Management and outcome of whole-spine epidural abscesses - institutional case series and systematic review

Whole-spine epidural abscesses (WSEAs) are rare infections with potentially devastating neurological sequelae. As common treatment guidelines remain absent, this study aims to describe the management and resulting outcomes of WSEAs through an institutional case series and systematic literature review.

What are the optimal management strategies for WSEAs based on current evidence?

We retrospectively analyzed a series of WSEAs treated at our tertiary spine center (2009-2024) and conducted a systematic review using PubMed, Ovid, and Web of Science (1990-2024). Data on patient demographics, clinical presentation, bacteriology, diagnostics, treatment strategies, and outcomes were extracted. Statistical analyses included multivariate and risk-factor analysis.

We identified 48 patients (mean age 54 years, 69 % male), including four from our institutional series. Staphylococcus aureus was the most common pathogen (67 %). Five patients (10.4 %) received only conservative antibiotic treatment, primarily those with mild neurological deficits or severe comorbidities. Surgical intervention was performed in 43 patients (89.6 %), mainly utilizing skip-laminectomies (68.8 %) with catheter-based epidural irrigation. Favourable outcomes were comparable between surgical (86.1 %) and conservative (80.0 %) treatment (p = 0.567). Initial neurological status strongly correlated with neurological outcome (Spearman's ρ = 0.563, p = 0.0001).

Both surgical and conservative approaches can achieve favourable outcomes in selected WSEA cases. The strong correlation between initial and final neurological status emphasizes the importance of early diagnosis and targeted intervention. Skip-laminectomies with catheter-based irrigation appear effective while minimizing surgical morbidity. Conservative treatment may be considered in neurologically intact patients under close surveillance. Larger multicenter studies are needed to establish reliable treatment algorithms.