Abnormal Anatomical Features Likely Responsible for Unintended Epidural Flow

During epidural injection, degenerative and pathologic facet joint changes were found to increase the risk for spread into the retrodural space, according to the results of a study published in Regional Anesthesiology & Pain Medicine.

Patients (N=452) who underwent epidural steroid injection for low back or radicular pain in 2020 at the Yonsei University College of Medicine in South Korea were included in this study. The epidural injections were administered by 1 of 2 operators using a fluoroscopy-guided approach with 1 to 2 mL of contrast media and a 20-gauge Tuohy needle. After confirmation of correct epidural flow, 5 to 6 mL of 0.5% lidocaine were injected with or without 5 mg of dexamethasone. The incidence of spread to the retrodural space was evaluated via imaging and clinical data. In addition, 16 cadavers and lumbar vertebral tissue blocks from 5 cadavers were used for anatomical study and sectional dissection, respectively.

The mean ages of patients with (n=33) and without (n=419) spread to the facet joint were 57.1 (range, 20-85) and 59.4 (range, 20-89) years, 60.6% and 58.9% were women, mean BMI values were 23.6 (range, 15.7-30.8) and 23.7 (range, 15.8-37.3) kg/m², and 9.1% and 7.6% had undergone prior spinal surgery, respectively.

The rate of spreading was similar between both operators (5.4% vs 6.3%; P =.729). The fluoroscopic images indicated that all but 2 injections were successfully administered at the intended spinal level.

Patients who had inadvertent spread had a higher prevalence of severe central stenosis, as well as degenerative and pathologic changes of the facet joint. The patients who experienced spread also tended to have more incidences of spondylolisthesis, compression fracture, and interspinous bursitis.

The injections with inadvertent spread tended to have left direction contrast flow (47.2%) compared with right (38.9%) or bilateral (13.9%) flow, and ipsilateral flow (63.9%) compared with bilateral (13.9%) or contralateral (5.6%) flow. Most of these epidural injections demonstrated correct positioning (88.9%).

In multivariate logistic regression analysis, facet joint pathologies identified on magnetic resonance imaging (MRI) were found to be associated inadvertent spread occurring during fluoroscopy-guided lumbar interlaminar epidural injection (odds ratio [OR], 4.382; 95% CI, 1.160-16.558; P =.029).

Evaluation of the cadaver specimens revealed a consistent radiolucent area posterior to the ligamentum flavum. Sagittal histologic sections revealed distinct ligamentum flavum, retrodural space, and interspinous ligament layers. The retrodural space was filled with adipose and loose connective tissues.

On MRI, facet joint effusion to the epidural space through the ligamentum flavum was observed, but it could not be seen as a direct route of communication. Degenerative facet joints with irregular cartilage subluxation were observed, however, with a narrowed joint space and partial ligamentum flavum detachment. Direct communication of spaces was observed in specimens with spondylolysis and segmental disease after fusion surgery.

This study may limited by selection bias due to its retrospective design.

These data indicate that inadvertent spread to the retrodural space during epidural steroid injection is not uncommon and likely due to abnormal anatomical features. The study authors conclude, “Based on our results, identifying high-risk patients based on anatomical considerations, and recognizing the imaging features of injection into the retrodural space, may assist clinicians in further investigating and refining performance of epidural injections.”