CNS injury following intra-arterial steroid injection: More than microembolization?

Transforaminal epidural steroid injections (TF-ESI's) are used to treat radicular pain. This approach is associated with devastating complications including stroke and death. While the mechanism is unknown, the leading hypothesis is that inadvertent intravascular injection of particulate steroids leads to microembolisation. We tested this hypothesis using a rodent model. In anesthetized rats, the internal carotid artery (ICA) was dissected and its branches ligated. The external carotid artery (ECA) was ligated and cannulated for injection. Five groups were tested: Depo-Medrol (40mg/mL, n=11), Depo-medrol carrier (N=6), Solumedrol (n=8), Decadron (4mg/mL, n=8) and normal saline (n=7). Drugs, in a total volume of 50 µL, were injected into the ICA via the ECA at 25uL/min. The animals were sacrificed on POD 3. The extent of CNS injury was quantitated by image analysis of coronal sections. In separate groups, the extent of injury to the blood brain barrier (BBB) was determined by Evan's blue dye leakage 2h after drug injection. Evaluation demonstrated 8 of 11 animals in the Depo-Medrol, 8 of 8 in the solumedrol group, and 3 of 6 animals in the Depo-Medrol carrier groups had cerebral hemorrhage; no lesions were identified in the dexamethasone and saline control groups (p<0.01). There was Evan's blue leakage detected in the Depo-Medrol and Solumedrol but not the Decadron or saline groups. This study presents the first in vivo evaluation of intra-arterial steroid injections. The data demonstrate that Depo-Medrol, its non-particulate carrier and non-particulate formulation Solumedrol can produce significant injury to the BBB when injected intra-arterially. These results suggest that Depo-Medrol induced injury is produced not only by particulate obstruction of the cerebral microvasculature, but also may involve BBB breakdown.