US vs FL-Guided Lumbar Nerve Root Injections

Background and Rationale Lumbosacral radiculopathy is a common and disabling condition resulting from irritation or compression of spinal nerve roots. When conservative treatments fail, epidural steroid injections are a frequently employed non-surgical intervention. Traditionally, selective lumbar nerve root injections are performed under fluoroscopic (FL) guidance to ensure precise needle placement. However, fluoroscopy exposes patients and healthcare providers to ionizing radiation and has limitations in soft tissue visualization. Ultrasound (US) guidance has emerged as a viable alternative, enabling real-time visualization of soft tissues, nerves, and vascular structures while entirely eliminating radiation exposure. This randomized clinical trial aims to compare the efficacy, safety, and functional outcomes of US-guided versus FL-guided selective lumbar nerve root injections in patients with refractory lumbosacral radiculopathy.

Study Design and Randomization This is a single-center, parallel-group, randomized clinical trial with a 1:1 allocation ratio conducted at Tanta University Hospitals. Participants are randomly assigned to either the US-guided or FL-guided group using a computer-generated randomization sequence created via RANDOM.ORG. The sequence utilizes permuted blocks of varying sizes (4 and 6) to ensure balanced group allocation. Allocation concealment is maintained using sequentially numbered, opaque, sealed envelopes managed by an independent research coordinator. While patients and proceduralists cannot be blinded to the intervention, the outcome assessors and data analysts remain blinded to group assignment.

General Procedural Protocol Prior to the intervention, patients are positioned prone or lateral with lumbar exposure. Skin sterilization is performed, and a single dose of broad-spectrum parenteral antibiotic (cefazolin 1 g intravenous) is administered. For both study arms, the standardized injectate consists of a 3 mL total volume containing 1 mL corticosteroid (dexamethasone 4 mg), 1 mL local anesthetic (2% lidocaine), and 1 mL 0.9% sodium chloride.

Image-Guidance Protocols

• Ultrasound-Guided Technique: A posterior paravertebral ultrasound scan is performed using a curved transducer, followed by a transverse sonogram at the target site to identify the spinous process, vertebral lamina, facet joints, and intervertebral foramen. A 22-gauge spinal needle is advanced using an in-plane technique until mild paresthesia in the affected dermatome is elicited, followed by medication injection. No contrast agent is utilized in this arm.
• Fluoroscopy-Guided Technique: A C-arm fluoroscopy unit is used for visualization. A 22-gauge spinal needle is inserted into the safe triangle near the nerve root. Iodinated contrast material (iohexol 240 mgI/mL, 1-2 mL) is injected to confirm extra-vasculature, extra-thecal needle placement before the therapeutic medication is administered.
• L5-S1 Modification: For L5-S1 level injections in both groups, a modified approach is employed with the needle directed more laterally to access the lateral recess of the spinal canal, targeting the L5 nerve root as it courses toward the L5-S1 foramen.

Post-Procedure Care Following the injection, patients are monitored for 30 to 60 minutes to observe for any immediate complications, such as allergic reactions, bleeding, or neurological deficits. Discharge instructions include mandatory rest for 48 hours and the avoidance of strenuous physical activity. Follow-up assessments are conducted at 1 week, 1 month, and 3 months by the blinded independent assessors.

Statistical Analysis Plan The planned sample size is 64 patients (32 per group), calculated to provide 80% power to detect a 50% reduction in pain scores, accounting for an anticipated 15% dropout rate. Statistical analysis is performed on an intention-to-treat (ITT) basis. Missing data for the primary analysis are handled using multiple imputations by chained equations (MICE) under the assumption of missing at random, generating 20 imputed datasets. Continuous outcomes will be analyzed using analysis of covariance (ANCOVA) with baseline values included as covariates. Categorical responder analyses will also be conducted to evaluate the minimal clinically important difference (MCID) and substantial clinical benefit (SCB).