Intervertebral disc disease, including intervertebral disc extrusions and protrusions, is the most common spinal cord disorder in dogs. Atypical and uncommon intervertebral disc herniations include intradural/intramedullary disc extrusion, intervertebral foraminal disc extrusion and intervertebral disc herniation (Schmorl's node). Intradural/extramedullary disc extrusion is the least common type of intervertebral disc herniation in veterinary medicine, characterized by extruded disc material within the intradural space. To date, only one study has been published in veterinary medicine reporting intradural/extramedullary disc extrusions. In this study, low field MRI was used, and the authors could not find any MRI features to diagnose with confidence an intradural/extramedullary disc location of the extruded disc material. The aim of this study was to describe the high field (1.5T) MRI characteristics of surgically confirmed intradural/extramedullary disc extrusions. This is a retrospective, multicentric and descriptive study. Inclusion criteria was surgical confirmation of intradural/extramedullary disc extrusion by durotomy and complete MRI study of the spine. Seven cases were included. Images were reviewed by a radiology resident and a certified radiologist, with emphasis on the following signs: “Golf-tee sign” (widening of the subarachnoid space cranial and caudal to the lesion), “Beak sign” (pointed and sharp compressive lesion) and “Y sign” (division of the dura and arachnoid layers). MRI showed a “Y sign” in all the cases (7/7) seen from the T2-weighted sagittal views, while “Golf-tee sign” was not recognized in any of the cases (0/7). Additionally, “beak sign” was present in half of the cases (4/7). “Y sign” maybe a reliable MRI feature for identifying intradural/medullary disc extrusions from the MRI study. As the arachnoid is peeled from the dura by the disc herniation there is a splitting of the arachnoid mater and the ventral dura. The intradural disc material will be surrounded by CSF signal intensity margin, giving the appearance of a Y, which can be identified from the T2-weighted sagittal images.

A 5‐month‐old, male, entire labrador retriever (case 1) and a 6‐year‐old, female, neutered greyhound (case 2) presented following chronic, progressive, non‐painful paraparesis in both cases with L4–Cd and L4–S1 neurolocalisation respectively. Magnetic resonance imaging of the lumbosacral region revealed multifocal muscular changes in both cases with patchy T2‐weighted and water‐weighted T2‐weighted Dixon hyperintensities, with moderate contrast enhancement throughout the gluteal and lumbar paraspinal muscles. Multifocal radiculoneuropathy in case 1 and meningomyelitis in case 2 were identified. Crucially, in case 1, the identification of abnormal muscles on magnetic resonance imaging allowed the diagnosis of neosporosis via polymerase chain reaction on targeted muscle biopsy when previously indirect fluorescent antibody test serology was inconclusive and polymerase chain reaction on lumbar cerebrospinal fluid was negative. These cases highlight the magnetic resonance imaging findings of canine lumbar neosporosis and offer an alternative route of diagnosis through targeted muscle biopsy.