Electrodiagnostic Medicine III: Case Studies

Published on 06/06/2015 by admin

Filed under Physical Medicine and Rehabilitation

Last modified 06/06/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 862 times

Chapter 11 Electrodiagnostic Medicine III

Case Studies

The goal of electrodiagnostic (EDX) testing is to assess the integrity of the nerves, the neuromuscular junction, and the muscle unit. As an extension of the neurologic examination, EDX testing identifies neuromuscular disease and provides the clinician with relevant information regarding pathophysiology, anatomic localization, severity, temporal profile, and prognosis.

Pathology can be identified at the level of the anterior horn cell, spinal nerve, plexus, peripheral nerve, neuromuscular junction, or muscle. Nerve conduction studies (NCSs) provide information on the integrity and function of motor and sensory peripheral nerve fibers, and can help differentiate muscle and nerve disorders. Repetitive nerve stimulation can identify disorders of the neuromuscular junction. Needle electromyography (EMG) identifies changes in motor unit morphology, which assists with differentiating nerve from muscle pathology, provides information on the temporal profile and severity of the disorder, and localizes the process. Identification of various forms of spontaneous activity can narrow the diagnosis in some cases, and in other cases provide information on disease activity. Even in upper motor neuron disorders, EDX testing can provide some insight. When a patient presents with weakness, if spontaneous activity and motor unit potentials (MUPs) are normal but there is poor voluntary activation, this can suggest an upper motor neuron problem as the etiology, assuming the patient is giving good effort.

The electromyographer tailors the EDX study based on the history and physical examination. A differential diagnosis is developed before the study that considers disease processes involving multiple levels of the central and peripheral nervous system. A deductive process is used in which each NCS and muscle examined with needle EMG should assist with narrowing the differential diagnosis until a conclusion is reached. This dynamic (rather than protocol or rote) process for the selection of nerves and muscles to study is more likely to ultimately provide an accurate diagnosis, helps limit the number of studies to the minimum required for the diagnosis, is required by the American Association of Neuromuscular and Electrodiagnostic Medicine guidelines, is required by the Current Procedural Terminology code used for billing these studies, and is the standard of practice.3

Once those data are interpreted, the clinical history and physical examination are again reviewed to ensure that the EDX conclusions are concordant with the clinical presentation. The central goal of the EDX evaluation is to reach this conclusion efficiently and reliably, while taking into account patient tolerance, which can occasionally limit the extent of the examination.

This chapter reviews a series of cases to illustrate a logical approach to the neurophysiologic workup of various clinical complaints. In clinical practice each case is unique, however, and will vary from patient to patient and clinician to clinician. The specific testing must be individualized based initially on the presenting complaint; however, during the study, the procedure changes based on the data obtained from initial NCS or needle EMG findings. Specific testing performed also depends on patient tolerance and other factors such as anticoagulation status; the presence of lymphedema, central lines, or pacemakers; and patient positioning. The cases reported below demonstrate how to apply the basic science and techniques described in Chapters 9 and 10 to a clinical situation. The EMG findings in the following cases are graded using the Mayo Clinic rating scale (Box 11-1).9 Insertional activity is described as increased if there is anything more than minimal electrical noise generated by the electrode movement itself. This can include positive waves, fibrillation potentials, myotonic discharges, myokymia, neuromyotonia, or the “snap-crackle-pop” insertional activity seen in some normal subjects. Fibrillation potentials and positive sharp waves are interpreted the same in our laboratory, both representing active or uncompensated denervation, which is reflected on the tables in a single column referred to as fibrillation potentials. The grading of fibrillation potentials, fasciculation potentials, and voluntary MUPs is otherwise defined in Box 11-1. Normative data for NCSs (based on the Mayo Clinic EMG laboratory database) are provided in Table 11-1.

BOX 11-1 EMG Grading Guidelines (Mayo Clinic EMG Reports)

Spontaneous Activity

Case 1: Upper Limb Paresthesia and Pain

A 45-year-old, right-handed female secretary presents with pain and aching in her right hand that wakes her from sleep and prevents her from typing for more than 5 to 10 minutes. She describes a pins-and-needles sensation in all five digits of her right hand, but on specific questioning admits there are also mild symptoms in the left hand. When the symptoms are severe, she feels pain throughout her entire hand, forearm, and arm as far as the shoulder. These symptoms have been present and getting gradually worse for at least 6 months. She also feels her right hand is weak, as she is dropping objects and is more clumsy than usual. She reports chronic neck pain that seems to be most problematic toward the end of the workday.

Examination reveals a relatively healthy appearing but mildly overweight 45-year-old woman. Neck range of motion is limited in extension and lateral rotation to the right, with reproduction of focal lower right cervical paraspinal pain. Strength testing is difficult because of giving-away weakness with testing of most muscle groups of the right upper limb, including the intrinsic hand muscles. Strength testing on the left is normal. Muscle stretch reflexes are normal throughout both upper limbs, and sensory testing reveals subjective decreased sensation to pinprick throughout the right hand.

Differential Diagnosis

The patient presents with predominantly right hand and arm pain and paresthesias, and a nonfocal physical examination. The most likely cause for these symptoms would be a mononeuropathy (which could involve the median, ulnar, and/or radial nerves based on the distribution of symptoms) or cervical radiculopathy. A plexopathy should be considered, and with the gradual onset of symptoms, the most likely cause of such would be an infiltrative or compressive lesion. More generalized pathology such as polyradiculoneuropathy or length-dependent peripheral neuropathy would be unlikely to present in this manner, particularly given the positional nature of the symptoms. A musculoskeletal cause for the symptoms is certainly a consideration; patients with myofascial pain often have associated paresthesias (typically in a nondermatomal distribution), and such patients are frequently referred for EDX evaluation to exclude a neurogenic cause for their pain. Lastly, it is always prudent to consider a more central process such as a syrinx or a cortical lesion, but the intermittent nature and gradual onset of the symptoms make these less likely.

When evaluating for a mononeuropathy, it is helpful to consider the potential etiologies because this will determine which EDX studies should be performed. Although focal compression, such as carpal tunnel syndrome, is by far the most common etiology for a mononeuropathy, a more generalized process involving multiple nerves (e.g., mononeuritis multiplex, hereditary neuropathy with tendency to pressure palsies, multifocal motor neuropathy with conduction block) needs to be considered. Cervical radiculopathy is also high on the list of differential diagnoses, and as she reports paresthesias throughout the entire hand, this could reflect involvement of the C6, C7, and/or C8 nerve roots. The NCS and needle EMG findings for this case are shown in Table 11-2.

Discussion

Given the differential diagnosis outlined above, it is important to perform median and ulnar motor and sensory NCSs. A median F-wave was also performed because this may be prolonged where proximal slowing is present, such as in radiculopathy. When median neuropathy at the wrist (carpal tunnel syndrome) is a clinical consideration, we perform the median motor study first and then determine the most appropriate median sensory study based on the motor results. If the motor study is normal, as it was on the left, there are a number of different orthodromic or antidromic sensory studies that can be performed. We chose the orthodromic palmar sensory study on the left because it is one of the most sensitive techniques available and can be abnormal in cases of very mild median neuropathy.29 Other options include radial-to-median distal latency comparison recording from the thumb, ulnar-to-median distal latency comparison recording from the fourth digit, or antidromic median sensory studies recorded from the most symptomatic digit.29 If the motor distal latency is prolonged, a less sensitive but technically less challenging study such as a median antidromic sensory study would be appropriate, and that was the choice on the right. The ulnar sensory response must also be obtained to ensure there is not evidence for a more generalized process. In this case, the NCSs show evidence of a median neuropathy at the wrist, bilaterally, with focal slowing demonstrated by prolonged median motor and antidromic sensory latencies on the right, and relative prolongation of the orthodromic median palmar sensory distal latency on the left, compared with the ulnar response (with a difference greater than 0.3 ms considered abnormal).29

Since the NCSs have confirmed the presence of a median neuropathy at the wrist, the purpose of the needle examination in this case is threefold: (1) to confirm the level of the median nerve lesion, (2) to provide additional information regarding severity, and (3) to exclude a superimposed process that could be contributing to the presenting symptoms, such as cervical radiculopathy. Examining a median-innervated thenar muscle provides the information on severity; in this case the finding of long-duration MUPs and some fibrillation potentials provides further evidence that this is a relatively severe lesion. Examining the pronator teres is helpful because this is a more proximal median-innervated muscle, and if normal, helps exclude a more proximal median neuropathy. Additionally, as the pronator teres is innervated by the C6 and C7 roots, a normal examination helps exclude a cervical radiculopathy at those levels. The deltoid also assesses for C5 and C6 root involvement, which can lead to similar symptoms. Examining the first dorsal interosseous helps exclude an ulnar neuropathy (unlikely given the normal ulnar NCS), but also excludes a C8 radiculopathy or lower trunk plexopathy as a cause for the abnormal findings in the abductor pollicis brevis muscle. Cervical paraspinal muscles were not examined in this case because all the findings on both NCSs and needle EMG pointed to carpal tunnel syndrome. However, if there is a strong clinical suspicion for recent-onset cervical radiculopathy, the paraspinals should be examined because they are often the first muscle to show changes on needle EMG.

Although the diagnosis in this case is bilateral median neuropathy at the wrist, consistent with carpal tunnel syndrome, the presentation highlights that many patients do not describe classic symptoms, and one must consider a broad differential diagnosis in such cases. In carpal tunnel syndrome, more than 50% of patients report sensory symptoms outside the median nerve distribution,30 and this is a reason that EDX examination is critical in refining the clinical impression.

Case 2: Aching and Numbness of the Left Thumb, Index Finger, and Middle Finger

A 53-year-old, otherwise healthy woman presents with a 7-week history of aching and numbness of the left thumb, index and middle finger. Although there was no clear injury before the onset, approximately 1 week before she had been painting a ceiling and had experienced mild neck pain. The symptoms are constant, graded at a level of 3/10, but do worsen with activity to 7/10. There is occasional aching in the dorsum of the forearm, and she also reports a dull, aching, intermittent pain in the left scapular region. She frequently awakens at night with pain and has associated hand and arm numbness. She reports no clear focal weakness but is having some functional difficulties she relates to pain. There are no lower limb symptoms, bowel or bladder complaints, and no constitutional symptoms. There is no history of carcinoma. The patient reports that a similar episode occurred approximately 5 years ago that resolved spontaneously after approximately 2 months. There was no medical evaluation at that time. Clinical examination shows mild paraspinal tenderness and point tenderness in the trapezius ridge and medial parascapular region. Strength testing shows 4/5 strength of the left triceps, as well as trace weakness in the finger extensors. The left triceps reflex was mildly reduced compared with the right. Pinprick sensation was mildly reduced in the index and middle finger. Lower limb examination was normal. Long tract signs were absent.

Discussion

The EDX findings are relatively straightforward, showing fibrillation potentials in radial- and some median-innervated muscles of the forearm as well as paraspinal muscles. Because these finding occur in two or more muscles innervated by the same root (C7) but different peripheral nerves, the diagnosis of a cervical radiculopathy is most likely. The presence of fibrillation potentials in cervical paraspinal muscles strengthens that conclusion. The patterns of findings do, however, raise some additional points. Because each muscle is innervated by two or more roots, there is significant variability in the pattern of findings, and this can make localization to a single root difficult. In this case, radial-innervated muscles seem more involved with relative preservation of the pronator teres. The flexor carpi radialis was clearly involved, however, suggesting that in this patient the pronator teres might have predominant C6 innervation. Although the radial sensory is not a routine study, it can be very helpful. If the findings are primarily in the radial distribution, a normal radial sensory study would be consistent with a preganglionic process and make radial neuropathy or brachial plexopathy much less likely. The sensory studies can be particularly helpful when there has been a previous cervical spine surgery, and paraspinal needle examination might be unreliable. The other inconsistency in this case is the large, stable MUPs noted, which would reflect longstanding reinnervation and would not be consistent with the relatively acute onset of symptoms. The history noted a previous episode several years ago, and these findings are more consistent with an underlying longstanding or old C7 radiculopathy with an acute exacerbation. The borderline median sensory distal latency raises the possibility of a superimposed median neuropathy at the wrist; further studies with either palmar orthodromic sensory responses or comparative sensory studies to the thumb or ring finger might be indicated.

Case 3: Shoulder Pain and Weakness

A 54-year-old woman with a history of breast cancer, status post right modified mastectomy, radiation, and chemotherapy 5 years previously, presents with a 4-week history of right shoulder pain and weakness as well as numbness in the right hand and forearm. She is otherwise healthy and has not been taking any medications except tamoxifen since completing chemotherapy. She denies previous episodes of similar symptoms, nor any history of neck or shoulder trauma. She had recently increased her activity with involvement in a weightlifting program, and had noted some generalized muscle aching in association with that. She states the pain came on suddenly during the night approximately 6 weeks previously with a severe burning and at times deep aching sensation, preventing her from sleeping. The severe pain lasted about 10 days and has now significantly improved, but as the pain improved, she noticed significant weakness that has persisted, such that she can barely lift her arm from her side. She still has good use of her hand.

On examination, she is a slim woman with mild lymphedema of the right upper limb. Passive shoulder range of motion is full, although she has some pain at end range of abduction and flexion. Active shoulder motion is markedly limited because of weakness. Cervical spine range of motion is full and pain free. Strength testing is notable for normal strength in the left upper limb. On the right there is marked weakness of shoulder external rotation and abduction, elbow flexion and supination, and mild weakness of pronation, elbow extension, and finger extension, with normal strength of the hand intrinsic muscles. There is moderate winging of the scapula at rest. Sensation is decreased to pinprick in the right lateral forearm. Muscle stretch reflexes are decreased at the right biceps and brachioradialis, but normal in the right triceps and left upper limb.

Discussion

NCSs are important in this case to determine which peripheral nerves are affected, whether both motor and sensory fibers are involved, and to assess the severity of axonal loss (which is helpful with regards to prognosis). If sensory nerve action potentials (SNAPs) are low amplitude, this reflects pathology distal to the nerve root at the level of the plexus or peripheral nerve. Needle examination is used to map out the distribution of involvement, using knowledge of anatomy, particularly the brachial plexus, to differentiate between peripheral nerve, plexus, and root involvement.

NCSs in this case show low-amplitude median antidromic and absent lateral antebrachial cutaneous sensory responses, which implies that the lesion is distal to the dorsal root ganglion at the level of the plexus or peripheral nerve. The normal median motor response (C8, T1, lower trunk) in the presence of a low-amplitude median sensory response (C6, C7, middle trunk) suggests the lesion is proximal to the median nerve and at the level of the plexus. The median sensory response on the left is mildly low amplitude with a prolonged distal latency, most likely representing a preexisting median neuropathy at the wrist of the type seen in carpal tunnel syndrome. The musculocutaneous compound muscle action potential (CMAP) is absent on the right but easily elicited on the left. This study was performed because the more easily obtained median and ulnar motor studies are both derived from C8–T1 axons and therefore are spared in this case.

Needle examination supports recent severe denervation involving most of the C5- and C6-innervated muscles examined, placing the lesion at the level of the upper trunk of the brachial plexus (involvement of the supraspinatus and the supinator localizes the lesion proximal to the lateral cord), with more mild involvement of some middle trunk muscles. The findings in the serratus anterior suggest the lesion is very proximal, because the long thoracic nerve forms directly off the C5, C6, and C7 ventral rami. Cervical paraspinal muscles showed a few fibrillation potentials, indicating some degree of nerve root involvement, but these were much less prominent than in some of the affected limb muscles.

The needle EMG findings, when taken in combination with the NCS results, are most consistent with a severe right upper, and to a lesser extent, middle trunk brachial plexopathy, with evidence to suggest mild cervical root involvement also. Given the somewhat patchy nature of the needle EMG findings, the absence of trauma, and the classic history of severe pain at onset followed by weakness that persists, the cause is almost certainly inflammatory and is typical of neuralgic amyotrophy or Parsonage-Turner syndrome.32 Often a particular peripheral nerve will be severely involved. For example, patients can present with sudden onset of severe dyspnea that is worse when recumbent, because of involvement of the phrenic nerve. They can also present with a classic anterior interosseous neuropathy with weakness only involving the flexor pollicis longus, flexor digitorum profundus to the second and third digits, and the pronator quadratus. Such cases are frequently misdiagnosed as entrapment neuropathy. Therefore it is critical when evaluating such cases to perform a careful needle examination, looking for patchy involvement outside the muscles that are clearly involved clinically. This includes a search for subclinical contralateral limb involvement, to avoid unnecessary surgery in such cases. A careful history should quickly alert the electromyographer to the likely diagnosis in such cases. Imaging of the cervical spine in this setting can be helpful in excluding a compressive process, but care should be taken to not assign clinical significance to mild spondylotic changes. Imaging of the brachial plexus can also reveal signal abnormalities, and in this case it would also be indicated to exclude an infiltrative process.

Case 4: Hand Numbness and Weakness

A 45-year-old physician presents with a 2-week history of numbness and tingling in the ulnar border of his right hand, with weakness of the hand such that he has difficulty opposing his fourth and fifth digits, turning the car key in the ignition, and opening jars. The medical history is unremarkable, although the patient admits to similar symptoms several months earlier. At that time he did not have the hand weakness, however, and the symptoms were less severe and resolved spontaneously. He is an avid road cyclist and developed these symptoms during a 500-mile, 5-day cycling trip. On examination there is decreased sensation to pinprick in the ulnar border of the right hand, as well as the fourth and fifth digits. Muscle stretch reflexes are normal. Strength testing reveals moderate weakness of the interossei and abductor digiti minimi muscles with preservation of the wrist flexors. Rare fasciculation is present in the first dorsal interosseous.

Discussion

NCSs showed a low-amplitude ulnar motor response on the right, with slowed conduction velocity and borderline distal latency. There was a significant reduction in the CMAP amplitude between the elbow and wrist stimulation sites suggesting a conduction block. This was confirmed with stimulation at the below-elbow site, followed by short-segment stimulation (inching) between the elbow and below-elbow sites (Figure 11-1). There was also significant slowing localized to the elbow segment (evident as a conduction velocity in the above-elbow to wrist segment that was more than 10 ms slower than the conduction velocity of the below-elbow to wrist segment). The left ulnar motor response was normal but the ulnar sensory response was low amplitude with a slowed conduction velocity. With normal median motor and sensory studies on the right, these findings are most suggestive of bilateral ulnar neuropathy at the elbow, more pronounced on the right.

The purpose of the needle examination in this case, where the diagnosis appears evident from the NCSs, is to provide further information on severity, temporal profile, and as always, to exclude coexistent problems such as plexopathy, radiculopathy, or more generalized peripheral neuropathy. Both ulnar- and non–ulnar-innervated C8 and T1 muscles must be examined. Proximal ulnar muscles such as the flexor carpi ulnaris and/or flexor digitorum profundus to digits 4 and 5 should be examined to confirm a lesion at the elbow. The examiner must keep in mind, however, that a fascicular process can spare these fibers even if the lesion is proximal. Intrinsic hand muscles such as the first dorsal interosseous and abductor digiti minimi can be examined to evaluate the severity of the lesion, and non–C8-innervated muscles such as the flexor pollicis longus and extensor indicis proprius can rule out plexopathy or radiculopathy.

Findings of fibrillation potentials and reduced recruitment in proximal and distal ulnar-innervated muscles suggest moderately severe axonal loss in this case, with the lesion at the level of the elbow (because of the finding of conduction block on ulnar motor NCSs and involvement of the flexor carpi ulnaris on needle EMG). The presence of long-duration MUPs indicates the process has been present to some degree for more than a few weeks, predating the 2-week history that the patient reports.

Ulnar neuropathies are relatively common, generally presenting with varying degrees of sensory symptoms in the fourth and fifth digits as well as weakness of ulnar-innervated hand and sometimes forearm muscles. Sensory fibers are usually the most vulnerable and can show abnormalities in very mild neuropathies. Motor responses can be recorded from both the abductor digiti minimi and first dorsal interosseous muscles to improve sensitivity, as a normal response can be obtained from one of the muscles or the other because of the fascicular nature of nerve topography. The diagnosis of a focal ulnar neuropathy at the elbow is suspected with slowing across the elbow segment of greater than 10 m/s compared with the below-elbow segment, of if there is an amplitude drop of 20% over a 10-cm segment. If a question remains, sequential 2-cm stimulation across the elbow can confirm an ulnar neuropathy if there is a 10% drop over a 2-cm segment or a latency difference of greater than 0.7 ms.4,5 An incidental median-to-ulnar crossover in the forearm can simulate an ulnar neuropathy and should always be excluded, particularly if there appears to be a conduction block in the forearm.

An ulnar neuropathy at the wrist (Guyon’s canal) will show prolonged motor and sensory distal latencies without evidence of conduction block at the elbow (and conduction velocities can be normal). The dorsal ulnar cutaneous sensory study can be useful in localizing more distal ulnar neuropathies; if the antidromic sensory response is low or absent but the dorsal ulnar cutaneous response is preserved, this implies the lesion is distal to the take-off of the dorsal ulnar cutaneous nerve, 5 to 8 cm proximal to the wrist. Side-to-side comparison of the dorsal ulnar sensory response is necessary, with a greater than 50% decrease in amplitude considered abnormal. Compression of the deep ulnar nerve can occur in the hand, and in such cases the motor response might be low amplitude with a prolonged distal latency, but the sensory responses will be spared. Needle EMG in that setting shows changes in ulnar-innervated intrinsic hand muscles only.

Case 5: Painless Weakness

A 37-year-old man with no significant medical history presents with a 6-month history of left leg weakness. His wife noted he was dragging his left foot, and in the last 2 months this has progressed with an inability to run. He reports frequent cramps in his calf and occasional tingling in his foot, but cannot be more specific about the distribution or timing of the sensory symptoms. He admits to occasional bouts of acute low back pain, without any recent exacerbations. There are no bulbar or upper limb complaints.

On examination, he has an obvious foot drop when ambulating, atrophy of the left calf, and some fasciculations noted at rest in the gastrocnemius muscle, as well as rare fasciculation potentials in the triceps. Strength testing reveals moderate weakness of the left ankle dorsiflexors, plantar flexors, toe flexors and extensors, and mild weakness of the proximal left lower limb muscles. In the right lower limb, he has mild weakness of the ankle dorsiflexors and toe extensors. Strength is normal in the upper limbs, and sensation is intact throughout. Reflexes are diffusely brisk but symmetric. Plantar responses are downgoing on the right and equivocal on the left. Cranial nerve examination is normal.

Differential Diagnosis

When a patient presents with progressive, painless weakness, one must always consider the possibility of amyotrophic lateral sclerosis (ALS), particularly when muscle stretch reflexes are normal or hyperreflexic in the presence of lower motor neuron weakness. Other causes for painless weakness include multifocal motor neuropathy with conduction block (MMNCB), which initially presents with weakness in a single peripheral nerve distribution, but progresses to involve multiple nerves. In early MMNCB, muscle bulk is typically preserved in the face of significant weakness, but with time patients develop axon loss with associated muscle atrophy and loss of reflexes. Reflexes would generally be reduced in the affected distribution as well. A lumbar plexopathy could explain the left lower limb weakness particularly given the history of mild sensory symptoms. In some cases of infiltrative or inflammatory etiology, bilateral involvement of the plexus can be seen. Although unusual in the absence of pain, multiple lumbosacral radiculopathies should be considered, particularly in the setting of diabetes, or as in this case with a background history of low back pain. When hyperreflexia is present, cervical or thoracic stenosis or intramedullary lesions are a possible explanation for brisk reflexes in an otherwise lower motor neuron problem at the level of the lumbar spine. Some myopathies can present with predominantly lower limb weakness; side-to-side asymmetry is common in inclusion body myositis, in which there is typically early involvement of the quadriceps and ankle dorsiflexors. His young age would generally argue against this. Inherited myopathies such as limb girdle dystrophy could also present with lower limb weakness and some asymmetry. Neuromuscular junction disease such as Lambert-Eaton myasthenic syndrome should be considered, but there is typically prominent proximal weakness with a fatigable component, depressed muscle stretch reflexes, and autonomic symptoms such as dry mouth. The NCS and EMG findings are presented in Table 11-6.

Discussion

ALS is a disorder of the anterior horn cell that presents with progressive, painless weakness combined with upper motor neuron findings and the absence of sensory symptoms. It is an almost universally fatal disease, with an average life span from the time of diagnosis of 3 to 5 years.28 Other disease processes that could potentially masquerade as ALS, such as polyradiculopathy, polyradiculoneuropathy, or MMNCB, must be entirely excluded because they can have treatment implications. EDX testing is helpful to confirm the diagnosis when it is clinically apparent, and to diagnose the condition when it is clinically suspected but too early to diagnose on clinical grounds alone (abnormal findings are usually present on needle examination long before they become clinically evident).

Buy Membership for Physical Medicine and Rehabilitation Category to continue reading. Learn more here