Utility of Digital Subtraction Angiography in Cervical Transforaminal Epidural Steroid Injections: Does Digital Subtraction Improve Safety Over Live Fluoroscopy?
Cervical transforaminal epidural steroid injections (CTFESI) have been used for many years in the conservative management of cervical radicular pain. Reports of severe neurological complications from CTFESI often implicate accidental intravascular injections as the mechanism of injury.i In order to reduce the risk of serious injury during CTFESI, our ability to reliably identify intravascular needle placement is paramount.
In our pain clinic, we perform CTFESI using real-time imaging and digital subtraction angiography (DSA). Like many practitioners, we have adopted several measures considered important in reducing the risks of neural injury. We perform CTFESI with a short bevel needle and use only the non-particulate steroid dexamethasone.
Additionally, we perform contrast dye studies using digital subtraction angiography to complement real-time fluoroscopy prior to injection. As a quality improvement measure, we track the vascular flow patterns obtained during CTFESI utilizing both live fluoroscopy and digital subtraction. The results are compared for each imaging technique in an effort to identify the most sensitive method for detecting intravascular injection.
IRB approval was obtained. We reviewed records from CTFESIs performed in an academic outpatient clinic from November 2008 to June 2015. Over 385 studies were performed; however, only 382 patients had complete documentation available for review. These injections were performed or supervised by fellowship trained pain physicians.
When aspiration revealed blood, the event was documented and the needle was immediately repositioned. Contrast dye flow studies were performed only if aspiration was negative.
Flow studies detected 52 intravascular needles despite negative aspiration of blood. Of the 52 positive cases, 20 (38.4%) were detected with standard real-time fluoroscopy without need for DSA. An additional 32 cases (61.5%) were found to be positive for intravascular flow using DSA despite negative blood aspiration and negative flow study using live fluoroscopy.
Furthermore, the overall rate of intravascular detection identified by live fluoroscopy (5.2%) (95% CI: 3.2% - 8.0%) and DSA (13.6%) (95% CI: 10.3% - 17.5%) were significantly different (p<0.0001) when analyzed utilizing a test of marginal homogeneity. Statistical analyses were generated using StatXact (ver4) (CYTEL Software Corporation, Cambridge MA).