392 Electroencephalography and clinical Neurophysiology , 89 (19931 392-398 '~ 1993 Elsevier Scientific Publishers Ireland, Ltd. 0924-980X/93/$06.00

EEM 93503

Sensory nerve findings by tactile stimulation of median and ulnar nerves in healthy subjects of different ages

G. Caruso a, j. Nilsson b C. Crisci b, M. Nolano a, R. Massini b and F. Lullo "Department of Clinical Neurophysiology, School of Medicine, University Federico 11, Naples (Italy), and b Fondazione Clinica

del Lavoro, Centro Medico, Campoli M.T. (BN) (Italy)

(Accepted for publication: 22 July 19931

Summary We studied orthodromic sensory conduction velocity along the distal and proximal segments t)f the median and ulnar nerves by tactile stimulation of the distal phalanx of the 3rd and 5th digits in 44 healthy subjects divided into 2 age groups: from 16 to 35 years and from 63 to 81 years. In the same nerves, we used selective electrical stimulation of the corresponding digital nerves to obtain sensory potentials. In both groups, responses to tactile stimuli had a longer latency and smaller amplitude than those to electrical stimulation, and they were distributed in a series of 6 -7 main deflections, apparently regardless of whether the recording site was distal or proximal. Moreover, irrespective of the nerve and of subject age, conduction velocity along both the digit-wrist and the wrist-elbow nerve segments was significantly slower with tactile stimuli than with electrical stimuli. However, independently of the stimulus used, conduction velocity along the proximal nerve segment was significantly faster than that measured along the digit-wrist nerve segment. In both the median and ulnar nerves, maximum potential amplitude, cumulative area and conduction velocity were significantly reduced in the older age group. This finding could reflect the smaller number of Meissner 's corpuscles in older subjects, and the loss of large nerve fibres in individuals over 611.

Key words: Tactile stimulation; Peripheral nerve conduction; Meissner 's corpuscle concentration

The study of sensory nerve conduction by supramax- imal electrical stimulation as routinely performed in clinical neurophysiology provides general information about impulse propagation along the faster conducting nerve fibres. In fact, the electrically evoked sensory action potentials are the result of an indiscriminate activation of morphologically and functionally different nerve fibres and so provide a general overview of the function of the nerve and indirect information about its anatomical state. Therefore, although this examination is rather specific, it may not always be sufficiently sensitive.

Study of the sensory responses to tactile stimulation is a more appropriate way of examining the peripheral nervous system (Pratt and Starr 1981; Buchthal 1982a,b; Schieppati and Ducati 1984; Hashimoto et al. 1989). With this technique one can explore, starting from the thinnest terminations, the function of a selected group of nerve fibres, which in some neuropathies may be primarily and specifically involved. Moreover, with this

Correspondence to: Giuseppe Caruso, M.D., Depar tment of Clini- cal Neurophysiology, Universit~ Federico II, "Nuovo Polielinico," Via S. Pansini 5, 80131 Naples (Italy). Tel.: 39 81 7462663; Fax: 39 81 5469861.

technique one can explore the specific mechanorecep- tors and so have an indirect measure of their threshold (Mackel 1988) and concentration (Johansson and Vallbo 1979; Caruso et al. 1993).

The aim of this study was to compare electrical and tactile sensory evoked potentials in healthy subjects of different ages, and to correlate the electrophysiological data with the density of some mechanoreceptors pres- ent in the glabrous skin of the hand.

Methods and materials

The median and the ulnar nerve of 44 healthy informed volunteers (20 males and 24 females) were studied. Subjects were divided into 2 age groups: from 16 to 37 years and from 65 to 78 years for the median nerve, and from 24 to 35 years and from 63 to 82 years for the ulnar nerve. In a quiet, temperature-controlled room (21-23°C), while the subject's right upper limb rested on a pillow, we stimulated digit I II for the median nerve and digit V for the ulnar nerve. Electri- cal stimulation was applied using two thin platinum electrodes (Dantec 13L70) on the distal phalanx along the lateral aspect of digit III , and along the medial aspect of digit V. Stimuli were delivered from a con-

TACTILE CONDUCTION VELOCITY IN HEALTHY SUBJECTS 393

TRIG ,_Ln_

I , .... I ' I'sur,eoe I

ffr--P I I

I " i

% Resulting stimulus

DuIse

~ i 4 lJv

- - - - ~ ~ 1 1 0 pV

i i 4 m s

Fig. I. Scheme for electrical stimulation of median digital nerve and for recording of sensory responses from wrist and elbow. To minimize the stimulus artifact, biphasic stimulation pulses were used when necessary. On the right, two separate averages were superimposed to ensure the reproducibility of the responses. Note the smaller amplitude and the more dispersed morphology of the elbow recorded potential. The subject

was a 20-year-old woman without history, signs or symptoms of peripheral nervous system disease.

stant-current isolated stimulator (Dantec 15E07). Stim- ulus duration was 0.2 msec, and stimulus strength, 9-18 mA. In some cases, we used biphasic stimulation pulses to minimize the stimulus artifact (Nilsson et al. 1988), and a large metal ground electrode (Dantec 13S97) was wrapped around the wrist between stimu- lating and recording electrodes (Fig. 1).

Tactile stimulation was applied to the fingertip at the vortex of the skin ridges, using a vibrator (Ling model 201) powered by a Hewlett-Packard power sup-

ply (model 6824 A), and triggered by a Tektronix square wave generator (model FG 501 A). The tactile probe was 2 mm in diameter, giving an area of 3.14 mm 2, at a rate of 150/zm/msec with a rise time of 5 msec, determined by an accelerometer (Briiel and Kj~er, model 8001) connected to a charge amplifier (Briiel and Kj~er, model 2635). The indentation, which was maximally 0.5 mm, was maintained for 100 msec to prevent an "off-response" interfering with the determi- nation of response latency and amplitude (Fig. 2). The

I T R I G ~ Amplifier

Vibrator

/ Displacement amplifier

Surface temperature:

37*C

Low-noise amplifier 2

Low-noise Output: (1-2)

Fig. 2. Scheme for applying tactile stimuli to the finger tip of digits III and V. Amplifier 2 was connected to a loop that picked-up the electromagnetic signal and compensated the magnetic interference.

394 (L C A R U S O ET AL.

A [

10 pV ]

t , , I ,,,, 0 10 20

C f

0.5 pV

• I ,, I o 1o 20

Fig. 3. Right median nerve of a 27-year-old man without history, signs or symptoms of peripheral nervous system diseases, Sensory potentials evoked at the wrist (A) and elbow (B) by selective supramaximal electrical stimulation of the lateral digital nerve of digit III, In C and D: sensory potentials at wrist and elbow evoked by tactile stimuli applied at the ridge vortex of the finger tip of digit lII, Both with electrical and with tactile stimulation, responses were unipolarly recorded with needle electrodes. A and B: average of 60 responses. C and D: average of 256 responses. E: displacement of stimulating probe. The contact area of the tactile probe was 3.14 ram2; the indentation was 300 izm and the rate of indentation.

150 ~tm/msec.

pre-indentation was from 0.01 to 0.1 mm. The shortest latency was measured from the peak acceleration, usu- ally 0.5-1 msec after the triggering of the vibrator. To reduce the magnetically induced artifact, a pick-up loop was placed near the recording electrodes (Fig. 2), and the amplified magnetic-induced artifact was sub- tracted on-line (Buchthal 1982a). 256-512 artifact-free sweeps were averaged, and duplicate averages were run and superimposed to ensure reproducibility.

With both electrical and tactile stimulation, re- sponses were recorded unipolarly from the wrist and the elbow, using needle electrodes (Dantec 13L60). The recording electrodes (3 mm bared tip) were placed as close to the nerve trunk as possible, and the refer-

ence electrode (5 mm bared tip) was inserted subcuta- neously at about 15-20 mm transverse distance. To ensure that the needles were placed optimally, the near-nerve needles were used as stimulating cathode and the needle pos i t ion was adjusted unti l stimuli of 0.5-0.9 m A were able to produce a response from either the abductor pollicis brevis muscle or the abduc- tor digit± minim± muscle. The evoked muscle action potential was recorded via a concentric EMG needle (Dantec 13L50) inserted in the muscle. The impedance was kept below 5 kS'2, and action potentials were amplified by a low-noise amplifier (Dantec 15C02)with a bandwidth from 20 Hz to 4 kHz ( - 3 dB). Averaged potentials were displayed on an X-Y plotter (Hewlett-

TABLE 1

Median nerve: mean values of orthodromic sensory conduction velocity and maximum amplitude (and area) of compound evoked action potential to electrical and tactile stimulation of the distal phalanx of digit lII. Note that in both age groups the velocity is significantly slower in the distal than in the proximal nerve segment, and that in older subjects, both the electrical and tactile parameters are lower than in younger subjects.

Mean age Subjects Conduction velocity Potential amplitude

(years) (no.) Dig. Ill-wrist Wrist-elbow At wrist At elbow (range) (m/ see ) (m / sec ) (#V) (txV)

Electrical stimulation 26.8 + 5.2 18 54.9 ± 4.3 64.3 ± 3.8 12.6 +_ 5.7 5.7 + 1 .t~ (16-37) 72.3+_4.5 13 47.2+6.1 57.4+4.1 4.7±3.,4 2.7± 1.7 (65-78)

Tactile stimulation p-p p ] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . p-p #v (area pc ' ~< msec) (area #V × msec)

26.8 + 5,2 I S 42.0 _+ 5.5 57.0 ff 53) 0.57 ± 0.25 0.46 _+ 0.23 (16-37) ( 1.54-f 0.64) ( 1.45 ± 0,44) 72.3 ± 4.5 13 36.7 ± 4.9 47.3 + 7,8 0,48 + (i.28 0.32 _+ O. 16 (65-78) ( 1. l 7 ± 0.44) (11.84 ± 0.31)

TACTILE CONDUCTION VELOCITY IN HEALTHY SUBJECTS 395

Packard 7090A) and, to obtain a quantitative value of the size of the tactile evoked sensory potential, the area was measured with a computer-assisted system of image analysis (Vidas, Zeiss).

Skin surface tempera ture was kept at 37°C by a feedback-controlled infra-red heating lamp (Dantec Tempera ture Control Unit 15H02) with the tempera- ture sensor placed at the thenar eminence of the hand.

The differences between groups were statistically analysed by means of Student 's t test, unpaired sam- pies, 2-tailed. P < 0.05 was regarded significant.

For histological procedures, skin blocks (approx. 1 cm 3) were obtained from the distal phalanx of digit I I I and digit V of 6 male subjects (age 20-66) who had suffered a violent death. Necropsy was performed within 30 h after death. The paraff in-embedded speci- mens were stained with a modified thrichrome tech- nique. The density of mechanoreceptors (Meissner's and Pacinian corpuscles) per mm 2 was determined by tallying them on 5 /~m thick longitudinal serial sec- tions, under light microscope.

Results

Electrophysiological findings

Electrical stimulation Orthodromically propagated sensory responses to

supramaximal electrical stimuli of the digital branches of both the median and ulnar nerves generally con- sisted of well-defined action potentials, almost always triphasic at the wrist, and of smaller amplitude, longer duration and often wi th a larger number of small components at the elbow (Figs. 1 and 3A, B). In both

nerves, maximum conduction velocity was higher along the proximal than along the distal nerve segment.

Median nerve. Mean values of maximum conduc- tion velocity and peak-to-peak amplitude to electrical stimulation, measured in younger and in older subjects, are listed in Table I. Values differed significantly ( P < 0.0004-P < 0.0001) both between the proximal and the distal nerve segments, and between the younger and the older subject groups.

Ulnar nerve. The results on the ulnar nerve ob- tained from electrical stimulation of digit V were simi- lar to those of the median nerve. Table I I shows the mean maximum conduction velocity and the peak-to- peak amplitude. Similar to the median nerve, the val- ues of the ulnar nerve differed significantly ( P < 0.0002-P < 0.0001) both between the nerve segments, and between age groups.

Tactile stimulation Regardless of recording site, tactile evoked poten-

tials consisted of a series of 6 -7 main spike compo- nents and of numerous minor deflections (Fig. 3C, D). For both nerves and age groups, maximum conduction velocity was significantly slower ( P < 0.001-P < 0.0001) and amplitude smaller ( P < 0.0001) than with electrical stimulation. Moreover, mean maximum conduction ve- locities were higher in the younger than in the older age group ( P < 0 .01-P < 0.001).

Median nerve. Table I includes the mean maximum conduction velocity and the area of response evoked by tactile stimulation. The tactile velocity values differed significantly ( P < 0.0001) between the distal and the proximal nerve segments and between the younger and the older age groups ( P < 0.009-P < 0.001). In addi- tion, while the area of the potentials recorded at the

TABLE II

Ulnar nerve: mean values of orthodromic sensory conduction velocity and maximum amplitude (and area) of compound evoked action potential obtained by electrical and tactile stimulation of the distal phalanx of digit V. Differences between distal and proximal nerve segments, and between age groups as in Table I.

Mean age Subjects Conduction velocity Potential amplitude

(years) (no.) Dig. V-wr i s t Wrist-elbow At wrist At elbow (range) (m/sec) (m/sec) (~V) (/~V)

Electrical stimulation 28.4 ± 2.9 13 50.2 ± 6.1 65.9 ± 4.8 12.6 ± 4.1 5.2 ± 1.4 (24-35) 71.6_+ 6.2 12 48.8 + 4.4 59.2 ± 6.7 5.8 + 4.4 2.3 ± 1.2 (63-82)

Tactile stimulation P'P btV p-p/.LV (area beV x msec) (area/zV × msec)

28.4+2.9 13 42.9±5.1 60.6±7.0 0.53±0.23 0.46±0.16 (24-35) (1.31±0.48) (1.39±0.25) 71.6±6.2 12 37.3±5.5 48.2±9.1 0.50±0.36 0.31±0.16 (63-82) (0.94 ± 0.38) (0.78 ± 0.35)

396 (L C A R U S O ET AI_,

wrist and elbow of the older persons was significantly smaller than that of the younger ones (P < 0.009-P < 0.001), no difference was seen when, within the same age group, the area of the proximal response was compared with that of the distal one. However, tactile velocity values differed significantly ( P < 0.002-P < 0.0001) when compared with electrical values.

Ulnar nen~e. Similar to the electrical stimulation, the values obtained along the ulnar nerve to tactile stimulation of digit V corresponded to those observed along the median nerve. The mean maximum conduc- tion velocity and the area of responses elicited by tactile stimulation are shown in Table II. Here too, conduction velocities differed significantly between the distal and the proximal nerve segments, and between the two groups (P < 0.01-P < 0.0001). And again, while in both age groups no significant differences were found between the areas of the potentials recorded at the wrist and at the elbow, statistically significant dif- ferences (P < 0.05-P < 0.0006) were observed when the older subjects were compared with the younger ones. Here too, finally, velocity values obtained with tactile stimulation were significantly lower (P < 0.07-P

TABLE 111

Concentrat ion of Meissners corpuscles (per mm ~) in specimens of skin taken from the distal phalanx of digit III and digit V of 6 male subjects (age 20-66) who had suffered a violent death.

No. Age Meissner 's corpuscles ( × m m 2 )

Digit III Digit V

1 66 18.4 31.0 2 41 23.8 26.5

35 26.0 4 24 10Ll 16.7 5 23 15.4 14.6 ¢~ 20 20.9 17.0

< 0.002) than those obtained with electrical stimula- tion.

Histological findings

In the 6 post-mortem subjects, Meissner's corpuscles were identified and counted in a single layer in the papillary ridges of the dermis of the distal phalanx of digit Ill and digit V (Fig. 4). The density was about 20 u n i t s / m m 2 (Table Ill), which is in keeping with previ- ous studies (Bolton et al. 1966), and there were no

Fig. 4. Post-mortem specimen of the finger tip of digit II1 of a 66-year-old man who hall suffered a violent death, but without history of diseases of the peripheral nervous system. Cross-section showing the different layers of the skin. Upper right insert: a Meissner corpuscle at higher

magnification. Lower right insert: a Pacinian corpuscle and a cross-section of a small nerve.

TACTILE CONDUCTION VELOCITY IN HEALTHY SUBJECTS 397

Right median nerve Surface temperature 37 ° C

Recording: Wrist Recording: Elbow

4-

2 ~tV

i

4 I.tV

~ - - - . - - - 20 I.tV 10 gV

I I I I

10 ms 10 ms Fig. 5. Right median nerve of a 30-year-old woman without history, signs or symptoms of peripheral nervous system diseases. Sensory potentials evoked at the wrist (left) and elbow (right) by supramaximal electrical stimulation of the lateral digital nerve of digit IlL In A, the electrical stimulation was performed at the distal phalanx, in B, at the middle phalanx, in C, at the proximal phalanx. Note that the latency becomes shorter and the potential amplitude becomes larger, as the stimulation point is moved in a proximal direction. However, besides the shorter distance between the stimulation and the recording points, here the number of stimulated fibres increases as the electrical stimuli are applied

more proximally.

significant differences among subjects. Because Pacinian corpuscles were not evenly distributed in the subcutaneous layers of the skin (Fig. 4), we were not able to calculate their density in a reliable manner.

Discussion

The smaller amplitude and longer duration of elec- trically evoked action potentials at the more proximal recording site is probably due to temporal dispersion of the impulse conduction along morphologically inhomo- geneous fibres activated simultaneously but aspecifi- cally by supramaximal electrical stimuli. Indeed, the more proximal the stimulation point, the more compact and of larger amplitude and shorter duration the com- pound evoked potential (Fig. 5).

The polyphasic aspect of compound sensory poten- tials to tactile stimulation, on the contrary, cannot be entirely attributed to temporal dispersion in impulse propagation along nerve fibres of different structure and function. Indeed, a similar dispersed morphology was observed both at the proximal and distal registra- tion points. Moreover, there were no significant differ-

ences in the maximum amplitude and the cumulative area between different segments of the same nerve of the same age group (Tables I and II). It is feasible that, because of their spatial distribution with respect to the probe and because of the viscoelastic nature of the medium (Buchthal 1982b), the rapidly adapting mechanoreceptors, constituted mainly by Meissner's and Pacinian corpuscles, are reached by stimuli with sequential delays. Each spike component of the tactile response would, thus, represent a number of contem- poraneously activated mechanoreceptors. However, also within nerve fibres of the same type, axon diame- ter and thus conduction velocity vary within a certain range (Knibest61 1973, 1975; Ochoa and Torebj6rk 1983; Light and Perl 1984; Mackel 1988; Kakuda 1992). Therefore, some temporal dispersion cannot be ex- cluded even when mechanical stimuli are used.

Along the distal segments of both the median and ulnar nerves, the mean tactile conduction velocity was significantly slower than the electrical one. This is partly due to the proximal shift of the stimulating cathode in relation to the electrical stimulus strength (Buchthal and Rosenfalck 1966; Wiederholt 1970), but it is mainly due to the receptor transduction time and

398 G. CARUSO ET AL.

to the conduction along the terminal axons, which probably undergo tapering and branching in the subcu- taneous and cutaneous layers of the skin on their way to the mechanoreceptors. However, also along the proximal segment of both the median and ulnar nerves of both age groups, maximum conduction velocity to tactile stimulation was significantly slower than that by electrical stimulation. This could be because the tactile stimuli activate specifically the Meissner mechanore- ceptors, whose innervating axons have a diameter of 6-12/xm (Light and Perl 1984).

The smaller size of tactile-evoked potentials in older subjects probably reflects the reduction in the number of Meissner's corpuscles that occurs with age (Bolton et al. 1966). However, the population of our histologi- cal study was not sufficient to confirm this hypothesis. In the elderly, nerve conduction velocity is slower and electrically evoked potential amplitude smaller than in younger subjects (Buchthal and Rosenfalck 1966; Rosenfalck and Rosenfalck 1975). This phenomenon is probably attributable to the loss of large nerve fibres and to changes in internodal length in subjects over 60 years (Lascelles and Thomas 1966).

Along the distal segment of the median and ulnar nerves, both the electrical and tactile maximum con- duction velocities are slower than in the proximal seg- ment. This is probably due to a shorter internodal distance of the nerve fibres situated in a body segment, which undergoes a relatively smaller post-natal growth. In fact, the lesser growth of the fingers and palm with respect to the forearm is significantly correlated with the shorter internodal distance and lower conduction velocity of their nerve fibres (Caruso et al. 1992).

Research partially supported by a MURST 60% grant, 1992.

References

Bolton, C.F., Winkelmann, R.K. and Dyck, P.J. A quantitative study of Meissner's corpuscles in man. Neurology, 1966, 16: 1-19.

Buchthal, F. Human nerve potentials evoked by tactile stimuli. Maximum conduction and properties of compound potentials. Acta Physiol. Scand., 1982a, Suppl. 502: 5-18.

Buchthal, F. Human nerve potentials evoked by tactile stimuli. Stimulus parameters and recruitment of components. Acta Phys- iol. Scand., 1982b, Suppl. 502: 19-32.

Buchthal, F. and Rosenfalck, A. Evoked action potentials and con duction velocity in human sensory nerves. Brain Res.. 1966, 3: t -122.

Caruso, G., Massini, R., Crisci, C., Nilsson, J., Catalano, A., Santoro, L., Battaglia, F., Crispi, F. and Nolano, M. The relationship between electrophysiological findings, upper limb growth and histological features of median and ulnar nerves in man. Brain, 1992, 115: 1925-1945.

Caruso, G., Nolano, M., Lullo, F.. Crisci, C., Nilsson, J. and MassinL R. Median nerve sensory responses evoked by tactile stimulation of the finger proximal and distal phalanx in normal subjects. Muscle Nerve, 1993, in press.

Hashimoto, 1., Yoshikawa, K., Sasaki, M., Gatayama, 3. and No- mura, M. Sensory nerve action potentials elicited by mechanical air-puff stimulation of the index finger in man. Electroenceph. clin. NeurophysioL, 1989, 72: 321-333.

Johansson, R.S. and VaUbo, ,~.B. Tactile sensibility in the human hand: relative and absolute densities of four types of mechanore- ceptive units in glabrous skin. J. Physiol. (Lond.), 1979, 286: 283-300.

Kakuda, N. Conduction velocity of tow-threshold mechanoreceptive afferent fibers in the glabrous and hairy skin of human hands measured with microneurography and spike-triggered averaging. Neurosci. Res., 1992, t5: 179-188.

Knibest61, M. Stimulus-response functions of rapidly adapting meehanoreceptors in the human glabrous skin area. J. Physiol. (Lond.), 1973, 232: 427-452.

Knibest61, M. Stimulus-response functions of slowly adapting mechanoreceptors in the human glabrous skin area. J. Physiol. (Lond.), 1975, 245: 63-80.

Lascelles, R.G. and Thomas, P.K. Changes due to age in internodat length in the sural nerve in man. J. Neurol. Neurosurg. Psychiat., 1966, 29: 40-44.

Light, A.R. and Perl, E.R. Peripheral sensory systems. In: P.J. Dyck, P.K. Thomas, E.H. Lambert and R. Bunge (Eds.), Peripheral Neuropathy. Saunders, Philadelphia, PA, 1984: 210-230.

Mackel, R. Conduction of neural impulses in human mechanorecep- tire cutaneous afferents. J. Pysiol. (Lond.), 1988, 401: 597-615.

Nilsson, J., Ravits, J. and Hallett, M. Stimulus artifact compensation using biphasic stimulation. Muscle Nerve, 1988, 11: 597-602.

Ochoa, J. and Torebj6rk, H.E. Sensations evoked by intraneural microstimulation of single mechanoreceptor units innervating the human hand. J. Physiol. (Lond3, 1983, 342: 633-654.

Pratt, H. and Starr, A. Mechanically and electrically evoked somato- sensory potentials in humans. Electroenceph. clin. Neurophysiol., 198l, 5t: 138-147.

Rosenfalck, A. and Rosenfalck, P. Electromyography. S e n t r y and Motor Conduction. Findings in Normal Subjects. Rigshospitalet, Copenhagen, 1975: 1-49.

Schieppati, M. and Ducati, A. Short-latency cortical potentials evoked by tactile air-jet stimulation of body and face in man. Electroen- ceph. clin. Neurophysiol., 1984, 58: 418-425.

Wiederholt, W.C. Stimulus intensity and site of excitation in human median nerve sensory fibres. J, Neurol. Neurosurg. Psychiat., 1970, 33: 438-441.