The anterior epidural space is mostly occupied by the dura, posterior longitudinal ligament, and periosteum of the vertebral body, which are closely adherent. Hence, SEA occurs in the posterior epidural space (4). The etiopathogenic mechanism of spinal involvement in our patient remains to be determined. However, our case is unique as the cervicothoracic abscess occurred in the anterior epidural space, while the lumbosacral abscess occurred in the posterior epidural space, suggesting a possible route of infection. Tetsuka et al. (5) believed that SEAs secondary to septic spondylitis or intervertebral discitis tend to be located anterior to the dural tube, while those due to hematogenous infections tend to be located posterior to the dural tube. The vertical sheath of the epidural space enables the spread of abscesses from the level of origin to multiple levels longitudinally. Infections can enter the epidural space through various mechanisms, including hematogenous spread (50%), direct extension from adjacent infection (33%), inoculation from spinal procedures (15%), and unknown mechanisms (6, 7). The incidence of infection after combined spinal-epidural anesthesia is approximately 1/2000 (5, 6). Other potential risk factors include diabetes, human immunodeficiency virus (HIV) infection, trauma, tattooing, acupuncture, and infection of the adjacent bone or soft tissue (8, 9). Additionally, degenerative disk disease, large osteophytes, and chronically hypertrophied facet joints may be targets of hematogenous spread (10). Hematogenous spread is commonly caused by methicillin-resistant Staphylococcus aureus (MRSA), and the exotoxins, extracellular enzymes, and cell surface substances of MRSA can cause ischemic changes and tissue destruction in the spinal cord through various mechanisms, ultimately leading to spinal cord injury (11). Other bacteria causing hematogenous spread include negative cultures (13.9%), Gram-negative bacteria (8.1%), coagulase-negative staphylococci (7.5%), and Streptococcus anginosus group (SAG, 6.8%) (12).

Blood culture following an admission of our patient indicated S. intermedius. S. intermedius is one of the three species found in the SAG, also known as the Streptococcus milleri group (SMG) (2). SAG is a commensal in the oral and gastrointestinal tracts. But SAG is notorious for causing invasive infections, including head and neck abscesses, bacteremia with endocarditis, liver abscess, thoracic empyema, brain abscess, and spinal epidural abscess (13, 14). Although rare, patients with underlying diseases such as cirrhosis, diabetes, and malignancy are susceptible to SAG infection (15). We reported a case of an extensive SEA treated conservatively, along with a review of relevant literature and discussions on SEA caused by S. intermedius. We conducted a PubMed search using the keywords “Streptococcus anginosus group abscess” and “Streptococcus intermedius epidural abscess,” to identify reports and studies on spinal abscess caused by S. intermedius. Only 6 cases were found, including the present case, with an average patient age of 61 (standard deviation: 18.01) years, and four of the seven patients were women. In contrast to the case report mentioned in Table 1 (14, 16–20), this patient had no known risk factors such as dental disease, surgery, or trauma. After 2 months of IV antibiotic treatment, the patient showed partial resolution of the epidural abscess on MRI. Gangone et al. (21) reported a case of an immunocompetent 72-year-old patient with a complaint of simple low back pain, who was found to have co-infection with S. intermedius and Streptococcus griseus. The patient was treated with IV antibiotics and recovered completely. Although Gangone et al. (21). found no such dental intervention in their patient’s medical history, poor dental hygiene is a known predisposing factor for spinal abscesses and infective endocarditis. Similarly, our patient was immunocompetent, only had lumbar disc herniation and lumbar spondylolisthesis located at L5 level. Although we have not yet determined the exact cause of the patient’s extensive SEA, the combination of imaging and medical history suggests that multiple confounding factors may have contributed to the condition. We believe this case highlights the importance of considering S. intermedius as a potential causative agent of SEA, particularly in patients with underlying diseases.

SEA is typically characterized by the triad of fever, back pain, and neurological dysfunction. However, only 0.8% of patients exhibit this classic presentation on admission, making the diagnosis of SEA challenging (22). The prevalence of back pain and the rarity of SEA create a diagnostic conundrum, resulting in delayed diagnosis in 75–89% of cases (23). In this case, the patient presented with typical symptoms, progressing through four stages of back pain, nerve root symptoms, muscle weakness, and paresthesia, ultimately leading to complete paralysis within 1 day of entering the third stage. This progression is consistent with Heusner’s description of the four progressive stages of clinical presentation of SEA, which are variable in duration and may result in unpredictable neurological deterioration, underscoring the importance of prompt diagnosis and treatment (24). Although abnormal laboratory values such as leukocytosis or elevated inflammatory markers and isolation of pathogenic pathogens from blood cultures may be predictive of disease severity in established diseases, they are not specific for diagnosis (25). Furthermore, in this case, the patient’s MRI performed before admission did not reveal SEA, further highlighting the diagnostic challenge associated with this condition.

In suspected cases of SEA, imaging should be promptly performed. While lumbar puncture results showing perimembranous inflammation can aid in the diagnosis, this method is invasive and carries the risk of spreading the infection to the subarachnoid space or causing meningitis, as is the case with CT myelography (26). Therefore, after two lumbar punctures were performed and failed, we did not attempt a third. Puncture failure may be caused by the patient’s obesity (27, 28). MRI is the most accurate diagnostic tool for SEA, revealing not only the presence and extent of the abscess but also the degree of spinal cord compression. However, MRI findings in the early phase of the clinical course can be insignificant or subtle. Typical features of spinal infections include the contiguous involvement of two vertebrae and inflammatory changes within the intervertebral disc, but these are relatively chronic changes that may take weeks or months to manifest (29). A diagnostic delay of 4 months is not uncommon, given the time lag between initial symptoms and MRI (30). Intravenous gadolinium-enhanced MRI has high sensitivity and specificity, allowing for the diagnosis of SEA as epidural masses with surrounding septic or necrotic material exhibiting linear enhancement on T1-weighted contrast-enhanced MRI or hyperintensity on T2-weighted MRI (31). In cases where the diagnosis is still unclear after enhanced CT and plain MRI, a gadolinium-enhanced MRI should be performed, as recommended by Dunbar et al. (30) He reported a case of an immunocompetent patient who underwent two enhanced CT and one plain MRI with no results until undergoing enhanced MRI, which provided a clear diagnosis of SEA due to Pasteurella multocida (30). In our case, The patient presented with neurological root symptoms but no fever before admission. Initial MRI and plain radiography diagnosed the patient with simple lumbar disc herniation, which may have been either underdiagnosed or caused by a rapidly developing abscess in the week, leading to subsequent neurological impairment. Therefore, repeat examinations or gadolinium-enhanced MRI should be considered when initial MRI findings are not diagnostic of SEA.

For SEA confirmed by imaging, the standard treatment has traditionally been urgent surgical decompression followed by 6 weeks of IV antibiotic therapy. The literature suggests that surgical intervention is necessary in cases of acute or progressive neurological deficits, spinal instability, progressive deformities, or disease progression despite antibiotic therapy (7). Kim et al. (32) identified four independent predictors of nonoperative management failure in a cohort of 142 medically managed patients: age > 65 years, diabetes, neurologic impairment, and MRSA. If all four risk factors are present, they report a 99% failure rate. In a similar study, Patel et al. (33)reported three additional independent predictors of failure: leukocytosis >12.5, positive blood cultures, and CRP >115. If all three risk factors are present, they report a 77% failure rate (33). However, urgent surgical decompression is not always beneficial, as it may result in significant surgical trauma, impair spinal stability, and require spinal fusion surgery, which can reduce the patient’s range of motion in the lower back (34). The timing of surgical intervention for SEA is also controversial (35). In this case, we considered the patient’s poor condition and concluded that open surgery involving multiple segments may not be well-tolerated, despite the presence of independent predictors of non-surgical treatment failure and a high rate of non-surgical treatment failure as reported by Kim and Patel. The epidural compression was primarily caused by granulation tissue, and the size of the epidural abscess was relatively small. Therefore, we determined that excising only a small portion of the granulation tissue and performing decompression would not be particularly beneficial (36). Additionally, the origin of the abscess, whether it is ventral or dorsal, affects spinal stability after decompression surgery. Karikari et al.’s (37) regression model showed that ventral abscesses can be approached ventrally, preserving the posterior longitudinal ligament and reducing postoperative morbidity. We believed that this patient’s lumbar abscess originated from the dorsal side and had a relatively poor prognosis compared to the ventral side.

MRI is a crucial diagnostic tool for SEA patients as it provides accurate and objective imaging, enabling conservative treatment options consisting of systemic antibiotics and CT-guided percutaneous needle aspiration. Adogwa et al. (38) examined 82 cases of SEA in patients over the age of 50 with multiple comorbid conditions and found that early surgical decompression combined with IV antibiotics was not superior to IV antibiotics alone in this small group of patients. Arko et al. (12) came to similar conclusions. He concluded that patients can usually be treated with intravenous antibiotics and do not always require surgery, even though the patient may deteriorate clinically at any time (12). Antimicrobial agents are best chosen for the causative organism identified in blood cultures. Without knowing the causative bacteria, treatment should be initiated empirically. However, it is likely that different diseases, such as spinal abscess, liver abscess, and cystic fibrosis, will display significant differences in antimicrobial susceptibilities relative to published reports, given the frequent exposure of patients with underlying diseases to chronic macrolide suppressive therapy, inhaled aminoglycosides, and frequent use of fluoroquinolones (39). The previous standard antibiotic regimen for the treatment of abscesses associated with S. intermedius was penicillin plus chloramphenicol, but increasing resistance of S. intermedius to some antibiotics and the possibility of antagonism between penicillin and chloramphenicol have been documented (2). The most potent empiric therapies for S. intermedius include vancomycin, teicoplanin, and imipenem (40, 41). The British Society of Antimicrobial Chemotherapy (BSAC) advises that abscesses may be treated with a combination of Cefotaxime (a beta-lactam antibiotic) and metronidazole parenterally for 3 to 4 weeks or between 4 and 6 weeks when they are aspirated (42). Our case involved a patient with extensive SEA caused by S. intermedius infection, who presented with type II respiratory failure upon admission and was immediately transferred to the ICU for tracheal intubation. She had no sinusitis, heart disease, dental procedures, or other known risk factors for abscesses caused by S. intermedius. Intravenous vancomycin was initiated before the results of the blood culture were available, and after the culture results became available, meropenem and moxifloxacin were added. Fortunately, the patient’s drug sensitivity test results suggested sensitivity to these antibiotics. The duration of antibiotic administration is supported by the literature. After 5 days of antibiotic therapy, the blood culture became negative, and a repeat MRI performed 6 weeks later revealed a decrease in abscess size compared to the previous scan. After 2 months of antibiotic treatment, the patient’s muscle strength was restored. These outcomes suggest that our conservative treatment approach was successful in managing the patient’s condition.