Ultrasound of Focal Neuropathies

Published on 06/06/2015 by admin

Filed under Physical Medicine and Rehabilitation

Last modified 06/06/2015

Print this page

This article have been viewed 2696 times

Chapter 5 Ultrasound of Focal Neuropathies

Michael S. Cartwright, MD

Key Points

• Focal neuropathies are common and cause significant morbidity and medical expense.

• Neuromuscular ultrasound addresses some of the shortcomings of electrodiagnostic studies for the assessment of focal neuropathies because it permits painless assessment of extrinsic and intrinsic causes of focal neuropathy and helps confirm localization inferences made from nerve conduction studies and electromyography.

• Ultrasonographic findings seen in focal neuropathies include focal nerve enlargement, decreases in nerve echogenicity and mobility, increases in nerve vascularity, and extrinsic structures such as cysts, vessels, tumors, and muscles compressing nerve.

• Neuromuscular ultrasound is used most commonly to assess carpal tunnel syndrome, and findings such as increased median nerve cross-sectional area proximal to the tunnel, decreased nerve echogenicity and mobility, and increased nerve vascularity result in diagnostic sensitivity and specificity near 90%.

• In ulnar neuropathy at the elbow, nerve enlargement seen with ultrasound helps localize the site of entrapment with sensitivity and specificity greater than 80%.

• Ultrasound identifies intraneural ganglia in 18% of otherwise unexplained cases of peroneal neuropathy at the knee, which often leads to successful surgical intervention.

• Published case reports and series demonstrate the usefulness of neuromuscular ultrasound for assessing focal neuropathy in the radial, suprascapular, sciatic, tibial, femoral, and lateral femoral cutaneous nerves.

Focal neuropathy is a broad term that refers to a discrete peripheral nerve lesion that causes pain, numbness, or weakness. Focal neuropathies are classified by either location or etiology. The location refers to the specific site along the peripheral nerve. The etiology can be further divided into extrinsic and intrinsic causes. Extrinsic causes include any structure compressing and damaging the nerve, such as ganglion cysts, inflamed tendon sheaths (tenosynovitis), anomalous muscles, aberrant or dilated vessels, and benign or malignant tumors. The most common extrinsic cause of focal neuropathy is entrapment of a nerve as it passes through a stiff fibro-osseous tunnel, such as the carpal tunnel in the wrist or the cubital tunnel near the elbow.¹ Extrinsic etiologies also include trauma to the nerve, which can result from prolonged compression, acute blunt injury, or penetrating injury. Intrinsic etiologies are less common than extrinsic. Examples of intrinsic causes include neuromas, nerve sheath tumors, and nerve infarct (often associated with vasculitis).

The epidemiology of all focal neuropathies taken together is not known because they are coded and evaluated separately, based on location or etiology. However, by assessing the epidemiology of only the most common types, it is clear that focal neuropathies are prevalent and cause significant morbidity and medical expense. Carpal tunnel syndrome (CTS), or median mononeuropathy at the wrist, is the most common entrapment neuropathy, with an age-adjusted incidence of 105 cases per 100,000 person-years and a prevalence of 3% to 10% in the general population.^1,² It accounts for more than $500 million annually in health-care and indemnity costs in the United States, and individuals with CTS have an average of 84 work days lost.^3,⁴ Ulnar neuropathy at the elbow is the second most common focal neuropathy, with a standardized incidence of 20.9 cases per 100,000 person-years.⁵ Other common focal neuropathies include ulnar neuropathy at the wrist, radial neuropathy in the spiral groove, and peroneal neuropathy at the fibular head.^6,⁷

The diagnostic approach for an individual with a potential focal neuropathy varies based on presentation, site of neuropathy, suspected etiology, local standard of care, and physician preference. For the most common focal neuropathies, such as CTS or ulnar neuropathy at the elbow, some physicians will make the diagnosis and treatment decisions based on history and examination alone, whereas most prefer to confirm their clinical opinion with nerve conduction studies (NCS) and electromyography (EMG) for any type of focal neuropathy. For cases with an atypical presentation, or neuropathy at a nonentrapment site, electrodiagnostic studies (NCS and EMG) are recommended.⁸

Although the information obtained from electrodiagnostic studies is helpful, there are important limitations to the current diagnostic approach for the evaluation of focal neuropathies. First, structures extrinsic to the nerve, such as tendons, bones, and vessels, are not assessed with NCS and EMG. Second, the etiology of the neuropathy (entrapment, compression from a mass, and so on) can be inferred with electrodiagnostic studies but not directly detected. Third, similar to the previous limitation, electrodiagnostic studies require inferences regarding localization because they do not directly visualize the lesion. Fourth, there are both false-positive and false-negative studies; and fifth, NCS and EMG are painful, which can limit the extent of a study, particularly in children. All of these limitations hamper the ability to detect the etiology and localization of focal neuropathies, but imaging can fill in these gaps.

High-resolution ultrasound has evolved over the past two decades such that small structures, like nerves, are now easily visualized.^9,¹⁰ It also allows visualization of surrounding structures, such as muscles, tendons, bones, and vessels, as discussed in Chapter 4. Ultrasound addresses the shortcomings of electrodiagnostic studies because it permits direct assessment of extrinsic and intrinsic causes of focal neuropathy and helps confirm localization inferences made from NCS and EMG. In addition, ultrasound is painless.

The first study to use ultrasound to assess a focal neuropathy occurred in the early 1990s and was in individuals with CTS.¹¹ Since then, ultrasound has been studied for the diagnosis of focal neuropathies at sites throughout the arm and leg. This chapter describes the techniques used to image focal neuropathies with ultrasound as well as the data supporting its use. The techniques described in this chapter are brief, because the chapter focuses more on pathologic conditions, but more detailed protocols are listed in the Appendix. In addition, it should be noted that the imaging sequences in this chapter often describe the longitudinal view of the nerve first, but the protocols typically suggest starting with a cross-sectional view. As is described later, changes in nerve cross-sectional area are often seen with focal neuropathies, so cross-sectional area reference values are provided (Table 5.1).

Table 5.1 Nerve Ultrasound Cross-sectional Area Reference Values

Median Nerve

Wrist (Carpal Tunnel Syndrome)

In addition to being one of the first areas explored with neuromuscular ultrasound, the median nerve at the wrist, and carpal tunnel syndrome in particular, is the area most commonly studied and described in the literature. This is because carpal tunnel syndrome is extremely common and ultrasound of the median nerve at the wrist is relatively straightforward from a technical standpoint.

Imaging starts with a sagittal view, by placing the linear array transducer across the wrist (Fig. 5.1). The image obtained in this position includes the median nerve in a longitudinal view, as well as the flexor digitorum superficialis and profundus tendons (Fig. 5.2). This view allows the physician to obtain a general overview of the median nerve and other wrist structures. It also provides an image in which pinching of the median nerve, as it passes under the rigid flexor retinaculum, can be assessed subjectively and objectively.¹² In this sagittal view, flexion of the fingers causes sliding of the tendons in a distal-to-proximal plane. In normal conditions, the median nerve also slides in this same plane¹³ but with less displacement than the tendons (Video 5.1). One study demonstrated decreased longitudinal sliding of the median nerve in those with CTS compared with controls,¹⁴ but this finding was not confirmed in a separate report.¹⁵

Fig. 5.1 A high-resolution linear array transducer is placed over the wrist to obtain a sagittal image of the median nerve.

Fig. 5.2 A sagittal image of the median nerve (N) at the wrist. The flexor tendons (T) are seen running deep to the nerve.

Next, the linear array transducer is rotated 90 degrees and a cross-sectional image is obtained at the level of the distal wrist crease (Fig. 5.3). In this image, the transverse carpal ligament, median nerve, flexor digitorum tendons, and carpal bones can be seen (Fig. 5.4). This is the view in which the majority of objective measurements have been reported. As will be outlined, many different techniques have been described for acquiring measurements that accurately differentiate those with CTS from healthy controls, but the universal finding is that the cross-sectional area of the median nerve is enlarged at the entrance to the carpal tunnel in those with CTS (see Chapter 2 for a list of pathologic processes known to result in nerve enlargement). Initial studies suggested that internal landmarks, such as the hook of the hamate or level of the pisiform bone be used to determine the site at which the cross-sectional area measurement is obtained.^11,¹⁶ However, subsequent studies have shown that finding the area of maximal median nerve enlargement proximal to the entrance of the tunnel and obtaining the cross-sectional area of the median nerve at this site (Fig. 5.5), are sensitive and specific and require only one cross-sectional area measurement be obtained.¹⁷ Some clinicians have suggested that demyelinating polyneuropathies, such as Charcot-Marie-Tooth, which cause diffuse nerve swelling (see Chapter 7), could be mistaken for CTS if only a single measurement of the median nerve cross-sectional area at the wrist is obtained. Therefore, Hobson-Webb and colleagues showed that a ratio of the median nerve area at the wrist to the area of the nerve in the forearm is an accurate method to diagnose CTS and handle the potential problem of diffuse nerve enlargement.¹⁸ Their study showed that a ratio greater than 1.4 resulted in a sensitivity of 100% for the diagnosis of CTS. (A similar ratio of greater than 1.4 is also used to diagnose aneurysmal enlargement of the arteries, as described in Chapter 4.)

Fig. 5.3 The transducer is placed over the distal wrist crease to obtain a cross-sectional image of the median nerve at the wrist.

Fig. 5.4 This is a cross-sectional view of the median nerve (N) at the wrist. The flexor tendons (T) and carpal bones (C) are seen deep to the nerve. The nerve has the typical honeycomb appearance with multiple fascicles.

Fig. 5.5 Cross-sectional views taken at the distal wrist crease. A, This individual does not have carpal tunnel syndrome and the area of the median nerve (outlined) is 10 mm²B, The individual in this image has carpal tunnel syndrome and the area of the nerve (outlined) is 18 mm². The nerve in image B is also more hypoechoic than in A, and the normal honeycomb nerve pattern seen in A is lost in image B.

In addition to the cross-sectional area of the median nerve at the wrist, other measurements have been proposed to diagnose CTS with neuromuscular ultrasound (Box 5.1). Nerve flattening, which can be numerically demonstrated by comparing the maximum with the minimum nerve diameter, is present in CTS.¹⁹ Bowing of the flexor retinaculum is also reported in CTS.¹¹ Similar to the findings in the longitudinal plane, decreased nerve mobility has been reported with dynamic imaging in a transverse view.²⁰ Median nerve hypervascularity, as detected with color flow Doppler, has been described as a sensitive (95%), and relatively specific (71%) technique for the detection of carpal tunnel syndrome.²¹ This finding has not yet been confirmed in subsequent studies, and anecdotal experience in the author’s laboratory suggests that this degree of accuracy may be somewhat overinflated, but the observation is valid. The finding of median nerve hypervascularity, which has also been noted with magnetic resonance imaging (MRI),²² may provide insight into the pathophysiology of CTS because it suggests there is a circulatory disturbance in the median nerves of those with CTS. Finally, the median nerve of those with CTS is frequently found to be hypoechoic; however, this finding has not been systemically studied or quantified.²³

Other anatomic findings have been detected using ultrasound to examine the wrists of those with CTS. One finding that is noted rather commonly, and is estimated to occur in 16% of wrists,²⁴ is the presence of a persistent median artery running through the carpal tunnel (Fig. 5.6; Table 5.2). The relevance of this patent vessel regarding CTS is not known, but there are case reports of thrombosed persistent median arteries leading to CTS.^25,²⁶ A finding that at times coexists with a persistent median artery is that of a bifid median nerve (Fig. 5.7). The median nerve typically branches distal to the carpal tunnel, but in 9% of healthy controls and 19% of those with CTS a bifid median nerve is present proximal to, or within, the carpal tunnel (see Table 5.2).²⁷ It is postulated that individuals with bifid median nerves are predisposed to the development of CTS, but this has not been confirmed in a prospective fashion. Of note, hand surgeons are often unable to confirm the presence of a bifid median nerve during open carpal tunnel release, and it is thought that perhaps the epineurium is not completely divided, so it obscures the presence of the early median nerve division.

Fig. 5.6 These images are taken from the same wrist and demonstrate a persistent median artery adjacent to the median nerve. A, An anechoic round structure medial to the nerve. B, Power Doppler flow signal consistent with an artery.

Table 5.2 Common Peripheral Nerve Findings

Finding	Frequency in Healthy Controls	Frequency in Disease State
Persistent median artery	16%	NA
Bifid median nerve	9%	19% in CTS
Ulnar subluxation	25%	NA
Peroneal ganglia	NA	18% in peroneal neuropathy at fibular head

CTS, Carpal tunnel syndrome; NA, not available.

Fig. 5.7 This image is taken from an individual with carpal tunnel syndrome. The median nerve is bifid, with each division of the nerve outlined. There is a hyperechoic area between the two branches of the nerve, but a persistent median artery was not detected. Each branch of the median nerve is hypoechoic, and the overall cross-sectional area of the two branches added together is 40 mm².

Other Sites

Although the majority of neuromuscular ultrasound research involving the median nerve has focused on the wrist and CTS, there are scattered reports of focal median nerve pathology detected at sites outside the wrist. The technique for imaging these proximal sites varies based on the location of interest, but in general a cross-sectional view is used. In the author’s lab, imaging typically starts with a cross-sectional view at either the wrist or the antecubital fossa, and then the median nerve is tracked farther distal or proximal to the area of interest (Video 5.2). As the median nerve is traced distal into the palm it splits into several branches, which can be challenging to image (Video 5.3). There are two case reports of mid-palm ganglion cysts causing isolated mononeuropathies of the motor branch of the median nerve, which were detected by ultrasound (Fig. 5.8).^28,²⁹ The median nerve is more easily seen as it is traced proximal from the wrist. Several case reports have described pathology in the forearm leading to median mononeuropathies, including traction injury, neurilemmoma, and a tendon sheath fibroma (Figs. 5.9 and 5.10).³⁰^–³² Other pathologic processes, detected with ultrasound, also have been reported to cause median mononeuropathies in the upper arm, proximal to the antecubital fossa. These include compressive vascular anomalies, neuromas, and hamartomas (Figs. 5.11 and 5.12).³³^–³⁷

Fig. 5.8 Ultrasound of the thenar eminence shows a spherical anechoic lesion (arrow), which is consistent with a mid-palm ganglion cyst. This ganglion cyst was compressing the median nerve and leading to a median mononeuropathy.

(From Kobayashi N, Koshino T, Nakazawa A, Saito T: Neuropathy of motor branch of median or ulnar nerve induced by midpalm ganglion, J Hand Surg Am 26:474–477, 2001.)

Fig. 5.9 A, The cross-sectional ultrasound image of the proximal forearm demonstrates the normal echo-texture of the median nerve (straight arrow). The hyperechoic and homogeneous ground-glass appearance of the flexor digitorum profundus muscle (curved arrows) is also shown. B, The intraoperative photo depicts a fibrotic band (straight line) across the anterior aspect of the median nerve (arrow). This process occurred after traction injury to the nerve. Arrowheads, Arteries; ∗, pronator teres muscle.

(From Ginn SD, Cartwright MS, Chloros GD et al: Ultrasound in the diagnosis of a median neuropathy in the forearm: case report, J Brachial Plex Peripher Nerve Inj 2:23, 2007.)

Fig. 5.10 Ultrasonography of neurilemmoma. A, Longitudinal ultrasound of a mass in the left forearm demonstrates an ovoid, well-defined solid mass with a heterogeneous echo pattern. The median nerve (arrow) is in direct contact with the mass. B, Cross-sectional ultrasound at the junction of the tumor and the normal median nerve demonstrates the origin of the tumor (black arrows) and displacement of the noninvolved fascicles (white arrows). C, Cross-sectional ultrasound of the proximal median nerve shows normal fascicles contained in a normal-sized nerve (arrows).

(From Kuo YL, Yao WJ, Chiu HY: Role of sonography in the preoperative assessment of neurilemmoma, J Clin Ultrasound 33:87–89, 2005.)

Fig. 5.11 A, Ultrasound of the arm in an individual with a high median mononeuropathy shows the median nerve (arrow) adjacent to a pseudoaneurysm (arrowhead). The hyperechoic area superior and lateral to the median nerve is dense inflammatory tissue that is encasing the nerve and binding it to the pseudoaneurysm. B, A normal median nerve (arrow) and brachial artery (arrowhead) from an unrelated control for comparison.

(From Cartwright MS, Donofrio PD, Ybema KD, Walker FO: Detection of a brachial artery pseudoaneurysm using ultrasonography and EMG, Neurology 65:649, 2005.)