It was a cross-sectional study conducted on 100 subjects aged 40 years and above after getting their informed written consent to participate. The consecutive patients clinically diagnosed to have cervical radiculopathy, referred from department of Orthopaedics were prospectively recruited for the study. Subject’s detailed history and thorough clinical examination parameters were recorded. Subjects with Diabetes mellitus, with clinical or electrophysiological evidence of polyneuropathy as well as subjects with symptoms of less than 3 weeks duration and having spinal surgery done within the preceding 15 years were excluded. In all the subjects spine MRI was performed. Institutional Ethics Committee’s approval was obtained and study was conducted at fixed room temperature of 30⁰C.

In all the subjects, nerve conduction study was done using RMS EMG EP Mark-II. Motor nerves tested were Median, Ulnar, Radial, Axillary, Musculocutaneus and Suprascapular nerve. Surface disc electrode was placed on corresponding muscles. Ground electrode was placed between stimulating and recording electrodes. Belly tendon montage was used with cathode and anode 3 cm apart. Filters were set at 2 Hz to 5 kHz and sweep speed was 5 ms per division for motor study and for sensory study, filters were at 20 Hz to 3 kHz and sweep speed was 2 ms per division. Duration for both motor and sensory study was at 100 μs. Parameters like Distal Motor Latency (DML), Compound muscle action potential (CMAP) amplitude, Conduction Velocity (CV) were evaluated for motor nerves. Sensory conduction study (antidromic) involved stimulation of sensory nerves proximally and recording Sensory Nerve Action Potential (SNAP) with electrodes placed distally over the dermatomic distribution. Distance between active electrode and cathode of stimulator was divided automatically by onset latency to give sensory conduction velocity. SNAP amplitude was taken from peak to base. Sensory nerves tested were Median, Ulnar and Radial nerves. Parameters like SNAP amplitude and conduction velocity were evaluated for sensory nerves.

Statistical analysis was done by using Statistical Package for Social Science (SPSS) version16. The study observations and results were noted and analysed to find the specificity, sensitivity, positive predictive value and negative predictive value.

Age and gender wise distribution of all the study subjects is depicted in [Table/Fig-1]. Age groups were not statistically different between male and females in study subjects. The sensitivity, specificity, positive and negative predictive values of motor and sensory nerve conduction study is shown in [Table/Fig-2 and 3]. Among various motor nerve conduction parameters CMAP was found to be more sensitive with high positive predicative value. CV was found to have greater specificity and DML had least negative predictive value. Among various nerves proximal nerves were found to have somewhat greater sensitivity and specificity as compared to distal nerves. Sensory nerve conduction parameters were found to have less sensitivity but higher specificity as compared to motor parameters. Positive and negative predictive values were comparable to motor parameters [Table/Fig-2 and 3].

Any process that causes impingement of exiting cervical nerve roots can lead to a radicular disorder like cervical radiculopathy. This impingement may be brought about by pathological or degenerative changes. Though MRI is considered as the gold standard for diagnosis of cervical radiculopathy, sometimes this disease may present itself even without any evident structural lesion seen on MRI. In such circumstances role of Electrophysiologic tests in the diagnosis is crucial. Therefore it is reasonable to evaluate diagnostic efficacy of nerve conduction studies and also there is a definite need to establish a cost effective, reliable, and accurate means for establishing the diagnosis of cervical radiculopathy.

Clinical examination is not always straight forward if the patient is in pain [8]. Plain radiographs in various projections may be helpful, but clinical symptoms often correlate poorly with the radiological findings [9]. Imaging techniques are mainly directed to localise the abnormality, identify compression of the spinal cord, nerve roots, and to exclude intraspinal lesions. Shafaie et al., [8] have reported that correlation between MRI and surgical findings is frequently unreliable. Furthermore abnormalities in MRI have been found in asymptomatic subjects. In this scenario, neurophysiological investigations are perceived by many as being helpful in the diagnosis of cervical radiculopathy and are also useful in excluding peripheral nerve lesions.

In present study, we documented that among various motor nerve conduction parameters Compound Muscle Action Potential (CMAP) was found to be more sensitive with high positive predicative value in diagnosing cervical radiculopathy. Conduction Velocity (CV) was found to have greater specificity and Distal Motor Latency (DML) had least negative predictive value. Among nerves, proximal nerves were found to have some what greater sensitivity and specificity as compared to distal nerves. Sensory nerve conduction parameters were found to have less sensitivity but higher specificity as compared to motor parameters. Positive and negative predictive values of sensory nerve conduction studies were comparable to motor parameters.

Nerve compression in CR may lead to motor, sensory, and/or autonomic changes. The usefulness of nerve conduction studies is dependent on their ability to detect motor changes occurring as a result of nerve compression. In radiculopathy, abnormalities in sensory-nerve action potentials (SNAPs) are uncommon. Typically, compression that leads to cervical radiculopathy occurs proximal to the dorsal root (sensory) ganglion. Unless the dorsal root ganglion at the distal most aspect of the neural foramen is involved, the SNAPs will remain normal. Compound muscle action potentials show a decrease in amplitude proportional to muscle atrophy. Significant alterations may be seen in polyradiculopathies with multiple muscle involvement. Nerve-conduction velocity and latency changes are not typically found in cervical radiculopathies unless there is extreme demyelination of axons [10]. We recorded the comparable results as these findings. Sensory radiculopathy can only rarely be reliably localized segmentally by electrodiagnostic (EDX) techniques for the following reasons: symptoms of pain and paraesthesia are primarily mediated through C-type sensory fibers, which are too small to be studied by routine EDX techniques; the peripheral processes of sensory root fibers remain intact with intraspinal lesions, so SNAPs remain normal [5]. Our results go hand in hand with these observations.

Ashkan et al., [11] reported that sensitivity of both MRI and neurophysiologic studies (NPS) for the diagnosis of cervical radiculopathy was 93 and 42% respectively, whilst the positive predictive values for MRI and NPS were similar 91 and 86%. MRI had a higher negative predictive value (25% vs 7%). However, agreement between MRI and NPS was highest in patients with clear radicular symptoms and abnormal clinical signs (60%), suggesting that MRI and NPS remain complementary modalities in the evaluation of cervical radiculopathy. This is co-existent with our findings.

Therefore, based upon above observations and discussion we are of the opinion that, nerve conduction studies are useful supportive diagnostic tool for suspected cervical radiculopathy as they are found to have reliable sensitivity and specificity. We further conclude that among motor nerve conduction parameters compound muscle action potential is more sensitive as compared to distal motor latency and conduction velocity and sensory nerve conduction parameters are less sensitive in diagnosing cervical radiculopathy.

It was cross-sectional study with quite lesser sample size, therefore future studies with larger sample size are recommended.