Emil Theodor Kocher—Valve surgery for epilepsy


Address correspondence to Werner Surbeck, Clinic for Neurosurgery, Kantonsspital St. Gallen, Rorschacher Strasse 95, 9007 St. Gallen, Switzerland. E-mail: werner.surbeck@kssg.ch


Emil Theodor Kocher (1841–1917) was a pioneering and versatile Swiss surgeon who played a decisive role in the surgical evolution on the threshold to the 20th century. Apart from conducting intense research and fostering the development of the surgical treatment of thyroid gland diseases (honored with a Nobel Prize in 1909), he remained a generalist and was active in orthopedic, genitourinary, and neurologic surgery. Even today, many surgical techniques and instruments are still named after him, thus providing evidence of his great impact. His neurosurgical ambitions included, in particular, cerebral and spinal trauma, the pathophysiology of elevated intracranial pressure, as well as etiological considerations and the operative treatment of epilepsy. This article aims to shed light on Kocher’s work on epilepsy, published exclusively in German, and illustrates the development of his idea on valve surgery for recurrent general convulsions.

Emil Theodor Kocher (1841–1917) was an internationally distinguished physician who played a decisive role in the surgical evolution on the threshold to the 20th century (Fig. 1). He was elected as Professor of Clinical Surgery in Berne, Switzerland, a position he held for 45 years (1872–1917). Specifically for his intense effort to lower mortality and morbidity rates in surgery of the thyroid gland, he was the first surgeon ever to be honored with the Nobel Prize in Physiology or Medicine in 1909 (Hildebrandt et al., 2012). Although he remained a generalist with a broad surgical spectrum, Kocher had great interest in neurologic surgery. His pathophysiologic considerations in the field, however, are less commonly known. It is our intention to shed light on Kocher’s early attempts in epilepsy surgery and to illustrate the development of his idea on valve surgery for recurrent general convulsions.

Figure 1.

Emil Theodor Kocher (1841–1917). From (Bonjour, 1950).

Kocher’s First Surgical Interventions for Epilepsy

Kocher lived in times when Dogma, speculation, and natural philosophy in medicine were increasingly replaced by scientific evidence (Hildebrandt et al., 2012). At the seventh International Medical Congress held in London in 1881, Kocher witnessed the well-known debate between Friedrich Goltz (1834–1902) and David Ferrier (1843–1928) over cerebral localization of brain functions. Goltz presented his studies on bilateral decortication in dogs with no permanent loss of motor strength or deficits in primary sensation, and he hypothesized the idea of a rather “diffuse” distribution of brain functions. In contrast, Ferrier put forward his striking experiments in monkeys, in which circumscribed lesions produced well-defined and reproducible deficits. After a selected committee had meticulously investigated the animals and arrived at the conclusion that the cortectomy in Goltz’s dogs was much less widespread than anticipated, the majority of the medical world was convinced of the existence of circumscribed cortical localization of function (Tyler & Malessa, 2000). Victor Horsley (1857–1916), another participant in this debate, established a lifelong friendship with Kocher, and they led an animated discussion about their common professional interest, including neurosurgery (de Quervain, 1918). As confidence grew about the localization of certain brain functions to circumscribed cortical areas, Jackson’s (1870, 1873) theory of focal epilepsy as a manifestation of cortical irritation opened a new option for epilepsy therapy: the surgical removal of the irritation.

Four months after the inspiring insights gained in London, Kocher performed his first cranial surgery with the intention to cure epilepsy. The patient, a 16-year-old farmworker, had sustained an impression fracture in the left frontoparietal region by falling from a hay loft 4 years before the intervention. Six weeks after the initial impact, focal motor seizures affecting the opposite leg, arm, and side of the face occurred with occasional generalization. Seizure frequency increased over the years before surgery from occasionally to twice a week. Kocher (1893) assumed that a bone fragment or osteophyte irritated the underlying brain “like a nail.” In December 1881, surgery was carried out, consisting of enlargement of the cranial bone defect and removal of a jagged bone fragment that had penetrated the dura. The bone fragment lay in a shallow groove above a thickened arachnoidea, which he left untouched. After an initial period of seizure freedom, seizures relapsed soon after the intervention. The patient died in 1888 after 9 hr of continuous seizures.

Soon after Horsley’s (1886) renowned publication on the first resections of cortical lesions followed by a period of time free of seizures, Kocher adopted this practice. In 1892 he performed a cortical lesionectomy in an 8-year-old patient with intractable epileptic seizures. The epilepsy in this patient had started soon after a head trauma with a right-sided frontal impact fracture and subsequent super infection at the age of 5 years (Kocher, 1893). The seizure semiology consisted of turning the head and eyes to the left (deviation conjugée), infrequently followed by motor seizures affecting the face as well as the left leg and arm. The resection of a cortical scar of the right frontal lobe led to immediate and permanent seizure freedom.

In cases without visible external defects from former head injuries, however, correct localization of the trepanation area was still difficult. It was necessary to correctly attribute the initial semiology to a known functional area of the brain. Furthermore, the suspected cortical area had to be precisely localized externally for direct approach. For this purpose, Kocher developed a new instrument called a craniometer, which made it possible to project the known functional cortical areas on the outside of the human skull (Fig. 2). Kocher (1892) developed this method on corpses and merged his findings with a cortical map created by Horsley that displayed the results of experimental stimulation in monkeys and intraoperative stimulation in men (Schultke, 2009). Although the primary motor area was studied extensively, other functional areas were still more speculative at that time.

Figure 2.

Kocher’s craniometer, which made it possible to project the known functional cortical areas on the outside of the human skull. From (Kocher, 1892).

The Origin of Kocher’s Thesis Pointing out the Important Role of Intracranial Pressure in the Pathogenesis of Epilepsy

In 1892, Kocher performed a trepanation above the central region in a boy with focal motor seizures. Although he did not find a pathologic cortical lesion and therefore terminated the operation without the intended cortical resection, the boy was initially seizure free. At that time, the wound was often not completely closed, in order to allow blood to drain out of the wound resulting in transient liquorrhea. After the postoperative liquorrhea ceased in the above-mentioned case, the seizures, however, relapsed. Kocher (1893) made similar observations in other patients with posttraumatic porencephaly where resection was not possible: Seizures relapsed with complete cessation of liquorrhea. Based on these experiences, Kocher postulated that focal epilepsy was caused by two main factors: Structural abnormality of the cerebral cortex and an individual’s disposition. He was convinced that the increased production of cerebrospinal fluid (CSF) or an augmented CSF tension constituted the most important factor for epileptic seizures. He believed that further circumstances lowering the threshold for cortical irritation by an unknown mechanism included mental illness, alcoholism, infections, and diffuse brain injuries (Kocher, 1893, 1899). Relying on the work of Kussmaul & Tenner (1857), who showed that rapidly occurring severe anemia of the brain (e.g., in exsanguination or after ligation of the carotid arteries) caused coma and generalized seizures, Kocher was confident that the underlying cause for epileptic activity was localized or generalized impairment of the capillary blood supply. A circumscribed or generalized increase in intracranial pressure (ICP) would therefore trigger seizures by compression of the capillaries. Kocher (1893) argued that altered capillary beds within structural lesions were particularly vulnerable to the influence of ICP fluctuations, leading to seizures originating in these areas.

Next to mass lesions causing a permanent ICP increase, Kocher particularly pointed out ICP fluctuations following changes of the cerebral blood volume (CBV) as demonstrated by Geigel (1890). Therefore, he concluded that strong emotions, physical effort, or changes in body position could lead to seizures by increased arterial or venous CBV and ICP. Within this context, Kocher (1893) mentioned the clinical observation of a higher incidence of seizures during sleep and referred to Hitzig, who measured the CSF pressure in former trepanation defects, showing that changes in mood and even mental activity would have an influence on the ICP.

Efforts to Find Evidence in Support of Kocher’s Theory

At Kocher’s suggestion, the Japanese guest assistant Ito carried out experiments with guinea pigs, in which he induced epileptic seizures by repeatedly tapping their skullcap after measuring the initial CSF pressure. Once the guinea pigs displayed convulsions, an increased ICP could be measured. Subsequent craniectomy disclosed a central hyperemia. Postoperatively, a significant decrease in seizure frequency and a verifiably reduced ICP was observed. Ito concluded that the tapping-induced functional hyperemia of the central nervous system (CNS) led to an increase in ICP, which finally caused the epilepsy (Ito, 1899). Because no structural brain damage was found in the epileptic guinea pigs, Kocher felt vindicated in his conviction of a key role for elevated ICP in the genesis of epilepsy.

Berezowsky, a Russian guest assistant, was involved with follow-up examinations of 29 patients who had been operated on for epilepsy at Kocher’s institute in Berne. In cured epileptics, he discovered that the trepanation defect had stayed open and the scalp above the defect could sink in or bulge out; therefore, acting like a valve for balancing out ICP fluctuations. In patients with recurrent seizures, however, the skull opening was entirely or partially closed through bone healing. Berezowsky pointed out that only one of 18 patients with remaining trepanation defect following surgery for severe head injury developed seizures postoperatively, even though cortical scars were not consequently resected in the series. In contrast, less severe cases of head trauma that had not been operated on and therefore not provided with such a valve, resulted in epilepsy. In experiments with animals, Berezowsky examined the bone healing of skull deficits. He identified a pivotal role of the dura with regard to the degree of the osseous defect recovery. Animals with an intact or readapted dura displayed new bone formation on its surface, and the trepanation defect subsequently closed. If the dura was excised, however, new bone formation did not occur (Berezowsky, 1899).

In 1899, Kocher analyzed a total of 175 surgeries for focal epilepsy by adding his own operated cases to the existing literature (Matthiolius, 1899). He postulated that ablation of the causative lesion offered the most favorable prognosis with regard to becoming free of seizures (48–89%). Furthermore, he examined the results of trepanations without intraoperative identification of a structural abnormality. Kocher compared cases where the dura was opened intraoperatively to those where it was not. He noticed that the opening of the dura was a decisive factor in surgical success: 55% of the cases with intraoperative opening of the dura remained seizure free, whereas this was achieved in only 14% of patients with preserved dural integrity (Kocher, 1899).

The Resulting Operative Strategy in Epilepsy Surgery

Based on his aforementioned observations, Kocher recommended the normalization of an assumed elevation in ICP as a general strategy for epilepsy surgery. In a speech at the Congress of the German Society for Surgery in 1899, he suggested improving pressure conditions with the help of a safety valve in addition to removing the anatomic alteration—if such an alteration was detected and its excision was possible. For this purpose, cranial bone flaps or the bone fragments of impression fractures were not to be reinserted before closure of the scalp. Furthermore, the dura beneath the osseous lesion was removed as far as possible (Fig. 3) according to Berezowsky’s experiments in order to prevent the trepanation defect from healing. This way a valve was created balancing out ICP fluctuations. Concerning the prevention of posttraumatic seizures, Kocher (1899) warned the audience: “With regard to the utmost important prophylaxis of epilepsy in the future, we have to acknowledge that it is normally not the opening of the injured skull but its closure that causes serious damage….”

Figure 3.

The dura beneath the skull opening was removed as far as possible to prevent the trepanation defect from healing. From (Berezowsky, 1899).

In cases of focal epilepsy and intraoperative findings of a causative porencephalic cyst, Kocher inserted a silver cannula into the cyst to achieve a local ICP decrease (Kocher & de Quervain, 1901). This cannula was frequently kept open for months with meticulous wound hygiene. Convinced by the paramount importance of an ICP increase in the genesis of epilepsy, Kocher went as far as proposing an operative valve even in case of recurrent general convulsions. He carried out his first five valve surgeries to prove the theory between 1896 and 1898 with persuasive results. Three of the surgeries led to seizure freedom; one intervention significantly reduced the seizure frequency. Kocher, however, refrained from extensive craniectomies in patients with epilepsy, because he was afraid that the concomitant disturbance of cerebral blood flow would cause a relapse of seizures. In three of the aforementioned cases, he therefore carried out a large craniotomy and downsized and reinserted the cranial bone flap in the second step of the surgery (Fig. 4).

Figure 4.

Operative valve in case of primary generalized epilepsy: a large craniotomy was carried out with subsequent downsizing and reinsertion of the bone flap. Additional dural removal was performed to prevent the osseous defect from healing. From (Berezowsky, 1899).

Kocher was well aware that the definition of success of epilepsy surgery was difficult and that it was practically impossible to compare operation outcomes, as indicated by an article of one of Kocher’s doctorate students (Schär, 1899). Berezowsky’s interpretation of Kocher’s surgery statistics was that the results of epilepsy treatment had significantly improved since the new decompressive method had been introduced. However, Kocher also desisted from clearly defining success despite his own distinct criticism (Kocher, 1893; Berezowsky, 1899).

Failure of the Valve Surgery for Epilepsy

It was rather Kocher’s authority and his plausible explanations than the actual results that caused a number of renowned German surgeons such as von Bergmann, Gussenbauer, Kümmell, and Lauenstein to adopt Kocher’s valve surgery for recurrent general convulsions (Tröhler, 1984). Fedor Krause (1852–1937), however, the founder of neurosurgery as a distinct subdiscipline in Germany, started carrying out valve surgeries only after positive results had been published by some of his colleagues. Kocher’s theory was already confronted with growing doubts at that time (Krause, 1910). In 1901, August Bier demonstrated that an increased ICP was rather a concomitant of epileptic seizures than an immediate cause by experimentally increasing the brain pressure in patients (Bier, 1900). However, Kocher’s critical mind, his ambition to reach a pathophysiologic understanding of epilepsy, and his rigorous implementation of experimental results into therapeutic practice were highly appreciated by his contemporaries in the field of neuroscience. For this reason, Kocher was recruited as editorial assistant of the new journal Epilepsia in 1908, 1 year before he received the Nobel Prize in Physiology or Medicine for his work on the thyroid gland (Shorvon et al., 2009).

In 1910, at the 39th Congress of the German Society for Surgery, Kocher was confronted with disillusioning postoperative results after valve surgery presented by Krause and others. He responded to the audience’s criticism of his epilepsy theory as followed: “It was never my intention to suggest the theory that a valve will cure everything, or that an increased pressure explains epilepsy. I only wanted to point out one factor we might have to consider … I want to emphasise the fact that some cases were cured thanks to a valve only. I don’t put much value to the theory of positive pressure; but I don’t want that … it prevents you from examining, in appropriate cases, if an augmented pressure exists in order to reduce it, if the more simple actions cannot be taken, i.e., removing a focus in the cortex such as a scar, a foreign object, etc.” (Kocher, 1910).

Until his death in 1917, Kocher remained convinced that his pathophysiologic theory was correct. Although he abandoned his practice of valve surgeries on recurrent general convulsions, he still decidedly took the view “…that in the case of every epilepsy, in which its origin can be localized to at least some extent, operative treatment is recommended from the point of view that even in those cases, in which the focus cannot be removed by surgery, trepanation will locally release pressure and have a soothing effect” (Kocher, 1916).


Vascular theories of various types were discussed throughout the 19th century to explain the origin of epilepsy. They were only abandoned in the 1930s, when the modern electromagnetic theories of epilepsy became widely accepted as a result of Hans Berger’s (1873–1941) discovery of the human electroencephalography (EEG) in the 1920s (Reynolds & Trimble, 2009). Among the advocates of a vascular origin of epilepsy were prominent figures in the world of epileptology such as Jackson (1870, 1873) and William Aldren Turner (1864–1945) (Turner, 1907). Wilder Penfield (1891–1976) was also considering a vascular mechanism in epilepsy (Penfield, 1933). However, Kocher’s thesis of increased ICP causing epilepsy by compression of the capillary bed is, to the author’s knowledge, unique.

Although valve surgery was abandoned when laboratory investigations failed to demonstrate that intracranial hypertension produced seizures and the results of surgery proved unsatisfactory, Kocher’s continued work on the physiology and pathophysiology of ICP had a major impact on the emerging discipline of neurologic surgery. He strongly influenced the neurosurgical career of Harvey Cushing (1869–1939), who spent half a year at the University of Berne in 1900. Kocher inspired Cushing’s experimental studies on the systemic effect of elevated ICP that finally led to the discovery of the vasopressor response (Fodstad et al., 2006).


None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with these guidelines.