Chapter 51. Ultrasound-Assisted Nerve Blocks in Adults

In recent years interest has been growing in the practice of regional anesthesia and, in particular, in peripheral nerve blocks for surgical anesthesia and postoperative analgesia. Peripheral nerve blocks have been found to be superior to general anesthesia1 because they provide effective analgesia with few side effects2 and can hasten patient recovery.3 Unfortunately, the practice of regional anesthesia does not enjoy widespread endorsement because of inconsistent success, varying from one anesthesiologist to another. Current methods of nerve localization (eg, paresthesia and nerve stimulation) are essentially “blind” procedures, since they both rely on indirect evidence of needle-to-nerve contact.4,5 Seeking nerves by trial and error and random needle movement can cause complications. Although uncommon, complications such as intravascular local anesthetic injection resulting in systemic toxicity, inadvertent spinal cord injury following interscalene block, pneumothorax following supraclavicular block, and nerve injury have all been reported.6,7

Imaging guidance for nerve localization holds the promise of improving block success and decreasing complications. Among imaging modalities currently available, ultrasonography seems to be the one most suitable for regional anesthesia. Perhaps the most significant advantage of ultrasound technology is the ability to provide anatomic examination of the area of interest in real time.8 Ultrasound imaging allows one to visualize neural structures (plexus and peripheral nerves) and the surrounding structures (eg, blood vessels and pleura), navigate the needle toward the target nerves, and visualize the pattern of local anesthetic spread.9

An ultrasound probe (transducer) has dual functions. It emits and receives sound waves, thus functioning both as a speaker and a microphone. As the name implies, ultrasound waves are high-frequency sound waves (≥20,000 cycles/sec, 20 kHz) that are not audible to the human ear. Ultrasound frequencies useful in clinical medicine are in the megahertz (MHz) range.10 When an electric current is applied to an array of piezoelectric crystals (quartz) within the ultrasound transducer, mechanical energy, in the form of vibration, is generated, resulting in ultrasound waves. As the ultrasound waves move through body tissues of different acoustic impedances, they are attenuated (lose amplitude with depth), reflected, or scattered. Waves reflected to the transducer are then transformed back into an electrical signal that is then processed by the ultrasound machine to generate an image on the screen.

Depending on the amount of wave returned, anatomic structures take on different degrees of echogenicity. Structures with high water content, such as blood vessels and cysts, appear hypoechoic (black or dark) because ultrasound waves are transmitted through the structures easily with little reflection. On the other hand, bone and tendons block ultrasound wave transmission, and the strong signal returned to the transducer gives these structures a hyperechoic appearance (bright, white) on the screen. Structures of intermediate density and acoustic impedance, such as the liver parenchyma or the thyroid gland, appear gray on the screen. Knowing the speed of sound in tissue (1540 m/sec on average) and the time of echo return, the distance between the probe and the target structure (depth) is calculated.

Ultrasound Equipment

Ultrasonography has many applications in clinical anesthesia. With appropriate probes, vascular imaging, echocardiography, and nerve imaging can be performed with the same unit. Compound imaging is an advanced feature in some of the cart-based units. The resolution of nerve images is enhanced with compound imaging when multiple lines of crystals on the transducer (as opposed to a single line) emit and receive ultrasound in multiple planes before final display of the image that is electronically reconstructed. Color Doppler is another useful feature that differentiates vascular from nonvascular structures (eg, nerves). Compact portable units currently available with many of these sophisticated features are also suitable for peripheral nerve imaging.

Transducers (Probes)

Ultrasound scanning of deep abdominal organs, such as liver, gallbladder, and kidneys, requires low-frequency probes (3–5 MHz). Scanning superficial structures such as the brachial plexus, on the other hand, requires high-frequency probes (10–15 MHz) that provide high axial resolution; however, beam penetration is limited to 3 to 4 cm. A lower frequency probe (4–7 MHz) is suited for scanning deeper structures, such as the brachial plexus in the infraclavicular region and the sciatic nerve in adults.

Probe Orientation

It is advisable to follow the tradition of pointing the premarked end of the probe toward the head when scanning in a sagittal or parasagittal plane, and pointing toward the patient's right when scanning in an axial plane, so that saved images will be correctly interpreted at a later time.

Scanning Technique

Patient positioning for each block is essentially the same as that used for standard, non-image–guided peripheral nerve blocks. Sterile technique should be followed, especially when a continuous catheter technique is performed, in which case a long sterile sheath covering the probe and the cord and sterile transmission gel are recommended.

Transverse and longitudinal views are most commonly used for nerve imaging. When the probe is perpendicular to the long axis of the nerve, the transverse (short axis, cross-sectional) view shows nerves in round to oval shape with internal hypoechoic nerve fascicles surrounded by the hyperechoic epineurium. When the probe is parallel to the long axis, nerves in longitudinal view appear tubular with linear hypoechoic fascicular components mixed with hyperechoic bands corresponding to the interfascicular epineurium.11,12 Nerves have different degrees of echogenicity. For example, nerve roots and trunks of the brachial plexus in the interscalene and supraclavicular regions appear mostly hypoechoic, whereas peripheral branches of the brachial plexus and the sciatic nerve are largely hyperechoic.

High-frequency linear probes, in the range of 10 to 15 MHz, are best suited for imaging the brachial plexus in most locations, except perhaps the infraclavicular region, where the cords may be more deeply located and thus probes in the 4- to 7-MHz range may be required.

The Interscalene Region

In the interscalene region, the cervical roots forming the plexus are located between the anterior and middle scalene muscles. They are best visualized when scanned in the lateral aspect of the neck in an axial oblique plane (Figure 51–1A). In this manner, the sternocleidomastoid muscle can be identified superficially. Deep to it are the anterior and middle scalene muscles where one or more roots are visualized in the interscalene groove.13 They appear mostly hypoechoic, with few internal punctuate echoes (Figure 51–1B). Deeper to this plane, the vertebral artery and vein are seen next to the vertebral transverse process. The carotid artery and internal jugular vein can be identified medially.

The Supraclavicular Region

In the supravlavicular region, the brachial plexus is best scanned with a linear probe in a coronal oblique plane (Figure 51–2A).14 The subclavian artery is the most prominent landmark identified immediately superior to the first rib (Figure 51–2B). The trunks, or divisions, of the plexus in this region are tightly arranged within what seems to be a single sheath, immediately lateral and cephalad to the subclavian artery. The anterior and middle scalene muscles can be identified as they insert on the first rib. The pleura can be seen immediately deep to the first rib.

The Infraclavicular Region

In the infraclavicular region next to the coracoid process, the cords of the plexus lie deep to the pectoralis major and pectoralis minor muscles. They can be best imaged with a linear probe in the range of 4 to 7 MHz, in a parasagittal plane, immediately medial to the coracoid process (Figure 51–3A).15,16 In this manner, a transverse view of the cords adjacent to the axillary vessels can be obtained (Figure 51–3B). The cords appear hyperechoic, with the lateral cord commonly cephalad and the posterior cord posterior to the artery. The medial cord in this region can often be seen between the artery and vein, but is not always visible.

The Axillary Region

In the axilla and the upper arm, the neurovascular bundle is located in the internal bicipital sulcus, which separates the flexor muscle compartment of the arm (biceps and coracobrachialis muscles) from the extensor compartment (triceps). At this level, terminal braches of the brachial plexus, such as the musculocutaneous, median, ulnar, and radial nerves, are located superficially, usually within 1 to 2 cm of the skin. A linear 10- to 15-MHz probe is therefore recommended. To obtain a transverse view of the neurovascular bundle with the arm abducted at 90 degrees and the forearm flexed, the probe is positioned perpendicular to the long axis of the arm, as close to the axilla as possible (Figure 51–4A). The round pulsatile axillary artery is easily identified in the bicipital sulcus and is distinguished from the axillary veins that are readily compressed. Nerves in the axilla are round to oval-shaped and hypoechoic with internal hyperechoic areas, presumably the epineurium. In the axillary region, the median and ulnar nerves are usually lateral and medial to the artery, respectively (Figure 51–4B). The radial nerve is often posterior or posteromedial to the artery, but nerve location is highly variable.17 The musculocutaneous nerve often branches off more proximally and can be seen as a hyperechoic structure. It can be found between the biceps and coracobrachialis muscles for a short distance before entering the body of the coracobrachialis muscle (see Figure 51–4B). When performing an axillary block, it is best to inject local anesthetic around each nerve individually to achieve consistent success. Local anesthetic spread within the sheath compartment may be presumably restricted by the septa when observed under ultrasound.18

The lumbar plexus (L1 to L5) and the sacral plexus (S1 to S4) provide innervation to the lower extremity. Unlike the brachial plexus, the lumbosacral plexus and its proximal branches are quite deep. Sonographic imaging can be more challenging except for the distal peripheral branches.

Paravertebral Anatomy & Lumbar Plexus Blocks

Ultrasound imaging of the lumbar plexus in the paravertebral region in adults is technically difficult because of its deep location. Using curved 4- to 5-MHz transducers, Kirchmair and associates identified the lumbar plexus within the psoas muscle and correlated ultrasound images with anatomic specimens.19 Scanning is performed with the patient either prone with a pillow under the abdomen to reduce lumbar lordosis or in the sitting position. The transducer is placed longitudinally, in a parasagittal plane, approximately 3 cm from the midline to determine the location of the lumbar transverse processes. Once accomplished, the transducer is turned 90 degrees to the transverse axial plane and positioned between two transverse processes so that bony interference to ultrasound beam penetration is minimized (Figure 51–5A). In the axial image, two muscles are identified deep to the subcutaneous plane: the erector spinae muscle immediately lateral to the spinous process and the smaller quadratus lumborum more laterally. The psoas muscle lies deep (anterior) to these two muscles and is adjacent to the vertebral bodies and intervertebral discs (Figure 51–5B). Previous anatomic studies demonstrate that the lumbar plexus most often lies between the posterior third and anterior two thirds of the psoas muscle; the average skin-to-plexus distance is 5–6 cm.20 For this reason it has been recommended that local anesthetic be administered in the posterior one third of the muscle. Ultrasound also identifies the inferior pole of the kidney (as low as the L3 to L4 level) and can potentially avoid renal hematoma due to inadvertent needle trauma.21

Femoral Nerve

The three main terminal branches of the lumbar plexus are the femoral, obturator, and lateral femoral cutaneous nerves. The femoral nerve derived from L2 to L4 is the largest branch and can be easily imaged in the inguinal region using a linear 8–12 MHz transducer (Figure 51–6A).22 The probe placed over the inguinal crease in the transverse axial plane shows the femoral nerve immediately lateral to the femoral vessels, often appearing oval or triangular in shape (Figure 51–6B). It lies deep to the ileopectineal arch and overlying the groove between the iliac and psoas muscles. The femoral nerve can be imaged further distally for a short distance until it divides into small terminal branches that become indistinguishable from the surrounding tissue. It is possible to image the saphenous nerve, which is next to the femoral vessels in the mid to distal third of the thigh.

Sciatic Nerve

The sciatic nerve also originates from the lumbosacral plexus (L4 through S3) and enters the gluteal region through the greater sciatic foramen, between two muscle planes. The anterior muscle plane is formed by the obturator internus and inferior gemellus muscles, and the posterior, more superficial muscle plane by the gluteus maximus muscle. In the gluteal region the sciatic nerve is not easily identified by ultrasonography because of its depth. Lower in the subgluteal region, the sciatic nerve is more superficial, usually within 5 cm from the skin surface, and can be blocked as described by Raj and coworkers23 and later by Sutherland24 and Sukhani and associates.25 With a curved 5–7 MHz transducer, a transverse view of the sciatic nerve can be obtained showing bony landmarks, the greater trochanter of the femur laterally, and the ischial tuberosity medially, when the patient is positioned semiprone with the limb to be blocked uppermost (Figure 51–7A). The approximate location of the sciatic nerve is in the midpoint of a line uniting both landmarks. The sciatic nerve often appears hyperechoic and elliptical deep to the distal gluteus maximus muscle and lateral to the biceps femoris muscle (Figure 51–7B).26 It is usually surrounded by a well-defined border, presumably the aponeurosis of the surrounding muscles.

Moving caudally, the sciatic nerve can be imaged using a 7-MHz probe up to the popliteal fossa, where it divides into the peroneal and tibial nerves (Figure 51–8A). Here, the sciatic nerve often appears round and hyperechoic and is located posterior to the femur, lateral to the popliteal artery, and deep (anterior) to the semitendinous and semimembranous muscles medially and the biceps femoris muscle laterally (Figure 51–8B and 51–8C). More distally, the peroneal nerve may be followed as far laterally as the head of the fibula.

Ultrasound-guided blocks for peripheral nerves follow several general principles.

1. The quality of ultrasonographic nerve images captured depends on the quality of the ultrasound machine and transducers, proper transducer selection (eg, frequency) for each nerve location, the anesthesiologist's familiarity and interpretation of sonographic anatomy pertinent to the block, and good eye-to-hand coordination to track needle movement during advancement.

2. Optimal patient positioning and sterile technique are encouraged. This is particularly important for the continuous catheter technique, for which it is necessary to use sterile transmission gel and a sterile plastic sheath to fully cover the entire transducer.

3. Nerve localization by ultrasound can be combined with nerve stimulation. Both tools are valuable and complementary, not mutually exclusive. Ultrasonography provides anatomic information, while a motor response to nerve stimulation provides functional information about the nerve in question.

4. Observing local anesthetic spread is another valuable feature of ultrasound in addition to real-time visual guidance to navigate the needle toward the target nerve.

5. Two approaches are generally available to block peripheral nerves. The first approach aims to align and move the block needle in line with the long axis of the ultrasound transducer, so the needle stays within the path of the ultrasound beam (Figure 51–9). In this manner, the needle shaft and tip can be clearly visualized. This approach is preferred when it is important to track the needle tip at all times (eg, during supraclavicular block to minimize inadvertent pleural puncture). The second approach places the needle perpendicular to the probe (Figure 51–10). In this case, the ultrasound image captures a transverse view of the needle, which is shown as a hyperechoic “dot” on the screen. Accurate moment-to-moment tracking of the needle tip location can be difficult, and needle tip position is often inferred indirectly by tissue movement. This approach, however, is particularly useful for continuous catheter placement along the long axis of the nerve.

Outcome Studies

Ultrasound-guided techniques may improve the accuracy, success, and safety of regional anesthesia. However, few prospective randomized outcome studies have been conducted and published so far. Williams and coworkers suggested that the addition of ultrasound guidance improves the quality of supraclavicular block27 when compared with neurostimulator guidance alone. Marhofer and associates also suggest that ultrasound guidance speeds the onset, improves the quality, and reduces the incidence of vascular puncture during three-in-one blocks.28 No study to date, however, has examined the effect of ultrasound on nerve injury. In summary, although preliminary experience has been encouraging, more outcome data are required to define the success and safety profile of ultrasound-guided peripheral nerve blocks.

Neuraxial anesthetic techniques can be challenging because of anatomic variability among patients29 and imprecise determination of the level of the vertebral interspace by physical examination alone (inaccurate 70–80% of the time).30,31 Spinal needle insertion and local anesthetic injection at the wrong lumbar interspace (ie, too cephalad) may have been implicated in previously reported injuries to the conus medularis.32 Potentially, imaging guidance may improve accuracy and safety of needle placement during neuraxial blocks.

Over two decades ago, attempts were made to image the ligamentum flavum using ultrasonography.33 Because the epidural and subarachoid spaces are surrounded by bones, anatomic assessment in this region is difficult since the majority of the ultrasound beam is reflected on contacting the bony spinous processes. With a linear or curved 4- to 7-MHz probe, limited ultrasound beam passage is possible only through the interspinous space (Figure 51–11A), especially in the paramedian region. The ligamentum flavum and the dura mater are dense tissues that appear hyperechoic on ultrasound, whereas the low-density epidural space and the cerebrospinal fluid in the intrathecal space appear hypoechoic (Figure 51–11B).

Ultrasound determination of the spinal level is more accurate than clinical examination. This has been confirmed in two recent studies showing accurate ultrasound determination in over 70% of patients when compared with MRI examination.31,34 The markers were always placed within one interspace of the intended level. Ultrasonography can also determine the depth of needle penetration to reach the epidural space35 and can help reduce the number of needle puncture attempts. The paramedian region has been suggested by some to be the optimal window for ultrasound imaging, especially in the thoracic spine36 because of a higher soft tissue to bone ratio.37 In contrast to peripheral nerve blocks, real-time, image-guided neuraxial techniques have not been reported. Ultrasonography has been used primarily to help define the anatomy, depth, and angle of needle penetration immediately prior to performing the technique.