Ultrasound guidance for performing interventional pain procedures is a relatively new approach when compared to traditional anatomic-landmark or fluoroscopy-based methods. Ultrasound allows interventionalists to enhance accuracy and safety of procedures performed to relieve pain. This chapter includes descriptions of ultrasound-guided techniques to perform injections on or in neuraxial joints, sympathetic ganglia, epidural space, and peripheral nerves. Each procedure is described under the following headings: Anatomy, Sonoanatomy, Indications, Technique, and Pearls.
Keywordsinjections, pain management, peripheral nerve, spine, ultrasound
Ultrasound (US) is a relatively new modality for guiding interventional procedures to provide pain relief. Traditionally, anatomic landmarks, fluoroscopy, computed tomography (CT), and nerve stimulation have been used for procedural guidance in pain management, but these modalities are associated with several limitations that can be overcome with US. US can enhance the accuracy and safety of procedures while eliminating exposure to ionizing radiation associated with fluoroscopy and CT and painful muscle contractions elicited by nerve stimulators. Imaging in real time, portability, and ability to see anatomic structures (e.g., nerves, muscles, blood vessels) are distinct advantages of US over other procedural guidance modalities.
There are also some limitations with US guidance. Image quality is compromised with increase in depth of scanning and bony structures, and there are no validated surrogates for contrast-enhanced visualization of injectate available with fluoroscopy-guided techniques. A limited field of view and dependence of image quality on skill of the operator, who has to focus simultaneously on both image acquisition and procedural performance, are some of the other challenges with US.
US-guided procedures for relieving pain can be anatomically classified broadly into two categories, axial (spine-related) and peripheral structures (nerves and muscles). We present information on procedures in each of these two categories under the following headings: Anatomy, Sonoanatomy, Indications, Technique, and Pearls. This is preceded by discussion of US physics and knobology. Knowledge of these is a necessary prerequisite for performing procedures accurately, safely, and with minimal discomfort to the patient.
Basics of Ultrasound
Medical US refers to use of sound waves at frequencies higher than the upper audible limit of human hearing. The usual range of medical US for procedural guidance is 1–18 MHz. US waves are generated by a transducer (also called a probe), a device that uses the piezoelectric effect to convert electrical energy to acoustic energy. The waves are propagated through the tissue and reflected back to the transducer from tissue interfaces that differ in density. A reverse piezoelectric effect is used to generate images from the reflected waves. The speed of sound waves through the tissues is a function of wavelength and frequency. Higher frequency is associated with shorter wavelengths and vice versa. Shorter wavelengths (i.e., higher frequencies) result in higher resolution (the ability to discriminate between two adjacent structures). Therefore high-frequency linear transducers (range 6–18 MHz) produce images with a better resolution of structures located in a relatively narrow area, which corresponds to footprint of the probe at a maximum depth of 6–7 cm, while low-frequency curved/curvilinear transducers (range 2–5 MHz) are used to visualize deeper structures with a wider field of view but with poorer resolution. Depth and focus of the ultrasonic waves can be adjusted by the operator using controls on the US machine panel as can be Gain, a feature that allows adjustment of intensity of the image. Time gain compensation (TGC) is an additional feature that reduces impact of wave attenuation by tissues through increased intensity of the received signal in proportion to the depth. This reduces artifacts in the image. US can also be used to assess tissue movement by using the color Doppler function (change in frequency of sound waves as an object moves toward or away from the receiver). This is a useful feature for US-guided pain procedures because it helps to detect the presence of small blood vessels, thereby reducing the probability of vascular puncture and injection. A “color” function is available on most US machines that provides directional information with red color indicating movement towards and blue color indicating movement away from the transducer. The power Doppler function is more sensitive to movement but it does not provide directional information.
The operator should be aware of various artifacts that can affect the interpretation of images produced by US machines. These include anisotropy and reverberation. Anisotropy refers to the change in characteristics of the image as a result of the change of the US probe angle relative to the structure of interest. It is often seen with tendons and can result in an incorrect diagnosis of a tear in the tendon. Reverberation is a phenomenon when a smooth structure (e.g., pleura, metallic implant, or fluid trapped within gas) reflects the sound beams back and forth between itself and the probe, causing linear echoes deep to the structure. It is often referred to as “comet tail” artifacts. Reverberation artifacts can be reduced by changing the angle of insonation (i.e., the angle between the probe and the skin).
In summary, the anesthesiologist or pain interventionalist should at least consider the following prior to performing a US-guided procedure: transducer type, depth, gain, and Doppler function. Angle of insonation can be adjusted by heel-toe or toggling movements to reduce reverberation. Finally, ergonomics in terms of positioning the patient, US machine, transducer grip, and injectate assembly (needle, extension tubing syringe, syringe) should be optimized.
Injection of the Cervical Facet Joint Nerve Supply
Cervical medial branch nerves that innervate the facet joints arise from the cervical dorsal rami and innervate the cervical facet joints (CFJs). CFJs receive dual innervation from two medial branches, one above and one below the joint. As an example, the CFJ between the inferior articular process of the third cervical vertebra and the superior articular process (SAP) of the fourth cervical vertebra (C3–C4) is innervated by the third and the fourth cervical medial branches. After their origin, the cervical medial branches travel transversely along the “waist” of the facetal pillar (approximately halfway between the joints) with the exception of innervation of the C2–C3 FJ. The C3 dorsal ramus has two medial branches, a deep medial branch that passes around the waist of the C3 articular pillar and supplies the C3–C4 FJ and a superficial medial branch, the third occipital nerve (TON). The TON curves around the lateral and then the posterior aspect of the C2–C3 FJ and contributes articular branches to the joint.
In the longitudinal plane on the lateral aspect of the neck, scanning is initiated behind the ear over the mastoid process. Sliding the probe caudally and posteriorly brings the lateral tip of the transverse processes (TPs) of C1 and, further caudad, C2 into view. The pulsation of the vertebral artery can also be visualized between these two levels. Caudad and posterior from the tip of the TP of C2, a wavy hyperechoic shadow with a series of “peaks and troughs” is identified ( Fig. 79.1 ). This represents the surface of the cervical articular pillars, with the peaks representing the CFJ and the troughs representing the “waist” of the articular pillars, where the cervical medial branches transverse in an anterior-to-posterior direction. The first of these peaks is the C2–C3 FJ; a “drop-off” toward the cranial side represents the downslope of the inferior articular process, and the absent SAP of the C2 can be identified. The medial branches can be visualized in the trough as round or oval structures under the overlying semispinalis capitis muscle. An artery (usually a branch of ascending cervical artery) can often be visualized adjacent to the nerve.
Cervical medial branch injections are performed for the diagnosis and treatment of cervicogenic headaches and/or CFJ-related neck and shoulder pain not responsive to conservative therapy. Patients often present with a history of whiplash injury or clinical and/or radiologic evidence of CFJ arthritis.
The patient is positioned in the lateral decubitus position with the neck in a neutral position (a small pillow under the head helps to achieve this goal). The side of the procedure is kept nondependent and the interventionalist faces the patient. It is advisable to perform a color Doppler scan of the injection site prior to inserting the needle to avoid penetrating any blood vessels during injection. The targets for these injections are medial branch nerves at the “troughs” on either side of the CFJ. A high-frequency linear US probe with a small footprint is oriented longitudinally over the cervical facet column with the cephalad edge of the probe over the tip of the mastoid bone behind the ear. For injection a 25-gauge needle (4 or 8 cm long) is inserted out of plane from the anterior side of the probe. The needle is then advanced slowly until it penetrates the fascia of the semispinalis capitis; its tip is then placed adjacent to the nerve (or, if it is not visualized, the deepest part of the trough). Turning the US probe to a longitudinal orientation visualizes the needle in plane. The tip of the needle should be over the most prominent aspect of the facetal pillar and dorsal to the posterior tubercle of the TP. Note that the TON is located at the surface of the C2–C3 FJ, and it is usually hyperechoic. The volume of local anesthetic injected per level is 0.5 mL.
The exiting cervical nerve root can be visualized anterior to the posterior tubercle; care should be taken to ensure that the injectate does not flow toward it.
This is an out-of-plane technique (at least during needle insertion), and the needle tip or shaft will be visualized only when it crosses the US beam. An extension tubing connected to a syringe of saline may be used for hydrodissection to locate the needle tip. Only small volumes (0.1 mL at a time) should be used for this purpose.
US-guided cervical medial branch injection is an “advanced” technique because of the proximity of the medial branch nerves to vertebral and other arteries and neural foramina. It is recommended that interventionalists gain experience with other, more superficial nerve or musculoskeletal US-guided procedures before attempting this procedure.
Injection of the Cervical Nerve Root
The cervical nerve roots emerge between the anterior and posterior tubercles of the TPs at each cervical level (except the C7, since there is no anterior tubercle at this level). The nerve roots occupy the lower part of the cervical intervertebral foramen with the epiradicular veins in the upper part. Vertebral, ascending cervical, and deep cervical arteries are also in proximity to the cervical spinal nerve roots, and these arteries often give branches that contribute to radicular arteries perfusing the spinal cord. One third of these vessels enter the foramen posteriorly, potentially forming a radicular or segmental feeder vessel to the spinal cord, making it vulnerable to inadvertent injury or injection.
It is vital to identify the appropriate level for injection. Two anatomic features of the cervical spine are useful in this regard. The transverse process of the C6 vertebra has a prominent anterior tubercle (usually larger than the posterior tubercle), and the C7 vertebra possesses only a posterior tubercle and no/rudimentary anterior tubercle. To locate the level of interest, one must slide the probe caudally until the TP of the C7 is identified (it appears as a gently sloping structure with the posterior tubercle at its peak) and then count up from there. At the C7 level, the emerging cervical nerve root can be identified immediately ventral to the posterior tubercle as a hypoechoic round structure, and the vertebral artery can be identified anterior to it by its pulsation. Moving the probe cephalad to the TP of the sixth cervical vertebra (C6) will bring the prominent anterior and smaller posterior hyperechoic tubercle into view. Similarly, the TPs of the fifth, fourth, and third cervical vertebrae can be identified by their tubercles, which are usually of equal size/prominence (unlike the C6) ( Fig. 79.2 ). The US probe may have to be rotated slightly from the transverse position to visualize both tubercles because the anterior tubercle is often slightly cephalad to the posterior tubercle.
Compression of an exiting cervical spinal nerve by a herniated intervertebral disc or foraminal stenosis can cause radicular pain. Injection of local anesthetics and/or steroids around compressed cervical nerve roots can provide analgesia; this intervention can also be used to prognosticate the analgesic outcome of cervical discectomy or decompression-fusion surgery.
The patient is placed in the lateral decubitus position with the neck in a neutral position with the procedure side nondependent. The operator should stand behind the patient. A high-frequency linear transducer is placed over the cricoid cartilage and then moved laterally to the side of interest. The lateral lobe of the thyroid gland is visualized wrapped around the lateral wall of the trachea, followed by the common carotid artery, internal jugular vein, and the muscles (superficial: sternocleidomastoid; deep: anterior and middle scalene, longus colli, and capitis). A more lateral and posterior position of the US probe will bring the prominent anterior and a smaller posterior tubercle of the cervical TP of the C6 and the emerging nerve roots between the two tubercles into view (see Fig. 79.2 ). The appropriate nerve root to be blocked is now identified. A color Doppler scan of the injection site is performed to avoid penetrating any blood vessels in proximity of the nerve root during the injection. A 22- or 25-gauge blunt needle (4–8 cm long) is introduced in plane from the dorsal side of the probe (i.e., the side closer to the operator) at an angle of 30 degrees to the skin. This approach is used because there is less probability of encountering blood vessels posterior to the nerve root as compared to the anterior aspect. The target for the tip position is immediately posterior to the emerging nerve root. Following a negative aspiration to rule out intravascular placement of the needle, up to 1 mL of the injectate is slowly injected around the nerve root.
The vertebral artery usually enters the foramen transversarium below the C6 in most patients, but it may be outside this foramen at the C6 level in up to 10% of patients. A thorough preprocedure scan to identify any aberrant blood vessels in proximity of the target nerve root is recommended.
Failure to visualize spread of the injectate around the nerve root during the injection may be due to intravascular injection or incorrect orientation of the US probe.
Injection of the Lumbar Facet Joint Nerve Supply
Lumbar medial branch nerves arise from the lumbar dorsal rami and innervate the lumbar facet joints (LFJs). Each LFJ receives dual innervation from two medial branches, one above and one at the level of the joint. As an example, the LFJ between the inferior articular process of the third lumbar vertebra and the SAP of the fourth lumbar vertebra (L3–L4) is innervated by the second and third lumbar medial branches that cross the grooves between the L3 SAP and TP and the L4 SAP and TP, respectively. The L5 dorsal ramus, located at the junction of the sacral SAP and the sacral ala, is targeted for blocking the nerve supply to L4–L5 or L5-sacral FJs. As opposed to the thoracic FJ, each LFJ is dorsal to the lumbar TPs.
The patient is placed in prone position with pillows underneath the abdomen. Minimizing the lordosis is important to open up the interlaminar space and enable the US beam to strike the LFJ perpendicularly. A low-frequency curvilinear probe is used to scan the lumbosacral spine. The lumbar spinous processes are large and tend to reflect most of the US beam. The upper margin of the spinous process of a lumbar vertebra corresponds to the lower level of the TPs of the level above. A US scan at the midline in the sagittal plane reveals an undulating shadow that represents the tip of the lumbar spinous processes with minimal transmission of the US beam deeper to the spinous processes. The most caudal (also continuous) shadow in this view is the sacral median crest. A paramedian sagittal scan with the probe tilted medially reveals laminae and interlaminar spaces where the posterior and anterior dura can be identified ( Fig. 79.3A ). The spinal level should be counted and indicated on the skin with a marker using this view. Orienting the US probe perpendicularly at this spot (i.e., removing the medial tilt) reveals the “sawtooth” appearance of the articular processes and the LFJ (see Fig. 79.3B ). Moving the probe further laterally while maintaining the perpendicular orientation allows the TPs to be visualized at a greater depth compared with the LFJ (see Fig. 79.3C ), while moving it slightly medially from this point shows the junction of the articular process and the TP (see Fig. 79.3D ). Rotating the probe transversely and scanning in the midline cranially from the level of the upper sacrum initially reveals the flat appearance of dorsal sacrum, with the prominent median sacral crest (“flying seagull” sign) followed by an interlaminar view between the sacrum and the fifth lumbar vertebra. Further cranial scanning helps confirm the counting of the intervertebral spaces, which was performed with the US probe in the sagittal orientation. The anterior and posterior dura can be visualized in the middle with the “two steps” of the inferior articular process (first step) and the TP (second step) (see Fig. 79.3E ). The SAP is often difficult to visualize because it is oriented parallel to the US beam. The visualized medial edge of the TP (signifying the groove between it and the SAP) is the target for the medial branch nerve block.