Acta Neurol Scand 1998: 98: 280-282 Printed in UK - all rights reserved

Copyright Q Munksgaard 1998 ACTA NEUROLOGICA

SCANDINAVICA ISSN 0001-631 4

Near-nerve versus surface electrode recordings of patients with

sensor carpal

y nerve conduction tunnel syndrome

Smith T. Near-nerve versus surface electrode recordings of sensory nerve conduction in patients with carpal tunnel syndrome. Acta Neurol Scand 1998: 98: 280-282. 0 Munksgaard 1998.

Objectives - The aim of the study was to compare the relative sensitivity of sensory nerve conduction (SNC) recorded with near-nerve needle electrodes and SNC recorded with surface electrodes in demonstrating focal slowing of the median nerve in patients with symptoms of carpal tunnel syndrome (CTS). Materials andmethods - Eighty-two consecutive patients with clinical symptoms and signs of CTS were studied prospectively by the same clinical neurophysiologist. Orthodromic near-nerve recording from digits 1 and 3, distal motor latency, and antidromic surface recording from palm to digit 2 and wrist to digit 2 were performed in all patients. Near-nerve recording of the ulnar nerve was done in patients with abnormal median nerve conduction. Patients were compared to controls. Results - Near-nerve SNC was abnormal (slowed velocity or absent response) in 52% of the patients from digit 1 to wrist, in 51% from digit 3 to wrist, and 40% had a prolonged distal motor latency. Surface antidromic SNC was abnormal in 49% from wrist to palm, and in 43% from wrist to digit 2. Statistical analysis revealed no significant difference between the near-nerve method and the surface method. Absent sensory potentials were more common with the surface method. Conclusion - As a routine study of CTS patients, the surface technique can be used and the near-nerve technique merely used for patients

1 with absent surface responses.

Standard electrophysiologic testing for carpal tunnel syndrome (CTS) uses orthodromic or antidromic sensory nerve conduction (SNC) and surface or near-nerve stimulating and recording electrodes to demonstrate focal median nerve slowing (1). Investigations with surface electrodes for stimu- lating and recording are more quickly performed and better tolerated by patients than investigations using near-nerve electrodes. On the other hand near-nerve recording is a very sensitive method of detecting CTS (2). In this report a comparison was made of the relative sensitivities of surface and near-nerve SNC in demonstrating focal slowing in patients with symptoms of CTS.

Material and methods

Eighty-two patients with clinical suspicion of CTS consecutively referred to the department of neuro-

in

T. Smith Department of Neurology. Odense University Hospital, Odense. Denmark

Key words: carpal tunnel syndrome; sensory conduction; near-nerve recording, antidromic recording; nerve compression syndromes

Torben Smith, Department of Neurology, Odense University Hospital, Ddense. DK-5000, Denmark

Accepted for publication May 6, 1998

physiology were studied. The patient population consisted of 50 women and 32 men, the mean age was 44 years, range 17-88.

Ninety-three percent of the patients complained of paresthesias, 50% had arm or wrist pain and 16% experienced weakness of the hand. The clini- cal examination revealed hypesthesia in 50% of the patients, hand weakness in 21% and atrophy of the thenar in 11%. Tinel’s sign was positive in 68% and Phalen’s sign in 72% of the patients. The duration of symptoms ranged from 1 month to 10 years (mean 14 months). Symptoms and signs were evalu- ated to be indicative of severe CTS in 25% and of mild C T S in 75% of patients. In patients with bilateral symptoms the hand examined was the most affected one. All electrophysiologic studies were performed by the same experienced clinical neurophysiologist. Near-nerve and surface SNC studies were done in all patients.

280

Near-nerve

Orthodromic sensory recording from digits 1 and 3 to wrist and distal motor latency to the abductor pollicis brevis muscle (APB) were studied accor- ding to the routine of the laboratory using near- nerve technique (2, 3). In short, sensory fibers of digits 1 and 3 were stimulated via bipolar ring electrodes, and the sensory action potentials were recorded via bipolar needle electrodes placed at the wrist. The recording electrode was adjusted near the nerve using the threshold of the motor response evoked in the APB, stimulus current < 1 mA. When necessary, electronic averaging of 100 to 1000 responses was used to record a sensory action potential. Motor fibers were stimulated with the same needle electrodes and the muscle action potential recorded via a bipolar surface electrode placed over the APB muscle. Near-nerve recording of the ulnar nerve was done in all patients with abnormal median nerve conduction. The con- duction parameters obtained were compared to normals (3). Abnormal values were defined as values outside the meanf2 standard deviations (SD) calculated for the normals.

Surface antidromic sensory recording from palm to digit 2 and from wrist to digit 2 were performed in all patients (4). In short, the median nerve was stimulated at palm and wrist with a surface stimulator and antidromic sensory potentials were recorded with bipolar ring electrodes placed around digit 2. If necessary electronic averaging was used. The SNC velocity was calculated in segments from digit 2 to palm, from palm to wrist and from digit 2 to wrist. The obtained velocities and amplitudes were compared to normals (4). Reference limits for the values were defined as the meank2 SD for the normals. As mentioned earlier the near-nerve technique with stimulating digits 1 and 3 was the routine method of the laboratory, whereas the surface technique with recording from digit 2 is known as the preferred method in other departments of neurophysiology.

Temperature of the hand was maintained at 34 degrees Celsius using a thermostat-controlled infra- red heating lamp.

For statistical analysis McNemar’s change test (5) and inter-observer variation with kappa- statistics (6) was used to analyze the difference and the agreement of the two methods. The study was approved by the local ethical committee and informed consent was obtained from the patients.

Results

The various conduction parameters determined are listed in Table 1. Decreased SNC velocity or absent sensory response from digit 1 to wrist was found in 52% and from digit 3 to wrist in 51 YO of the patients

Table 1. Sensory and motor conduction parameters in patients with symptoms of CTS

SensCV near-nerve digit 1 -wrist digit 3-wrist

SensCV surface digit 2-wrist palm-wrist

Distal motor latency wrist- APE

~~ ~~

n Decr.CV Absresp Abnorm %

82 42 1 43 52 78 36 4 40 51

82 22 13 35 43 82 22 18 40 49

82 32 1 33 40 Increased

n number of patients. Decr decreased, CV conduction velocity, Abs resp absent response. Sens sensory. Abnorm abnormal

with the near-nerve recording method. Forty per- cent had a prolonged distal motor latency from wrist to the APB muscle. All patients had normal sensory and motor conduction in the ulnar nerve.

The sensory nerve conduction using surface elec- trodes was abnormal (decreased SNC velocity or absent sensory response) from digit 2 to wrist in 43% and from palm to wrist in 49% of the patients. Absent sensory nerve potentials were more com- mon with surface recording (16%) than with near- nerve recording (5%). Out of 13 patients with no sensory response from digit 2 using surface recor- ding it was possible to record a response from digit 1 in 11 and from digit 3 in 7 with near-nerve electrodes. All 13 patients had decreased sensory conduction velocity. Near-nerve SNC velocity (mean+SD) for these 13 patients was 2 5 f l l m/s (digit 1 to wrist) and 35f9 m/s (digit 3 to wrist) compared with 33f6 m/s and 38+7 m/s for those 22 patients who had abnormal but present sensory nerve potentials using surface techniques. Five patients had no response from digit 2 when stimulating the palm but had a response when stimulating wrist revealing decreased sensory con- duction velocity from wrist to digit 2.

Seven patients with abnormal near-nerve sensory recordings had normal findings with surface rec- ording, and 3 patients with abnormal surface recordings had normal near-nerve recordings (Table 2). Statistical analysis with McNemar’s test for correlated proportions demonstrated no signi- ficant difference between the near-nerve method

Table 2 Number of patients with abnormal or normal sensory conduction according to near-nerve and surface recording

Near-nerve

Normal Abnormal

Surface Abnormal Normal

37 7

3 35

28 1

Smith

and the surface method (0.3<P<0.5). Inter- observer variation with kappa-statistics revealed a very good agreement between the two methods (kappa =0.757, 95% confidence limits: 0.615- 0.898). Patients with abnormal near-nerve recor- dings but normal surface recordings had mild symptoms and conduction velocities close to the lower borderlines of controls. Near-nerve SCN velocity (mean? SD) for these patients was 41f3 m/s (digit 1 to wrist) and 46f2 m / s (digit 3 to wrist) compared with 32+9 mls and 38f8 mls for the 37 patients with abnormal surface SNC velocity.

Discussion

This study was performed prospectively and on consecutive patients investigated with both methods by the same examiner. With surface electrodes it was found that 43% of the referred patients with symptoms of CTS had abnormal sensory conduc- tion between wrist and digit 2. This percentage of abnormalities was less than but still in accordance with previous published studies (7-9) that met all the literature review criteria of AAEM (1). The finding of 52% of patients with abnormal sensory conduction in the wrist-digit segment using near- nerve recording was significantly less compared to what was found in other studies using near-nerve technique (2,10,11). The high incidence of abnor- mal findings in these studies was probably due to the fact that the population of patients was biased with patients with advanced CTS (1).

In the present study 49% of the patients had abnormal sensory conduction in the palm-wrist segment using surface technique. This was also in agreement with published studies (8, 9, 12), although others found higher percentages of abnor- malities (13, 14). Many different electrodiagnostic methods are used in demonstrating abnormal conduction in patients with symptoms of CTS. Comparison of the sensitivities of the different methods has demonstrated that techniques which evaluate median nerve conduction over a short distance across the carpal tunnel (palmar studies) are more sensitive than techniques which measure the conduction over a longer segment (wrist to digit) (1). In this study the reverse appeared to be the case. This was probably due to the fact that the comparison was made between a near-nerve method and a surface method, and this was not the case in previous studies which all were using surface electrodes.

It is the author’s firm belief although it was not quantitated and documented, that the patients

preferred the surface technique in nearly all cases. Also the surface technique was less time con- suming. It can be concluded that as a routine study of sensory nerve conduction in CTS patients, sur- face technique can be used and near-nerve tech- nique merely used for patients with clinical symptoms of CTS but with normal surface record- ing. The most important merit in the near-nerve technique against the surface technique is less absent responses.

References 1) 1. JABLECK CK, ANDARY MT, So YT, WILKINS DE, WILLIAMS EH.

(AEM Quality Assurance Committee). Literature review of the usefulness of nerve conduction studies and electro- myography for the evaluation of patients with carpal tunnel syndrome. Muscle Nerve 1993: 16: 1392-1414.

logic findings in entrapment of the median nerve at wrist and elbow. J Neurol Neurosurg Psychiatry 1974: 37: 340-60.

3. ROSENFALCK P, ROSENFALCK A. Electromyography, Sensory and Motor Conduction: Findings in Normal Subjects. Laboratory of Clinical Neurophysiology, Copenhagen: Rigshospitalet, 1975.

4. KIMURA J. Assessment of individual nerves. In: KIMURA J. Electrodiagnosis in Diseases of Nerve and Muscle: Princi- ples and Practice. Philadelphia: F.A. Davis Company, 1989: 103-38.

5. KIRKWOOD BR. Further methods for contingency tables. In: KIRKWOOD BR. Essentials of Medical Statistics Oxford: Blackwell Scientific Publications 1988: 94- 105.

6. FLEISS JL. The measurement of interrater agreement. In: FLEW JL. Statistical Methods for Rates and Proportions,

7. CARROLL G. Comparison of median and radial nerve sensory latencies in the electrophysiological diagnosis of carpal tunnel syndrome. Electroencephalogr Clin Neuro- physiol 1987: 68: 101-6.

8. JACKSON D, CLIFFORD JC. Electrodiagnosis of mild carpal tunnel syndrome. Arch Phys Med Rehabil 1989: 70: 199-204.

9. KIMURA J. The carpal tunnel syndrome: Localization of conduction abnormalities within the distal segment of the median nerve. Brain 1979 102: 619-35.

10. CASEY EB, LEQUESNE PM. Digital nerve action potentials in healthy subjects, and in carpal tunnel and diabetic patients. J Neurol Neurosurg Psychiatry 1972: 35: 612-23.

11. CIONI R, PASSERO S, PARADISO C, GIANNINI F, B A ~ S T I N I N, RUSHWORTH G. Diagnostic specificity of sensory and motor nerve conduction variables in early detection of carpal tunnel syndrome. J Neurol 1989: 236: 208-13.

12. KIMURA J. A method for determining median nerve con- duction velocity across the carpal tunnel. J Neurol Sci 1978:

13. F~LSENTHAL G, SPINDLER H. Palmar conduction time of median and ulnar nerves of normal subjects and patients with carpal tunnel syndrome. Am J Phys Med 1979: 58:

14. STEVENS JC. AAEE minimonograph #26: The electro- diagnosis of carpal tunnel syndrome. Muscle Nerve 1987:

2. BUCHTHAL F, ROSENFALCK A, lkOlABORG W. Electrophysio-

1981: 211-36.

38: 1-10.

131-8.

10: 99-113.

282