Motor and Cognitive Outcomes in Children After Functional Hemispherectomy

Samargia, Sharyl A. MA, CCC-SLP; Kimberley, Teresa Jacobson PT, PhD

Pediatric Physical Therapy:
doi: 10.1097/PEP.0b013e3181bf710d
Review Article

Purpose: Medically intractable epilepsy is a chronic recurrence of seizures that often requires surgery to reduce or eliminate them. Although a reduction of seizures is the primary goal of hemispherectomy, the effect of surgery on motor and cognitive skills is also of importance. This review will provide a discussion of (1) evidence regarding motor and cognitive outcomes, (2) predictors of these outcomes, and (3) neural mechanisms responsible for preservation of function after hemispherectomy.

Summary of Key Points: Motor and cognitive outcomes after hemispherectomy are variable and depend on many predictors including etiology and duration of seizure disorder, age at the time of surgery, premorbid status, and postsurgical seizure control. A refined ipsilateral pathway may explain the preservation of motor function in some children.

Conclusions: A clear understanding of outcome predictors is important for planning effective rehabilitative programs after surgery.

In Brief

This review of outcomes following hemispherectomy enabled the authors to identify predictors of outcomes important for planning treatment following surgery. They also propose that ipsilateral pathways may explain preservation of motor and cognitive skills following hemispherectomy.

Author Information

Department of Communicative Disorders (S.A.S.), University of Wisconsin, River Falls, Wisconsin; and Program in Rehabilitation Science (S.A.S., T.J.K.) University of Minnesota, Minneapolis, Minnesota

Address Correspondence to: Sharyl Samargia, MA, CCC-SLP, Department of Communicative Disorders, University of Wisconsin-River Falls, 410 South Third Street, River Falls, WI 54022. E-mail:

Sharyl Samargia was a doctoral student at the University of Minnesota, Program in Rehabilitation Science, when this report was prepared.

Article Outline
Back to Top | Article Outline


Seizures are defined as sudden, uncontrolled episodes of excessive electrical neuronal discharges, which may cause differences in behavior, movement, or sensation. Raimorch1 reported that epilepsy, a chronic recurrence of seizures, has a prevalence of approximately 1% in the general population. Of that 1%, approximately 2% to 3% of individuals have a diagnosis of medically intractable epilepsy (MIE), seizures that are unable to be controlled pharmacologically or nutritionally. Seizures cause metabolic changes in the brain, including an increase of 2 to 3 times the cerebral metabolic rate of oxygen consumption and cerebral metabolic rate of glucose, as well as an increase in blood flow at the seizure site. These metabolic phenomena can take 24 hours or longer to return to baseline postictally.2 Brain changes secondary to seizure activity can include neuronal cell loss in the neocortex, hippocampus, and cerebellum. Other sequelae include epileptic necrosis and the degeneration of neurons both focal and distal to the site of seizure onset.2 Holmes2 cited the study by Wasterlain and Plum, who compared the behavior of epileptic rats with a control group. They concluded that frequent, brief seizures in pubescent animals resulted in significant detrimental differences in learning, memory, activity level, and behavior. They also found that the younger the animal is at the time of seizure onset, the more detrimental the seizures were to the development of neuronal connections. This phenomenon has also been observed in humans. Children with a history of infantile spasms without effective seizure management have a poorer developmental outcome than those children with a later onset of seizures.3,4

Eliminating or even reducing seizures may not be able to be accomplished pharmacologically. Epilepsy surgery is often the only viable option for greatly reducing or eliminating seizures in children with MIE. Functional hemispherectomy has been used as a treatment option for children with MIE since the 1970s and involves removal of the affected temporal and parietal lobes and disconnection of the affected frontal and occipital lobes from the unaffected hemisphere. In the past, success of this procedure was measured primarily in terms of seizure control and there has been a high success rate of minimizing and even eliminating seizures in many infants and children.6–10 Historically, functional outcomes after surgery have not been fully examined. Only recently have researchers begun to quantify functional outcomes in addition to seizure reduction after hemispherectomy to determine whether children demonstrated long-term deficits in the areas of motor skills, cognition, speech, and language. The question of importance to rehabilitation providers is if seizure control is attained, will a child who has undergone a disconnection of the cerebral hemispheres have better functional outcomes than a child with continued seizure activity but both hemispheres connected? Information regarding functional outcomes after functional hemispherectomy will assist in postsurgical planning. The purpose of this review is to discuss the literature regarding (1) motor and cognitive outcomes of children who have undergone hemispherectomy, (2) predictors of these outcomes, and (3) neural mechanisms responsible for preservation of function in some children.

Back to Top | Article Outline


Study Identification and Selection

Electronic databases (PubMed, MEDLINE, and CINAHL) were searched using the keywords: hemispherectomy in children and were combined with additional keywords including motor skills, motor outcomes, cognitive skills, and cognitive outcomes. The search was limited to articles written in English and those published between 1988 and 2008. The reference lists of the identified articles were also reviewed resulting in the location of additional relevant studies. Articles that were chosen for review met the following inclusion criteria: (1) pediatric population, (2) status post hemispherectomy, and (3) measure of motor and/or cognitive outcomes (Table 1). Some articles provided a discussion and comparison of outcomes after multiple types of epilepsy surgical procedures. In these cases, the data from the hemispherectomy population were extracted.

Back to Top | Article Outline
Motor Outcomes after Hemispherectomy.

Several researchers have investigated motor outcomes in children after hemispherectomy (Table 2). The authors who measured preoperative and postoperative motor skills reported motor impairment in the majority of children before surgery, which was typically characterized by decreased strength, increased muscle tone, and decreased active range of motion contralateral to the affected hemisphere. The primary motor outcomes reported postoperatively included severity of hemiparesis, muscle tone, and ambulation status using a variety of manual tests (Table 2).8–11 In only one of the studies reviewed were children with MIE who underwent hemispherectomy compared with a control group.4 Results from studies that reported preoperative and postoperative motor skills are summarized in Table 3.

Hemiparesis remained unchanged postoperatively in approximately half of the children. Of the children who presented with more significant hemiparesis postoperatively, the upper extremity (UE), distal more than proximal, contralateral to the disconnected hemisphere was more affected than the lower extremity.3,5,8,9,11

In the 2 studies that investigated ambulation, all children who were ambulatory before surgery remained ambulatory after surgery.5,8 The authors of one study reported that of 11 children who were not ambulatory before surgery, 4 were able to walk postoperatively.8 However, no timeline was given related to when ambulation returned nor was the type of ambulation specified.

In 2 studies,9,10 increased muscle tone in the proximal and distal arm was observed in all children before surgery. During the first year postoperatively, muscle tone increased in the UE contralateral to the affected hemisphere, distally more than proximally. This increase was not statistically significant. However, in one study at 24 months postsurgery, there was a statistically significant increase in muscle tone observed distally in the UE.9 Importantly, these children demonstrated significantly higher scores across all domains of the Pediatric Evaluation of Disability Inventory between 12 and 24 months postoperatively compared with their preoperative Pediatric Evaluation of Disability Inventory scores.9 Thus, although their muscle tone had increased, some areas of functional skills had improved significantly over time.

Back to Top | Article Outline
Cognitive Outcomes after Hemispherectomy.

It might be hypothesized that children with improved seizure control who require less antiseizure medication could be more alert and therefore able to attend, remember, and respond to various motor tasks required for self-care. The evidence to date suggests no clear answer to that question. Unfortunately, a wide variety of measures were used to report cognitive outcomes; thus, it is difficult to compare outcomes across studies (Table 4). However, for the purpose of this review, improved cognitive performance on any measure was considered and reported here. Studies that reported preoperative and postoperative cognitive skills are summarized in Table 5. In terms of number of children, approximately half demonstrated no change in cognitive skills. The remaining children were fairly evenly divided between those who improved and those who declined in cognition postoperatively.4–8,11–13

The results of the majority of the studies that included measures of cognitive outcomes were solely based on intelligence quotient (IQ) scores; however, in 2 studies, differences in specific areas of cognition were measured, including IQ, visual-spatial skills, perceptual-motor skills, visual attention, verbal fluency, adaptive functioning, language, and behavior in children after hemispherectomy.12,13 In one study, the majority of children demonstrated stronger performance IQ, when compared with verbal IQ.12 The other areas of reported deficit included visual spatial skills, attention, verbal fluency, and visual memory skills. Although this descriptive information is beneficial, reports of quantitative outcomes regarding specific areas of cognition such as attention, memory, and executive functioning are needed to better understand the cognitive outcomes of this population.

Back to Top | Article Outline
Predictors of Motor and Cognitive Outcomes.

Children with MIE are not a homogeneous group, and there are many variables that could affect postoperative motor and cognitive outcomes.

Back to Top | Article Outline
Seizure Control

Seizure control was positively correlated with improved motor and cognitive outcomes after surgery, but the significance of the results was variable.6,8,13–15 Postoperative cognitive outcomes were significantly better in the children who were seizure free, when compared with children with residual seizures, which emphasizes the importance of seizure control to development of functional skills.10,11 Maehara et al8 found that 5 of the 6 children in the seizure-free group showed qualitatively marked improvement in their developmental quotient (DQ) over an average period of 47 months after surgery. Six of the 8 subjects with residual seizures showed no significant gains or improvement in DQ over time and 2 with cortical dysplasia demonstrated a decline in DQ over time. These findings suggest that continued seizures “interrupt” development, causing the gap between chronological age and developmental age to increase over time.

Back to Top | Article Outline

Authors who compared motor and cognitive outcomes based on etiology reported variable results. One study included a comparison of motor outcomes after hemispherectomy among 4 groups using the Vineland Adaptive Behavior Scale. Postoperatively, 66% of the hemimegalencephaly group had worsening hemiparesis. Between 40% and 50% of the cortical dysplasia group demonstrated improvement in hemiparesis. Groups with cerebral infarction and Rasmussen encephalitis also demonstrated an improvement in hemiparesis postoperatively.6 Another study included a comparison of the differences in postoperative motor skills between children with perinatal stroke, Rasmussen encephalitis, and cortical dysplasia. Based on the scores from the 74-point Fugl-Meyer Assessment of Motor Recovery Scale, the perinatal stroke group scored significantly better than the other 2 groups in overall motor skills, arm function, and hand and wrist function.3 This evidence conflicts with the findings reported by Jonas et al,6 who suggested that children with cortical dysplasia had similar outcomes as those with cerebral infarction. No significant differences in motor skills were reported between the cortical dysplasia and Rasmussen encephalitis groups.3 Another author used a “burden of illness score,” which was based on motor disability, intellectual disability, and seizure outcome and found the Rasmussen encephalitis group to have a decline in motor skills postoperatively. In comparison, more than half the group with a diagnosis of cortical dysplasia had improved motor outcomes postoperatively.11

Analysis of cognitive outcomes based on etiology indicated that children with cortical dysplasia had significantly poorer postoperative cognitive skills than children with Rasmussen encephalitis.13 However, it should be noted that the majority of children with cortical dysplasia had a seizure onset before 1 year of age and had residual seizures postoperatively, which could have had a negative effect on cognitive outcomes. The vascular group showed the greatest cognitive improvement and was reportedly seizure free after surgery.13

Overall, the only consistent finding was that the children with vascular etiology seem to have more favorable motor and cognitive outcomes after hemispherectomy than children with cortical dysplasia or Rasmussen encephalitis. This suggests that acquired etiologies may be more indicative of better motor and cognitive outcomes than congenital etiologies.4,5

Back to Top | Article Outline
Postsurgical Duration Period

According to the authors of one study, the duration of the postsurgical period was the only significant predictor of motor outcomes after hemispherectomy. These authors did not find a significant correlation between duration of seizure disorder, etiology of seizure disorder, type of surgery, or Engel classification of seizure severity and motor outcomes as measured by the Movement Assessment Battery for Children or the Gross Motor Function Measure.10 Using an analysis of variance for repeated measures, vanEmpelen et al10 analyzed children’s motor skills at 6, 12, and 24 months postoperatively and found scores in all domains of the Gross Motor Function Measure significantly increased by 24 months after surgery. However, distal UE strength and tone remained poorer than baseline 2 years postoperatively. This information indicates improvements in motor skills occur over time, which has important implications in clinical management of these patients. Improvements in motor skills can be observed up to 2 years after surgery, with the exception of the distal portion of the contralateral UE, which may not return to baseline postoperatively.

Back to Top | Article Outline
Duration of Seizure Disorder, Presurgical Development, and Age at the Time of Surgery

A shorter duration of seizure disorder and higher level of overall development presurgically were associated with better postoperative outcomes in cognitive and motor skills compared with those with a longer presurgical seizure history and more significantly impaired development presurgically.5,6,8

Authors of one study reported correlations among these predictors and performance on the Skills of Independent Behavior-Revised (SIB-R) using correlation coefficients with an alpha level of 0.05. Statistically significant correlations were reported between the age of the patient at the time of surgery and the broad independence and social communication sections of the SIB-R. The authors reported that the younger the patient was at the time of surgery, the better the outcome. Also, significant correlations were reported between duration of seizure history and performance on all sections of the SIB-R; the longer the duration of seizures, the poorer the performance.5 Children who underwent hemispherectomy before normal brain maturation had more positive motor outcomes than those who underwent hemispherectomy later in life.16 Normal brain maturation was a qualitative term to distinguish those who underwent hemispherectomy during early childhood compared with those who underwent surgery in early adulthood.

Back to Top | Article Outline


Of relevance to the discussion of motor and cognitive outcomes after hemispherectomy is the mechanism of cortical reorganization after surgery. What mechanism in the brain accounts for preservation of motor and/or cognitive skills after disconnection of one hemisphere? The authors of the articles reviewed did not speculate on the mechanisms responsible for the preservation of cognitive skills after hemispherectomy beyond seizure reduction. However, evidence suggests that the use and refinement of the ipsilateral pathways may explain the preservation of motor skills in the paretic limbs.

The age at surgery and at seizure onset could influence the integrity and use of the ipsilateral pathway.17 This phenomenon could offer an explanation of motor skill preservation after hemispherectomy that is observed in some children.17–19 During development, ipsilateral pathways become inhibited by approximately 10 years of age, as cortical maturation takes place. It has been reported that ipsilateral motor-evoked potentials are not detected after the age of 10 years in children who are developing typically.16,18 However, some ipsilateral pathways can persist after cortical maturation; in fact, approximately 10% to 25% of all ascending fibers in the healthy brain are thought to be ipsilateral.17 These findings could be due to voltage spread or interhemispheric transmission that occurs in the healthy brain, but cannot account for the use of the ipsilateral pathway in children after hemispherectomy because there is a complete disconnection between hemispheres. Using functional magnetic resonance imaging in the population that has undergone hemispherectomy, direct ipsilateral pathways have been shown to be active during active and passive movements of the paretic limb.18 There are 2 possible explanations for this phenomenon: (1) ipsilateral pathways in the diseased brain are strengthened and refined due to functional demand or (2) cortical reorganization of the ipsilateral pathways by axonal sprouting occurs to develop a new functional pathway. Additionally, contralateral descending motor tracts may cross at the level of the spinal cord, which may account for bilateral lower extremity control or preservation after surgery.17–19 Evidence exists that the ipsilateral compound muscle action potentials evoked by transcranial magnetic stimulation correlate with residual motor skills. When stimulating the ipsilateral pathway, children with an early seizure onset who underwent hemispherectomy at an early age demonstrated transcranial magnetic stimulation-evoked responses that were short in latency and large in amplitude.16 In comparison, young adults who had a later seizure onset and underwent hemispherectomy at a later age demonstrated longer latency and smaller amplitude responses. This suggests that if the hemispherectomy is performed before 10 years of age, the ipsilateral pathways may not yet be inhibited, and because of the seizure activity in the opposite hemisphere, those pathways could be strengthened. Use of the ipsilateral pathways seems to be independent of etiology because ipsilateral activation has been observed in children with Rasmussen encephalitis and vascular etiologies.18

The distal-proximal gradient, which refers to the differences in proximal versus distal motor performance, can account for the preservation of proximal motor skills post hemispherectomy. Evidence suggests proximal muscles receive more bilateral cortical input than distal muscles,20 indicating the unaffected hemisphere continues to provide input to those muscle groups after unilateral lesion or damage. This supports the observable difference between the preservation of proximal versus distal motor skills after hemispherectomy.

However, the ipsilateral pathway theory does not account for children who demonstrate increased motor impairment after hemispherectomy. The only argument that may support this phenomenon is Wallerian degeneration that most likely occurs in the ipsilateral pathways after hemispherectomy, which may account for motor skill impairment postoperatively.21

Discussion of the ipsilateral mechanism related to recovery has been exclusive to recovery of motor tasks; no research included discussion of the neurological mechanism for the preservation of cognitive skills in this population. It could be speculated that the cortical region responsible for cognition was not affected by the seizures because of its dispersed network or alternatively cognition migrated to the healthy hemisphere before surgery. The reduction of seizure activity after surgery alters the need for antiepileptic medications, which can often affect cognitive skills. This change in or elimination of medication may also account for the preservation or improvement of cognitive skills in this population.

Back to Top | Article Outline


In the majority of cases, functional hemispherectomy has been shown to be an effective procedure to attain seizure reduction in some children.1,5–7,9,12,13 Reports indicate that children who were seizure free had better motor and cognitive outcomes than children with residual seizures.6,7,11 This suggests that an absence or decrease in seizure activity is a good indicator of functional improvement and may further suggest that a reduction or elimination of residual seizure activity promotes cortical reorganization possibly through the use and refinement of ipsilateral pathways. Further investigation is needed regarding the timing of inhibition of the ipsilateral pathway, what variables affect strengthening and refining the ipsilateral corticospinal pathway, and how this mechanism affects functional outcomes after hemispherectomy. This could be accomplished using neuroimaging both preoperatively and postoperatively. Functional neuroimaging techniques could also be used to gain a better understanding of the neural mechanisms responsible for preservation of cognitive skills in children after hemispherectomy. Additional clinical research may help define the nature of cognitive deficits exhibited by this population. None of the studies reviewed measured the effect of antiseizure medications on motor and cognitive outcomes in those with residual seizures, which is another variable that may contribute to the child’s functional performance.

Additional variables that can influence functional outcomes included the following: presurgical development, premorbid spasticity, postsurgical duration, age at surgery, duration, and etiology of the seizure disorder. The evidence suggests that a child undergoing hemispherectomy who has (1) acquired developmental milestones before surgery, (2) demonstrated no presurgical hemiparesis, and (3) a short seizure history may have a more positive motor and cognitive outcome after surgery.

We believe that the timing of ipsilateral pathway inhibition is also an important consideration in predicting outcomes after hemispherectomy, particularly with regard to the age of the child at surgery. An understanding of these variables and their effect on motor and cognitive outcomes can assist in creating effective rehabilitative programs maximizing neuroplasticity. Although predictors affecting postsurgical outcomes in motor and cognitive function should be considered when assessing a child for surgical candidacy, the benefits of seizure reduction or elimination far outweigh the risks of functional decline.

Comprehensive preoperative and postoperative cognitive and motor testing protocols should be clinically implemented to provide further evidence regarding specific outcomes in this population.

Back to Top | Article Outline


1. Raimorch A. Pediatric Neurosurgery. 2nd ed. New York: Springer-Verlag; 1987.
2. Holmes G. Do seizures cause brain damage? Epilepsia. 1991;32:S14–S28.
3. de Bode S, Firestine A, Mathern GW, et al. Residual motor control and cortical representation of function following hemispherectomy: effects of etiology. J Child Neurol. 2005;20:64–75.
4. Devlin A, Cross A, Harkness W, et al. Clinical outcomes of hemispherectomy for epilepsy in childhood and adolescence. Brain. 2003;126:556–566.
5. Basheer S, Connolly M, Lautzenhiser A, et al. Hemispheric surgery in children with refractory epilepsy: seizure outcome, complications and adaptive function. Epilepsia. 2007;48:133–140.
6. Jonas R, Nguyen B, Hu B, et al. Cerebral hemispherectomy: hospital course, seizure, developmental, language, and motor outcomes. Neurology. 2004;62:1713–1721.
7. Lettori D, Battaglia D, Sacco A, et al. Early hemispherectomy in catastrophic epilepsy: a neuro-cogntivie and epileptic long-term follow-up. Seizure. 2008;17:49–63.
8. Maehara T, Shimizu H, Kawai K, et al. Postoperative development of children after hemispherectomy. Brain Dev. 2002;24:155–160.
9. van Empelen R, Jennekens-Schinkel A, Buskens E, et al; Dutch Collaborative Epilepsy Surgery Programme. Functional consequences of hemispherectomy. Brain. 2004;127:2071–2079.
10. van Empelen R, Jennekens-Schinkel A, Gorter JW, et al. Epilepsy surgery does not harm motor performance of children and adolescents. Brain. 2005;128:1536–1545.
11. Vining E, Freeman J, Pillas D, et al. Why would you remove half a brain? The outcome of 58 children after hemispherectomy—the Johns Hopkins experience: 1968–1996. Pediatrics. 1997;100:163–171.
12. Battaglia D, Chieffo D, Lettori D, et al. Cognitive assessment in epilepsy surgery of children. Childs Nerv Syst. 2006;22:744–759.
13. Pulsifer M, Brandt J, Salorio C, et al. The cognitive outcome of hemispherectomy in 71 children. Epilepsia. 2004;45:243–254.
14. Tinuper P, Andermann F, Villemure J, et al. Functional hemispherectomy for treatment of epilepsy associated with hemiplegia: rationale, indications, results, and comparison with callosotomy. Ann Neurol. 1988;24:27–34.
15. Duchowny M. Hemispherectomy for epilepsy: when is one half better than two? Neurology. 2004;64:1664–1665.
16. Benecke R, Meyer BU, Freund HJ. Reorganisation of descending motor pathways in patients after hemispherectomy and severe hemispheric lesions demonstrated by magnetic brain stimulation. Exp Brain Res. 1991;83:419–426.
17. Holloway V, Gadian D, Vargha-Khadem F, et al. The reorganization of sensorimotor function in children after hemispherectomy. A functional MRI and somatosensory evoked potential study. Brain. 2000;123:2432–2444.
18. de Bode S, Mathern GW, Bookheimer S, et al. Locomotor training remodels fMRI sensorimotor cortical activations in children after cerebral hemispherectomy. Neurorehabil Neural Repair. 2007;21:497–508.
19. Graveline CJ, Mikulis DJ, Crawley AP, et al. Regionalized sensorimotor plasticity after hemispherectomy fMRI evaluation. Pediatr Neurol. 1998;19:337–342.
20. Dijkerman HC, Vargha-Khadem F, Polkey CE, et al. Ipsilesional and contralesional sensorimotor function after hemispherectomy: differences between distal and proximal function. Neuropsychologia. 2008;46:886–901.
21. Wakamoto H, Eluvathingal TJ, Makki M, et al. Diffusion tensor imaging of the corticospinal tract following cerebral hemispherectomy. J Child Neurol. 2006;21:566–571.
22. Korkman M, Granstrom ML, Kantola-Sorsa E, et al. Two-year follow-up of intelligence after pediatric epilepsy surgery. Pediatr Neurol. 2005;33:173–178.

child; cognition; epilepsy; functional hemispherectomy; motor skills; outcomes; review article

© 2009 Lippincott Williams & Wilkins, Inc.