Inpatient Electromyography: Benefits and Limitations for the Hospital Evaluation of Neuromuscular Disorders
Inpatient evaluation of neuromuscular disorders often requires electrodiagnostic testing at the bedside. In our experience, the ordering physician does not always know how to appropriately select patients or critically interpret the results of the requested studies. We review the basic principles of electromyography and nerve conduction studies, including safety, limitations in the hospital setting, goals of testing, and scenarios where inpatient electrodiagnostic testing is often used. With this knowledge in tow, we hope to improve the ability of the ordering physician to select appropriate patients for testing and to foster realistic expectations about the contribution of electrodiagnostic studies to the diagnosis and treatment of inpatients.
Electromyography (EMG) is a valuable tool for characterizing disorders of muscle and the peripheral nervous system and is frequently used to evaluate inpatients with sensory loss and weakness. Electromyography is, however, very user-dependent interms of performance and interpretation and is fraught with significant limitations in the hospital setting. Armed with a basic knowledge of these shortcomings, the ordering physician can better tailor questions for the consulting electromy?ographer. What follows is by no means an exhaustive review for the seasoned electrodiagnostic consultant. Rather, it is a guide for reasonable expectations and an understanding of procedures performed at the request of the ordering physician.
What Is EMG?
Electrodiagnostic testing is commonly referred to as EMG, but such use is a misnomer; EMG is only one part of the electrodiagnostic examination, which typically consists of a nerve conduction study, followed by needle EMG. Nerve conduction testing involves stimulation of motor or sensory nerves using relatively small voltages. (By contrast, the voltage discharged from a commercially available stun gun is several hundred times stronger.) An action potential is propagated along the nerve and recorded distal to the stimulation. The distance and time between the stimulation and the recording pickups are noted. Three parameters are calculated:
? Distal latency (time, in milliseconds, between distal stimulation and recording)
? Baseline to peak amplitude of the action potential (in millivolts or microvolts)
? Conduction velocity (distance over time, in meters/second).
These measures are compared with normal laboratory values. Patterns of abnormality suggest axonal or demyelinating disease (Table 1).
Needle EMG measures the instability of the muscle membrane. A 24- to 30-gauge needle is inserted into the muscle, irritating its membrane. No electricity is delivered; the intrinsic electrical activity of relaxed and activated muscle fibers is recorded. Relaxed muscle should be quiescent, with activity only noted on needle insertion and movement (?sampling?).
Abnormal findings seen in both myopathic and neurogenic disorders include increased insertional activity, positive waves, and fibrillations. These events denote instability of the muscle membrane brought out by needle irritation. Activation of the muscle evaluates the voluntary motor unit. The amplitude, duration, size, and pattern of recruitment of additional muscles with increased contraction is noted. An abnormal pattern may denote a neurogenic or myopathic process (Table 2).1
Safety and Pain Concerns
Given that electricity is used during nerve conduction studies, safety is a concern. Although the electric current used is relatively weak, the patient should always be properly grounded during the study. This has the added benefit of reducing shock artifact, allowing for accurate interpretation of the findings. Even with proper grounding, several conditions should give the electromyographer pause. In a patient with a catheter leading to the right atrium of the heart, electricity may be conducted from the stimulator to the cardiac conduction system, inducing an arrhythmia.2 Hence, performing electrodiagnostic testing on a limb with a central line should be avoided whenever possible. Likewise, since an external pacemaker or defibrillator could malfunction or misfire when an outside electrical source is added, nerve conduction studies should be avoided in such patients.
Needle EMG can also be dangerous in certain situations. Puncturing the skin over a site of infection is unwise, because the needle can introduce the infection into deeper structures. Some electromyographers advocate cleaning the skin with an alcohol swab before every insertion; however, their reasoning appears to be related more to legal liability than to any true risk of infection. Any bleeding is usually self-limited, even in patients taking antiplatelet agents or therapeutic anticoagulation (although noncompressible deep sites should usually be avoided).3 Needlesticks are generally a greater threat to the electromyographer than to the patient, with HIV, hepatitis viruses, and other bloodborne pathogens rampant in today?s hospital population.4
The electromyography/nerve conduction study is usually very well-tolerated, with minimal lingering effects. We warn patients who come to our laboratory that they may experience some degree of discomfort. Stimulations have been likened to a rubber band being snapped against the skin. Needle EMG causes some sharp pain with insertion, and patients may complain of a cramping sensation in the muscle. A skilled electromyographer uses small needle excursions to effectively sample muscle and minimize pain. Still, the running joke in our laboratory is that the most frequently uttered word of the electromyographer is ?sorry.?
Localizing Nerve and Muscle Problems
Electrodiagnostic testing evaluates the integrity of the peripheral nervous system. Needle EMG can detect abnormalities in muscle caused by denervation or intrinsic muscle disease. Motor nerve conduction studies examine the motor unit, which includes the nerve cell body in the anterior horn of the spinal cord, the nerve axon (as the neural impulse moves from nerve root, plexus, and distal nerve), the neuromuscular junction, and the muscle fibers innervated by the nerve. Sensory nerve conduction studies test large-diameter nerve fibers that carry proprioceptive information and vibrations from receptors in the skin to the cell bodies in the dorsal root ganglion. Since the dorsal root ganglion is outside the nerve root entry zone of most compressive radiculopathies (ie, lesions involving the nerve root), sensory nerve conduction studies are usually normal in such cases. Figure 1 depicts the elements of sensory and motor nerve conduction studies.
Electrodiagnostic testing is not a substitute for a thorough neurologic examination. Weakness and numbness can result from stroke, multiple sclerosis, or other lesions of the central nervous system. A careful examination concentrating on muscle power, reflexes, and sensation should be performed by the ordering physician, and results should be consistent with peripheral pathology before ordering electrodiagnostic studies. For example, unilateral ankle clonus and a Babinski sign would be inconsistent with sacral nerve root compression.
Electromyographers attempt to answer the questions of the consulting physician by a careful application of peripheral neuroanatomy. The principles of commonality and exclusion are essential in localizing problems. The savvy electromyographer will try to surround abnormal findings with normal findings to rule out one of several possibilities suggested by the physical examination. For instance, in distinguishing an L5 radiculopathy from a peroneal nerve injury, abnormal peroneal sensory and motor nerve conduction studies with normal nonperoneal L5-innervated muscle findings on needle electromyography would suggest a peroneal neuropathy.
Electromyographers Performing Other Studies
The most common electrical studies other than standard nerve conduction studies and needle EMG are late responses and repetitive nerve stimulations. Late responses include F-wave latencies and H-reflexes (Figure 2). Technical performance of F-waves is essentially identical to motor nerve conduction studies; stimulation of a motor nerve causes nerve depolarization both distally and proximally. The action potential depolarizes the motor nerve root to the anterior horn cell, causing the motor neuron to propagate an action potential distal to the motor root and nerve(s). The latency is measured and compared with laboratory standards. A prolonged latency is typically seen in demyelinating disorders of the proximal motor nerve or nerve root. The H-reflex is usually only reliably performed on the tibial nerve, recording from the gastro?cnemius/soleus muscles. It is the electrophysiologic equivalent of the ankle jerk tendon reflex, representing the same sensory and motor arc. Comparison between the left and right legs is usually indicated.5
Repetitive nerve stimulations are used to evaluate dysfunction of the neuromuscular junction using a train of 4 nerve stimulations, at a stimulation frequency of 2 to 3 per second on a motor nerve. The compound muscle action potential amplitudes of the first and fourth stimulations are compared; a pathologic result is considered to be a decrement greater than 10%. The pattern of decrement before and after exercise varies, based on the disorder (ie, myasthenia gravis, Lambert-Eaton myasthenic syndrome, or botulism).6
The Limits of Inpatient EMG
Electrodiagnostic testings can be very helpful in identifying neuromuscular disorders in hospitalized patients, but they have significant limitations. Bedside nerve conduction studies and needle EMG are likely to be contaminated by electrical artifacts in the presence of drips, lines, ventilators, or other sources of electrical interference. Although good technique and grounding can reduce the effects of interference from outside electrical devices, studies in the intensive care unit (ICU) can be difficult to perform and interpret.
|Suspected diagnosis||Special studies||Comment|
|Critical illness neuropathy/myopathy||Sensory and motor NCS Needle EMG||Extent of myopathy vs neuropathy may be difficult to distinguish|
|Guillain-Barr? syndrome||Sensory and motor NCS F- wave late responses||Routine studies may be normal early in disease course|
|Motor neuron disease||EMG of the bulbar, thoracic paraspinal muscles||Establishes multinerve, multilimb distribution|
|Myasthenia gravis||RNS||Distal RNS may be normal|
|Traumatic nerve injury Needle EMG||Sensory and motor NCS not suspected||Wait up to 3 weeks postinjury if transection|
|EMG = electromyography; NCS = nerve conduction study; RNS = repetitive nerve stimulation.|
Another issue that can interfere with hospital EMG is the ability of the patient to cooperate. A delirious individual may rip off the leads and have exaggerated responses to the discomfort of electrodiagnostic studies. Evaluation of voluntary motor units during needle EMG is impossible in the patient with impaired consciousness. We would caution against the use of sedatives to allay procedure-related anxiety.
Other issues that are not unique to inpatient electrodiagnosis have already been alluded to. Since nerve conduction studies are performed on distal nerve segments, more proximal lesions may not be detected. Late responses may be particularly disrupted by an electrical artifact. Patients with central lesions in the spinal cord, brainstem, or cerebrum will likely have normal EMG and nerve conduction study results.
Indications for Inpatient Electrodiagnostic Testing
Failure to wean patient from the ventilator
Intubation for respiratory insufficiency is generally a reversible process. Failure to wean a patient from the ventilator after resolution of the primary process should prompt a search for other causes, including neuromuscular diseases, either acquired in the ICU or existing before intubation. Critical illness polyneuropathy or myopathy occurs in up to 60% of patients with acute respiratory distress syndrome,7 likely because of the prolonged use of steroids and neuromuscular paralytics.8 The pattern of abnormality seen on electromyography/nerve conduction studies may distinguish between myopathy and polyneuropathy. Guillain-Barr? syndrome is often preceded by surgery, pregnancy, or bacterial infection9 and can also be associated with HIV seroconversion.10 In early Guillain-Barr? syndrome, distal nerve conduction studies and EMG may be normal, with prolonged F-wave responses only. Generalized myasthenia gravis may be unmasked by the stresses of the ICU,11 and repetitive nerve stimulation should show an amplitude decrement of more than 10% from the first to the fourth stimulations. Bulbar-predominant motor neuron disease will demonstrate widespread denervation, with reduced compound muscle action potential amplitudes and evidence of muscle membrane instability and abnormal voluntary units in a multinerve, multilimb distribution (Table 3).12
Traumatic nerve injury
Patients with long-bone injuries secondary to crush or penetrating trauma often have concomitant nerve damage. It usually takes 2 to 3 weeks before muscle membrane instability develops after axonal injury; in addition, nerve conduction study results may be normal if stimulations do not include the injured nerve segment.13 Early studies may only be able to rule out complete nerve lesions. Although the presence of voluntary units in muscle innervated by a damaged nerve excludes nerve transection, their absence does not always indicate a complete injury. We recommend waiting 3 weeks after injury before performing electrodiagnostic studies.
Patient who cannot walk
Older patients often have functional leg weakness or gait instability as a result of pain, deconditioning, or central or peripheral nervous system lesions. The presence of sensory ataxia can be readily determined by evaluating for a Romberg sign, tuning fork testing, and assessment of joint position sense. The absence of some sensory nerve action potentials may be normal in the elderly population. True peripheral neurogenic weakness may be the result of myopathy, polyneuropathy, focal mononeuropathy, radiculopathy, or spinal stenosis. Electrodiagnostic studies can confirm the localization of gait abnormalities of peripheral nervous system origin.14
Patients often develop weakness after prolonged surgery from nerve stretching during positioning, nerve compression caused by hematoma, or direct surgical ligation. The same comments regarding traumatic nerve injury apply here, and a 3-week waiting period is usually indicated before testing.
Patients with or without evidence of secondary gain may complain of weakness despite normal muscle tone and reflexes. In such individuals, the results of imaging and laboratory studies are either normal or equivocal. The number of muscles and nerves tested during electrodiagnosis in an attempt to exclude abnormality, however small, can be extensive and may include repetitive nerve stimulations and late responses.15 Unfortunately, the likelihood of a false-positive result is increased as the number of tests increase.16
If the ordering physician insists on an electrodiagnostic test, we recommend a limited nerve conduction study and a minimal number of muscles tested by needle EMG. The patient?s failure to cooperate is another potential problem. In our opinion, the less credence that can be given to somatic causes of evident psychogenic complaints, the greater the likelihood that the patient will find appropriate psychologic or psychiatric treatment.
Fasciculations arouse the suspicion of motor neuron disease among health care providers.17 In the absence of lower motor neuron signs, such as atrophy or weakness, and upper motor neuron signs, such as hyperreflexia, amyotrophic lateral sclerosis is unlikely. Causes of fasciculations include excessive caffeine intake, sleep deprivation, the benign fasciculation cramping syndrome, and muscle overuse.18 Electrodiagnostic studies may be ordered to quell patient or physician anxiety; there is little role for inpatient EMG, and our recommendation is that EMG studies in such patients should only be performed electively, if at all.
Acute pain may be related to injury to the bone, muscle, joint, or ligaments and only occasionally to injury to the nerves. Low back pain from an acute herniated disk resolves within 6 to 8 weeks in 90% of patients.19 As with traumatic nerve injury, results of EMG of the relevant muscles affected by nerve root compression will likely be normal for up to 3 weeks postinjury. Patients with severe, acute pain often have little tolerance for the additional discomfort of electrodiagnostic studies, so we may postpone them until the acute pain has subsided.
Follow-up testing is indicated when management would be affected by their results. Months after traumatic nerve injury, an inability to generate voluntary units in affected muscles may indicate the failure of axonal regeneration, necessitating surgical intervention. In contrast, electromyography for a muscle biopsy-proven myopathy or repetitive nerve stimulation for acetylcholine receptor antibody-positive myasthenia gravis is merely redundant. After a positive electrodiagnostic study in these cases, repeat evaluations rarely add information to what can be gained from clinical examination.
The electrodiagnostic evaluation of inpatients can be a useful adjunct to clinical examination and appropriate imaging studies. With focused questions and reasonable expectations from the consulting physician, needle EMG and nerve conduction studies can provide valuable information about diseases of the peripheral nervous system. Knowledge of the shortcomings of these tests is crucial to their interpretation as well as to the care of the patient.
The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Department of the Army or the US Department of Defense.
1. All the following statements about electrodiagnostic testing are true, except:
A. Needle EMG delivers low-voltage electricity
B. Needle EMG measures the stability of the muscle membrane
C. Patients should be grounded during nerve conduction studies
D. Sensory nerve conduction studies test large-diameter nerve fibers
2. Which of these nerve conduction study findings is most characteristic of axonal injury?
A. Reduced amplitude
B. Prolonged distal latency
C. Reduced conduction velocity
D. Increased conduction velocity
3. Which of these statements about late responses is NOT true?
A. They are used to evaluate proximal nerve segments
B. They can be used to diagnose polyneuropathies such as Guillain-Barr? syndrome
C. F-waves help evaluate the sensory component of the peripheral nervous system
D. The H-reflex is evoked by stimulation of the tibial nerve
4. All the following conditions are relative contraindications to electrodiagnostic testing, except:
A. External pacemaker
B. Needle EMG over cellulitis
C. Therapeutic oral anticoagulation
D. Limb with a central line
5. Electrodiagnostic testing should be delayed for 3 weeks in the following situations, except:
A. Gait instability in an elderly woman
B. Acute-onset low back pain in an elderly man
C. Postoperative weakness
D. Penetrating leg injury
1. Krarup C. Pitfalls in electrodiagnosis. J Neurol. 1999;246: 1115-1126.
2. Barry DT. AAEM minimonograph #36: basic concepts of electricity and electronics in clinical electromyography [published correction appears in Muscle Nerve. 1991;14:1153]. Muscle Nerve. 1991;14:937-946.
3. Al-Shekhlee A, Shapiro BE, Preston DC. Iatrogenic complications and risks of nerve conduction studies and needle electromyography. Muscle Nerve. 2003;27:517-526.
4. American Association of Electrodiagnostic Medicine Professional Practice Committee. Risks in electrodiagnostic medicine. Muscle Nerve. 1999;22(suppl 8):S53-S58.
5. Bischoff C. Neurography: late responses. Muscle Nerve. 2002;26 (suppl 11):S59-S65.
6. Shapiro BE, Preston DC. Repetitive nerve stimulation and exercise testing. Phys Med Rehabil Clin N Am. 2003;14:185-206.
7. Bercker S, Weber-Carstens S, Deja M, et al. Critical illness polyneuropathy and myopathy in patients with acute respiratory distress syndrome. Crit Care Med. 2005;33:711-715.
8. Bolton CF. Neuromuscular manifestations of critical illness. Muscle Nerve. 2005;32: 140-163.
9. Govoni V, Granieri E. Epidemiology of the Guillain-Barr? syndrome. Curr Opin Neurol. 2001;14: 605-613.
10. Dalakas MC, Pezeshkpour GH. Neuromuscular diseases associated with human immunodeficiency virus infection. Ann Neurol. 1988;23(suppl):S38-S48.
11. Marinelli WA, Leatherman JW. Neuromuscular disorders in the intensive care unit. Crit Care Clin. 2002;18:915-929, x.
12. Krivickas LS. Amyotrophic lateral sclerosis and other motor neuron diseases. Phys Med Rehabil Clin N Am. 2003;14:327-345.
13. Robinson LR. Traumatic injury to peripheral nerves. Muscle Nerve. 2000;23:863-873.
14. Dillingham TR. Electrodiagnostic approach to patients with weakness. Phys Med Rehabil Clin N Am. 2003;14: 163-184.
15. Wilbourn AJ. The electrodiagnostic examination with hysteria-conversion reaction and malingering. Neurol Clin. 1995;13:385-404.
16. Rivner MH. Statistical errors and their effect on electrodiagnostic medicine. Muscle Nerve. 1994;17:811-814.
17. Blexrud MD, Windebank AJ, Daube JR. Long-term follow-up of 121 patients with benign fasciculations. Ann Neurol. 1993;34:622-625.
18. Kincaid JC. Muscle pain, fatigue, and fasiculations. Neurol Clin. 1997;15:697-709.
19. Deyo RA, Weinstein JN. Low back pain. N Engl J Med. 2001; 344:363-370.