Subacute anterior spontaneous spinal epidural hematoma at the thoracolumbar junction: A diagnostic confounder following necrotizing staphylococcal pneumonia Authors: Anwar Kalali, Soufiane Kaddouri, Jalil E. Fakir, Achraf Saidi, Hamza Retal, Najwa E.-Cherif Kettani, Meriem Fikri, Firdaous Touarsa Published: 2026 ABSTRACT Spontaneous spinal epidural hematoma (SSEH) is a rare cause of acute spinal cord compression, accounting for less than 1% of spinal space-occupying lesions. Its ventral localization is exceptionally uncommon due to the firm adherence of the dura to the posterior longitudinal ligament. We report the case of a 61-year-old woman with type 1 diabetes and a remote history of necrotizing Staphylococcus aureus pneumonia (treated 5 months prior) who presented with a 1-week history of progressive paraparesis and marked biological inflammatory markers (Leukocytes: 19,800/mm³; Eosinophils: 6800/mm³). A multimodal radiological evaluation was performed. Chest CT revealed a sequelar right apical cavity, while brain MRI was unremarkable. Spinal MRI demonstrated a D11-L1 anterior biconvex epidural collection (hyperintense on T1 and hypointense on T2-weighted images), associated with L4-S1 intrathecal hematic infiltration. These findings are pathognomonic for intracellular methemoglobin in a late subacute hematoma. Despite the pulmonary cavity and the biological “red herrings,” MRI signal analysis correctly ruled out a late-onset epidural abscess. Emergency laminectomy confirmed the diagnosis, and the patient showed significant recovery at 3 months. This case highlights the role of MRI biophysics in deconstructing diagnostic traps and the importance of recognizing the methemoglobin signature in complex clinical scenarios. Introduction Spontaneous spinal epidural hematoma (SSEH) is a rare neurosurgical emergency [1]. While posterior locations are more frequent due to the greater anatomical space in the dorsal spinal canal, anterior SSEH is reported in fewer than 20 cases in the literature [2,3]. Diagnosing anterior SSEH is particularly challenging as its ventral location and clinical presentation frequently overlap with infectious processes such as spinal epidural abscess (SEA), neoplastic lesions, or sequestered disc herniations. This diagnostic dilemma is further amplified in patients with systemic inflammatory markers and significant pulmonary sequelae from prior necrotizing infections. In such scenarios, the clinician is often led toward a disseminated infectious etiology (eg, Pott’s disease). However, the temporal evolution of hemoglobin degradation products on MRI provides a unique biophysical “signature” that allows for a confident diagnosis. We report a rare case where a sequelar lung cavity, marked leukocytosis, and a remote history of necrotizing Staphylococcus aureus pneumonia acted as major diagnostic confounders, and where precise MRI signal analysis was pivotal in identifying a subacute ventral SSEH. Case report Clinical presentation A 61-year-old woman with a long-standing history of type 1 diabetes mellitus presented to our emergency department with a 1-week history of progressive bilateral lower limb weakness and associated upper limb pain. Over the 48 hours preceding admission, she developed urinary retention and fecal incontinence. Her medical history was notable for a severe episode of necrotizing Staphylococcus aureus pneumonia 5 months prior, which had been successfully treated with intravenous antibiotics and bronchoalveolar lavage. She denied any recent trauma or use of anticoagulant/antiplatelet medications. On examination, she had spastic paraparesis (MRC grade 3/5) and a T11 sensory level. Sphincter tone was markedly diminished. Laboratory findings Biological workup revealed significant leukocytosis (19,800/mm³) with neutrophilia (10,300/mm³) and marked hypereosinophilia (6800/mm³). The platelet count was mildly elevated at 466,000/mm³. Importantly, there was no coagulopathy (normal PT, aPTT, and INR). These inflammatory markers, combined with her medical history, initially suggested a recurrent or late-onset septic process. Imaging findings A comprehensive multimodal imaging workup was performed:•Chest CT: Revealed a sequelar right apical cavitary image with thickened walls (Fig. 1), consistent with the history of necrotizing pneumonia. No active parenchymal infiltrates were noted.Fig. 1Chest computed tomography (axial view, lung window). Residual right apical cavitary image with thickened walls (red arrow), appearing as a sequela of a necrotizing Staphylococcus aureus pneumonia 5 months prior.Fig 1 – dummy alt text•Brain MRI: Remained unremarkable, ruling out intracranial hemorrhage, septic emboli, or secondary sedimented subarachnoid blood.•Spinal MRI (Dorso-lumbar):○D11-L1 Level: An extensive anterior epidural collection was identified, extending from D11 to L1. The lesion was biconvex (lenticular) in the sagittal plane, measuring 6 mm in maximum thickness, causing significant cord compression. The collection was homogeneously hyperintense on T1-weighted images (Fig. 2) and markedly hypointense on T2-weighted images (Fig. 3). Postcontrast images showed a thin, smooth peripheral rim enhancement with no central enhancement.Fig. 2Sagittal T1-weighted MRI of the thoracolumbar spine. Extensive anterior epidural collection from D11 to L1 showing a spontaneous homogeneous hyperintense signal (white arrow), characteristic of methemoglobin.Fig 2 – dummy alt textFig. 3Sagittal T2-weighted MRI of the thoracolumbar spine. The anterior epidural collection appears markedly hypointense (white arrow), causing significant posterior displacement and compression of the spinal cord. This “T1-high/T2-low” signal mismatch identifies the late subacute stage.Fig 3 – dummy alt text○L4-S1 level: Extensive intrathecal hematic infiltration was noted from L4 to S1, appearing hyperintense on T1 and isosignal on T2 sequences (Fig. 4).Fig. 4Sagittal T1 and T2 weighted MRI of the lumbo-sacral spine. Spontaneous T1 hyperintense and T2 isosignal within the intrathecal space from L4 to S1 (arrowheads), consistent with sedimented hematic infiltration associated with the primary epidural event.Fig 4 – dummy alt text○Vertebral analysis: No disc or vertebral signal abnormalities were found, ruling out spondylodiscitis or vertebral metastasis. •Chest CT: Revealed a sequelar right apical cavitary image with thickened walls (Fig. 1), consistent with the history of necrotizing pneumonia. No active parenchymal infiltrates were noted.Fig. 1Chest computed tomography (axial view, lung window). Residual right apical cavitary image with thickened walls (red arrow), appearing as a sequela of a necrotizing Staphylococcus aureus pneumonia 5 months prior.Fig 1 – dummy alt text Chest CT: Revealed a sequelar right apical cavitary image with thickened walls (Fig. 1), consistent with the history of necrotizing pneumonia. No active parenchymal infiltrates were noted.Fig. 1Chest computed tomography (axial view, lung window). Residual right apical cavitary image with thickened walls (red arrow), appearing as a sequela of a necrotizing Staphylococcus aureus pneumonia 5 months prior.Fig 1 – dummy alt text Chest computed tomography (axial view, lung window). Residual right apical cavitary image with thickened walls (red arrow), appearing as a sequela of a necrotizing Staphylococcus aureus pneumonia 5 months prior. •Brain MRI: Remained unremarkable, ruling out intracranial hemorrhage, septic emboli, or secondary sedimented subarachnoid blood. Brain MRI: Remained unremarkable, ruling out intracranial hemorrhage, septic emboli, or secondary sedimented subarachnoid blood. •Spinal MRI (Dorso-lumbar): Spinal MRI (Dorso-lumbar): ○D11-L1 Level: An extensive anterior epidural collection was identified, extending from D11 to L1. The lesion was biconvex (lenticular) in the sagittal plane, measuring 6 mm in maximum thickness, causing significant cord compression. The collection was homogeneously hyperintense on T1-weighted images (Fig. 2) and markedly hypointense on T2-weighted images (Fig. 3). Postcontrast images showed a thin, smooth peripheral rim enhancement with no central enhancement.Fig. 2Sagittal T1-weighted MRI of the thoracolumbar spine. Extensive anterior epidural collection from D11 to L1 showing a spontaneous homogeneous hyperintense signal (white arrow), characteristic of methemoglobin.Fig 2 – dummy alt textFig. 3Sagittal T2-weighted MRI of the thoracolumbar spine. The anterior epidural collection appears markedly hypointense (white arrow), causing significant posterior displacement and compression of the spinal cord. This “T1-high/T2-low” signal mismatch identifies the late subacute stage.Fig 3 – dummy alt text D11-L1 Level: An extensive anterior epidural collection was identified, extending from D11 to L1. The lesion was biconvex (lenticular) in the sagittal plane, measuring 6 mm in maximum thickness, causing significant cord compression. The collection was homogeneously hyperintense on T1-weighted images (Fig. 2) and markedly hypointense on T2-weighted images (Fig. 3). Postcontrast images showed a thin, smooth peripheral rim enhancement with no central enhancement.Fig. 2Sagittal T1-weighted MRI of the thoracolumbar spine. Extensive anterior epidural collection from D11 to L1 showing a spontaneous homogeneous hyperintense signal (white arrow), characteristic of methemoglobin.Fig 2 – dummy alt textFig. 3Sagittal T2-weighted MRI of the thoracolumbar spine. The anterior epidural collection appears markedly hypointense (white arrow), causing significant posterior displacement and compression of the spinal cord. This “T1-high/T2-low” signal mismatch identifies the late subacute stage.Fig 3 – dummy alt text Sagittal T1-weighted MRI of the thoracolumbar spine. Extensive anterior epidural collection from D11 to L1 showing a spontaneous homogeneous hyperintense signal (white arrow), characteristic of methemoglobin. Sagittal T2-weighted MRI of the thoracolumbar spine. The anterior epidural collection appears markedly hypointense (white arrow), causing significant posterior displacement and compression of the spinal cord. This “T1-high/T2-low” signal mismatch identifies the late subacute stage. ○L4-S1 level: Extensive intrathecal hematic infiltration was noted from L4 to S1, appearing hyperintense on T1 and isosignal on T2 sequences (Fig. 4).Fig. 4Sagittal T1 and T2 weighted MRI of the lumbo-sacral spine. Spontaneous T1 hyperintense and T2 isosignal within the intrathecal space from L4 to S1 (arrowheads), consistent with sedimented hematic infiltration associated with the primary epidural event.Fig 4 – dummy alt text L4-S1 level: Extensive intrathecal hematic infiltration was noted from L4 to S1, appearing hyperintense on T1 and isosignal on T2 sequences (Fig. 4).Fig. 4Sagittal T1 and T2 weighted MRI of the lumbo-sacral spine. Spontaneous T1 hyperintense and T2 isosignal within the intrathecal space from L4 to S1 (arrowheads), consistent with sedimented hematic infiltration associated with the primary epidural event.Fig 4 – dummy alt text Sagittal T1 and T2 weighted MRI of the lumbo-sacral spine. Spontaneous T1 hyperintense and T2 isosignal within the intrathecal space from L4 to S1 (arrowheads), consistent with sedimented hematic infiltration associated with the primary epidural event. ○Vertebral analysis: No disc or vertebral signal abnormalities were found, ruling out spondylodiscitis or vertebral metastasis. Vertebral analysis: No disc or vertebral signal abnormalities were found, ruling out spondylodiscitis or vertebral metastasis. Management and follow-up The patient underwent emergency D11-L1 laminectomy. Intraoperatively, a dark, “curd-like” organizing hematoma was evacuated from the ventral space. No pus was found. Histopathology confirmed an organizing hematoma with degraded erythrocytes and fibrin; microbiology was negative. At 3-month follow-up, the patient demonstrated significant recovery (MRC 4+/5) and regained full sphincter control, ambulating with minimal assistance. Discussion Anatomical challenges and localization The spinal epidural space is anatomically narrower ventrally than dorsally. Ventrally, the dura mater is firmly anchored to the posterior longitudinal ligament and the vertebral bodies by fibrous bands known as Hofmann’s ligaments [2]. These anatomical barriers restrict the expansion of ventral collections, explaining why anterior SSEH is rare and typically presents with a focal, lenticular (biconvex) morphology that causes severe cord compression [2,3]. The diagnostic trap: “Red herrings” and biomarkers The most striking feature of this case was the major diagnostic trap created by the laboratory and CT findings. The leukocytosis (19,800/mm³) and neutrophilia, combined with the sequelar lung cavity, were highly suggestive of a spinal epidural abscess (SEA). In the literature, Staphylococcus aureus is the leading cause of pyogenic spinal infections, often requiring a high index of suspicion in postpneumonia patients [4,5]. However, leukocytosis in SSEH is increasingly recognized as a physiological stress response to acute cord compression and severe pain rather than a mandatory sign of infection [3,6]. The marked hypereosinophilia observed here acted as an additional clinical “red herring”. MRI biophysics: the methemoglobin signature The diagnostic cornerstone of this case lies in the interpretation of the paramagnetic effects of hemoglobin degradation. The “T1-high/T2-low” signal mismatch is pathognomonic for the late subacute phase (3-7 days) [1,7].1.T1-hyperintensity: This results from the paramagnetic effect of methemoglobin, which shortens the T1 relaxation time of water protons [6,7].2.T2-hypointensity: This signal indicates that methemoglobin remains intracellular (within intact red blood cells). The compartmentalization of paramagnetic iron causes magnetic field inhomogeneities and proton dephasing, leading to a marked drop in T2 signal [1,6].Furthermore, the thin, smooth peripheral enhancement represents vascularized granulation tissue forming around the organizing clot, which must be distinguished from the thick, irregular wall of an abscess [3,6]. 1.T1-hyperintensity: This results from the paramagnetic effect of methemoglobin, which shortens the T1 relaxation time of water protons [6,7]. T1-hyperintensity: This results from the paramagnetic effect of methemoglobin, which shortens the T1 relaxation time of water protons [6,7]. 2.T2-hypointensity: This signal indicates that methemoglobin remains intracellular (within intact red blood cells). The compartmentalization of paramagnetic iron causes magnetic field inhomogeneities and proton dephasing, leading to a marked drop in T2 signal [1,6].Furthermore, the thin, smooth peripheral enhancement represents vascularized granulation tissue forming around the organizing clot, which must be distinguished from the thick, irregular wall of an abscess [3,6]. T2-hypointensity: This signal indicates that methemoglobin remains intracellular (within intact red blood cells). The compartmentalization of paramagnetic iron causes magnetic field inhomogeneities and proton dephasing, leading to a marked drop in T2 signal [1,6]. Furthermore, the thin, smooth peripheral enhancement represents vascularized granulation tissue forming around the organizing clot, which must be distinguished from the thick, irregular wall of an abscess [3,6]. Dual-compartment involvement and pathogenesis A unique feature of this case was the associated intrathecal blood (L4-S1). This “dual-compartment” hemorrhage suggests that the initial rupture of the valve-less internal vertebral venous plexus (Batson's plexus) created enough hydrostatic pressure to cause a minor dural tear or a concomitant subarachnoid extension [8]. Predisposing factors such as diabetic microvascular fragility and the systemic stress of a prior necrotizing infection likely favored this venous rupture [3,9]. Prognosis and surgical timing While early intervention is critical, our patient showed significant recovery despite being operated on 7 days postonset. This underscores that in subacute cases with progressive deficits, late surgical decompression can still prevent permanent myelomalacia and lead to favorable functional outcomes [1,9]. Conclusion Anterior SSEH is a rare but formidable “mimicker” of spinal infectious processes. This case demonstrates that even in the presence of marked leukocytosis, hypereosinophilia, and sequelar pulmonary cavitation, a meticulous analysis of MRI signal biophysics—specifically the intracellular methemoglobin signature—remains the most reliable tool to identify a subacute hematoma. Recognizing these pathognomonic imaging patterns allows the radiologist to bypass clinical “red herrings,” preventing diagnostic delays and guiding prompt surgical decompression. Furthermore, this report underscores that surgical evacuation remains highly beneficial even when performed in the late subacute phase, offering a significant chance for neurological recovery. Ethics approval Our institution does not require ethical approval for reporting individual cases. Patient consent A written informed consent was obtained from the patient for the publication of this case report.