemoglobin (g/dL)13.88.0–14.0Haematocrit (%)37.824.0–45.0Mean cell volume (fL)43.940.0–55.0Mean cell haemoglobin (pg)16.013.0–17.0Mean cell haemoglobin concentration (g/dL)36.530.0–36.0Platelets (×104/L)24.330.0–80.0Blood urea nitrogen (mg/dL)19.917.6–32.8Creatinine (mg/dL)1.080.90–2.10Total protein (g/L)7.35.7–7.8Ammonia (µg/dL)2623–78Alanine aminotransferase (U/L)20022–84Alkaline phosphatase (U/L)210–58Aspartate aminotransferase (U/L)6918–51Creatine kinase (U/L)31387–309Glucose (mmol/L)28471–148Total cholesterol (mg/dL)11795–259Calcium (mg/dL)10.28.8–11.9Phosphorus (mg/dL)3.52.6–6.0Magnesium (mg/dL)2.31.6–2.8Sodium (mmol/L)152147–156Potassium (mmol/L)3.93.4–4.6Chloride (mmol/L)112107–120Serum amyloid A (µg/mL) <3.750–5.49 Blood examination results on day 1 Based on these findings, we suspected a disorder involving both the trigeminal nerve and cerebellum. MRI was performed using a 3.0 T scanner (Vantage Titan 3T®, Canon Medical Systems, Otawara, Japan) with 0.2 mL/kg of gadodiamide hydrate (Omniscan®; GE HealthCare Pharma, Tokyo, Japan) administered as an MRI contrast agent. On the MRI, T2-weighted (T2W), T1-weighted (T1W), and postcontrast T1W images were acquired in the transverse, sagittal, and coronal planes. FLAIR, T2*-weighted (T2*), and diffusion-weighted images, as well as apparent diffusion coefficient (ADC) maps, were acquired only in the transverse plane. MRI revealed a well-defined, ovoid, solitary mass measuring 8.9 × 10.8 × 9.6 mm that extended from the left side of the fourth ventricle to the cerebellum. The mass was mildly hyperintense on T2W images, isointense on T1W images, hyperintense on FLAIR images, mildly hyperintense on diffusion-weighted images, mildly hypointense on the ADC map, and strongly enhanced on post-contrast T1W images. Some contrast defect areas were observed internally. As the contrast defect area showed hyperintensity on T2W images and hypointensity on T1W images, the lesion was suspected to be necrotic. However, because only a portion of the tissue could be removed by the subsequent surgery, necrosis could not be confirmed. The cerebellum appeared diffusely and mildly hyperintense on T2W images, isointense on T1W images, and hyperintense on FLAIR images. The arbor vitae were unclear on transverse images. The choroid plexus from the third to the fourth ventricle through the cerebral aqueduct was contrast-enhanced on contrast T1W images, particularly around the fourth ventricle (Fig. 1Fig. 1.Transverse and sagittal magnetic resonance images showing the mass (8.9 × 10.8 × 9.6 mm) in the fourth ventricle. Imaging included T2-weighted (T2W) (A), T1-weighted (T1W) (B), post-contrast T1W (C), fluid-attenuated inversion recovery (FLAIR) (D), diffusion-weighted images (E), and apparent diffusion coefficient (ADC) map (F) at the level of the fourth ventricle in the transverse plane, post-contrast T1W image at the level of the medulla oblongata in the transverse plane (G), and the midline sagittal plane (H). The post-contrast T1W image showed a well-defined, ovoid, solitary mass (white arrow) with an internal folded contrast defect. The mass was mildly hyperintense on T2W images, isointense on T1W images, hyperintense on FLAIR, mildly hyperintense on diffusion-weighted images, and mildly hypointense on the ADC map. The choroid plexus from the third to fourth ventricle through the cerebral aqueduct was contrast-enhanced on T1W post-contrast images, particularly around the fourth ventricle.). Ventriculomegaly of the lateral and third ventricles, interthalamic adhesion deformation, elevation of the mesencephalic tectum, and narrowing of the cerebral sulcus were also observed. The intraventricular area was hyperintense on T2W images and hypointense on FLAIR images (normal CSF signal intensity). The periventricular area showed hyperintensity on FLAIR images, hyperintensity on diffusion-weighted images, hypointensity on the ADC map, and was not contrast-enhanced on contrast T1W images. Based on these findings, cytotoxic edema was suspected (Fig. 2A–DFig. 2.Images of post-contrast T1-weighted (A), fluid attenuated inversion recovery (FLAIR) (B), diffusion-weighted (C), apparent diffusion coefficient map (D) at the level of interthalamic adhesion in the transverse plane, T2-weighted (E) in the sagittal plane, and T2-weighted (F), T1-weighted (G), and post-contrast T1-weighted (H) at the level of second cervical vertebra in the transverse plane. The intraventricular area showed no contrast enhancement on T1W postcontrast images and was hypointense on FLAIR images (normal CSF-like signal intensity). (D) The periventricular region tends to show hypointensity on ADC maps, even in normal cases. However, in the present case, the region indicated by the yellow arrowhead shows hypointensity compared to the surrounding white matter. (B, D) Hyperintense regions (red arrows) with bilateral symmetry were observed; this is thought to represent the partial volume effect. A filamentous area of hypointensity was observed on the T1W images and hyperintensity was observed on the T2W images in the dorsal part of the dorsal columns from the level of the second cervical vertebra to the level of the first thoracic vertebra (white arrow). The vessel wall of the internal vertebral venous plexus showed mild hyperintensity on T1W and T2W images and clear contrast enhancement on post-contrast T1W images (white arrowhead).). On the ADC map, the white matter, including the periventricular region, tended to show hypointensity compared with the gray matter, even in normal areas. However, in this case, the periventricular region showed hypointensity, even compared with the white matter (Fig. 2D, yellow arrowhead). The FLAIR and ADC maps showed continuous bilateral symmetrical hyperintense regions in the ventricles, which were thought to represent a partial volume effect (Fig. 2B, 2D, red arrow). Based on findings in the right olfactory bulb, a dilated olfactory recess or olfactory bulb cavity dilation (9.1 × 4.3 × 10.9 mm), with normal CSF-like signal intensity, was suspected. In the cervical spinal cord, the meninges of the spine, spinal nerve roots, and vessel wall of the internal vertebral venous plexus were strongly enhanced on contrast-enhanced T1W images (Fig. 2F–H). The spinal cord parenchyma at the level of the first cervical vertebra was ill-defined and diffusely hyperintense on T2W and FLAIR images, but showed no contrast enhancement on contrast T1W images; as a result, edema was suspected. We also identified a filamentous area of hypointensity on T1W images and hyperintensity on T2W images in the dorsal part of the dorsal columns, from the level of the second cervical vertebra to the level of the first thoracic vertebra (Fig. 2E). This finding suggested expansion of the central canal or edema. No enlargement of the mandibular, medial retropharyngeal, or superficial cervical lymph nodes was observed on MRI. CSF lines were observed in the rhomboid fossa of the fourth ventricle, but not in the roof of the fourth ventricle. As the arbor vitae were clearly visible on the sagittal plane, the mass was determined to originate from the fourth ventricle. Based on these imaging findings, neoplastic diseases such as ependymoma, lymphoma, or choroid plexus tumors were suspected [6, 10, 11]. Given the cat’s age and previous reports, ependymoma was considered the most likely diagnosis [10]. Despite attempts, CSF samples could not be collected for analysis. Transverse and sagittal magnetic resonance images showing the mass (8.9 × 10.8 × 9.6 mm) in the fourth ventricle. Imaging included T2-weighted (T2W) (A), T1-weighted (T1W) (B), post-contrast T1W (C), fluid-attenuated inversion recovery (FLAIR) (D), diffusion-weighted images (E), and apparent diffusion coefficient (ADC) map (F) at the level of the fourth ventricle in the transverse plane, post-contrast T1W image at the level of the medulla oblongata in the transverse plane (G), and the midline sagittal plane (H). The post-contrast T1W image showed a well-defined, ovoid, solitary mass (white arrow) with an internal folded contrast defect. The mass was mildly hyperintense on T2W images, isointense on T1W images, hyperintense on FLAIR, mildly hyperintense on diffusion-weighted images, and mildly hypointense on the ADC map. The choroid plexus from the third to fourth ventricle through the cerebral aqueduct was contrast-enhanced on T1W post-contrast images, particularly around the fourth ventricle. Images of post-contrast T1-weighted (A), fluid attenuated inversion recovery (FLAIR) (B), diffusion-weighted (C), apparent diffusion coefficient map (D) at the level of interthalamic adhesion in the transverse plane, T2-weighted (E) in the sagittal plane, and T2-weighted (F), T1-weighted (G), and post-contrast T1-weighted (H) at the level of second cervical vertebra in the transverse plane. The intraventricular area showed no contrast enhancement on T1W postcontrast images and was hypointense on FLAIR images (normal CSF-like signal intensity). (D) The periventricular region tends to show hypointensity on ADC maps, even in normal cases. However, in the present case, the region indicated by the yellow arrowhead shows hypointensity compared to the surrounding white matter. (B, D) Hyperintense regions (red arrows) with bilateral symmetry were observed; this is thought to represent the partial volume effect. A filamentous area of hypointensity was observed on the T1W images and hyperintensity was observed on the T2W images in the dorsal part of the dorsal columns from the level of the second cervical vertebra to the level of the first thoracic vertebra (white arrow). The vessel wall of the internal vertebral venous plexus showed mild hyperintensity on T1W and T2W images and clear contrast enhancement on post-contrast T1W images (white arrowhead). The cat returned to the hospital on day 6 due to worsening respiratory condition. As increased intracranial pressure was suspected, decompressive craniectomy was performed the next day. Surgery was performed with the cat in a prone position with the neck flexed. A suboccipital craniotomy was performed. After removing part of the occipital bone, the cerebellum was gently lifted to expose the lesion; however, the entire lesion could not be sufficiently exposed. The lesion was well visualized by fluorescein angiogram, with a clear border between the lesion and cerebellum. In contrast, the border between the lesion and the dorsal aspect of the medulla oblongata was unclear. Because the lesion was fragile and adhered to the adjacent brain tissue, only partial resection of the mass was possible. Postoperative MRI was not performed. The cat did not resume spontaneous respiration after surgery and received ventilatory support until day 9. As the cat’s respiratory condition did not improve, the owner requested euthanasia. Autopsy was not performed because the owner’s consent could not be obtained. Histopathological examination of the resected tissue revealed inflammatory cell infiltrates, mainly consisting of plasma cells and lymphocytes. Immunohistochemistry demonstrated granular positivity for anti-coronavirus antibodies within the cytoplasm of macrophages (Fig. 3Fig. 3.Representative images of (A) hematoxylin and eosin staining and (B) immunohistochemistry for anti-coronavirus antibodies. Numerous inflammatory cells—predominantly plasma cells and lymphocytes— infiltrated the tissue obtained from the fourth ventricular lesion (A). Granular cytoplasmic positivity was observed in a subset of macrophages (red arrows) (B).). A mouse antifeline coronavirus antibody (clone FIPV3-70; Bio-Rad Laboratories Inc., Hercules, CA, USA) was used at a dilution of 1:500. Antigen retrieval was performed by heating tissue sections in a citric acid buffer solution at 121°C for 10 min. Due to the limited amount of resected tissue, comprehensive assessment of the lesion architecture was not possible. The specimen consisted of inflammatory and cerebellar tissue, and no abnormalities were identified in the cerebellar component. Based on the intraoperative and histopathological findings, the lesion was determined to be located in the fourth ventricle. Furthermore, postmortem serum antibody tests for Toxoplasma gondii immunoglobulin (Ig) G, feline leukemia virus, and feline immunodeficiency virus were negative. However, FCoV IgG was positive with a high titer (1:3,200) (reference interval: 1:400). The serum protein fraction (albumin/globulin ratio) was within the reference range (Table 2Table 2.Results of serum and plasma examinations after euthanasiaParameterResultToxoplasma gondii serologyIgG antibody not detectedFeline leukaemia virus serologyAntigen not detectedFeline immunodeficiency virus serologyAntibody not detectedFeline coronavirus serologyPositive (antibody titer ≥3,200)Albumin:globulin ratio1.08). Based on these findings, CNS-FIP was diagnosed. Representative images of (A) hematoxylin and eosin staining and (B) immunohistochemistry for anti-coronavirus antibodies. Numerous inflammatory cells—predominantly plasma cells and lymphocytes— infiltrated the tissue obtained from the fourth ventricular lesion (A). Granular cytoplasmic positivity was observed in a subset of macrophages (red arrows) (B). Table 2.Results of serum and plasma examinations after euthanasiaParameterResultToxoplasma gondii serologyIgG antibody not detectedFeline leukaemia virus serologyAntigen not detectedFeline immunodeficiency virus serologyAntibody not detectedFeline coronavirus serologyPositive (antibody titer ≥3,200)Albumin:globulin ratio1.08 Results of serum and plasma examinations after euthanasia Herein, we describe a young cat with a solitary lesion in the fourth ventricle that was diagnosed as FIP postmortem. Previous studies have reported inflammatory lesions in the meninges, choroid plexus, ependymal layer of the fourth ventricle, solitary mass lesions in the fourth ventricle, and vasculitis and/or perivasculitis as pathological findings in CNS-FIP [3, 4, 7, 13, 21]. However, to the best of our knowledge, this is the first report of a solitary mass lesion in the fourth ventricle revealed by MRI. Recent reports have described various MRI findings of CNS-FIP [4, 7, 9, 11, 14]. Of the 24 reported CNS-FIP cases, the CSF signal intensity was not suppressed on FLAIR images in 31.6%, ventricular enlargement was observed in 86.9%, periventricular interstitial edema in 78.9%, syringomyelia in 73.9%, compression and distortion of interthalamic adhesions in 30.4%, and contrast enhancement of the ependyma in 83.3%. Contrast enhancement of the ependymal layers of the third and fourth ventricles was reported in most cases [4]. Consistent with these previous reports, the present case showed syringomyelia and ventricular enlargement, with contrast enhancement of the ependymal layers of the third and fourth ventricles and cervical meninges. Contrast enhancement of the ependymal layers of the third and fourth ventricles has been reported to be associated with ventricular enlargement, which may be caused by altered CSF flow due to physical obstruction by inflammatory lesions in the fourth ventricle [4]. In the present case, we suspected that contrast enhancements were caused by meningeal carcinomatosis, although there are limited reports of leptomeningeal spread of ependymomas [8]. Furthermore, the cerebral parenchyma around the lateral ventricles was hyperintense on T2W and FLAIR images without contrast enhancement. This feature is consistent with interstitial edema and presumably occurs secondary to pressure from the lateral ventricles of the brain parenchyma [12]. Thus, some MRI findings in this case were similar to previous reports. However, the presence of a mass lesion with well-defined contrast enhancement in the fourth ventricle and some MRI findings (i.e., suppression of the CSF signal intensity in the ventricle on FLAIR images) differed from previous reports of CNS-FIP. The cause of the lack of contrast enhancement around the lateral ventricles was difficult to determine as an autopsy was not performed. Administration of prednisolone may have had an anti-inflammatory effect in this case. A previous report of pathological findings in cats with CNS-FIP described three types based on the location of inflammation; however, the causes of inflammation at different locations remain unknown [21]. Based on this classification, the present case is the most common type, which accounts for 46% of cases, and is accompanied by inflammation around the ventricles. The same report also describes the presence of fibrinous lesions in the fourth ventricle of a 7-month-old cat diagnosed with CNS-FIP [21]. These findings suggest that CNS-FIP should be included in the differential diagnosis when mass lesions are detected on MRI in the fourth ventricle of young cats. Due to differences in the prognosis and treatment of FIP and other neoplastic diseases [4], it is important to obtain informed consent for prognosis and treatment from the owner after MRI. The median survival time for CNS-FIP is 14 days, although treatment with the antiviral drugs remdesivir and nupiravir has been recently attempted [19]. Surgical excision and radiation therapy are the most common treatments for brain tumors in cats, and a combination of radiation therapy and ventriculoperitoneal shunting has been reported to result in the longest survival time, of 1,103 days [1]. In this case report, we describe what is, to the best of our knowledge, the first MRI findings of a cat with CNS-FIP presenting a solitary mass in the fourth ventricle. Additionally, we believe this is the first study to report MRI findings using DWI and ADC mapping in cats with CNS-FIP. Given these findings, we propose that CNS-FIP should be considered alongside other neoplastic diseases in young cats with mass lesions in the fourth ventricle observed on MRI. Conflicts of Interest All authors have no conflict of interest.