Department of Clinical Neurophysiotogy, Prince of Wales Medical Research, Institute ... University of New South Wales, Sydney, Matraville, NSW 2036 Australia.
Exp Brain Res (1992) 88:219-223
Experimental BrainResearch © Springer-'v%rlag1992
Research Note
Selective temporal shift in the somatosensory evoked potential produced by chronic stimulation of the human index finger S.C. Gandevia and K. Ammon Department of Clinical Neurophysiotogy, Prince of Wales Medical Research, Institute and School of Medicine, University of New South Wales, Sydney, Matraville, NSW 2036 Australia Received February 27, 1991 / Accepted August 28, 1991
Summary. The present study determined whether the cortical potential from the human index finger changed with chronic nerve stimulation. Cerebral potentials were repeatedly recorded to stimulation o f the ulnar nerve and the digital nerves o f thumb, index and middle fingers, before and during a 7-day period in which the index was electrically stimulated (80 Hz) for 8-10 h daily. Cerebral potentials were recorded at three scalp sites over the contralateral "hand" area. Chronic stimulation produced no significant changes in the amplitudes or distribution of the cerebral potentials from the individual digits or the ulnar nerve. However, for the stimulated index finger there was a significant, progressive increase in latency o f N20 and P25 without a detectable change in conduction velocity of distal peripheral axons. Timing in human central somatosensory pathways may be altered by the previous pattern o f peripheral nerve inputs. Key words: Cutaneous afferents - Plasticity ........ H a n d function
1990) and that its nerve supply is commonly damaged, it is surprising that no electrophysiological studies have yet attempted to assess the degree to which the somatosensory projection from a normal human hand may adjust to a marked change in the sensory input from one finger (cf. Sica et al. 1984). The latency and amplitude o f the N20 component o f the upper limb somatosensory evoked potential change during the day and with arousal (Emerson et al. 1988; D o w n m a n and Wolpaw 1989), and the sizes of multi-unit receptive fields change in the hours following intense stimulation of the mixed ulnar nerve in the cat (Recanzone et al. 1990). The present study examined the short-latency cerebral from the thumb, index and middle fingers under controlled conditions, and when the digital nerves of the index finger were stimulated for several hours per day for a week. Although there was no change in the amplitude of the cerebral potential produced by stimulation o f the index finger, the latency of the initial cortical negativity increased significantly during the period of chronic stimulation.
Introduction
Methods
Cortical sensorimotor representations can be remodelled in primates (Merzenich et al. 1983; Wall et al. 1983; Wall et al. 1986; Donoghue and Sanes 1988; for review see Kaas et al. 1983; Jenkins and Merzenich t987). In particular, the zone of somatosensory cortex representing a digit increases significantly if adjacent digits are amputated (Rasmusson 1982; Merzenich et al. 1984; see also Calford and Tweedale 1988) or the digit is used in a precise task for hours per day (Jenkins and Merzenich 1987). Also, the cortical zone for a digit may change in size and location after infarction of that for adjacent fingers (Jenkins and Merzenich 1987). Given that m o t o r function of the hand is dependent upon cutaneous feedback (e.g. Johansson and Westling 1984; Gandevia et al.
Repeated electrophysiological studies
Offprint requests to: S.C. Gandevia
Data were obtained from 18 recording sessions in one subject. He was then aged 35, free from neurological disturbance with normal sensory and motor conduction in peripheral nerves. Cerebral potentials were recorded to stimulation of the digital nerves of the thumb, index and middle fingers, and also to stimulation of the ulnar nerve. They were recorded with needle electrodes in the contralateral scalp at sites marked with indelible ink: 6.5 cm lateral and 1.0 cm posterior to the vertex (termed "lateral" or "hand" area electrode), 3.5 cm lateral to the vertex along a line joining the vertex to the lateral electrode ("medial" electrode), and anterior to the two electrodes forming an equilateral triangle ("anterior" electrode). A frontal reference was used. The digital nerves were stimulated using ring electrodes soaked in saline and covered in electrode paste and fastened around the mid-point of the proximal and middle phalanx with the cathode proximal. The afferent volley was monitored with fixed bipolar electrodes (interelectrode distance 40 mm) with the
220 distal electrode 30 mm proximal to the wrist. The cerebral potential produced by stimulation of the ulnar nerve at the wrist was recorded and its afferent volley monitored just abowe the elbow. The stimuli were delivered from a constant current stimulator (pulse width 100 las; stimulus rate 3 Hz) at an intensity supramaximal for the initial components of the afferent volley. The first 50-100 responses were discarded and then duplicate averages (50 ms at 10 kHz) were made of 500 responses for digital nerves and 250 responses for the ulnar nerve. The order in which the digits were studied varied randomly, with the ulnar nerve studied last in each session.
Experimental sequences Two control sets of data were obtained, one for 5 consecutive days and the other for 3 days immediately prior to chronic stimulation (see below). Stimulation of the index finger was achieved using a commercially available transcutaneous nerve stimulator. This constant voltage source delivered stimuli at 80-100 Hz (pulse width 100 t~s) at a level producing strong continuous paraesthesiae (100-140 volts). Strip electrodes (20-25 mm length, 4-5 mm width) lined with conductive jelly were secured around the proximal and middle phalanx. Stimuli ran for 8.-10 h per day, 2-3 h each evening with the balance while the subject was asleep, To eliminate the possibility that the electrodes might be dislodged the hand and forearm were loosely strapped in a light cast. There were no failures of the electrodes or the stimulator and, provided that the digit remained comfortably flexed (rest position), the stimulus intensity remained constant. To ensure cessation of any paraesthesiae following the stimulation (e.g. Ng et al. t 987; Applegate and Burke 1989), all experimental sessions were begun mid-morning at least two hours after the repetitive stimuli had been stopped. In the session after the 7th day of chronic stimulation the relationship between the amplitude of the afferent volley and the cerebral potential was studied. Stimulus intensity was adjusted to evoke a peripheral nerve volley of maximal size and then of 50% and 10% of maximum. The amplitude of the nerve volley was subsequently re-checked. Previous studies have documented this relation in detail (Gandevia et al. 1982; Gandevia and Burke 1984; see also Eisen et al. 1982).
Data storage and analysis All averages were stored on disc. Data from the initial control study (5 days) were analyzed prior to the main study. However, data from the control period of 3 days, and stimulation period of 7 days were not retrieved until completion of the study. The amplitude and latency of the cerebral potentials were measured using cursors (100 gs steps). As there was minimal difference between each pair of means, values were averaged (see Fig. 1). The following latencies were analyzed: initial cortical negativity - N20, subsequent positivity - P25, N30 and the associated amplitudes N20 P25 and P25-N30. While a frontal reference will distort the amplitude of the N20 P25 component (e.g. Desmedt and Cheron 1981 ; Mauguitre et al. 1983), the present study examined any changesin the cerebral potential. Values outside two standard deviations from the mean were considered statistically significant.
Table 1. Summary of control data for the latency and amplitude of cerebral potentials recorded at lateral scalp electrode overlying the contralateral hand area (mean, S.D. in brackets, n = 5)
Nerve input
Results
Control sequences Examples of the cerebral potentials are shown in Fig. 1. The amplitude and latency of cerebral potentials produced on consecutive days were relatively reproducible (Table 1). The coefficient of variation (standard deviation/mean) for the first set of 5 control days ranged from 3.0%-10.5% (mean 6.7%) for the amplitude of N20-P25 across the sets of nerves tested (digital nerves of thumb, index and middle fingers, and the ulnar nerve). For the amplitude of P25-N30 this coefficient ranged from 1.8%-13.7% (mean 6.8%). For measurements of latency the coefficients of variation were 0.61% (range: 0.41%-0.91%), 0.59% (range: 0.38%-0.75%) and 1.25% (range: 0.95%-1.55%) for N20, P25 and N30 respectively. For data collected on three consecutive days six weeks later, all latencies did not differ statistically from the initial control values. There were slight changes in the amplitudes in some of the potentials presumably due to minor differences in location of scalp electrodes. For subsequent comparison the control range was that defined immediately prior to the main study. However, the observed changes in latency (see below) would have remained statistically significant had either control range been used. The distribution of the initial component was slightly more posterior for the ulnar and middle finger inputs (Fig. I).
Changes in amplitude and latency during chronic stimulation The absolute amplitude of the cerebral potentials did not alter during 7 days with continuous stimulation of the index for 8-10 h per day at any of the scalp sites. Data for amplitudes of cerebral potentials recorded at the "hand" area electrode to stimulation of the index are shown in Fig. 2. No change in the amplitude of the index cerebral potentials was evident either when the amplitudes at the medial and anterior sites on the contralateral scalp were expressed relative to those for the lateral hand area site, or when all amplitudes were expressed relative to those for the ulnar nerve. These analyses provided no evidence that the site or size of the cortical projection from the index finger had altered as a result of chronic stimulation. However, there was a progressive change in the latency for the cerebral responses to index stimulation with successive periods of chronic stimulation (Fig. 1). No such changes were observed for potentials from the adja-
N20
P25 (ms)
(ms) Digit I Digit II Digit III Ulnar nerve
21.7 22.2 22.1 19.6
(0.09) (0.20) (0.10) (0.12)
24.3 25.1 24.8 22.3
N30 (ms) (0.13) (0.09) (0.18) (0.17)
30.5 31.1 30.7 30.0
(0.29) (0.29) (0,47) (0.46)
N20-P25 (laV)
P25-N30 (gV)
1.81 1.93 2.12 6.24
1.94 1.76 1.34 5.44
(0.19) (0.18) (0.09) (0.19)
(0.04) (0.12) (0.18) (0.25)
221
A
Di•gitl
Lateral
Digit II
Fig. 1. A Cerebral potentials recorded
DigIIIitf ~
U ~ 10 m s
22.5
B
23.3
from stimulation of digit I, digit II, digit III and the ulnar nerve during one experimental session prior to commencement of chronic stimulation• Recordings were made from three scalp sites (lateral [hand area], medial, and anterior) over the contralateral sensorimotor cortex as shown diagramatically in the inset. A frontal reference was used. Duplicate averages are shown superimposed, each of 500 responses following digital nerve stimulation and 250 following stimulation of the ulnar nerve. Stimulus intensity was supramaximal for the surface-recorded afferent volley. Vertical calibration: digits, 4 ~V; ulnar, 10 I~V. B An expanded view of the cerebral potentials from stimulation of the index finger during the control sessions prior to chronic stimulation. Mean latency for N20 marked with an arrow. The response from stimulation of the index at day 7 is shown below. The traces have been plotted for the interval 15-35 ms following the stimuli. Calibration: horizontal, 5 ms; vertical, 2.5 gV
5 ms 106 -
N 20 104 -
102 -
Control
(%)
100 -
98-
96
I
I
I
i
I
I
i
I
C
1
2
3
4
5
6
7
106"
P t04"
cent digits or the ulnar nerve. Thus, the latency o f N20 (at the " h a n d " area) increased from an initial control value of 22.49 ms ( ~ 0.20 ms, SD) to a final value of 23.28 ms at the end of chronic stimulation (p
