Median Neuropathy Workup

Blood tests to screen for underlying rheumatologic or inflammatory disease or other treatable systemic conditions known to be associated with CTS, if suspected:

• Fasting blood glucose and hemoglobin A1c
• Thyroid function (thyroid stimulating hormone [TSH])
• Vitamin B-12 level
• Erythrocyte sedimentation rate (ESR)
• Rheumatoid factor

• Wrist magnetic resonance imaging (MRI) has the greatest sensitivity and specificity of the available imaging modalities for the diagnosis of CTS but is usually not necessary.^[30] Current MRI research is focusing on the use of diffusion tensor imaging (DTI) and tractography of the median nerve in the diagnosis of carpal tunnel syndrome.^{[31, 32]} Tractography in patients with chronic compression of the median nerve typically shows that mean fractional anisotropy values in the median nerve are significantly lower within the carpal tunnel than in locations proximal to the carpal tunnel. Preliminary results suggest that quantitative evaluation of the median nerve with DTI is precise, and that in patients with unilateral involvement, the healthy contralateral nerve can be used as an internal control.^[33]
• Computed tomography (CT) scan of the wrist may be helpful in patients with subtle bony trauma.
• Wrist ultrasonography may evaluate the soft tissues of the carpal tunnel including the tendons and the median nerve and allow cross-sectional measurement of the carpal tunnel.
• Wrist radiograph can be used if a fracture (old or recent) is suspected.
• Cervical spine radiograph and/or MRI can be used if cervical radiculopathy is suspected.
• Chest CT scan, chest radiograph, or brachial plexus MRI can be used in patients with suspected brachial plexopathy or thoracic outlet syndrome.

Electrodiagnosis: The clinical bedside examination, including diagnostic provocative tests, have low validity, and patients with CTS symptoms should be referred directly for neurophysiologic examination. Electrodiagnostic studies remain the criterion standard for diagnosis of CTS, but should always be interpreted in combination of the clinical symptoms and signs.

Nerve conduction study (NCS): NCS measures the sensory and motor nerve conduction velocity (latency) and amplitudes across the wrist. Any focal median nerve conduction delay implies a demyelinative lesion of the median nerve. In mild or early CTS, there is usually conduction delay of sensory fibers only, without prolongation of distal motor latency. In more severe CTS, focal conduction block or secondary axon loss results in decreased median nerve sensory and motor amplitudes. Routine NCS may miss the diagnosis of CTS in up to 25% of cases. The sensitivity is greatly improved by measuring the median nerve latency within a shorter segment across the wrist in comparison to an adjacent nerve for the same distance. The clinical diagnosis of CTS may thus be confirmed with a high degree of sensitivity (>85%) and specificity (>95%).^[34]

Multiple techniques are used to diagnosis CTS. A typical electrodiagnostic protocol may include the following:

• Antidromic sensory NCS are recorded from a median innervated digit, typically by placing ring electrodes on the index finger, with electrical stimulation at the wrist at a distance of 14 cm.
• For internal median nerve comparison, the sensory potential from the index finger is also recorded by stimulation in the palm of the hand, at a midpoint between the wrist stimulation site and the recording ring electrode on the index (7 cm to each). The upper normal limit for the peak latency of the distal segment is 1.9 ms. The upper limit for the calculated peak latency difference between the wrist and palm stimulations is 1.6 ms.
• Antidromic sensory NCS of the ulnar nerve is recorded with ring electrodes on digit 5 and stimulation at the wrist (on the ulnar aspect).
• The median motor response is recorded with surface electrodes over the abductor pollicis brevis muscle (APB) and stimulation at the wrist and elbow.
• If the above routine median NCS is not diagnostic, the median nerve latency across the transcarpal segment is measured in comparison to an adjacent nerve (ulnar or radial).^{[35, 36]} This strategy may be particularly helpful in mild CTS cases and in patients with underlying polyneuropathy to detect superimposed focal median conduction delay. Sensitivity and specificity of these internal comparison studies depends greatly on the upper normal limit (cutoff) values. In patients with ulnar neuropathy, the comparison should be to the radial nerve or by internal median nerve comparison of the wrist segment versus the finger segment.

  • Median versus ulnar palmar mixed nerves studies (orthodromic) with stimulation at the palm and recording from the wrist at a distance of 8 cm. A palmdiff of 0.4 ms or greater is abnormal.
  • Median versus ulnar nerve distal sensory latencies to the ring finger, with stimulation at the wrist, at a distance of 14 cm (antidromic). A ringdiff of 0.5 ms or greater is abnormal.
  • Median versus radial nerve distal sensory latencies to the thumb with stimulation at the wrist, at a distance of 10 cm (antidromic). A thumbdiff of 0.5-0.7 ms or greater is abnormal.
• Needle electromyography (EMG) testing is optional for the diagnosis of CTS. It may be needed to differentiate CTS from cervical radiculopathy. In cases where surgery is being considered, it may document severity of CTS by documenting denervation to the APB muscle.

Robinson et al^[35] recommended the use of the combined sensory index (CSI) defined as the sum of the 3 latency differences listed above under 5) with higher sensitivity and reliability than the individual tests. Sensitivity for the tests was palmdiff 69.7%, ringdiff 74.2%, thumbdiff 75.8%, and CSI 83.1%. Specificity was 95.4-96.9%. Requiring 1, 2, or 3 tests to be abnormal yielded sensitivities of 84.8%, 74.2%, or 56.1%, respectively, but specificities of 92.3%, 98.5%, and 100%, respectively.

In a follow-up retrospective report on a larger patient group (300 hands), the same authors determined endpoints for individual tests that confidently predicted the results of the CSI; for ranges between these endpoints, further testing was required. These ranges were palmdiff 0-0.3 ms, ringdiff 0.1-0.4 ms, and thumbdiff 0.2-0.7 ms.^[37] A smaller prospective study of the same technique documented the overall superiority of the SCI versus individual tests for diagnostic accuracy, but when individual tests were markedly abnormal, it was not necessary to perform all 3 nerve conduction studies.^[38]

Electrodiagnostic studies in carpal tunnel syndrome

A report of the American Association of Electrodiagnostic Medicine, American Academy of Neurology, and the American Academy of Physical Medicine and Rehabilitation published in 2002 recommended the following electrodiagnostic studies in patients with suspected CTS (see list below for sensitivity and specificity of Techniques A-K):^[34]

• Perform a median sensory NCS across the wrist with a conduction distance of 13-14 cm (Technique G). If the result is abnormal, compare the result of the median sensory NCS to the result of a sensory NCS of 1 other adjacent sensory nerve in the symptomatic limb (Standard).
• If the initial median sensory NCS across the wrist has a conduction distance greater than 8 cm and the result is normal, 1 of the following additional studies is recommended:

  • Comparison of median sensory or mixed nerve conduction across the wrist over a short (7-8 cm) conduction distance (Technique C) with ulnar sensory nerve conduction across the wrist over the same short (7-8 cm) conduction distance (Technique D) (Standard)
  • Comparison of median sensory conduction across the wrist with radial or ulnar sensory conduction across the wrist in the same limb (Techniques B and F) (Standard)
  • Comparison of median sensory or mixed nerve conduction through the carpal tunnel to sensory or mixed NCSs of proximal (forearm) or distal (digit) segments of the median nerve in the same limb (Technique A) (Standard)
• Perform a motor conduction study of the median nerve recording from the thenar muscle (Technique H) and of 1 other nerve in the symptomatic limb to include measurement of distal latency (Guideline).
• Supplementary NCS: Comparison of the median motor nerve distal latency (second lumbrical) to the ulnar motor nerve distal latency (second interossei) (Technique J), median motor terminal latency index (Technique I), median motor nerve conduction between wrist and palm (Technique E), median motor nerve compound muscle action potential (CMAP) wrist to palm amplitude ratio to detect conduction block, median sensory nerve action potential (SNAP) wrist to palm amplitude ratio to detect conduction block, short segment (1 cm) incremental median sensory nerve conduction across the carpal tunnel (Option).
• Perform needle electromyography of a sample of muscles innervated by the C5-T1 spinal roots, including a thenar muscle innervated by the median nerve of the symptomatic limb (Option).

Comparison of pooled sensitivities and specificities of electrodiagnostic techniques to diagnose CTS ^[34]

For each electrodiagnostic technique to summarize results across studies, sensitivities were pooled from individual studies by calculating a weighted average. In calculating the weighted average, studies enrolling more patients received more weight than studies enrolling fewer patients. Specificities were similarly pooled by calculating the weighted average.

• Technique A. Median sensory and mixed nerve conduction: wrist and palm segment compared with forearm or digit segment: sensitivity 0.85; specificity 0.98
• Technique B. Comparison of median and ulnar sensory conduction between wrist and ring finger: sensitivity 0.85; specificity 0.97
• Technique C. Median sensory and mixed nerve conduction between wrist and palm: sensitivity 0.74; specificity 0.97
• Technique D. Comparison of median and ulnar mixed nerve conduction between wrist and palm: sensitivity 0.71; specificity 0.97
• Technique E. Median motor nerve conduction between wrist and palm: sensitivity 0.69; specificity 0.98
• Technique F. Comparison of median and radial sensory conduction between wrist and thumb: sensitivity 0.65; specificity 0.99
• Technique G. Median sensory nerve conduction between wrist and digit: sensitivity 0.65; specificity 0.98
• Technique H. Median motor nerve distal latency: sensitivity 0.63; specificity 0.98
• Technique I. Median motor nerve terminal latency index: sensitivity 0.62; specificity 0.94
• Technique J. Comparison of median motor nerve distal latency (second lumbrical) to the ulnar motor nerve distal latency (second interossei): sensitivity 0.56; specificity 0.98
• Technique K. Sympathetic skin response: sensitivity 0.04; specificity 0.5.