(1)
Department of Neurosurgery, St Elisabeth-Tweesteden Hospital, Tilburg, The Netherlands
It is now well known that the brain is electrically excitable. Physicians frequently make use of electromagnetic properties to monitor or localize brain functions in patients (e.g. EEG, MRI or electrical stimulation of the brain). However, in the nineteenth century, it was accepted as a fact that the cerebral hemispheres were non-excitable ‘by all common psychologic stimuli’ [1]. This dogma prevailed to such an extent that studies that challenged this concept initially had to be performed outside of the universities [2].
6.1 Fritsch and Hitzig
In 1870, Gustav Fritsch (1838–1927) and Eduard Hitzig (1839–1907) electrically stimulated the cortex of a dog. Their experiments were done at home, on the dressing table of Frau Hitzig [3]. They were able to demonstrate that stimulation of certain parts of the cortex induced (contralateral) motor responses, and thereby they presented for the first time convincing experimental evidence for localization of function (Fig. 6.1) [1]. The motivation for their experiments followed from several developments in the eighteenth and nineteenth centuries. Galvani, and many predecessors, had already induced movements in animals by stimulation of the muscles, nerves and brain [4]. Animals seemed to contain ‘animal electricity’, something that was distributed by the nerves and secreted by the brain [4]. Volta constructed the first battery, which proved useful for electrical stimulation (Fig. 6.2). By the early nineteenth century, medical applications of electricity were commonly advocated, and a unique field of study, called galvanism, had developed [4]. Boling describes some of the experiments of Aldini, a nephew of Galvani and professor of physics at the University of Bologna [4]. Aldini used a voltaic pile for his experiments on animals and humans (Fig. 6.2). He could elicit strong muscular contractions with stimulation of the dura and the cortex of a trephined ox. He had also noted that electrical stimulation of the scalp significantly improved the mood in his human patients. In 1802, Aldini stimulated the cortical surface of the left hemisphere of a decapitated criminal and observed contractions of the right face [5]. Fritsch and Hitzig’s experiments thus stand in a particular ‘scientific’ tradition. Their work also followed from personal experience:
Fig. 6.1
Drawing after Fritsch and Hitzig’s (1870) figure of stimulation sites on the dog’s cortex. Δ twitching of neck muscles, + abduction of foreleg, † flexion of foreleg, # movement of foreleg, ♢ facial twitching (Figure taken from Gross, 2007 [3])
Fig. 6.2
Volta’s electrolytic battery, as used in a modified form for the stimulation experiments of Aldini and Rolando (circa 1800). (Upper figure) Volta’s chain of cups. (Lower figures) Pile that Volta named the artificial electric organ, columnar apparatus and electromotor apparatus (Text and Figure taken from Boling, 2002 [4])
Hitzig had tried electrical stimulation of the human head for therapeutic purposes and had noticed it caused eye movements. He then tried rabbits and also elicited movements. Fritsch, while working as a battle field surgeon, had apparently noticed that the contralateral limbs twitched while dressing an open head wound. [3]
The dog’s cortex was stimulated with platinum wires with brief pulses of monophasic direct current from a battery (i.e. galvanic stimulation). A current was used that was just sufficient to elicit sensations on the experimenter’s own tongue (these measurements were at that time seen as the best means for regulation of the stimulation current) [6]. In the animal, usually a muscle twitch or spasm [Zuckung] was observed. According to Gross (2007), who reviewed Fritsch and Hitzig’s experiments, the central findings were that:
(a) the stimulation evoked contralateral movements, the crossed laterality confirming observations dating back to Hippocrates in the 5th century BC [7], (b) only stimulation of the anterior cortex elicited movements, (c) stimulation of certain parts of the cortex consistently produced the activation of specific muscles, and (d) the excitable sites formed a repeatable, if rather sparse, map of movements of the body laid out on the cortical surface. [3]
Fritsch and Hitzig confirmed their stimulation findings by lesioning the areas whereby electrical stimulation had led to muscular responses. They often also observed some recovery of function, which led them to suggest that there was more than one centre involved in motor control [8]. Their experimental results are meanwhile seen as a major step in the development of modern neuroscience [9]. At the time, however, findings were met with great scepticism. Contemporary opinion was that motor functions were controlled from the basal ganglia, in particular the corpus striatum (remember that, for instance, Broca explained Leborgne’s hemiparesis by a lesion of the striate body; see Chap. 1). The cortex was thought to be rather insignificant for this purpose (or, in fact, for any purpose). One of the chapters in Young’s book gives an excellent overview of the concepts and dogmas on sensorimotor functions that prevailed at the turn of the nineteenth century:
The corpora striata seem to have held the loyalties of all as the major motor organs. When, in 1865, Luys assigned discrete motor functions to cortical cells on histological grounds, he still held that the corpora striata were the effective motor organs. Carpenter’s standard text on Physiology held in 1869 that the corpus striatum was the motor ganglion and the thalamus the sensory. William Carpenter’s writings provide a clear picture of the orthodox view, and an opportunity to contrast this with the new approach of Spencer and Jackson. [9]
Young tries to reconstruct why the nineteenth-century investigators failed to see that there were anatomical connections that ran from the cortex to the spinal cord:
It is perfectly understandable that the investigators of the brain in the nineteenth century related the sensory tracts to the optic thalamus and the motor tracts to the corpus striatum. Todd and Bowman were quite right in tracing the posterior columns of the spinal cord to the thalami and the anterior columns to the corpora striata. But why did they stop there? It appears that their preconception allowed them to see this far and no farther. Neither the thalami nor the corpora striata are the termini of tracts which are seen to pass into them. [9]
The corpus striatum consists of both the caudate and lentiform nucleus. Ontogenetically, these nuclei are single structure, divided by the internal capsule. Modern neuroanatomy teaches that there are direct connections from the motor cortex to the spinal cord, passing through the internal capsule. Sensory information is indirectly passed on to the cortex, via a relay station in the thalamus. This ‘passing on’ of sensorimotor tracts was not seen or better perhaps not acknowledged by most researchers before 1870. Even after Fritsch and Hitzig’s experiments, it took years to convince the scientific community that the cerebral cortex was directly implicated in muscular movements:
As late as 1886, Jackson indicated that most physicians thought epilepsy to be a dysfunction of sub-cortical and medullary centres [10]. It is not until 1890 that one finds, in Foster’s standard Text Book of Physiology, the modern view which sees the fibres of the cortico-spinal tract merely passing through the corpora striata, structures whose functions are unknown. [9]
Hughling Jackson’s work played an important role in cerebral localization theories, although—as stated by Young—‘no claim is feasible that Jackson predicted Fritsch and Hitzig’s findings’ [9]. Hughling Jackson adopted the view that the nervous system is an aggregate of distinct functional organs and was convinced that all neurological functions were exclusively of sensorimotor origin: ‘The psychical, like the physical, can only be functions of complex combinations of motor and sensory nerves’ [11]. In 1873 Hughling Jackson wrote that epileptic discharges had their origin in the cortex. Later, Charcot would honour him with the term ‘Jacksonian epilepsy’, referring to his description of the ‘march’ of seizures over different body parts, that implicated a somatotopic representation of motor functions [12]. To capture some of the Zeitgeist, it is illustrative to quote a famous footnote that Hughling Jackson wrote in 1870:
It is asserted by some that the cerebrum is the organ of mind, and that it is not a motor organ. Some think the cerebrum is to be likened to an instrumentalist, and the motor centres to an instrument; one part is for ideas, and the other for movements. It may then be asked, How can discharge of part of a mental organ produce motor symptoms only? I say motor symptoms only, because, to give sharpness to the argument, I will suppose a case in which there is unilateral spasm without loss of consciousness. But of what ‘substance’ can the organ of mind be composed, unless of processes representing movements and impressions; and how can the convolutions differ from the inferior centres, except as parts representing more intricate co-ordinations of impressions and movements in time and space than they do? Are we to believe that the hemisphere is built on a plan fundamentally different from that of the motor tract? What can an ‘idea’, say of a ball, be, except a process representing certain impressions of surface and particular muscular movements? What is recollection, but a revivification of such processes which, in the past, have become part of the organism itself? What is delirium, except the disorderly revival of sensori-motor processes received in the past: What is a mistake in a word, but a wrong movement, a chorea? Giddiness can be but temporary loss or disorder of certain relations in space, chiefly made up of muscular feelings. Surely the conclusion is irresistible, that ‘mental’ symptoms from disease of the hemisphere are fundamentally like hemiplegia, chorea and convulsions, however specially different. They must all be due to a lack, or to disorderly development, of sensori-motor processes. [9]
6.2 Ferrier
In the years following the publication of Fritsch and Hitzig’s paper, the literature on cerebral localization became exhaustive [9]. Cerebral centres for various functions were described in various animals, and techniques for stimulation and ablation were further refined. Among the many investigators, Sir David Ferrier (1843–1928) stood out. He was the first to confirm the experiments of Fritsch and Hitzig, and, according to Sherrington, was the main figure to provide the basis for a ‘scientific phrenology’ [13]. He also paved the way for intracranial surgery (Fig. 6.3). The principal inspiration for Ferrier’s research was Hughlings Jackson’s ideas on functional localization in the cortex. Ferrier’s book The Functions of the Brain (1876) was dedicated to Hughlings Jackson, ‘who from a clinical and pathological standpoint anticipated many of the more important results of recent experimental investigation into the function of the cerebral hemispheres’ [6]. Another motivation was to follow-up on the discovery of the electrical excitability of the hemispheres by Fritsch and Hitzig. Ferrier was initially not very explicit in acknowledging the work of his predecessors, probably because they had not referenced Hughlings Jackson in their paper.1 Ferrier’s method differed considerably from that of Fritsch and Hitzig. He was careful to use the minimal current necessary to obtain responses and to guard against conduction to neighbouring structures ‘by insulation of the electrodes, and careful removal of fluid which is apt to collect on the surface’ [6]. Still, Ferrier observed considerable differences between different animals with respect to the excitability of the hemispheres. He also observed a significant variation between different regions in the brain:
Fig. 6.3
(a, b) Ferrier’s composite results of his monkey experiments, showing areas whereby stimulation results in motor responses. Figures and quotes are taken from Ferrier’s book The functions of the brain (1886) [6]. Note that motor responses are found well behind the central sulcus and also in the temporal region. Ferrier states that there are no exact boundaries for these areas and that stimulation results are dependent on the duration and intensity of the current. For example (compare figure d), stimulation of area 6 results in ‘Flexion and supination of the forearm—the completed action bringing the hand up to the mouth. The movement is essentially the same as that which occurs on stimulation of the sixth cervical root of the brachial plexus.’ (c) ‘Lesion of the left hemisphere, causing motor paralysis of the right leg and right hand and wrist, and of some of the movements of the right arm, and loss of sight of the right eye.’ Ferrier believed that the paralysis that followed a large ablation was permanent. Others observed that there was a considerable recovery when the animals were kept alive for a longer period [8, 162]. (d) ‘Lesion (f) of the left hemisphere, causing paralysis of the action of the biceps on the right side.’ (e) Ferrier, following the nomenclature and anatomical scheme of Ecker, translated his results in monkeys to the human brain. He acknowledged that ‘An exact correspondence can scarcely be supposed to exist, inasmuch as the movements of the arm and hand are more complex and differentiated than those of monkey; while, on the other hand, there is nothing in man to correspond with the prehensile movements of the lower limbs and tail in the monkey.’ (f) Diagram showing the relationship between the convolutions and the skull. This sort of information guided surgeons in their trephination and cortical exploration. Macewen was the first surgeon to use Ferrier’s data for localizational purposes and operated on a patient guided solely by the motor phenomena of the patient [4]
A current sufficient to cause decided contraction of the orbicularis oculi will frequently fail to produce any movements of the limbs. By arbitrarily fixing a standard of stimulation which they thought sufficient, Fritsch and Hitzig failed to elicit the most important positive result of deep significance in regions of the brain which they choose to call inexcitable. There is no reason to suppose that one part of the brain is excitable and another not. The question is, how the stimulation manifests itself. [6]
Ferrier used a faradic current (via an induction coil) which resulted in complex muscular movements of longer duration. Remember that Fritsch and Hitzig used direct and monophasic (galvanic) stimulation whereby they induced muscle twitches. Ferrier believed that these more complex movements resembled intentional movements better than the short muscular reactions that were produced by the galvanic method:
The closing and opening shock of the galvanic current, applied to the region of the brain, from which movements of the limbs are capable of being excited, causes only a sudden contraction in certain groups of muscles, but fails to call forth the definite purposive combination of muscular contractions, which is the very essence, and key to its interpretation. Fritsch and Hitzig, in their description of the results of their experiments with the galvanic stimulus, did not, in my opinion, sufficiently define the true character of the movements. If the galvanic current is applied for a longer period than necessary to cause the momentary closing or opening shock, electrolytic decomposition of the brain substance ensues at the points of contact with the electrodes; an objection from which the faradic stimulus is entirely free. I have in my possession the brains of monkeys and other animals, on which experimentation by the induced current was maintained for many hours, which, with the exception of some degree of hyperaemia consequent on exposure as much as stimulation, are entirely free from structural lesion. [6]
In his book, Ferrier meticulously presents arguments against the criticism that the elicited movements ‘are in reality due to conduction of currents to the real motor centres situated at the base of the brain’ [6]. That dispute was certainly still not settled at the time:
Areas in close proximity to each other, separated by only a few millimetres or less, react to the electrical current in a totally different manner. If there were no functional differentiation of the areas under stimulation the diverse effects would be absolutely incomprehensible on any theory of mere physical conduction, which would, under the circumstances, be practically to the same point in all cases. Movements of the limbs can only be excited from certain points, all others being ineffective. No current applied to the prefrontal or occipital regions will cause movements of the limbs, yet physical conduction to supposed motor centres and tracts at the base is just as easy from these points as from the parietal regions, which react invariably and uniformly. The supposition that it is mere conduction to the corpus striatum and motor tracts which accounts for the movements is further absolutely contradicted by the simple experiment of placing the electrodes on the island of Reil, which immediately overlies the lenticular nucleus. Here we get in nearest proximity to the corpus striatum and internal capsule, and yet no reaction whatever can be induced by currents which are highly effective when applied to the more distant parietal regions. [6]
Ferrier could elicit movements from a rather wide cortical territory in monkeys, including the parietal and temporal lobe (Fig. 6.3). However, he commented that:
The mere fact that movements result from stimulation of a given part of the hemisphere does not necessarily imply that the same is a motor centre in the proper sense of the term. [6]
For Ferrier, motor reactions could also be secondary related to stimulation of a sensory centre, ‘being of the character of associated or reflex indications of sensation’ [6]. When, for instance, stimulation of the temporal regions induced movements of the contralateral ear (area 14 in Fig. 6.3), these findings could also point to this region’s involvement in hearing [8]. The central sulcus did not form a boundary for motor function in Ferrier’s studies. In fact, he (wrongly) assumed that tactile sensation had no representation within the peri-Rolandic region:
In my earlier experiments, which I have since abundantly confirmed, I could discover no sign of impairment or loss of tactile sensibility after the most extensive lesions involving the convex aspect of the cerebral hemisphere. And yet, considering the definite localization of the centres of sight, hearing, smell, and probably taste, as well as the respective motor centres, no conclusion seems a priori better warranted than that there must be a definite region for the various forms of sensibility included generally under the sense of touch (contact, pressure, temperature, &c.). [6]
This region for tactile sensory functions was, according to Ferrier, located in the falciform lobe (i.e. hippocampal, parahippocampal and cingulate gyrus), a view adopted also by others (see Fig. 6.4). Ferrier rejected, based on experimental and pathological findings, the hypothesis that motor and sensory tracts could become ‘jumbled together indiscriminately in the cortical areas’ [6]. Instead he conducted several experiments whereby he selectively destroyed areas in the hippocampal region of monkeys to prove that ‘beyond all doubt (…) the falciform lobe is the centre of common and tactile sensations.’ In order not to damage the brain areas that surround the deeper-lying mesiotemporal regions, which would have undoubtedly occurred during surgical exploration, Ferrier inserted a wire underneath or through the occipital lobe to cauterize the deeper parts of the brain. In his chapter
Fig. 6.4
Diagram and quotes taken from Mills’ Cerebral localization in its practical relations (1888). Mills’s findings were ‘based upon the investigations of Ferrier, Horsley and Schäfer, and others, and upon a study of cases, personal and collected from the literature of the subject’ [163]. ‘By the sensorial area is meant that for the senses of touch, pain and temperature, and modification of these senses, and it has been made to include the gyrus, fornicatus, hippocampal convolution, precuneus, and also portions of the superior and inferior parietal convolutions. This sensorial area has therefore been extended to the external surface of the cerebrum so as to include the general postero-parietal region’. Note the naming centre in the posterior part of the third temporal convolution
‘The sensory centres’, Ferrier concluded that the falciform lobe is the cortical centre for the sensory fibres in the internal capsule. He was unable to differentiate the lobe further into various different sensory centres, as he had observed for cortical motor function. Still, he assumed that the ‘various motor centres are each anatomically related by associating fibres with corresponding regions of the falciform lobe. This association would form the basis of a musculo-sensory localization’ [6].
Although the evidence against Ferrier’s view gradually mounted, it took considerable time before the modern view was accepted that sensory functions were located in the postcentral region. See, for a historical review on this topic, the paper of Boling (2002) [4].
6.3 Sherrington and Grunbaum: The Primate Motor Cortex
Ferrier’s observations that motor areas were located both anteriorly and posteriorly to the central sulcus were initially widely supported by other clinical and experimental studies. But this view changed, albeit only gradually. Horsley and Beever, for instance, observed that electrical stimulation of the postcentral region was usually not very successful in evoking motor responses [14]. Then again, they had also observed recovery of function from small lesions in the precentral gyrus and concluded that ‘the pre-central gyrus is not in man the only out-going motor centre for voluntary movements of the upper limb’ [8]. Convincing evidence for the modern view that motor functions are located exclusively in the precentral gyrus was given in the animal studies of Charles Scott Sherrington (1857–1952) and Albert Grunbaum (1869–1921) [15, 16].2 Their paper, published in 1917, had a lasting influence for many years. The various observations and conclusions are still well worth reading and remain among the most informative investigations to date. It is worth mentioning that Cushing and Campbell cooperated in some of the experiments and were acknowledged in the paper. Lemon (2008), in a historical perspective on the paper, summarizes the findings as follows:
Leyton & Sherrington (1917) provided the first detailed proof that there was indeed localization of function within the cerebral cortex. The durability of their report probably owes most to the fact that Leyton & Sherrington (1917) were the first to establish precisely the true extent of the motor area, and to provide the first detailed ‘motor map’ of the primate motor cortex. In addition, they showed that surgical extirpation of the cortical tissue that, when stimulated, gave rise to movement of a particular body part, resulted in a widespread weakness and loss of use of that same body part. There was, however, substantial recovery in the weeks that followed, recovery that was not lost on lesioning either the adjacent tissue in the same hemisphere or the equivalent cortical area of the opposite hemisphere. Finally, they were able to trace the course of the degenerating corticofugal and corticospinal fibres. They observed widespread degeneration in the cervical cord after a lesion of the hand and arm cortical area and noted that after such a lesion in the chimpanzee (p. 185), ‘the whole of the cross-area of ventral horn has scattered through it many degenerating fibres…’, which I think is the first report of the direct cortico-motorneuronal projection, a projection whose existence was confirmed physiologically by Bernard & Bohm (1954) and one that appears to be unique to primates (Porter & Lemon, 1993). [16]
The fact that Leyton and Sherrington did not evoke motor responses from postcentral areas contrasted with the observations of previous investigators such as Ferrier. Ferrier seems to have been very careful in his stimulation procedures, and yet Leyton and Sherrington obtained different results and conclusions. This discrepancy must be explained by their further refinement of experimental methodology [16]. It is likely that Ferrier and previous experimenters used currents that were ‘too strong’ so that motor responses were produced from areas outside the true motor cortex [4, 17, 18]. Leyton and Sherrington had realized that several factors could affect stimulation results, notably ‘the depth of narcosis, freedom of blood supply, local temperature, and such effects of experimental exposure of the cortex as “drying”.’ They went to great lengths to control experimental conditions as much as possible:
For stimulation of the cortex we have used faradisation, applied for the most part by the unipolar method [15, 19]. For this a broad copper plate was strapped over a pad wetted with strong sodium chloride solution lying against the sole of the foot contralateral to the hemisphere under examination. The pattern of electrode used was that figured in the Journal of Physiology, vol. xxviii. p. 16 [19]. It has the advantage of being easily applied with a light and fairly constant pressure against the cortex surface without risk of pricking the cortex or its pia; also of being easily sterilised by the flame, and of being readily bent to any appropriate curve when surfaces not otherwise easily reached have to be explored. The inductorium was of the usual physiological pattern, worked by a single Daniell cell. In many instances we have used also the bipolar method, the electrode tips being 2 mm. apart. The unipolar method is preferable, and gives minuter localisation. Especially where, as in certain experiments, a cut surface is to be explored for fibres running at right angles to that surface.
The animals were in all cases deeply anesthetised with chloroform and ether mixture for the whole of the operation by which the cortex is exposed. During the actual exploration with faradism the anaesthesia was lightened, since in profound anesthesia the cortex becomes inexcitable.
After the dura mater was opened it was always necessary to prick or tear some small holes in the arachnoid to let out the subarachnoid fluid. If that is not done, localization in the neighborhood of the suIci is almost or quite impracticable.
A precaution found necessary for success in a prolonged examination of the cortex is prevention of a fall in temperature of the exposed cortical surface. The temperature of the room was therefore always kept high, usually fully 30 °C; and the cortex was kept as far as possible covered with cotton-wool swabs wrung out after being soaked with Locke’s fluid at 38 °C. [20]
The peri-Rolandic cortices of 22 chimpanzees, three orang-utans and three gorillas were probed and studied with electrical stimulation. Stimulation usually lasted 1–2 s [21]. The results are extensively described in a paper that would nowadays probably not be accepted for publication in such a lengthy format (88 pages). However, the amount of detail and data that is provided is well justified given the many insightful observations and conclusions. More than 400 (!) different movements are listed. These movements often consist not only of a first but also a second, third or even fourth movement. Each unique response is numbered and indicated on a scale drawing of the cortex. For instance, observation no. 187 consisted of flexion of the fingers without the thumb (first movement), wrist flexion (second movement), wrist supination (third movement) and elbow flexion (fourth movement). Movement no. 192 consists of flexion of all fingers and the thumb (first movement), thumb adduction (second movement) and elbow flexion (third movement). In seven animals, precentral areas where specific movements had been elicited with electrical stimulation were lesioned to study the resulting neurological deficits. This resulted in severe motor deficits, as expected from the stimulation results. However, these animals made a remarkable and fast recovery, up to the point where there was almost no deficit detectable. One animal (see Fig. 6.5 for an extensive description) was studied with stimulation mapping on six different occasions. In consecutive sessions, surgical lesions were made in both hemispheres specifically to test whether these areas were involved in recovery of motor functions:
Fig. 6.5
(a A) First operation, January 3—partial resection of left precentral gyrus. Shown is part of the exposed left hemisphere of a chimpanzee during surgery. Numbers denote the various motor responses of the wrist and hand that were obtained along this part of the precentral gyrus. ‘The part indicated by the enclosure within the dotted line was then extirpated: care was taken to include the whole anterior wall of the sulcus centralis, i.e. down to the bottom of the sulcus.’ What followed was a partial paresis of hand musculature. ‘Fifteen days after the operation great improvement had occurred in the use of the limb; a cursory examination would hardly detect any defect of movement in it; the wound had completely healed.’ Second operation, March 3—extension of previous resection in left hemisphere. ‘Faradisation of the cortex along the lower edge of the old lesion evoked no movement in hand, but retraction and raising of right angle of mouth, and at one point quite regularly a brisk turning of neck and head toward the opposite side. Faradisation by plunging the unipolar electrode into the soft scar, even when the penetration amounted to 1 cm, failed to evoke movement. Precentral gyrus above the lesion was explored up to the upper genu; it gave the same results as at the previous examination 2 months before.’ ‘The old scar was then entirely cut away, and the old lesion deepened everywhere by further ablation; and the old lesion was increased upward by removing part of the gyrus previously uninjured as far as the line marked 3. iii. in the map.’ ‘Next day, the animal doing very well and being very active, the movements of right arm were thoroughly examined. No difference was detected between its existing motility and that obtaining before the last operation.’ (a B) Third operation, April 2—resection of homologue motor area in right precentral gyrus. ‘Careful search was made for evidence of movement in the right arm on stimulation of this cortical area for the left arm, in order to test the supposition that the recovery of the right arm movements might be explicable by supplementary functions for right arm taken over by the cortical field for left arm. Even with very strong and diffuse (widespread bipolar electrodes) stimulation, let alone moderate and weak with the unipolar electrode, never was any trace of movement of right arm evoked by excitation of the motor arm area of the right hemisphere. The movements elicited in left arm were, however, very various and vigorous. Finally, the whole of the area which under faradisation had provoked “leading” (primary) movement in fingers, thumb, wrist, and elbow was then extirpated by the knife to a depth of about 8 mm, and the floor of the ablated area cauterised superficially with the electro-cautery.’ Lines of lesion shown in figure AB. This resulted in a clear impairment of left hand and wrist functions. ‘Not the slightest recrudenscence of symptoms of paresis and clumsiness in the right arm was detected.’ Fourth operation, April 3—extension of previous resection in right hemisphere. Given the good condition of the animal and the fact that it ‘climbed actively about the cage’, it was decided to ablate more of the arm area of the right hemisphere. ‘The area of cortex indicated on the figure as bounded by fissura centralis and the dotted line above (…, limit line marked 3, iv.,) was then excised to the same depths as yesterday.’ Directly after surgery, the paresis of the left arm had increased. However, improvement was already seen the day after surgery. Four days after the last operation, there is ‘Further improvement in motility of left arm. Since the right cortex operation, which impaired the motility of left arm, the motility of right arm has notably increased, and right arm has been much more frequently employed than before. No paresis remains detectable in it; and hand, and the whole arm, are now repeatedly employed for all their usual purposes. This morning, after its breakfast, the animal sat and picked its teeth with the isolatedly extended index finger of right hand. It was seen also to pick and scoop out the furrows of the pinna of the left ear with right index finger. When making an effort to take with the left hand a small object, e.g. maize-grain, there occurs frequently an accompanying strong contraction (flexion) of the fingers of right hand. The converse has not been noticed to occur.’ (a C) Fifth operation, April 8—third resection in left hemisphere. ‘Gyrus post-centralis was then tested by faradisation to see if, especially at the levels opposite the excised portion of arm area, it had acquired motor responsiveness to the electric stimuli, but no motor responses could be elicited from it. From precentralis above the lesion, from the edge of the lesion right up to the trunk area, between arm area and leg area, repeated excitation elicited, and elicited easily, movements of shoulder, but of no other part of arm. Movements thus evoked in shoulder were never on any occasion accompanied by or followed by movements of elbow, wrist, fingers, or thumb. (…) The strip of cortex above the former lesion was then excised to the limit shown (…) by the broken line marked 8, iv. From the old lesion the electrodes never obtained responses, although plunged deeply into the tissue, and although both the single and double electrodes with strong stimuli were used. On recovery from the operation narcosis the animal showed no impairment in the motility of right arm. (…) May 4.—The animal now uses both hands and arms well. Employs either hand in feeding himself with banana or grapes. Peels banana, holding it in one hand and stripping off the peel with the other.’ Left and right hemisphere were then again examined with similar stimulation results as in previous sessions. The animal was then killed. (b) Microscopic examination of bulb and spinal cord revealed degeneration of pyramids both on the right and left side. (c) Stimulation responses from one gorilla ‘grouped diagrammatically’ (Figures and quotations taken from Leyton and Sherrington (1917) [20])
Improvement in the willed actions of the limb set in very early, and progressed until the limb was finally used with much success for many purposes even of the finer kind. Thus after destruction of the greater part of the arm areas of both hemispheres the two hands were freely and successfully used for breaking open a banana and bringing the exposed pulp of the fruit to the mouth. And again, after considerable destruction of one leg area the foot was successfully used for holding on the bars when climbing about the cage. [20]
Leyton and Sherrington were impressed by the fact that it took the animals many hours to ‘realize’ that their arm or leg had lost its particular function. They kept on using the limb as if it was not affected. This made them wonder whether the function of the ablated motor cortex could be ‘infra-mental’:
The impression given us was that the fore-running idea of the action intended was present and as definitely and promptly developed as usual. All the other parts of the motor behaviour in the trains of action coming under observation seemed accurate and unimpeded except for the role, as executant, of the particular limb whose motor cortex was injured. [20]
The paper has ‘a great many key points of lasting value’ [16]. Of course, there is the first convincing somatotopic representation of motor functions along the precentral gyrus (Fig. 6.5c). The drawing of the gorilla’s brain is an important precursor of the ‘homunculus’ images that—20 years later—would become one of the most pervasive pictures in the neurological and neuroscientific literature. Leyton and Sherrington emphasized that their image was only a simplified and diagrammatical representation of their observations; it was not an accurate depiction of an individual’s functional topography. Despite these cautions, these visually appealing images are frequently cited and (mis)interpreted outside the context of the original publication. Consequently, important details and nuances are ‘lost.’ For instance, these diagrams do not show the anatomical and functional variability between animals, a variability that was much larger than what Leyton and Sherrington expected:
The dissimilarity of the convolutional pattern of the hemispheres even in individuals of the same species (…), and the seemingly variable relation of analogous functional points to sulci of corresponding name, makes it practically impossible to decide with sufficient exactitude what point on the hemisphere of one individual is identical with a given point upon another hemisphere. [20]
The animal studies showed that there was no strict anatomico-functional correlation and certainly no invariant somatotopic order of functions along the precentral gyrus. Of equal importance was Leyton and Sherrington’s observation that the areas from which responses of a particular body part were evoked overlapped with those that controlled other body parts, a fact that would later also be confirmed in humans [22]. The fact that there is no pointlike representation for muscles or movements and that functional borders are diffuse and not strict is also something that simply cannot be adequately visualized by a drawing.
Then there was another important issue that was broadly questioned at the time: what exactly was represented by the cortical motor areas? If one would again simply look at Fig. 6.5c, one is apt to think that each part of the motor cortex controls the muscles of the body parts that are schematically written on the different parts of the cortex. However, Leyton and Sherrington stressed that movements, and not muscles, were represented in the primary motor cortex (an opinion that was also favoured by Hughlings Jackson and Ferrier). Sherrington’s observations are in line with current neurophysiological findings that each part of the motor cortex is involved in the control of multiple muscles and that, conversely, individual muscles are controlled from a wide cortical territory. What the exact role of the primary motor cortex is in movement control, and how the information is coded in the cortex, remains to be determined. The fact is that motor control involves a great many more areas outside the primary motor cortex, and there are multiple functional representations, probably also within the primary motor cortex.
Leyton and Sherrington noted in particular that the movements that resulted from electrical stimulation were fractional in their nature, consisting of more or less elementary movements from which other and more complex movements were ‘constructed’, or as Leyton and Sherrington eloquently put it themselves:
that the individual movements, elicited by somewhat minutely localized stimulations, are, broadly speaking, fractional, in the sense that each, though co-ordinately executed, forms, so to say, but a unitary part of some more complex act, that would, to attain its purpose, involve combination of that unitary movement with others to make up a useful whole. In evidence of this ‘fractional’ character it is only necessary to note the predominantly unilateral character, as elicited from the cortex, of movements that under natural circumstances are symmetrically bilateral. [20]
Electrical stimulation thus cannot evoke ‘natural’ movements. Then there is another very important aspect of electrical stimulation mapping, one that is surprisingly seldom discussed in the modern (clinical) literature, that:
the cortical motor point, or many of them, are within limits functionally unstable. The chart obtained from a motor region examined at one time and by one series of stimulations may not agree in detail with that obtained from the same motor region at another time and under another series of stimulations. [20]
So the responses that result from stimulation of a certain cortical area are not ‘fixed’ and can be altered by preceding stimulations of the adjacent cortex. This phenomenon of ‘facilitation’ was also studied by others, notably by Brown [23, 24]. Leyton and Sherrington assumed that facilitation of responses was a physiological phenomenon that reflected the ‘rich mutual associations of the cortical motor points’ [20]. Facilitation was needed to compose more purposeful movements out of the partial and fractioned ones:
Phenomena, such as (…) the functional instability of cortical motor points, are indicative of the enormous wealth of mutual associations existing between the separable motor cortical points, and those associations must be a characteristic part of the machinery by which the synthetic powers of that cortex is made possible. The motor cortex seems to possess, or to be in touch with, the small localized movements as separable units, and to supply great numbers of connecting processes between these, so as to associate them together in extremely varied combinations. The acquirement of skilled movements, though certainly a process involving far wider areas (cf. v. Monakow) of the cortex than the excitable zone itself, may be presumed to find in the motor cortex an organ whose synthetic properties are part of the physiological basis which renders that acquirement possible. [20]
Leyton and Sherrington’s observational work is truly a landmark in the study of functional brain topography. This was not only because of the ordered motor maps that were the precursor of later human findings, but in particular because of the careful practical approach and more theoretical considerations of the underlying physiological processes. The latter is often forgotten, or, even worse, not known.
6.4 Krause, Foerster and Penfield: The Human Motor Cortex
We know that at the end of the nineteenth century, several surgeons began to use electrical stimulation to map motor areas in human patients. Feodor Krause (1857–1937) reported stimulation of the central area to map function and identify seizure foci as early as 1893 [25]. He may have been the first to do so, although Horsley, Sherrington and Keen are also named [26, 27]. What is beyond dispute is that Krause and Schum’s map of 1911, based on monopolar (faradic) stimulation in 142 patients, is the first detailed representation of the human motor cortex [28]. In an impressive series (even measured by current standards), they systematically found all motor areas to lie on the precentral gyrus and in a somatotopic order that resembles that of the well-known homunculus (see Fig. 6.6). A similar map of the precentral gyrus, albeit more schematic, was published in 1930 by Foerster and Penfield [29]. Both Krause and Foerster made monumental contributions to neurosurgery including the introduction of surgery for epilepsy. They greatly contributed to the development of electrocorticography [30] as well as intraoperative electrocortical stimulation for localization of functions and localization of the epileptogenic focus (see for details the historical chapter in Lüders and Comair, 2001) [31]. Foerster was strongly influenced by Wernicke, with whom he cooperated and published an anatomical atlas. Following Wernicke’s suggestion, he went to Paris to study with Dejerine, Marie and Babinksi, before returning to Breslau [32].
Fig. 6.6
Krause’s map of motor functions as a result of (faradic) stimulation in 142 patients, published in 1911. Note the amount of detail and the resemblance to the latter maps of Penfield. (Figure taken from Devinsky, 1992 [21]). ‘All the foci which are found belong to the precentral convolution. They lie on the cortex, so arranged that the centers for the lower extremities are situated above, near the longitudinal sinus, and, as has been determined on experimental animals, they extend down to the median side of the hemisphere also. About the upper fourth of the precentral convolution is taken up with the lower extremity of the opposite side of the body. About half of the middle portion contains the foci for contractions in the upper extremity, from shoulder to fingers. In the lower fourth are situated the foci for the muscles of the face and the muscles of mastication; here the centers for muscles of the larynx, the platysma myoides and the tongue should also be found’ (Krause 1934) [164]
The somatotopic order of the primary motor cortex was only to become famous, however, with the introduction of the homunculus by Wilder Penfield (1891–1976), first published in Brain with Boldrey in 1937 [33] and later, in a more final graphical form, in the seminal monograph The cerebral cortex of man in 1957 [34]. More than 30 years after Leyton and Sherrington’s landmark paper, Penfield and Rasmussen devoted six pages of their monograph to their studies of the anthropoid cortex [34]. At that time, Penfield was already a world-renowned neurosurgeon and famous for his awake surgical procedures. The monograph describes Penfield’s extensive experience with electrocortical mapping of the human cortex and begins with an overview of historical studies on brain mapping. Penfield was a great admirer of Sherrington, whom he knew as a teacher from his (under)graduate studies (1915–1919) in Oxford (see Fig. 6.7). Sherrington had a lasting influence on his clinical and scientific work [35]. In the monograph, the various types of ‘instability of a motor point’ that were mentioned by Leyton and Sherrington are summarized by Penfield. Penfield had previously shown himself that the rules of facilitation and deviation of response also applied to motor and sensory responses in humans [33, 36]:
Fig. 6.7
In 1915, Penfield (middle row, third from the left) enrolled in the course on mammalian physiology, directed by Sherrington (top row, left) at Oxford University. The photo shows the graduating class of 1916. For Penfield, Sherrington always remained his ‘scientific hero.’ In a tribute (1952) he stated: ‘It was not the example of Horsley or Cushing that led me into surgery of the nervous system. It was the inspiration of Sherrington. He was, so it seemed to me from the first, a surgical physiologist, and I hoped then to become a physiological surgeon’ [35]
Facilitation
Suppose stimulation is carried out at any given point on the precentral gyrus, for example at point A, which produces finger flexion. If now the stimulation is regularly repeated, advancing the electrode step by step across the cortex anteriorly to A, the same response continues to follow each stimulus until the electrode is a considerable distance anterior to what was otherwise the anterior limit of motor response.
Reversal of Response
If the electrode stimulates point A and then, after time is allowed for the movement to subside, the stimulation is carried step by step downward along the precentral gyrus, flexion of the digit continues to result from each stimulus. Thus when a point B is reached from which at a previous time extension of the digit had been produced, flexion instead of extension results. Consequently, the response from B has been reversed by antecedent stimulation.
Graham Brown and Sherrington (1912) [24] found reversal of response to occur so frequently that they concluded that reversal is “one of the specific offices of the cortex cerebri.”
Deviation of Response
If the electrode begins stimulating again at point A and progresses step by step along the motor cortex, producing finger flexion each time, it may happen that a point C is reached which had previously moved the wrist. Stimulation of C now produces finger flexion instead of wrist movement. This is deviation of response at point C. After little time has elapsed, the points B and C will go back to their former state and will yield their original response and not that which they were caused to yield by facilitation. [34]
Modern clinical opinion agrees with Leyton and Sherrington’s work that in humans the central sulcus is the border between primary motor and sensory representations. However, the matter has never been completely settled, and questions remain to what extent both gyri are conjointly involved in sensorimotor functions. Sherrington was unable to evoke motor responses from the postcentral gyrus in his animals. However, there was one exception—when postcentral stimulation was ‘facilitated’ by an immediate previous and positive stimulation of the precentral gyrus. In this way Sherrington—indirectly—demonstrated a functional connection between the peri-Rolandic areas:
When the centralis posterior near to the central fissure is faradised immediately after elicitation of a motor response from centralis anterior at a point in the latter lying about opposite the point faradised in centralis posterior, the motor response obtained from the centralis anterior may reappear, and this even a few times in succession, though not for many unless centralis anterior be restimulated. This ‘echo-response’ is a phenomenon of considerable constancy. Our observations on it were made chiefly in the region of the inferior genu and below that, and with motor responses in lips, thumb, or index finger. Graham Brown [23, 24] has, independently of us, observed the phenomenon in regard to flexion of the arm, and in small monkeys macacus and cercopithecus as well as in chimpanzee. [20]
From an anatomical perspective, the precentral and postcentral gyri have direct connections via short, U-shaped fibres. Intergyral connections were first described by Meynert in the second half of the nineteenth century [37]. In 1906, Jakob specifically described connections between ‘homologous’ parts of the precentral and postcentral gyrus (a ‘brachial centre’ and a ‘facio-lingual centre’) [37]. Jakob’s work was published in Spanish and therefore at the time had scant diffusion in the English literature [37]. Foerster described motor disturbances with lesions in the postcentral region (afferent paresis), whereby in severe cases these resembled pareses [38, 39]. Although power was preserved, the required movements could not be performed or were insufficient because of diffuse contraction of agonists and antagonists. Peri-Rolandic anatomico-functional connections are nowadays well established in animals and human [40–42]. With non-invasive MRI techniques, it has very recently become possible to visualize the various connections of the precentral and postcentral gyrus, as well as several other connections to neighbouring cortical and subcortical areas (see the beautiful images of Catani, 2012, in Fig. 6.8). Penfield, who systematically confirmed Sherrington’s findings in humans, already considered the primary sensorimotor cortex a ‘functional unit’ (Fig. 6.9). Although he demonstrated that the ‘primary representation of movement’ was to be found in the precentral gyrus, he stated that:
Fig. 6.8
MRI-based fibre tractography enables non-invasive visualization of white matter connections in the brain (i.e. virtual dissection). The method requires (manual) placement of at least two regions to select individual tracts. The white region corresponds to the central sulcus (cs). Letters a–f indicate the level of the axial MRI slices as shown on the right side of the figure. (Figure taken from the paper of Catani and Stuss, 2012 [37]). These authors state in the discussion of their paper that ‘The exact functional role of the short U-shaped connections remains to be explained. Overall our study suggests that the distribution of the U-shaped fibres follows a functional division rather than a purely anatomical pattern. The three tracts of the central sulcus, for example, whose distribution and relative volume have a precise correspondence with the homunculus regions (Penfield, 1937) [33], are probably in relation to the importance of sensory information for motor control of skilful movements of the hand, mouth/tongue and foot’ [165]. Given the dense peri-Rolandic anatomical connectivity, it is ‘surprising that direct connections between primary sensory and motor cortices are not considered to play a significant role in current models of sensory-motor integration, for example, in relation to grasping’ [37]
Fig. 6.9
Figures and quotations from Penfield and Ramussen’s monograph, The cerebral cortex of man (1957) [34]. (a, b) Photograph and corresponding drawing (upside down) of the exposed cortical right hemisphere of an 18-year-old boy with a history of focal epilepsy. The epilepsy started with sensation in the left side of his body, followed by clonic movements of the left arm and leg. During the awake procedure, electrocortical recordings were made. Tickets A and B mark abnormal spontaneous cortical activity. A small tumour was seen anterior to ticket B. Motor and sensory areas were mapped out by stimulation. Observed motor responses or patient’s reported subjective experiences are recorded, and the site of stimulation is indicated with a numbered ticket. ‘For this purpose a bipolar electrode was used with the points separated about 3 mm. Occasionally we find it useful to employ a monopolar electrode. The current used was from a stimulator built by Rahm [166] and modified by Jasper. It is our custom to begin stimulation with a frequency of 60 cycles per second and a voltage of 1/2 a volt. The voltage is gradually increased until the first response is obtained.’ In the book, clinical and electrocortical mapping results of numerous cases are described and illustrated with photographs and drawings, similar to the example given above. Chapters are devoted to specific functions (e.g. Head and Eye Movement, Vocalization, Arrest of Speech) and end with general inferences on brain function and related cortical areas. (c) Penfield observed that as much as 25% of sensory responses was elicited from stimulation of the precentral gyrus. Conversely, motor response (d) was obtained in 20% of stimulations of the postcentral gyrus. (e, f) The sensory (left) and motor homunculus show the order and comparative extent of functional cortical regions. The homunculus is laid upon a (coronal) cross section of the hemisphere. The bars denote more accurately the relative proportion of the area from which responses in the corresponding body part were evoked. These maps show the evidence that Penfield collected from a large number of neurosurgical patients
the study of the cerebral cortex of man indicates (…) that there is a subordinate motor representation in the postcentral gyrus. Conversely, the primary somatic sensory representation is postcentral but there is a corresponding representation of sensation in the precentral gyrus. [34]
Penfield observed in his patients that as much as 20% of motor response was evoked from stimulation of the postcentral gyrus. Vice versa, sensory responses could be evoked from the motor cortex in 25% of stimulations (see Fig. 6.9). Further evidence of a more intimate functional connection between the peri-Rolandic areas came from resections within this area. Penfield observed that, when part of the (diseased) postcentral gyrus was surgically removed:
Stimulation of the exposed precentral gyrus may still cause the patient to feel a sensation in the part that has lost its postcentral representation. The reverse is true for precentral excision. Paralysis follows the removal of the precentral gyrus alone. But this is followed in turn by partial recovery, and after recovery stimulation of the postcentral gyrus produces limited movement. [34]
Penfield’s sensory and motor homunculi have become iconic images that are consistently cited in medical and scientific textbooks (see Fig. 6.9). In some respects, however, these images have done more harm than good, in a manner similar to the Broca–Wernicke models [33, 43]. As was the case with Sherrington’s drawing of the gorilla’s motor functions, the complexity of Penfield’s experimental findings was greatly simplified to obtain the graphical representations. This was clearly acknowledged by Penfield and his co-authors. They warned their readers that ‘the exact position of the parts must not be considered topographically accurate. They are aids to memory, no more’ [43]. Penfield was well aware of the inconsistencies of previous mapping studies and certainly realized the complexity of sensorimotor representations in the brain. He denied a simple one-to-one mapping of structure and function:
The cortical motor sequence of man shows little preservation of the segmental representation of muscles found in the spinal cord and brain stem. There was no evidence of separation of the movement of primitive flexors and extensors. Movements produced by cortical stimulation are gross, awkward. They involve multiple joints and numerous muscles. [44]
In the sensorimotor strip there is an orderly succession of responses to electrical stimulation, but physiologically speaking there is no representation in points or centres. Instead, there is a succession of nerve circuits in which precentral and postcentral gyri are closely related to each other. [34]
Several authors elaborated on the dangers of cartoons such as the homunculus. Some considered it a misleading model of cortical functions that persisted for many decades [45–47]. Others, for instance, Schott (1993), even worried that the homunculus had impeded scientific advance:
Penfield’s homunculus was a deceptively simple and yet naive concept. This type of illustration, a form of map, was a highly original attempt to portray graphically the observations of brilliant and painstaking research and one which has had a lasting influence as a mode of representation. It is memorable and useful. It has, however, been of limited and even doubtful scientific value, since fact and fancy have been confused. Illustration of brain function by projected drawings may best be reserved for those rare instances where true images can be derived and recorded. (…) Representation of everything else may best be served by an unambiguous diagram or words. [43]
Somehow this critique never made it to mainstream science or clinical practice. As of today, the view prevails that the order of primary cortical motor functions is fixed and invariant for every individual, either healthy or diseased.
6.5 Bartholow and Cushing: First Experiences from Conscious Patients
The first record of the use of electrical stimulation in an awake patient dates from 1874 [48]. The case was published by Robert Bartholow (1831–1904) as ‘Experimental investigations into the functions of the human brain’ in the American Journal of the Medical Sciences [49]. The patient was a young woman with a carcinoma that had eroded the skin and skull beneath it. The lesion encompassed both hemispheres and unfortunately had led to brain abscesses that required surgical drainage. Given these circumstances, Bartholow applied faradic stimulation to the dura and the brain of the left hemisphere:
(…) when the circuit was closed, distinct muscular contractions occurred in the right arm and leg. The arm was thrown out, the fingers extended, and the leg was projected forward. The muscles of the neck were thrown into action, and the head was strongly deflected to the right. [21]
No pain was noted upon stimulation. Bartholow then repeated his experiments on the right side with similar results [21]. He used higher currents to produce ‘more decided reactions.’ This resulted in a generalized seizure (with focal onset in the left hand) that lasted 5 min. Following the experiments, there were several recurrent seizures; the patient died 4 days later. Autopsy revealed ‘needle tracts from the electrodes, extensive thrombus in the longitudinal sinus, and a thick layer of pus covering the left hemisphere’ [21]. Bartholow’s experiments, which obviously did not serve any medical purpose, met with fierce criticism. He responded in a letter to the editor with an explanation of the case and his considerations. The (dying) patient had given consent, he wrote, and he had expected that the small electrodes would have caused no injury and that the procedure would be safe [4]. He regretted that his experimental results, which he hoped would progress knowledge, ‘were obtained at the expense of some injury to the patient’ [21]. Bartholow acknowledged that injury was done, but that this was not the cause of the fatal outcome.
It can be difficult to judge whether or not ethical borders are encountered when treating patients. This is obviously always a concern when doing clinical research. Progress inevitably means the use of new methods that have yet to prove their clinical effectiveness and safety. Such methods may be labelled ‘experimental’ and are especially sought for when patients are suffering, and conservative treatments do not provide significant relief. Good examples can be found in the practice of Harvey William Cushing (1869–1939), who is considered by many the ‘father of neurosurgery’ and honoured both as a neurosurgeon and a physiologist. When he founded a school of neurosurgery in Johns Hopkins Hospital, he was keen to integrate laboratories and post-mortem examinations in clinical practice. He was convinced that he needed to be both a clinician and a scientist and that medical progress required both laboratory experiments and surgical experiments [50]. His innovations demonstrate that there is really no clear-cut border between conservative and new ‘experimental’ treatment of patients. One of many examples was his approach to patients with trigeminal neuralgia. The facial pain that is caused by this disease can be excruciating, as was demonstrated by the first patient to be operated on by Cushing and Walker. This former sea captain felt ‘a devil twisting a red-hot corkscrew into the corner of the mouth.’ He was:
very near the end of his rope after years of seeking relief from the malady. (…) The slightest movement of his face or beard could set off an attack. Drugs were useless. His teeth had long been extracted [still today, a dental or mandibular origin for these complaints is frequently suspected first, GR]. He could barely eat or talk, and in the summer of 1899 he appeared at Johns Hopkins threatening suicide if the surgeons couldn’t help him. Walker was emaciated and shrunken, unwashed and red-eyed from sleeplessness, drooling and writhing and crying out in pain. Two previous operations for his nerve trouble had given him only short-term relief. Now he did not much care if he died on the table. [51]
In Cushing’s time several surgeons had attempted to relieve the pain by cutting out parts of the peripheral or central nervous system. Cushing modified a surgical technique that was developed by Hartley and Krause and significantly contributed to the safety and effectiveness of the procedure (i.e. extirpation of the trigeminal ganglion via a craniotomy and subtemporal approach). As always, Cushing was well prepared. He first built on his observations by studying the literature and by practicing on cadavers. Exemplary for his determination and level of preparation is the fact that he found practice on ordinary anatomic material unsatisfactory and therefore performed a great many operations on fresh cadavers; the toughening of the bodily structures gave him markedly different sensations than those of fresh tissue [51]. Cushing carefully documented clinical findings in order to better understand the pathophysiological basis of the disease and to further improve his surgical procedures. He published extensively on his experiences in the literature [52]. Clinical documentation meant that the patients were subjected to time-consuming, daily investigations for a period of weeks. These rigorous and sometimes uncomfortable examinations were of no direct benefit to the patient. Nevertheless, patients participated in the experiments because they thought they were part of their treatment or perhaps felt obliged to cooperate to their surgeon who had relieved them of their pain. Contemporary surgeons such as Halsted and Keen stressed that surgeons had a moral obligation to perform experiments and to pioneer new techniques. Hospitals like the Johns Hopkins Hospital had an important function in education and research. In Cushing’s time antiseptics and anaesthesia greatly improved, and this substantially lowered the surgical risks [53, 54]. As a consequence, surgery became more elective, and surgeons routinely gained access to the human body which revealed information that could not be extracted from laboratory experiments or clinical bedside teaching.
In 1900–1901 Cushing made a tour visiting several of Europe’s leading clinics and surgeons, an exercise that was at that time frequently done by (young) American doctors. He visited, among others, Kocher, Kroneck, Horsley and Sherrington. On the instigation of Kocher, Cushing studied the relationship between intracranial pressure, vascular dynamics and respiration. Cushing’s name was eventually given to the phenomenon (‘reflex’) of increased systolic pressure, bradycardia and irregular respiration in patients with elevated intracranial pressure. Although he carefully studied the brain’s reaction to compression and made significant contributions, the pathophysiological mechanisms were in fact already known for decades [55]. At the University of Pavia, Cushing saw Riva-Rocci’s machine for measurement of blood pressure. Cushing was given an inflatable armlet to take home, and 4 months later he would introduce blood pressure measurements for his anaesthetized patients in the Johns Hopkins Hospital [56]. These are some of the more beneficial examples of his tour around Europe. However, in his opinion, there were also disappointments. Cushing was generally not very impressed by the quality of the surgical and anaesthesiological procedures. Cushing’s meticulously precise and perhaps even neurotic style of operating was in strong contrast with most contemporary surgical procedures. He witnessed Horsley’s operation on a trigeminal ganglion that was done within an hour. Cushing claimed he saw nothing more than ‘blood and swabs’ [57]. This all made him even more determined to develop new methods for his own patients.
Despite the criticism that had followed the Bartholow case (and other cases), electrocortical mapping gradually gained acceptance as a clinical tool for localization of both the epileptic focus and sensorimotor functions. In 1909 Cushing was the first to demonstrate that the human postcentral gyrus contained sensory representations and that upon electrical stimulation patients reported some kind of sensation [58]. Although it had been repeatedly confirmed in animal experiments that the precentral gyrus was involved in motor functions, the role of the postcentral gyrus was still a matter of debate at that time. Animals obviously could not report their subjective feelings and were also anaesthetized during the stimulation procedure. Cushing had already used electrical stimulation for motor mapping in several patients, a technique he had learned during his stay with Sherrington in 1901 [59]. In his 1909 publication, which was widely lauded for his surgical and electrophysiological accomplishments, Cushing reported on two patients with epilepsy [60]. After the craniotomy had been performed under morphine and chloroform, these patients were awoken. In the first patient, a 15-year-old boy, mapping along the postcentral gyrus resulted in sensation in various body parts. In his second patient, Cushing was also able to map the adjacent gyri, to confirm that these gyri did not induce sensations (see Fig. 6.10). With Horsley and Krause, Cushing became one of the pioneers of electrical stimulation in humans.
Fig. 6.10
(Top left) Cushing’s drawing of his results of electrocortical mapping in one of the first patients that were studied under awake conditions. Results from both motor and sensory mapping are shown, as well as ‘negative’ mapping results of posterior parietal areas. (Top right) Cushing operating in the Brigham and Women’s Hospital, Boston. (Bottom) Cushing was also a gifted artist and often made drawings of his anatomical studies or intraoperative findings
6.6 Penfield’s Speech and Brain Mechanisms
Penfield was the first to study systematically cortical language organization from the perspective of both electrical brain stimulation and resection of cortical areas (Figs. 6.11 and 6.12). He did so in patients with traumatic brain lesions or tumours who suffered from epileptic seizures. Penfield was much indebted to Foerster, whom he had visited and worked with for 6 months in Breslau in 1928. Foerster was an extremely innovative neurophysiologist, neurologist and neurosurgeon. According to Tan and Black (2001), ‘he published more than 300 scientific monographs encompassing every aspect of the nervous system, including tabes, movement disorders, spasticity, extrapyramidal diseases, dermatomes, epilepsy, cortical localization, brain tumors, peripheral nerve injuries, and pain’ [61]. Penfield learned Foerster’s method of cortical stimulation under local anaesthesia that was aimed at localization of both functional (motor) cortex and epileptogenic tissue. Electrocorticographic recordings and galvanic stimulation of the cortex helped Foerster to delineate the epileptogenic region during surgery [32]. Whenever possible from a functional point of view, Foerster performed a radical excision of traumatic ‘scar tissue’ to cure the patient from invalidating epileptic seizures. The many veterans from World War I with cerebral injuries and resulting epilepsy gave him enormous experience. Foerster and Penfield conjointly performed a number of studies on this subject [29, 62]. Penfield eventually established the Montreal Neurological Institute, which was modelled on Foerster’s institute in Breslau [4]. Penfield’s perioperative approach became known as the Montreal or Penfield procedure and is still the basis for modern awake surgical procedures [63]. He refined contemporary methods that were needed to study his patients under local anaesthesia and carefully documented operative and clinical findings. Penfield’s many experiences and ideas on the neurophysiological underpinnings and localization of language functions culminated in Penfield and Roberts’ book Speech and Brain Mechanisms (1959). Here is the beginning of the first paragraph of a chapter entitled ‘Forbidden Territory’:
Fig. 6.11
Wilder Penfield (1891–1976) as a student at Princeton University in 1913. Two years later, Penfield received a Rhodes scholarship and went to Oxford University to study medicine. There his clinical and surgical thinking was inspired by men like Osler, Holmes and Sherrington. In 1924 Penfield went to Spain to investigate the histological aspects of brain cells with Rio Hortega and Cajal, culminating in publications on oligodendroglioma [167]. Later, he would write and edit a textbook on neuropathology, Cytology and Cellular Pathology of the Nervous System (1932) [168]. In 1928 Penfield visited Foerster in Breslau, who educated him on epilepsy surgery and surgical procedures in awake patients. With Foerster he published a topographical functional map that was based on 100 patients; they also studied damaged brain tissue under the microscope in order to understand better the process of scar formation [62]. Penfield was the founder and first director of the Montreal Neurological Institute, which he modelled on Foerster’s institute in Breslau. Photograph taken from Wikipedia (https://en.wikipedia.org/wiki/Wilder_Penfield)
Fig. 6.12
Penfield’s case C.H. (left) Photograph showing a large craniotomy that exposes parts of the frontal, parietal and temporal lobe. The order of (positive) stimulations is indicated by the numbers on the tickets. Patient was a 37-year-old male with focal and secondary generalized seizures that had started 3 months after a head trauma. Clinical semiology, as well as preoperative electroencephalography and pneumoencephalography, suggested pathological changes within the anterior part of the temporal lobe, and operative cortical excision was recommended. During surgery, a traumatic scar and dense gliosis were found under the tip of the temporal lobe (i.e. a lesion had resulted from the head trauma). Electrocorticography found a focus of high-voltage sharp waves under the surface of the anterior end of the temporal lobe. (Right) Schematic drawing of intraoperative results. Dotted line indicates the part of the anterior temporal that was eventually removed. Language disturbances were produced at points 26, 27 and 28. Anarthria (motor speech arrest) was produced at points 23 and 24. In the book, all the patients’ responses are documented. Here is an excerpt: 24—patient tried to talk and mouth moved to the right, but he made no sound. 25—the patient hesitated and then named ‘butterfly’ correctly. Stimulation was carried out then below this point and at a number of points on the two narrow gyri that separate 25 from 24, but the result was negative—no interference with the naming process. The points of negative stimulation are shown by the small circles in the figure. 26—the patient said, ‘Oh, I know what it is. That is what you put in your shoes.’ After withdrawal of the electrode, he said, ‘foot.’ 27—unable to name tree which was being showed to him. Instead he said, ‘I know what it is.’ Electrode was withdrawn and then he said, ‘tree.’ Stimulation at point 28 and 30 led to a naming problem and speech arrest, respectively, but the electrograph also recorded (widespread) afterdischarges; thus, the stimulation results were not considered of localizing value. There was no evidence of aphasia until 20 h after operation. Following that, there was progressive development of profound aphasia. This began to improve at the end of 2 weeks and cleared up finally several weeks later (Figures and text taken from Penfield and Roberts’s Speech and Brain Mechanisms (1959) [36])
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
