Elsevier

Pediatric Neurology

Volume 32, Issue 5, May 2005, Pages 295-299
Pediatric Neurology

Review article
Rasmussen’s Syndrome: Progressive Autoimmune Multi-Focal Encephalopathy

https://doi.org/10.1016/j.pediatrneurol.2004.12.002Get rights and content

Rasmussen’s encephalitis, originally thought to be a chronic form of viral encephalitis, is now thought to be an autoimmune disease of the brain and is more properly termed Rasmussen’s syndrome. Starting in one area of one side of the brain, the disease appears to gradually and progressively involve that side of the brain causing progressive and intractable focal seizures, a hemiparesis, and expressive aphasia when the left hemisphere is involved. Immune therapy with steroids, immunoglobulins, or plasmaphoresis provide only temporary relief from seizures. Neither antibodies to Glu-R3 nor cortical biopsy are helpful in the diagnosis. Hemispherectomy of one form or another is the only curative therapy, and there is no evidence that one form of hemispherectomy is preferable to another. Immuno-ablative therapy may be a therapy of the future.

Introduction

Focal seizures due to “chronic encephalitis” were first reported by Rasmussen and his team in 1958 [1]. Twenty years later, Rasmussen reported 27 cases [2]; most children appeared to have a syndrome which we would now term Rasmussen’s syndrome. This syndrome occurs primarily in younger children and is characterized by intractable seizures, often with epilepsia partialis continua; a slowly progressive hemiparesis; hemianopia; progressive mental retardation and progressive cortical atrophy. Originally treated by resection of the affected area, Rasmussen stated that, “Focal cortical resection early in the course of the illness seldom reduced the seizure tendency significantly, unless it was practicable to carry out extensive excision such as a complete or subtotal hemispherectomy.” He further hoped that the etiology could be identified “… that treatment could [be instituted] to prevent the development of seizures, mental retardation, hemiparesis in these patients doomed to be severely handicapped.”

In the ensuing almost one-half a century, major progress has been made in understanding this then baffling disease, and we are perhaps on the verge of fulfilling Rasmussen’s desire to prevent the progressive nature of this disorder and the disability associated with the disease and its treatment. It is the purpose of this review to note where we have come from, where we are now, and where we are, perhaps, going.

“Chronic encephalitis” was initially identified by Rasmussen in children with medically refractory, focal epilepsy. The histology of the focal resections revealed, “active encephalitis,  perivascular cuffing with lymphocytes and glial nodules scattered throughout the grey and white matter.” “Laminar necrosis lead to spongy degeneration and ultimately to cortical destruction and gliosis.” [2]. Standard viral studies were negative, as were studies for slow and latent viruses. In 1988 Piatt stated that 36 patients with chronic focal encephalitis had been reported, and that this picture constituted 7.4% of the focal cortical resections performed at the Hospital for Sick Children in Toronto [3].

Although the diagnosis of Rasmussen’s syndrome was initially based on its pathologic picture, the acute childhood onset of difficult-to-control, focal seizures which gradually spread though the adjacent cortex has become suggestive of its classic clinical progression. The advent of computed tomography and later magnetic resonance imaging scans made it possible to exclude tumors or vascular abnormalities which might have mimicked the disease. Scans also demonstrated the atrophy and gliosis often associated with its early phases. The diagnosis of Rasmussen’s syndrome can now almost be made over the telephone, once vascular lesions and tumors have been ruled out. Brain biopsy is rarely indicated to make the diagnosis, may prove to be misleading (see below), and indeed may be contraindicated.

The seizures associated with this condition are characteristic, and for reasons unknown, are focal and remain focal or multi-focal, and unilateral. Epilepsy partialis continua is common at some stage of the disease, but not universal. Even with spread of seizures throughout one side of the body, the seizures are not Jacksonian in nature, but remain focal, in one muscle group, then slow down and stop. They may then pick up again as the focal seizures progress to involve an adjacent muscle group. The seizures of Rasmussen’s virtually never spread either to become diffuse unilateral or generalized seizures [4], [5], [6].

The seizures associated with Rasmussen’s syndrome are among the most intractable of all seizure types. While medications may at times result in temporary decrease in the seizures, patients inevitably become refractory to each. Medication only is thus but a temporizing measure.

Immunomodulatory therapies with steroids, plasmaphoresis, and immunoglobulins often provided transient improvement of the seizures and to some degree of the other symptoms, but relapse is inevitable [7], [8]. These therapies remain only a delaying tactic before the inevitable hemispherectomy. Immuno-ablative therapy, however, may ultimately prove more useful in halting this autoimmune disease.

Lobar and even multi-lobar resections, as originally demonstrated by Rasmussen, are ineffective in controlling either the seizures or the progressive nature of the disease. The condition inevitably recurs in other locations within the same hemisphere. The only treatment which has been demonstrated to be effective in children with Rasmussen’s syndrome is hemispherectomy [5].

In earlier eras, neurosurgeons were reluctant to perform hemispherectomy before completion of the hemiparesis by the disease itself, afraid of being accused of creating the otherwise inevitable hemiparesis Although the disease and its seizures may begin in different areas of the brain, early motor involvement is common and the disease inevitably involves the ipsilateral motor cortex, resulting in a contralateral hemiparesis.

The Johns Hopkins group was an advocate of early surgery [5], [9], [10] to avoid the handicap due to the continued seizures and the medication. Gradually, earlier surgery has become accepted by others [11]. We define “early” surgery as occurring when the family has come to recognize the inevitable progressive nature of the disease, and when the older, informed child and parent ask us for the surgery. The preexisting degree of hemiparesis, and the laterality of the condition are not determining factors in the decision. We have performed successful left hemispherectomies in individuals as old as 14 years. They have ultimately gone on to college.

The concept of removing one half of an individual’s brain is daunting to the families and older children, and to the physicians and surgeons involved [8], although once the diagnosis of Rasmussen’s syndrome is made, hemispherectomy remains the only current solution. Serum and spinal fluid assays for Glu-R3 antibodies are not needed and indeed may be misleading in either a positive or negative fashion. Glu-R3 may be absent in Rasmussen’s disease, and positive in other forms of intense seizures [12, personal observations].

Diagnosis of Rasmussen’s syndrome is made clinically by the progressive nature of the focal seizures in a child, and the absence of a structural or vascular lesion to account for them. We review the appropriate studies including the electroencephalographic and magnetic resonance imaging scans. Focal atrophy on scans may be present even at the time of the initial visit, but is not required in the early stages of the disease [9], [10].

Our approach to the decision-making process regarding surgery involves personally examining the patient and talking with the family and the child. Once the diagnosis is suspected, and confirmed by the individual’s course, discussions should be initiated about the inevitable hemispherectomy. This discussion requires an experienced team, familiar with children with hemispherectomies and their families; familiar with their fears, concerns, and outcomes. We introduce prospective patients to families and children who have previously undergone the procedure. Ideally, these patients are closely matched by sex, age, and locale so that they can meet each other and physically see what a child is like after removal of one half of the brain. Those families often become the support system for those deciding about the surgery.

The surgery is never done as an emergency, but, as the patients have been told on their initial visit, it will be performed only when the family has come to the conclusion that hemiparesis from the disease is inevitable; that there is no other cure for the constant seizures and the handicap they impose. The family must fully understand that medications and treatments are only temporizing procedures, and that life with a permanent hemiparesis brought about by the surgery is better than life with the ongoing seizures because the hemiparesis is inevitable. It is only when the disease involves the left hemisphere, and when the child is approaching the teen-age years that we actively encourage earlier surgery. The family, both mother and father, must ask us to perform the procedure, and the older child must acquiesce.

Initially, hemispherectomy was performed by the complete removal of the cortex and white matter, sparing the basal ganglia. Because of the late complication of what was termed “superficial cortical hemosiderosis” (see below), “functional hemispherectomies” were devised, removing the motor and parietal cortex and leaving behind the disconnected frontal and occipital cortex. This procedure was thought to avoid the shifting and micro-bleeding [4] thought to result in the hemosiderosis. Hemi-decorticetomy, removing the gray matter but leaving white matter covering of the ventricles, has been the method of choice at our institution [13]. Others have devised “keyhole” procedures [14], removing a portion of the sylvian cortex and disconnecting the hemispheric interconnections. It would appear that the outcome after hemispherectomy depends more on the experience of the surgeon and the team, than on the type of the surgery. Although different surgeons advocate and practice these different techniques, there are no convincing studies that one is better than another in eliminating the seizures or in avoiding either perioperative or long-term complications [9], [10], [11], [15].

The reports of outcomes of hemispherectomy procedures often combine children with unilateral developmental abnormalities or with vascular abnormalities with those who have Rasmussen’s disease. Therefore it is often difficult to sort outcomes, one from another. However, outcomes of the hemispherectomy depend on the etiology of the cortical dysfunction [10], [11], [16], [17]. Children with Rasmussen’s syndrome appear to have normal brains before the onset of the disease, and therefore have better outcomes when the diseased hemisphere is removed. Of 111 hemispherectomies performed at Hopkins [17], 46 were for Rasmussen’s, 65% of these became seizure-free after surgery, and most were off medications. Most of those with residual seizures had mild episodes which no longer interfered with their quality of life. Only four had significant residual seizures as a result of tissue inadvertently left behind. There was no mortality and no significant morbidity in those initially operated at our institution.

After hemispherectomy for Rasmussen’s, all children will walk, and most ultimately walk and run without bracing, but with a mild, but noticeable limp. None recover a useful arm, although most can use the paretic arm and hand as a helper. Although most have gross sensation, the lack of two-point discrimination, asterognosis, and fine motor control of the fingers limit the hand’s usefulness. All children have a persistent hemianopia, but they appear to accommodate to this with time, and few show evidence of neglect.

Twenty-one of the 46 patients with Rasmussen’s syndrome had left hemispherectomies, 8 after the age of 10 years. At follow-up all are talking, although 5 of the 21 have mild dysfluency and 2 have difficulty with naming and reading. Two operated as teen-agers have gone to college [16].

Although intraoperative bleeding occurs during hemispherectomy, and can be a problem, this complication, in our experience, occurs most commonly in young children with cortical dysgenesis. Although used as the reason for the development of “keyhole” surgery [14], severe bleeding appears to occur rarely in children undergoing hemispherectomy for Rasmussen’s disease.

Hydrocephalus occurs in approximately 25% of children postoperatively [18]. Children are often asymptomatic, and the ventricular enlargement is observed on computed tomographic scans routinely done 2–3 weeks postoperatively. Rarely symptoms appear in the weeks or months after surgery. It is thought that the hydrocephalus is the result of the red cells and protein clogging the pacchionian granulations. As the cerebrospinal fluid clears, the shunts may no longer be needed. Rarely, patients may have later symptoms of shunt failure, but this should always be kept in mind in children who have been shunted.

Superficial cortical hemosiderosis was thought to be due to small bleeding with shifts and minor trauma to the empty side, resulting in hydrocephalus and shifting of the brain [19]. It was considered a late complication of total hemispherectomy occurring 10–20 years after the original surgery. Causing death if not shunted emergently, it was the reason for the development of the functional procedures discussed above. Whatever its pathogenesis, it appears to have disappeared as a complication regardless of the procedure used since the advent of computed tomographic and magnetic resonance imaging scanning, which have allowed the detection and early treatment of the hydrocephalus.

In normal individuals, the left hemisphere is usually dominant for speech. When Rasmussen’s syndrome involves the left hemisphere, usually after speech dominance has been established, speech is progressively and inevitably involved. Although age and the severity of the seizure disorder makes the preoperative performance of a Wada test difficult or impossible, we have documented that children, presumably left hemisphere speech dominant before surgery, will lose word-finding skills after left side surgery, but retain word understanding [16], [20], [21], [22]. The rate of recovery of word-finding skills seems partially proportional to age at onset of the disease and its duration. The speed and quality of language recovery appear primarily related to the age at which the child undergoes left hemispherectomy [16], [20], [22]. Even in the early teen-age period, speech will recover sufficiently for the individual to attend and graduate from college and appears to be impaired by the continued presence of seizures. The upper age limit at which speech will transfer is not known, but we have observed young teen-agers reacquire speech after left hemispherectomy. We strongly believe that left hemispherectomy at an earlier age results in better and earlier speech recovery.

The cognitive outcome of 37 children with Rasmussen’s syndrome was assessed a mean of 5.5 years after surgery (Table 1) [16]. Seventy-three percent of the children were seizure-free at the time of follow-up, and an additional 22% had rare, nonhandicapping mild seizures. Seventy percent were mainstreamed in school, half with supportive services. Of the 31 children tested both preoperatively and postoperatively, there were no statistically significant changes in cognitive function after surgery, regardless of the side removed. Hemispherectomy did not impose a significant additional burden to the child, and in the family’s judgment had improved the child’s quality of life. In view of the motor dysfunction and visual impairment, the author does not believe that the intelligence quotient represents the true function of these children.

Chronic focal encephalitis was the name originally given to what is now called Rasmussen’s syndrome because the perivascular cuffing with lymphocytes and the gliosis were suggestive of a chronic viral process. No virus has been consistently related to the condition [23]. In a comprehensive pathologic evaluation of tissue from 45 Hopkins hemispherectomy patients with Rasmussen syndrome [24], the pathology was found to be both multi-focal and progressive. The progression appeared to be immune-mediated involving both neuroglial and T-lymphocytes. T-lymphocytes appear to initiate the process and ultimately produce neuronal injury. This observation suggests that the autoimmune process is cellular mediated rather than humoral, and tends to confirm the lack of a role for Glu-R3 in the pathogenesis [24], [25]. These pathologic studies emphasize three points: (1) that biopsy in noneloquent cortex can be misleading, because it may readily miss the signature pathology. It is also unnecessary since the clinical picture and progression are sufficiently unique to recognize the disease. (2) Age of onset significantly influences the burden of pathology. (3) The duration of the disease also influences the burden of pathology.

Thus, early and aggressive therapy appears indicated. However, because the only effective therapy is hemispherectomy, this is clearly constrained. As discussed previously, immunotherapy with steroids, plasmaphoresis, and intravenous immunoglobulin provide temporary relief to many, but always relapse. Immune-ablation therapy, in which the individual’s immune cells are wiped out with cyclophasphamide and the immune system is reconstituted by the child’s own cyclophasphamide-resistant progenitor cells, has been reported to more permanently alter the immune system and has been used in severe myasthenia gravis [26]. As of this writing (9/2004), immuno-ablative therapy has been used in two children at our institution with Rasmussen’s syndrome, with promising results. One severely affected child had a dramatic response lasting 5 months, then with recurrence of seizures underwent a hemispherectomy at the parents’ request. The second child has had no progression of his disease in 3 years. We consider these results to be preliminary and the procedure to be highly experimental.

Section snippets

Summary

Rasmussen’s syndrome is now known to be a severe autoimmune process leading to destruction of a single hemisphere. For unclear reasons, it never affects the other hemisphere. Although immunomodulatory therapy provides transient relief for some, the only definitive therapy is hemispherectomy in one of its varied forms. However, a therapy that could be used early in the condition, that is less invasive than hemispherectomy, and that will completely halt this progressive disease is eagerly

References (26)

  • E.P.G. Vining et al.

    Progressive unilateral encephalopathy of childhood (Rasmussen’s syndrome)A reappraisal

    Epilepsia

    (1993)
  • E.P.G. Vining et al.

    Why would your remove half a brain? The outcome of 58 children after hemispherectomy

    Pediatrics

    (1997)
  • R. Jonas et al.

    Cerebral hemispherectomyHospital course, seizure, developmental, language, and motor outcomes

    Neurology

    (2004)
  • Cited by (54)

    • Diagnosis and treatment of Rasmussen's encephalitis pose a big challenge: Two case reports and literature review

      2021, Annals of Medicine and Surgery
      Citation Excerpt :

      RES now is termed as Rasmussen's encephalitis (RE). The disease typically begins in childhood with refractory focal motor seizures which gradually evolve into epilepsia partialis continua (EPC) which is followed by progressive hemiparesis and mental deterioration [5,6]. To date, there is no definitive consensus on treatment, with proposed strategies ranging from acute and chronic immunotherapy to hemispherectomy.

    • Principles and Practice of Movement Disorders

      2021, Principles and Practice of Movement Disorders
    • Diagnostic Pathology: Neuropathology

      2017, Diagnostic Pathology: Neuropathology
    • Autoimmune Epilepsies

      2017, Seminars in Pediatric Neurology
      Citation Excerpt :

      Similar guidelines have been published for use in children.34 Rasmussen’s Encephalopathy is a neuroinflammatory disease affecting one hemisphere of the brain from which children progressively develop intractable seizures, hemiparesis, and cognitive deterioration over months to years with eventual severe impairment of quality of life.35–37 Seizures typically begin as focal or unilateral but subsequently evolve into multifocal seizures, epilepsia partialis continua, or more rarely, bilateral convulsive seizures.35,37,38

    View all citing articles on Scopus
    View full text