E. MARG, W. W. ALBERTS and B. FEINSTEIN
School of Optometry, University of California, Berkeley, Calif. and Mt. Zion Neurological Institute, Mt. Zion Hospital and Medical Center, San Francisco, Calif., U.S.A.
* Supported by PHS Research Grant No. 05061 from the National Institute of Neurological Disease and Stroke.
The development of micro-electrodes for use in human neurosurgical procedures (Marg, 1964) and their subsequent adaptation to chronic, cortical recordings (Marg and Adams, 1967; Marg et al., 1968, 1970; Marg and Adams, 1970) has increased the scope of both basic and clinical information obtainable on the function of the human brain. However, these fine, chronic recording tungsten electrodes, which are only about 75 microns in shank diameter, do not have the necessary stiffness to be pushed deep into the brain in a straight trajectory. This can be visualized directly when they are pushed through transparent gelatin instead of opaque animal brain tissue. Some kind of introducer is required if they are to be employed deep below the cortex for chronic unit recording. It appears that rigid shafts cannot be used chronically because any relative movement of the brain tissue and the electrode will cause loss of the unit and may induce necrosis of the tissue.
It has been shown in laboratory animals that if the micro-electrodes are introduced to the brain in such a way that the brain tissue is significantly damaged in the area of the microtips, no single neurons will be recorded (Marg and Adams, 1967). Presumably this is because the microtips will record only within about 100 microns of an active neuron. Relatively distant populations of functional neurons can be recorded by gross electrodes introduced to the brain by using a stainless steel shaft introducer (Chatrian et a/., 1959) which disrupts the brain tissue along its pathway. Even when micro-electrodes are introduced in a stiff bundle with a cross-sectional area of about I square mm, the disruption at the tips is enough to prevent unit recording. However, when a bundle of fine wires is introduced as a free cluster, effective recording of single neurons is practicable (Marg and Adams, 1967). A single, fine micro-electrode will also work, but a bundle increases the probability of recording single units under chronic conditions where manipulation of the position of the micro-electrode tip is not practicable.
METHOD
Our attempt to solve the problem of introducing the free-cluster bundle of micro-electrodes deep into the brain without their splaying uncontrollably took two stages. First, a new kind of introducer consisting of a 19 gauge (0.94 mm), 21 cm long stainless steel tube was made with a slit in its side just large enough to allow the fine micro-electrodes to pass singly through it (Fig. 1a, b, c). Since the electrodes can pass through the open slit before they are intended to, it was necessary to have two concentric tubes with displaced slits so that the electrodes could not pass through the double slits. The outer tube was 16 gauge (1.092 mm) and 23 cm long. Micro-electrodes in bundles of five to eight were placed inside the inner concentric tube introducer for subsequent stereotaxic insertion into the brain (Fig. 1c).
In the second stage, the bundle of micro-electrodes was then extended beyond the tip of the introducer into the brain tissue itself. Since the short segments of the tungsten wire micro-electrodes protruding from the tubes are relatively stiff, the electrodes will extend from the introducer in a straight line for 1 or 2 cm (as can be demonstrated in transparent gelatin). Then the internal introducing tube was aligned with the electrode shanks at the top of the slit and gently removed from the brain. The external and now remaining introducer tube was similarly removed. The electrodes were then in place with the microtips in the desired location, in which tissue had not been disrupted by the introducer. (If the microelectrode bundle has no electrical connector, a single tube without a slit might be used. There is, however, the likelihood of displacement of the bundle when the tube is being withdrawn. A connector would have to be attached after implantation.)
Fig. la - Cut-out drawing of the concentric
electrode introducer showing slits. A key on the inner tube fits
into the slit of the outer one so that the two slits are not in
alignment but the radial handles are. Lower right, a
cross-section of the key. Upper left, micro-electrode shanks
escaping from a tube slit.
Fig. 1b - Concentric slit tubes separated
and micro-electrode bundle before assembly. When the radial
handles are aligned, the slits are in their normal unaligned
position.
Fig. 1c - Concentric tubes assembled for
introduction to the brain with electrode in place. A slit segment
of plastic tube is placed over the protruding end of the distal
(plug) end of the inner tube above the radial handles but is not
visible.
RESULTS
Single neural unit activity was successfully recorded in three patient volunteers undergoing stereotaxic surgery for Parkinson's disease and other movement disorders. Figure 2 shows two single units of different amplitude recorded from a 61-year-old female patient with Parkinson's disease. The electrode tips were in the right nucleus ventrointermedius externus, an area where subsequently a therapeutic thermal lesion was made. X-rays showed the bundle to be exactly where it was aimed. Recordings were taken two days after the implantation and the figure shows monopolar activity recorded from one of the bundle of eight micro-electrodes with reference to a Beckman silver-silver chloride electrode on the forehead. The ears were grounded. Amplifier gain was 5 K with a passband of 10 Hz to 5 kHz and the activity was recorded with an FM system on 2-inch magnetic tape (Ampex FR 100A). The calibration, seen as a sine wave in the upper right of Figure 2, was 200 uV referred to the input (peak to peak) and at I kHz. A display was provided on a PDP-12 computer sampling every 15 u sec.
Fig. 2 - Responses of two neural units
recorded by a single micro-electrode in the nucleus
ventrointermedius externus. Upper right, calibration sine wave
200 uV and 1 kHz. Sample rate by PDP-12 computer: 15 usec/point.
When there is no longer any need for further recording, the electrode bundle can be simply removed by withdrawal under the usual sterile conditions without reopening the wound, just as a drain tube is removed.
CONCLUSIONS
While we have not yet had enough volunteer patients with this method to be able to discuss the findings in regard to movement disorders, we do feel that the basic problem of recording chronically from single neurons deep within the human brain has been solved. It is anticipated that this method may not only provide better diagnostic and prognostic information in neurosurgery but also help us to better understand brain function.
SUMMARY
Fine bundles of hair-like micro-electrodes attached to an electrical connector can be implanted in the brain stereotaxically by means of a special pair of concentric, longitudinally slit stainless steel tubes. The brain is not disturbed by the introducer in the vicinity of the microtips and single neural units can be chronically recorded from any desired locus, even deep within the brain.
RESUME
Un faisceau fait d'un grand nombre de tres fines microelectrodes reliees a un connecteur peut etre implante stereo-taxiquement dans le cerveau au moyen de deux tubes concentriques. Ces tubes en acier inoxydable vent fendus longitudinalement ce qui permet de les retirer ne laissant que les electrodes dans le cerveau. Le fonctionnement de ce dernier n'est pas affecte par l'introduction des microelectrodes et l'activite de neurones individuals peut etre enregistree chroniquement a partir de couches plus ou moins profondes du cerveau.
ACKNOWLEDGEMENTS
We thank Dr. Curtis Gleason for writing the computer program used to display unit activity and Mrs. Nancy Uyemura for making the micro-electrodes.
REFERENCES
CHATRIAN, G. E., DODGE JR, H. W., PETERSEN, M. C. and BICKFORD, R. G. (1959): A multi-electrode lead for intracerebral recordings. Electroenceph. clin. Neurophysiol., 11, 165.
MARG, E. (1964): A rugged, reliable and sterilizable microelectrode for recording single units from the brain. Nature (Lond.), 203, 601.
MARG, E. (1970): A neurological approach to perceptual problems. In: Early Experience and Visual Information Processing in Perceptual and Reading Disorders, pp. 151-156. Editors: F. A. Young and D. B. Lindsley. National Academy of Sciences, Washington, D.C.
MARG, E. and ADAMS, J. E. (1967): Indwelling multiple microelectrode in the brain. Electroenceph. clin. Neurophysiol, 23, 277.
MARG, E. and ADAMS, J. E. (1970): Evidence for a neurological zoom system in vision from angular changes in some receptive fields of single neurons with changes in fixation distance in the human visual cortex. Experientia (Basel), 26/3, 270.
MARG, E., ADAMS, J. E. and RUTKIN, B. (1968): Receptive fields of cells in the human visual cortex. Experientia (Basel), 24/4, 313.
MARG, E., CHOW, R., PERKINS, R. K., OLSEN, E. and WILSON, C. B. (1970): Measurement of uni activity in various systems of the brain postoperatively. Confin. neurol. (Basel), 32, 53.