Aetna considers nerve conduction velocity (NCV) studies medically necessary when both of the following criteria are met:

1. Member has any of the following indications:

   1. Diagnosis and prognosis of traumatic nerve lesions (e.g., spinal cord injury, trauma to nerves); or
   2. Diagnosis and monitoring of neuromuscular junction disorders (e.g., myasthenia gravis, myasthenic syndrome) using repetitive nerve stimulation; or
   3. Diagnosis of muscle disorders (e.g., myositis, myopathy); or
   4. Diagnosis or confirmation of suspected generalized neuropathies (e.g., uremic, metabolic or immune); or
   5. Differential diagnosis of symptom-based complaints (e.g., pain in limb or joint, weakness, fatigue, cramps, twitching (fasciculations), disturbance in skin sensation or paresthesias [numbness or tingling]) provided the clinical assessment supports the need for a study; or
   6. Localization of focal neuropathies or compressive lesions (e.g., carpal tunnel syndrome [see selection criteria below], tarsal tunnel syndrome, nerve root compression, neuritis, motor neuropathy, mononeuropathy, radiculopathy, plexopathy); and

2. For indications other than carpal tunnel syndrome, myasthenia gravis and Lambert-Eaton myasthenic syndrome, member has had a needle electromyographic (EMG) study to evaluate the condition either concurrently or within the past year.  The requirement for needle EMG with NCV may be waived for persons on anti-coagulant therapy with warfarin (Coumadin), direct thrombin inhibitors (e.g., dabigatran (Pradaxa), desirudin (Iprivask)), or heparins that can not be interrupted.

Aetna considers NCV studies experimental and investigational when these criteria are not met.

Carpal Tunnel Syndrome Selection Criteria:

For evaluation of individuals suspected of having carpal tunnel syndrome, Aetna considers the following services to be medically necessary: 

1. Sensory conduction studies across the wrist of the median nerve, and if the results are abnormal, of one other sensory nerve in the symptomatic limb; and

    

2. If the initial median sensory nerve conduction study across the wrist has a conduction distance greater than 8 cm, and the results are normal, additional studies as listed below:

    

   1. Comparison of median sensory conduction across the wrist with radial or ulnar sensory conduction across the wrist in the same limb; or
   2. Median sensory conduction across the wrist over a short (7 to 8 cm) conduction distance.

    

3. Motor conduction studies of the median nerve recording from the thenar muscle and of one other nerve in the symptomatic limb to include measurement of distal latency.

Frequency of Testing:

The following table lists the American Association of Neuromuscular & Electrodiagnostic Medicine's (formerly known as American Association of Electrodiagnostic Medicine) recommendations concerning a reasonable maximum number of NCV, needle EMG and other EMG studies per diagnostic category needed for a physician to render a diagnosis:

Source: AMA, 2010.

Utilization of motor or sensory nerve conduction velocity studies at a frequency of 2 sessions per year would be considered appropriate for most conditions (e.g., unilateral or bilateral carpal tunnel syndrome, radiculopathy, mononeuropathy, polyneuropathy, myopathy, and neuromuscular junction disorders).  Nerve conduction velocity studies performed more frequently than twice a year may be reviewed for medical necessity. 

• F-waves and H-reflex studies are performed to evaluate nerve conduction in portions of the nerve more proximal (near the spine) and, therefore, inaccessible to direct assessment using conventional techniques.  Electrical stimulation is applied on the skin surface near a nerve site in a manner that sends impulses both proximally and distally.  Characteristics of the response are assessed, including latency.  Late responses provide information in the evaluation of radiculopathies, plexopathies, polyneuropathies (especially with multifocal conduction block or in suspected Guillain-Barré syndrome or chronic inflammatory demyelinating polyneuropathy), and proximal mononeuropathies.  In some cases, they may be the only abnormal study.  
• Motor and sensory NCV studies and late responses (F-waves and H-reflex studies) are often complementary and performed during the same evaluation.

H-Reflex Studies: 

• Typically, only 2 H-reflex studies are performed in a given examination.
• H-reflex studies usually must be performed bilaterally because symmetry of responses is an important criterion for abnormality.  When a bilateral H-reflex study is performed, the entire procedure must be repeated, increasing examiner time and effort; there are no economies of scale in multiple H-reflex testing.
• H-reflex studies usually involve assessment of the gastrocnemius/soleus muscle complex in the calf.  Bilateral gastrocnemius/soleus H-reflex abnormalities are often early indications of spinal stenosis, or bilateral S1 radiculopathies.
• In rare instances, H-reflexes need to be tested in muscles other than the gastrocnemius/soleus muscle, e.g., in the upper limbs.   In conditions such as cervical radiculopathies or brachial plexopathies, an H-reflex study can be performed in the arm (flexor carpi radialis muscle).  Other muscles that may be tested, although rarely, are the intrinsic small muscles of the hand and foot.

F-Wave Studies: 

• Although the set-up for an F-wave study is similar to the set-up for a motor NCV study, the testing is carried out separately from motor NCV study, utilizing different machine settings and separate stimulation to obtain a larger number of responses (at least 10).
• The number of F-wave studies, which need to be performed on a given person, depends on the working diagnosis and the electrodiagnostic findings already in evidence.  It may be appropriate in the same person to perform some motor NCV studies with an F-wave and others without an F-wave.

Blink Reflexes:

• Aetna considers blink reflex testing medically necessary to evaluate disease involving the 5th or 7th cranial nerves or brainstem.  Blink reflexes are considered experimental and investigational for all other indications.  The blink reflex is an electrodiagnostic analog of the corneal reflex.  The latency of the responses, including side-to-side differences, can help localize pathology in the region of the 5th or 7th cranial nerves, or in the brainstem.  The latencies and amplitudes of directly elicited facial motor responses should be determined to exclude a peripheral abnormality if the blink reflexes are abnormal.
• Recordings should be made bilaterally with both ipsilateral and contralateral stimulation.

Experimental and Investigational: 

• Examination/NCV studies using the Brevio NCS monitor, NC-stat monitor, VT3000, XLTEK Neuropath, and other automated devices are considered experimental and investigational.
• F-wave (F-reflex) study for carpal tunnel syndrome is considered experimental and investigational since there is no proven value to performing an F-wave study for this condition.
• NCV studies are considered experimental and investigational for screening for polyneuropathy of diabetes or end-stage renal disease.
• NCV studies are considered experimental and investigational for the sole purpose of monitoring disease intensity or treatment effectiveness for polyneuropathy of diabetes or end-stage renal disease.
• The Medi-Dx 7000™ and Neural-Scan are considered experimental and investigational, and are discussed in CPB 0357 – Quantitative Sensory Testing Methods.

Note: Surface electrodes are usually employed for both stimulation and recording. Needle electrodes may be used when there is a need to evaluate a nerve that is deep in the tissue, such as the sciatic nerve in the thigh, or the femoral nerve in an extremely obese individual.

See also CPB 0357 - Quantitative Sensory Testing Methods.

Nerve conduction velocity (NCV) studies are usually carried out to (i) evaluate the integrity of, and (ii) diagnose diseases of, the peripheral nervous system.  These studies specifically measure the conduction velocity, latency, amplitude, as well as shape of the response following electrical stimulation of a peripheral nerve through the skin and underlying tissue.  Abnormal findings include conduction slowing, conduction blockage, lack of responses, and/or low amplitude responses.  Results of NCV studies can reveal the degree of demyelination and axonal loss in the segment of the nerve examined.  Demyelination results in prolongation of conduction time, while axonal loss generally leads to loss of nerve or muscle potential amplitude.

Nerve conduction velocity studies are performed by recording and studying the electrical responses from peripheral nerves or the muscle they innervate, following electrical stimulation of the nerve.  Usually surface electrodes are employed for both stimulation and recording because of their reproducibility and ease of use.  Needle electrodes may be used when there is a need to evaluate a nerve that is deep in the tissue, such as the sciatic nerve in the thigh, or the femoral nerve in an extremely obese individual.

In standard NCV testing, the stimulating, recording and ground electrode placement and the test design should be individualized to each patient's specific anatomy.  Nerves tested should be limited to the specific nerves and conduction studies needed for the particular clinical question being investigated.  The stimulating electrode is placed directly over the nerve to be tested, and stimulation parameters are adjusted to avoid stimulating other nerves or nerve branches.  In most motor nerve conduction studies, and in some sensory and mixed nerve conduction studies, both proximal and distal stimulation are used.  Motor nerve conduction study recordings are made from electrodes placed directly over the motor point of the specific muscle to be tested.  Sensory nerve conduction study recordings are made from electrodes placed directly over the specific nerve to be tested.  Waveforms should be reviewed on site in real time, and the technique (stimulus site, recording site, ground site, filter settings) should be adjusted as the test proceeds in order to minimize artifact, and to minimize the chances of unintended stimulation of adjacent nerves and the unintended recording from adjacent muscles or nerves.  Reports are prepared on site by the examiner, and consist of the interpretation of test results, using established techniques to assess the amplitude, latency and configuration of waveforms elicited by stimulation at each site of each nerve tested.  This includes the calculation of NCV, sometimes including specialized F-wave indices, along with comparison to normal values, summarization of clinical and electrodiagnostic data, physician interpretation, generation of a differential diagnosis, and, when appropriate, suggestions for additional testing.  Electromyoraphic recording is usually performed during the same patient encounter in order to carry out a more in-depth evaluation of the clinical question being investigated.

Standard NCV testing includes safeguards and procedures to assure proper performance and interpretation.  Many of those are not used in the automated nerve testing systems.  Therefore, literature about NCV testing of clinical efficacy does not necessarily apply to these automated devices.

Automated NCV testing is similar to standard NCV testing in that both involve electrical stimulation of peripheral nerves, and recording of electrical responses from the same peripheral nerve or from a muscle.  Automated devices, however, have a number of differences with standard NCV tests.

With standard NCV studies, the physician specialist and a registered technologist perform the testing.  With automated devices, the office staff typically perform the test.  With automated devices, only several specific nerves can be tested.  Whereas standard NCV tests can stimulate and record both proximally and distally, automated devices can only stimulate and record distally.  With automated devices, only one direction of conduction is available, whereas with standard NCV tests, orthodromic and antidromic conduction is available.  The technique of standard NCV tests varies according to the patient's situation, whereas with automated devices, a single specific technique is pre-determined.  With automated devices, electromyography (EMG) is generally not available at the point of service, although new automated devices are being developed that also have EMG capabilities.  Stimulator and recording sites are placed at pre-determined anatomic locations with automated devices, whereas with standard NCV testing, stimulator and recording sites can be moved around to find optimal locations.

With standard NCV tests, a trained clinician evaluates patient’s history and examination findings, determines what electrodiagnostic testing is needed to answer the clinical question at hand.  The clinician can consider the differential diagnosis as testing is conducted, and change the test as needed as it proceeds to narrow the differential diagnosis.  The clinician asks further history and checks further examination findings, and integrates those with test findings in developing an interpretation.  By contrast, automated NCV devices test preset nerves only.

With standard NCV testing, a trained clinician scores peaks, latencies, determines if tests are normal, adjusted to clinically relevant factors.  The clinician assesses latencies, amplitudes, configurations, and conduction velocities.  The clinician critiques tracings, and determines if repeat recordings needed.  The clinician takes into account the patient’s history, physical, NCV and EMG as needed when interpreting the results.  The clinician also considers normal variants.  By contrast, with automated devices, a computer scores amplitudes and latencies, and determines if tests are normal according to a look-up table.  The computer prints an automated interpretation statement for the physician to sign; the computer’s statement is taken from a programmed list of statements.

The NC-Stat Monitor (NeuroMetrix Inc., Waltham, MA) is an automated hand-held device using proprietary technology for conducting NCS.  According to the manufacturer, the NC-stat System is equivalent to larger, more expensive NCS/EMG instruments.  The monitor is intended to measure standard nerve conduction parameters such as amplitude, latency, and conduction velocity of the motor as well as sensory nerves.  The NC-stat System has been on the market since 1999; and recently received an updated Food and Drug Administration (FDA) 510(k) clearance.  The NC-Stat was initially cleared for marketing by the FDA as a device to measure neuromuscular signals as an adjunct to, and not a replacement for, conventional electrodiagnostic measurements.  The updated intended use language is "[t]he NeuroMetrix NC-stat is intended to stimulate and measure neuromuscular signals that are useful in diagnosing and evaluating systemic and entrapment neuropathies".  However, the Code of Federal Regulations clarifies that clearance for marketing under Section 510(k) does not in any way denote official approval of the device.  Clearance for marketing does not involve approval for the specific usefulness, or evidence of net health outcomes, in any specific patient population or disease categories.  These health care considerations generally depend on published literature.  The NC-stat System is designed to perform non-invasive NCS for patients with suspected upper and lower extremity disorders/diseases (e.g., carpal tunnel syndrome, low back pain/sciatica, and diabetic peripheral neuropathy).  

The available evidence for the NC-stat monitor is limited in comparison with standard NCV studies and needle EMG.  In the largest study of the NC-stat technology published to date, Katz (2006) established a normal data set for median nerve studies in industrial workers using NC-stat technology.  A total of 1,695 individuals applying for employment at a single heavy industry plant without symptoms of carpal tunnel syndrome (CTS) were studied.  Values for median distal motor latency (DML), amplitude, and F-waves were recorded in the dominant limbs.  The DML was 3.81 +/- 0.57 milliseconds, with a 95 % cut-off value of 4.75 milliseconds.  Amplitude of the compound muscle action potential was 0.95 +/- 0.46 mV, reflecting the use of volume conduction by this technology.  Most of the workers who were characterized as having borderline, prolonged, or very prolonged distal motor latencies according to NeuroMetrix automated report actually fell below the 95 % cut-off of this independent data analysis.  The author concluded that the NC-stat technology using DML appears to be no more sensitive or specific than a traditionally performed DML for the diagnosis of CTS.  Until recently promoted sensory studies using NC-stat technology are better defined, this technology can not be recommended for screening or diagnosis of CTS in an industrial population.

A technology assessment of this device, prepared by the Washington State Department of Labor and Industries (Morse, 2006): "The evidence evaluating the use of NC-stat is most abundant for nerve testing that may be useful to diagnose or screen for conditions at the wrist (i.e., Median and ulnar nerve studies).  There is very little or no available evidence (high quality, peer-reviewed) supporting the use of NC-stat and specific biosensors for testing of nerves in the lower extremities .... At this time there is not adequate scientific evidence to conclude that NC-stat is equivalent to traditional nerve conduction study methods for use in evaluating the functioning of the median, ulnar, peroneal, sural or tibial nerves.  The diagnostic accuracy of NC-stat is not yet demonstrated in the scientific literature to be equivalent to traditional or gold-standard testing methods.  NC-stat is therefore considered experimental and investigational .... NC-stat is considered controversial as the performance of testing at the point-of-service may not be supported by recommendations of the American Association of Neuromuscular & Electrodiagnostic Medicine."

Work-Loss Data Institute evidence-based guidelines on CTS (2006) stated that NC-stat monitoring is "not currently recommended."

The Brevio NCS-Monitor (NeuMed Inc., West Trenton, NJ) is a hand-held automated device designed to assess peripheral nerves for conditions such as CTS, diabetic peripheral neuropathy, and tarsal tunnel syndrome.  The latest version of the Brevio has a graphical user interface that integrates on-screen prompts to guide a user during an examination.  The device utilizes sophisticated firmware with algorithms that seek a maximal compound muscle action potential (CMAP) or sensory nerve action potential (SNAP), plots waveform images in real-time, marks all cursors, and indicates whether or not the latency and amplitude are within normal limits.  Upon completion of a full examination, the device can generate a report.  The whole examination process (including printing of the report) will take about 15 mins.  There is insufficient evidence to establish the clinical value of this automated NCV studies device.

Schmidt et al (2011) noted that automated hand-held NCS devices are being marketed for use in the diagnosis of lumbosacral radiculopathy (LSR).  In this study, these researchers compared the specificity and sensitivity of a hand-held NCS device for the detection of LSR with standard electrodiagnostic study (EDX).  A total of 50 patients referred to a tertiary referral EMG laboratory for testing of predominantly unilateral leg symptoms (weakness, sensory complaints, and/or pain) were included in the investigation; 25 normal "control" subjects were later recruited to calculate the specificity of the automated protocol.  All patients underwent standard EDX and automated testing.  Raw NCS data were comparable for both techniques; however, computer-generated interpretations delivered by the automated device showed high sensitivity with low specificity (i.e., many false-positives) in both symptomatic patients and normal controls.  The authors concluded that automated device accurately recorded raw data, but the interpretations provided were overly sensitive and lacked the specificity necessary for a screening or diagnostic examination.

The American Association of Neuromuscular & Electrodiagnostic Medicine (AANEM, 2005) stated that based on the literature, there are no contraindications to needle EMG in patients with lymphedema or prosthetic joints.  In patients with lymphedema, clinical judgment in each individual circumstance should be used in deciding whether the risk of complication is greater than the value of the information to be obtained from the needle electrode examination.  Thus, chronic lymphedema (from breast cancer surgery) is not a contraindication to the EMG requirement with nerve conduction velocity studies.

The AANEM (2006) states that "the standard of care in clinical practice dictates that using a predetermined or standardized battery of NCSs for all patients is inappropriate".  "It is the position of the AANEM that, except in unique situations, NCSs and needle EMG should be performed together in a study design determined by a trained neuromuscular physician".  The AANEM explained that standardized nerve conduction studies performed independent of needle EMG studies may miss data essential for an accurate diagnosis.  The AANEM position statement (2006) explains that "[t]he performance of or interpretation of NCS separately from the needle EMG component of the testing should clearly be the exception.  Nerve conduction studies performed independent of needle EMG may only provide a portion of the information needed to diagnose muscle, nerve root, and most nerve disorders.  When the NCS is used on its own without integrating needle EMG findings, or when an individual relies solely on a review of NCS data, the results can be misleading and important diagnoses may be missed.  Moreover, individuals who interpret NCV data without patient interaction or who rely on studies that have delayed interpretation, who have interpretation made off-site, and who interpret results without complementary information obtained from EMG studies are not meeting the standards outlined in the AANEM policy recommendations."

Needle EMG is relatively contraindicated in persons on anti-coagulant therapy with coumadin (Warfarin) or heparins that cannot be interrrupted.  Oh (2003) observed that patients with a variety of bleeding disorders may be referred for needle EMG.  Oh recommended that the referring physician and the electromyographer examine each case individually, carefully weighing the potential risks and benefits.  An increased potential for bleeding may be expected in patients with thrombocytopenia who have a platelet count of less than 50,000/cm3, with a prothrombin time of more than 1.5 to 2 times the control values (International Normalized Ratio of 1.5 to 2.0), or with a partial thromboplastin time greater than 1.5 to 2 times the control value when intravenous heparin therapy is administered (Oh, 2003).  Oh (2003) advised that, if the decision is made to perform a needle EMG in such a patient, the clinician should first to examine the small, superficial muscles and to watch for bleeding problems.  The author noted, however, that additionally prolonged local pressure is usually sufficient for hemostatis.  This is also the case with patients who have other coagulopathies or who are receiving anti-coagulants (Oh, 2003).  Oh (2003) stated that needle examinations should be avoided in patients with hemophilia and other hereditary coagulation disorders unless clotting functions have first been appropriately corrected.  Oh (2003) noted that there has been 1 report of a complication of subcutaneous bleeding secondary to the needle EMG in a patient receiving anti-coagulants and with a partial thromboplastin time greater than twice the control value.

In a discussion of complications from EMG examinations, Kuminga (2001) stated that bleeding tendencies deserve special mention in screening patients for electromyographic examination.  Kuminga (2001) stated that specific inquiry in this regard often reveals pertinent information that the patient may not volunteer.  To prevent unnecessary complications, the author recommends that the electromyographer consult with the referring physician to weight the diagnostic benefits against the risks.  A patient taking anti-coagulants should have appropriate laboratory tests for bleeding tendency prior to a needle study (Kuminga, 2001).  With heparin infusion, partial thromboplastin time should not exceed 1.5 of control value.  With warfarin (Coumadin) therapy, patients should have an international rating (INR) less than 2.0.  Kouminga (2001) stated that the same precautions should apply to those with other coagulopathy, such as hemophiia.  For thrombocytopenia, unless the platelet count falls below 20,000/mm, local pressure can usually counter the minimal hemorrhage.  The author noted that testing the degree of bleeding tendency with a superficial muscle helps determine the feasibility of further study of deeper muscles, which can not be compressed adequately to accomplish hemostasis.

Authorities recommend that patients on warfarin should stop taking warfarin 3 days prior to EMG and resume taking it immediately after the test, if permitted by their primary physician.  The use of aspirin, or aspirin-like medications (such as Plavix (clopidogrel) or Aggrenox (aspirin and dipyridamole)) is not a contraindication to the test.

Appendix

Documentation Requirements

The member's medical records must clearly document the medical necessity for the test. It is not necessary to include documentation with each claim submission. Data gathered during NCS, however, should be available which reflect the actual numbers (latency, amplitude, etc.), preferably in a tabular (not narrative) format. The reason for referral and a clear diagnostic impression are required for each study. In cases where a review becomes necessary, either a hard copy of waveforms or a complete written report with an interpretation of the test must be submitted upon request.

Normal findings and abnormalities uncovered during the study should be documented with the muscles tested, the presence and type of spontaneous activity, as well as the characteristics of the voluntary unit potentials and interpretation.