Intramedullary cervical spinal cord abscess mimicking spinal cord hematoma: a case report
Article information
Abstract
Intramedullary spinal cord abscess of the cervical cord is extremely rare and may present without typical signs of infection, creating a substantial diagnostic challenge. An 85-year-old female patient presented with progressive right hemiparesis and a normal C-reactive protein level. Magnetic resonance imaging revealed a ring-enhancing intramedullary lesion at the C3 level with restricted diffusion, characterized by hyperintensity on diffusion-weighted imaging and hypointensity on the apparent diffusion coefficient map. Emergency surgery revealed a purulent abscess, and cultures yielded Streptococcus constellatus. After surgical drainage and a 6-week course of targeted intravenous antibiotics, the patient showed marked neurological improvement. This case highlights that intramedullary spinal cord abscess can mimic spinal cord hematoma or tumor and that diffusion-weighted imaging with apparent diffusion coefficient mapping is critical for accurate differential diagnosis. Maintaining a high index of suspicion is essential for timely intervention and favorable outcomes.
Introduction
Intramedullary spinal cord abscess (ISCA) is an exceptionally rare infectious disease of the central nervous system. Since its first description by Hart in 1830, over 200 cases have been documented in the literature [1,2]. Despite its rarity, ISCA is associated with devastating consequences, including high rates of permanent neurological morbidity and mortality if not diagnosed and treated promptly [3]. The spinal cord tissue has a remarkable resistance to infection, and its unique vascular anatomy is thought to be a protective factor contributing to the low incidence of this condition [2].
The diagnostic process is frequently complicated by a nonspecific clinical presentation that can mimic a wide range of other spinal pathologies, such as spinal cord tumors (astrocytoma, ependymoma), acute transverse myelitis, or infarction [4]. The classic triad of fever, localized pain, and neurological deficits, while characteristic, is often absent, and studies indicate that fever is present in only 40%–50% of patients at the time of examination [1]. This diagnostic challenge is particularly pronounced when differentiating ISCA from an intramedullary hematoma, especially in cases with acute onset and no systemic signs of infection. Both conditions can present as neurological emergencies requiring urgent intervention, yet their management strategies differ fundamentally. Here, we report a rare case of cervical ISCA in an elderly patient, initially misdiagnosed as a spinal cord hematoma, and provide a detailed discussion of the radiological features crucial for distinguishing between these two critical pathologies.
Case
Ethical approval: This study was approved by the Institutional Review Board (IRB) of Pusan National University Hospital (IRB No. 2512-020-158). Written informed consent was obtained from the patient.
An 85-year-old female patient with a history of hypertension, diabetes mellitus, and an unruptured middle cerebral artery aneurysm was admitted to the emergency room with a 4-day history of progressive right hemiparesis and dysesthesia. She had no fever, recent trauma, or cervical skin abnormalities. On neurological examination, severe right-sided weakness was noted, with Medical Research Council grades 1/5 in the arm and 2/5 in the leg. Hoffmann sign was positive on the right side, while ankle clonus, Babinski sign, and Chaddock sign were absent. Laboratory tests revealed leukocytosis (14.76×10³/μL), but the C-reactive protein (CRP) level was within the normal range (0.24 mg/dL).
Cervical spine magnetic resonance imaging (MRI) demonstrated a distinct intramedullary lesion. Contrast-enhanced T1-weighted images revealed a ring-enhancing mass in the right posterolateral cord at the C3 level, accompanied by cord edema from C2 to C5 and expansion of the spinal cord. Diffusion-weighted imaging showed marked hyperintensity in the core of the lesion, with a corresponding hypointensity on the apparent diffusion coefficient (ADC) map, consistent with restricted diffusion (Fig. 1). Despite these findings, the initial diagnosis favored a spinal cord hematoma or tumor, influenced by the acute presentation without fever or elevated CRP.
Initial magnetic resonance imaging findings. (A) Sagittal T2-weighted image shows an intramedullary lesion (arrow) with extensive edema. (B) Sagittal and axial contrast-enhanced T1-weighted images demonstrate a distinct ring-enhancing lesion (arrows). (C) Sagittal diffusion-weighted image shows marked hyperintensity in the lesion core (arrow), indicating restricted diffusion. (D) Sagittal apparent diffusion coefficient map shows corresponding hypointensity in the lesion core (arrow), confirming restricted diffusion.
Because of rapid neurological deterioration, an emergency C2–4 hemilaminectomy and midline myelotomy were performed the day after admission. Pus drained from the cord, confirming an intramedullary abscess, and the cavity was thoroughly evacuated (Fig. 2). Microbiological culture identified Streptococcus constellatus. The patient received a six-week course of targeted intravenous antibiotics. Follow-up T2-weighted MRI demonstrated near-complete resolution of the abscess and surrounding edema (Fig. 3). After a period of intensive rehabilitation, motor strength improved to grade 3/5 in the right arm and leg.
Intraoperative photograph showing pus drainage after myelotomy. After C2–4 hemilaminectomy and midline myelotomy, purulent material is seen draining from the abscess cavity (arrow) within the right posterolateral cord at the C3 level. Retention sutures are placed along the myelotomy edges to facilitate exposure and evacuation of the abscess.
Discussion
This case exemplifies the diagnostic challenge posed by ISCA because of its atypical presentation. The patient was afebrile with a normal CRP level, which diverted attention from an infectious etiology despite leukocytosis. This highlights the unreliability of the classic infectious triad and indicates that many ISCA cases, particularly those with a subacute or chronic course, may lack systemic inflammatory signs [2,3]. A high index of suspicion is therefore essential, even when inflammatory markers are not elevated.
The causative agent in this case, S. constellatus, belongs to the S. anginosus group, which is known for its strong propensity to form abscesses throughout the body, including the central nervous system [5]. While Staphylococcus aureus is also a common pathogen, the isolation of a Streptococcus species is consistent with the existing literature on ISCA microbiology [3]. In many modern-era reports, the source of infection is cryptogenic, as was likely in this case [2].
Differentiating ISCA from an intramedullary hematoma is particularly difficult, as both conditions may present with acute pain and rapidly progressive neurological deficits. On conventional MRI, both can appear as intramedullary masses with cord expansion and edema. Ring enhancement may be present in either entity, reflecting a vascularized capsule. The signal characteristics of hematomas on T1- and T2-weighted sequences vary with the stage of hemoglobin degradation, which adds to the complexity of diagnosis [6].
Diffusion-weighted imaging with ADC mapping provides crucial functional information and was the key to the correct diagnosis in this case. Pus within an abscess cavity contains inflammatory cells, bacteria, necrotic debris, and proteinaceous material with high viscosity, which restricts water motion and lowers ADC values [7]. This produces the characteristic imaging pattern of hyperintensity on diffusion-weighted imaging with corresponding hypointensity on the ADC map. While some acute hematomas may also demonstrate restricted diffusion, it is usually less pronounced, more heterogeneous, and associated with susceptibility artifacts not typical of abscesses [8]. In contrast, necrotic tumors generally exhibit increased diffusion with high ADC values [7].
The distinction between ISCA and spinal hematoma has immediate therapeutic implications. Confirmed ISCA requires urgent surgical drainage followed by prolonged, targeted antibiotic therapy to prevent irreversible spinal cord injury, sepsis, or death [3]. In contrast, some hematomas can be managed conservatively, and unnecessary surgery carries its own risks [9]. As summarized in Table 1, all three previously reported cervical ISCA cases with documented diffusion-weighted imaging (DWI) findings demonstrated the same characteristic pattern of hyperintensity on DWI with corresponding restricted ADC values, and all were ultimately managed with surgical drainage followed by intravenous antibiotic therapy [3–5]. This consistent imaging pattern across reported cases supports the role of DWI/ADC mapping as the most reliable noninvasive discriminator between ISCA and other intramedullary lesions, including hematoma. Importantly, our case demonstrates that this imaging signature can be present even when the classic systemic signs of infection are entirely absent—a diagnostic pitfall that has not been uniformly emphasized in prior reports. Regarding surgical management, the need for myelotomy and abscess evacuation was consistent across all comparable cases, confirming that conservative management is not appropriate once intramedullary abscess is confirmed. The intraoperative photograph (Fig. 2) clearly demonstrates purulent drainage from the abscess cavity following midline myelotomy, providing direct visual confirmation of the diagnosis. In conclusion, this case underscores that ISCA is a rare but life-threatening neurological emergency that can closely mimic other intramedullary pathologies, most notably spinal cord hematoma. Since reliance on clinical signs such as fever or inflammatory markers can be misleading, DWI with ADC mapping serves as an invaluable, noninvasive diagnostic tool. Recognizing restricted diffusion within a ring-enhancing lesion should prompt strong suspicion for an abscess, and incorporating such advanced MRI techniques is crucial for timely intervention and optimizing neurological outcomes.
Notes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Funding
This research was supported by the Basic Research Support Program of Pusan National University (2026).
Author contributions
Conceptualization: IHH, BKC. Data curation: DHK, SK. Formal analysis: DHK. Investigation: DHK, IHH. Methodology: IHH, BKC. Supervision: BKC. Validation: IHH, BKC. Visualization: SK. Writing-original draft: DHK. Writing-review and editing: IHH, SK, BKC. All authors have read and approved the final manuscript.
