The structural basis of traumatic epilepsy and results of radical operation. <i>By</i> O. Foerster, Breslau, and Wilder Penfield, Montreal. <i>Brain</i> 1930; 53: 99–119

Formation and final appearance of the dural flap that accommodates putative local increases in pressure of the cerebrospinal fluid. From Krause, Surgery of the brain and spinal cord, 1911, Vol II, Figs 90 and 91, page 462). Working with (John) Hughlings Jackson (1835–1911), (Sir) Victor Horsley (1857–1916) had already stumbled on the fact that the electrically excitable surface of the brain could be used to identify a target for subpial resection of an irritable focus in cases of post-traumatic epilepsy. James B, aged 22 years, was the first of many cases in whom focal seizures were cured or much alleviated by excision of a vascular scar adherent to the meninges. Horsley was successful because Jackson was brilliant at suggesting the site of the lesion. Others depended for anatomical localization more on external landmarks. Krause had also recognized that the epileptogenic zone may be slightly remote from the macroscopic scar; and he assessed systematically the outcome of the valve procedure and local excision of injured brain. Various surgical techniques were then used to minimize the secondary scarring that often might follow removal of the epileptic cicatrix. Writing in Brain, Wilder Penfield (1891–1976) indicates that he has already presented observations on the structural basis of traumatic epilepsy and radical surgery to the Association for Research in Nervous and Mental Disease in New York on 2 December 1929; and Otfrid Foerster (1873–1941) is planning a more extensive version of the manuscript for publication in Zeitschrift für die gesamte Neurologie und Psychiatrie. American by birth, Penfield trained at the Johns Hopkins School of Medicine and the Peter Bent Brigham Hospital in Boston, before taking positions elsewhere. He persuaded the Rockefeller Foundation to create a neurological centre in New York but was unable to reconcile the competing interests of local specialists and moved to McGill in 1928. There he established the Montreal Neurological Institute which he directed from 1934 to 1960. Penfield had been influenced by European medicine and culture; and he learned much of his trade in England and continental Europe on various visits between 1915 and 1928. Sir Charles Sherrington (1857–1952), being too old to enlist and working incognito in a munitions factory during part of World War I (1914–18), must have warmed to Penfield’s decision, soon after he had arrived in Oxford as a Rhodes Scholar in 1915, to serve in France as a dresser. Back in Oxford from 1919, Penfield worked for several months on the classification of glia with Pio del Rio Hortega (1882–1945). Penfield made clear that it was the inspiration of Sherrington that led him into neurosurgery: ‘he was, so it seemed to me from the first, a surgical physiologist, and I hoped to become a physiological surgeon’. For Penfield, speaking in 1921, brain surgery was a terrible profession to be disliked ‘if [I] did not feel it would become different in my lifetime’. It did: and Sir George Pickering (1904–80) considered Penfield to have been ‘a great surgeon, a great scientist, and an even greater human being’. He mapped the cerebral cortex, defining and depicting the motor and sensory homunculi, and advanced the use of surgical treatments for epilepsy. Foerster was born, bred, worked and died in Breslau (now Wrocław, Poland). After graduation in 1896, Carl Wernicke (1848–1905) suggested that Foerster should study in Paris with Jules‐Joseph Déjérine (1849–1917), Pierre Marie (1853–1940) and Joseph François Felix Babinski (1857–1932). Foerster was also much influenced by Jackson and Sherrington. Using physiological principles, he devised the surgical procedures of posterior rhizotomy and anterolateral chordotomy for relief of pain and spasticity. Foerster characterized function in the cerebral cortex during surgery carried out under local anaesthesia. Using the techniques pioneered by Hans Berger (1873–1941), Foerster and Hans Altenburger (1902–38) introduced per-operative electrocorticography in 1935; and Foerster used hyperventilation as a means of temporarily reducing the threshold for seizures. As a neurologist Foerster advised general surgeons on how best to alleviate disorders of the brain and spinal cord; but during World War I he started to perform his own surgical procedures and became a self-taught neurosurgeon. As the late complications of their war wounds began to manifest, veterans with cerebral injuries resulting in epilepsy presented to Foerster. He identified the epileptogenic area using galvanic cortical stimulation under local anaesthesia. This made possible detailed mapping and maximal excision of scarred cortex whilst preserving intact vital areas. His patients might cry out during the procedure, only to be told ‘don’t be so touchy!’ Foerster attended Vladimir Ilyich Lenin (1870–1924) after his stroke but the patient was irascible and dismissed his doctors. Only Foerster remained, often having to observe his patient through the keyhole and manage the numerous Jacksonian seizures that Lenin suffered. Asked to suggest a scientist who could examine Lenin’s brain after the autopsy, Foerster explained that only Oskar Vogt (1870–1959) could locate the neurons responsible for genius. Foerster’s links to Russia as Lenin’s physician, and his wife’s Jewish ancestry, made for difficulties in his last years during the Nazi regime. His finances were restricted and his professional activities curtailed. But the Rockefeller Foundation eventually established a neurological institute, with Foerster as director, in Breslau. For Robert Wartenberg (1897–1956), Foerster was the foremost neurologist and neurosurgeon of his generation: ‘this was a man!’ Sherrington had performed serial section of posterior nerve roots in primates in order to map dermatomes. In his Schorstein Lecture delivered at the London Hospital on 13 October 1932 (The dermatomes in man. Brain 1933; 56: 1–39), Foerster drew on 30 years’ neurosurgical experience of similar studies in man, illustrating his text with 93 images of cases many of whom had undergone multiple posterior nerve root sections. Wartenberg describes this procedure as ‘for the relief of pain’. But, of Foerster’s methods, and his justifications for surgery, we learn directly only that: ‘I need not discuss here the circumstances under which such a selected procedure may be undertaken … vaso-dilatation following the electrical stimulation of a single posterior root is limited to … an area similar in shape … to the dermatome … I have obtained vaso-dilatation by stimulation of nearly all cervical, thoracic, lumbar and sacral posterior roots.’ Foerster and Penfield remind readers of Brain that epilepsy is often focal in onset even if generalized in appearance. The Jacksonian march and associated adversive features make this clear. But even when seizures start from a focus, the state of the brain parenchyma determines the threshold for secondary spread. In Jacksonian terms, the attack and its aftermath involve both positive and release phenomena. Much has been learned from stimulation of the exposed cerebral cortex under local anaesthesia from which can be mapped the sites of origin for spontaneous seizures that are focal in onset. The trigger zones reveal regional differences in excitability of the human brain. Stimulation at particular sites faithfully reproduces clusters of symptoms that reflect clinical syndromes. Stimulation of area 6a leads to turning of the eyes, head and trunk to the opposite side, but without aura. Area 17, the occipital pole, yields light, flames and stars in the contralateral visual field, whereas stimulation nearby produces an hallucination of formed figures and gaze to the opposite side. Parietal stimulation over area 5b and 5a induces contralateral turning and limb movements with or without sensory auras. Temporal stimuli result in auditory experiences; and grunting and pharyngeal responses are illicited from the operculum. It follows that the preliminary assessment of the patient with focal epilepsy should include both encephalography and electrical exploration of the cerebral cortex in order to map the probable site of initiation for the discharge. Whether or not excision then follows depends on estimating the effects of surgery compared to those of the lesion itself. Evidently, treatment restricted to surgical laceration of the offending site has recently been tried. But this merely exacerbates scar formation, and encourages development of the vascular network that anastomoses meningeal and cortical vessels. Laceration alters the architecture of the entire hemisphere with ‘wandering’ of the lateral and third ventricles towards the site of the wound as the cicatrix contracts. By contrast generous excision of cerebral tissue leads to minimal further vascularization and astroglial reaction, leaving a fluid-filled cavity and without additional distortion of the surrounding brain parenchyma. In the cases to be described—some the result of previous gunshot injuries and others relating to birth trauma—the histological evolution of focal brain injury, and the clinical consequences of excising the area that gives rise to epilepsy, are each considered. Case 1. Shaded area shows the position of the lesion at operation and the second line the extent of operative removal. The pattern of the attacks indicated that the stimulus might arise in front of or behind the lesion, as indicated by the arrows. Case 1. Cross section of cicatrix attached to the thickened dura. Case 2 (Wolf) suffers permanent partial left hemiplegia and Jacksonian seizures starting 5 years after a right parietal gunshot wound in 1914. Encephalography shows a cyst with ventricular enlargement and displacement towards the affected side. Peroperative cerebral stimulation produces convulsions with movements similar to his spontaneous Jacksonian seizures. No further attacks have occurred in the 2 years since surgical removal of the cyst and its associated cicatrix. Microscopically, this is characterized by meningeal vessels passing close to the area of electrical excitability with tissue destruction and reaction of neuroglia and microglia. Case 3. Area of lesion seen at operation indicated by shading. Dots show points stimulated by electrode. The symptoms experienced by the patient following each stimulus are indicated on the chart. Four years after gunshot injury of the right frontal lobe, managed in 1914 by immediate surgery with subsequent closure of the cranial defect, Case 4 (Laurenz) develops adversive seizures to the left with movements of both limbs. Encephalography demonstrates ventricular enlargement and displacement towards the lesion. Craniotomy in 1925 shows a scar with meningeal adherence to the underlying brain parenchyma. Electrical stimulation results in seizures reminiscent of his clinical attacks. Removal of the lesion has been completely successful to date. The sample contains fragments of shell embedded in the cicatrix with gliosis and a vascular plexus within the scar. Case 5 (Werner) suffers an injury to the vertex in 1916, managed by immediate surgery, resulting in bilateral spastic paraparesis and sensory deficits—both more marked on the left. Seven years later, he develops focal seizures suggestive of origin in the right parietal region with adversive and left-sided movements spreading to the right. Encephalography shows enlargement of the lateral ventricles with upward displacement on the right. Surgery is in two stages: first to free the cicatrix from surrounding tissues, resulting in several per-operative seizures; then stimulation and wider excision of the scar tissue. Histology shows glial and vascular changes. No further seizures have occurred during the 4 years since surgery. Following birth injury with infantile right hemiplegia, Case 6 (Zimmerman) becomes intellectually retarded from the age of 12 and soon after develops adversive turning of the head and eyes with movements of the right sided limbs followed by generalized convulsions. Encephalography shows a large defect with ventricular displacement towards the lesion. In fact, the cyst is paired and stimulation between the two elicits a typical convulsion. After removal of both components and the cicatrix that joins the two, the patient is free from seizures for the next 3 years (Fig. 5). Histology shows microglia having long parallel fibres coated with collagen, resembling the architecture of peripheral nerve. Case 6. Sketch of the excised tissue after formalin fixation, showing the two cysts. Case 7. Encephalogram (brow up) 19 years after birth injury, showing enormous enlargement of the right lateral ventricle and pulling of the third ventricle across [and] towards the right side. Case 7. High-power sketch of one of the strands. Some of the astrocytes can be seen to apply large processes to the collagen lining. The fine fibres are all neuroglial and arise from these cells. The dots indicate numerous such fibres cut across. Silver carbonate astrocyte stain, Hortega. Schematic drawing of three levels in any meningo-cerebral cicatrix of longstanding. Left panel shows blood vessels and bundles of collagen penetrating the brain beneath the meninges. Middle panel from area farther below surface shows intimate relationship of collagen strands and fine-fibred neuroglia, all parallel and pointing in the direction of the cicatricial pull. Right panel from an area deeper in the brain shows the structure of astrocytes and blood vessels which make up the vaso-astral framework and transmit the cicatricial pull. Modified from Fig. 23 p. 116. The lessons from this series are that surgery must be carried out under local anaesthesia in order to allow per-operative electrical stimulation followed by adequate excision and fascial repair of the defective dura. Too timid an approach merely leads to crowding of the affected areas with further scar formation and worsening of the seizures. Penfield, working in Montreal, prefers to replace the bone-flap; whereas, in Breslau, Foerster does not. Their paper addresses seven of twelve cases studied, the remaining five merely rehearsing the experience already gained. All cases show enlargement and displacement of the ventricular system towards the lesion. The mechanism of seizure induction is presumed to be direct tugging of the cicatrix on surrounding healthy tissue rather than passive displacement of the neighbouring parenchyma to fill space created by the loss of tissue. Support for this hypothesis comes from the appearance at surgery of the brain retracting by as much as 1.5 cm once the adherent meninges are freed from the over-lying skull. This interpretation is also consistent with the histology showing fibrous tissue originating from the adherent meninges and a plexus of blood vessels mingled with reactive astrocytes organized in parallel arrays that reach down to the advancing ventricle. Nerve fibres are sparse within the lesion and nerve cells rarely seen. Yet the presence of residual pigment within microglia (phagocytes) suggests that the organization of tissue continues for many years after the original insult. Indeed, the alteration of brain tissue, aberrant coating of astroglial fibres with collagen, and loss of nerve cells and their fibres is most evident in the two cases of birth injury. Furthermore, the intense vascular network, connecting to the meningeal vessels above and to those of the cerebrum below, suggests to the authors that the onset of focal seizures—after an interval of between 5 months and 14 years, and with an average of 5.5 years—relates more to vascular changes than the progressive scarring that continues to evolve over many years. That said, tension within the tissue may be an important contributing factor in the development of convulsions: ‘The blood vessels form in one sense the woof of the contracting network. Traction … upon the vessels [is] inevitable. The hypothesis at one suggests … that a vaso-motor reflex secondary to this traction is responsible for the initiation of the convulsive seizures.’

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.