Approach to Pediatric Electromyography

Neuromuscular Diagnoses are Different in Children Compared with Adults

The most common referral diagnoses to the typical electromyography (EMG) laboratory include radiculopathy, polyneuropathy, and carpal tunnel syndrome. However, adults are more commonly studied in the EMG laboratory, so this group of diagnoses reflects neuromuscular conditions seen in the adult age group. In contrast, the neuromuscular disorders seen in children often are different. For example, entrapment neuropathies are very common in adults but are extremely rare in children. Likewise, radiculopathy, probably the most common of all EMG referral diagnoses, is virtually unheard of in children, except in cases of trauma. Although peripheral neuropathies occur in children, they are most often genetic, whereas most peripheral neuropathies in adults referred to the EMG laboratory are acquired disorders, usually toxic, metabolic, inflammatory, or associated with other coexistent medical illnesses. Unlike adults, the more common diagnoses in children referred to the EMG laboratory are inherited disorders of the motor unit, including the anterior horn cell (e.g., spinal muscular atrophy), peripheral nerve (e.g., Charcot-Marie-Tooth [CMT]), or muscle (e.g., muscular dystrophy).

Children with neuromuscular disorders often present clinically as a delay in motor milestones. In many cases, it may not be clear from the symptoms and signs whether the etiology is central or peripheral. One of the best examples of this predicament is that of the floppy infant, in whom the differential diagnosis includes the entire length of the neuraxis, from brain to muscle. In this regard, EDX studies often are helpful in differentiating peripheral from central etiologies and, accordingly, guiding the subsequent evaluation in a useful and logical direction.

Maturation Issues

When studying children, it is essential to appreciate what is normal for what age . This is especially important when interpreting conduction velocities and differentiating a normal conduction velocity from axonal loss or demyelination. Most adult electromyographers who study adults are well versed in the EDX criteria for demyelination:

• Conduction velocities less than 75% the lower limit of normal

• Distal latencies and late responses greater than 130% the upper limit of normal

• Conduction block, which signifies not only demyelination but acquired demyelination

However, infants and young children often have slowed conduction velocities that would be considered in the “demyelinating range” for adults. In most cases, this is not because infants and young children have demyelinated nerves; rather, they have nerves that have yet to be myelinated in the first place. The process of myelination is age dependent, beginning in utero, with nerve conduction velocities in full-term infants approximately half that of adult normal values. Accordingly, nerve conduction velocities of 25–30 m/s are normal at birth . Conduction velocity rapidly increases after birth, reaching approximately 75% of adult normal values by age 1 year, and the adult range by age 3–5 years, when myelination is complete. Accordingly, when a child is studied in the EMG laboratory, it is essential that age-based normal control values are used ( Tables 41.1 and 41.2 ).

One interesting aspect of myelin maturation is often observed during nerve conduction studies in the pediatric population. Many are familiar with the fact that different white matter tracts in the central nervous system myelinate at different times. Indeed, one can often use the pattern of myelination on a brain MRI scan to correctly predict the age of a young child. Similarly, different fibers in the peripheral nervous system myelinate at different times as well. In the EMG laboratory, this often manifests as a bifid morphology (i.e., two separate peaks) on sensory nerve action potentials (SNAPs) in infants and children ( Fig. 41.1 ). This bifid morphology is due to some fibers having already been fully myelinated (the first peak), whereas others have not and trail behind (i.e., the second peak). It is not unusual to see bifid SNAPs between the ages of 3 months and 4–6 years. These bifid SNAPs are a completely normal finding. Eventually, as the fibers in the second peak fully myelinate, the second peak moves to the left and merges with the first peak. This forms a larger sensory response, as is typically seen in adults.

As with adults, the F response can be easily studied in children. Although the F response often is thought of as evaluating the proximal nerve segments, it assesses the entire length of the nerve, from the stimulation point to the spinal cord and back, and then past the stimulation point to the muscle. Thus, the F response latency depends not only on the conduction velocity and distal latency but also on the length of the limb. Because infants and children have slower conduction velocities than adults, one would expect the F responses to be very long. However, counterbalancing this is the very short limb length of a child compared with an adult. Thus, there are two opposing influences on the F response in children: limb length and conduction velocity. In infants and young children, the influence of the limb length is more overriding, resulting in F-wave latencies that are much shorter in children than adults (typically in the range of 16–19 ms in the upper extremities). Thus, whenever an F response is performed on a child, it is essential to compare it with normal control values for the child’s age or height.

The most important maturation issue for the needle EMG portion of the examination is the size of the motor unit. It is no surprise that the physical size of a motor unit of a newborn is much smaller than that of an adult. Transverse motor unit territory increases greatly with age, doubling from birth to adulthood, mostly because of the increase in individual muscle fiber size. Thus, normal motor unit action potentials (MUAPs) in infants typically are very small, representing the physical size of the motor unit . Indeed, in infants, it is often difficult to differentiate normal MUAPs from myopathic one s. This once again underscores that when one interprets EDX findings in children, including MUAPs, it is essential to use age-based normal control values ( Table 41.3 ).