Background and objective: The technique and validity of ultrasound (US)-guided periradicular injection in the middle to lower cervical spine has been well established by previous cadaveric studies. However, no US guided cervical selective nerve root block studies have been conducted in symptomatic patients. Accordingly, the objective of this study was to validate US guided selective nerve root block in symptomatic patients.

Study design: Retrospective study with prospectively collected data.

Methods: Forty-six consecutive patients with cervical radicular pain, diagnosed by clinical examinations and by radiologic studies, were recruited for this study. Study subjects had failed a prior course of conservative treatment, which included at least one month of physical therapy and an oral anti-inflammatory agent/analgesic trial. Ninety-two cervical extra-foraminal selective nerve root block (CESNRB) sessions were performed under ultrasound (US) guidance. Pre- and post-treatment visual analogue scores were compared prior to injection and at three month follow-up visits. Data collection and analysis were performed by an independent clinical reviewer.

Techniques: With a patient positioned supine with his head turned approximately 45 degree to the opposite side (within a comfortable range). Before planning the needle trajectory and injection procedure, pre-scanning of deep neck structures was performed to define the relationship of adjacent critical structures and to select the appropriate cervical level and its root. The pre-scanning sequence was as follows; 1) scan anatomical landmarks, 2) determine cervical levels, 3) locate the anterior and posterior tubercles and intervening nerve roots at each level, and 4) display the vertebral artery and several small critical arteries. Initially, the US probe was positioned in the axial plane to identify the following anatomical landmarks; the thyroid gland with isthmus, the common carotid artery and its bifurcation, the anterior and middle scalene muscles, the anterior tubercle of each cervical vertebra and its adjacent longus colli muscle (figure 1). To prevent inadvertent penetration of the vertebral artery, ascending cervical artery, or the deep cervical artery, these vessels were also identified prior to injection (figure 2). The ascending cervical artery, the deep cervical artery, and the V1 segment of the vertebral artery appeared as anechoic vessels anterior to the longus colli muscle in a bare region inferior to the thyroid lobe. Adjuvant Color or power Doppler US was utilized to identify these vessels. The following anatomic landmarks were used to determine cervical nerve root levels: carotid bifurcation ; the C4 root, the upper lobe of the thyroid gland ; the C5 root, the inferior thyroid lobe ; the C6 root, the bare region inferior to the thyroid gland without an anterior vertebral tubercle ; the C7 root, and inferior to the C7 transverse process ; the C8 root (Figure 1). For C7 and C8 nerve blocks, extra-caution was needed to avoid inadvertent penetration of the vertebral, ascending cervical, and deep cervical arteries. Next, the anterior and posterior tubercle of targeted cervical vertebrae was identified using their hyperechoic signal and postacoustic shadowing, and subsequently, intervening nerve root was identified. In transverse view, nerve root runs in the inter-tubercular groove and appear as a distinct homogeneous round to oval hypoechoic structures (figure 3). When a nerve root was traced distally, its trunk could be observed as a hypoechoic signal in the interscalene groove, sandwiched between the anterior scalene and middle scalene muscles. After the desired nerve root had been identified, a needle trajectory was planned on transverse scan images and a free hand technique was used to insert the needle without local anesthetic infiltration (Figure 4). A 1.5 inch, 25 gauge needle was used to puncture the skin, approximately 5-10 mm lateral to the edge of the ultrasound probe, and then advanced transversely at an angle of 0 to 20 degree with respect to the surface of probe (an angle of 70 to 90 degree with respect to the insonation beam) using an in-plane technique which allows the entire needle and its tip to be visualized. If the needle tip supposedly encounters the jugular vein or a small artery along its trajectory, its advancement is stopped in front of the vessel, and the needle position is adjusted under real-time image guidance, either superficially or deeply with respect to the vessel, to avoid penetration. After having passed such vessels, the needle is re-aligned to the targeted nerve root and meticulously advanced to its final destination. Movements of the needle shaft and tip were obvious in real time images and were visualized as bright linear echo patterns. Finally, with the needle tip located between the nerve root and the posterior tubercle outside the intervertebral foramen, a 0.5mL normal saline test injection is administered to check the fluid spreading pattern and to rule out the possibility of intravascular infiltration. When properly administered a focal accumulation of anechoic fluid could is observed by real time imaging. Bupivacaine (2mL of 0.125%) and 2mg/mL of Limethason (dexamethasone palmitate 4mg/mL, Welfide Korea, Seoul) is then injected incrementally under real time US guidance over 20 seconds, and are observed to encircle the nerve root as hypoechoic rim in real time (the donut sign) (figure 5).

Results: The study group included 22 males and 24 females. Average symptom duration was 10.6 months, and patients received an average of 2.02 injection sessions. Two patients were lost to follow up, and thus, only 44 patients had complete data available for analysis. The pre-injection average VAS score reported was 5.89, and at 3-month follow-up visits, the average score was 2.93. Overall, good or excellent results were achieved in 30 patients (68%). No complications were observed.

Conclusion: The findings of this study suggest that US guided extra-foraminal selective cervical root block is a safe and effective interventional treatment for patients with cervical radicular pain.

1. Filename: figure 1.jpg

2. Filename: figure 2.jpg

3. Filename: figure 3.jpg

4. Filename: figure 4.jpg

5. Filename: figure 5.jpg