Background and Objectives: Subdural injection is a well-known but often poorly recognized complication of neuraxial anesthesia/analgesia. This report aims to further describe the clinical presentation of subdural injection by analyzing radiographically proven cases. A new diagnostic algorithm is then proposed.

Methods: A literature search identified 70 radiographically confirmed cases of subdural injection. The prevalence of numerous presenting characteristics and their relationship to the volume of injected local anesthetics were examined. The ability of 2 previously published diagnostic paradigms to detect proven subdural injection was compared with that of a newly proposed algorithm.

Results: The dermatomal distribution of sensory blockade was excessive in 74% of cases, restricted in 17%, and neither in 9%. Motor blockade and respiratory depression were associated with larger local anesthetic injection volumes (median volume = 14 vs. 8 mL [P <.009] and 15 vs. 10 mL [P <.035], respectively), but segmental spread and cardiovascular depression were not. Only 33% of cases were positive for 2 or more of Collier's criteria; Lubenow et al.'s diagnostic paradigm detected 71% of cases. We propose a diagnostic algorithm structured as a "roadmap," whereby the clinician inputs the assumed neuraxial block (epidural vs. subarachnoid), and distribution of sensory blockade (excessive, restricted, neither). Specific minor criteria are then applied to diagnose subdural injections. This algorithm detected 93% of subdurals with excessive sensory block distribution, and all of those with restricted and normal distribution.

Conclusions: Radiographically proven subdural injections were used to further define the clinical presentation of subdural analgesia/analgesia and a new diagnostic algorithm is proposed.