Answers are grounded in this paper and cite their sources.
Background/Objectives: This study evaluated the retrospective visibility of metastatic epidural extension on unenhanced computed tomography (CT) in patients with known spinal metastases. Methods: This retrospective single-center cohort included 98 patients. Epidural extension, with or without neural compression, was determined from clinical MRI (79 patients) or contrast-enhanced CT alone (19 patients). Two spine surgeons independently reviewed the unenhanced CT closest in date to the reference examination, blinded to the imaging diagnosis and lesion location. CT findings were positive if either reader identified a lesion. Earlier CTs were reviewed only after a positive initial assessment, with knowledge of the identified lesion. Surgical and survival comparisons were secondary, exploratory, and unadjusted. Results: Epidural lesions were retrospectively visible in 39/50 reference-positive patients (78.0%), including 19/23 with Bilsky grades II–III (82.6%). One of the 48 reference-negative patients had a positive CT finding. Interobserver agreement was 92.9% (κ = 0.847). The interval from initial CT to reference imaging was a median of 1 day (range, 0–58) in CT-visible cases and 3 days (range, 0–160) in CT-nonvisible cases. Within 7 days, lesions were visible in 25/33 patients (75.8%). In the selected backward review, lesions were visible before reference diagnosis in 24/50 patients (48.0%); the interval from the earliest positive CT was a median of 30 days (range, 1–542). Exploratory comparisons showed surgical proportions of 48.3% versus 14.3% for Bilsky Ic–III versus Ia–Ib (p = 0.0277), without a statistically significant unadjusted survival difference (p = 0.0911). Conclusions: Metastatic epidural extension was retrospectively visible on unenhanced CT in this selected cohort. Earlier retrospective visibility does not establish prospective detectability or improved treatment or neurological outcomes from earlier recognition.
No takes yet. Share an insight, caveat, or question.
Takatori et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: