OUTCOME AFTER HEMISPHERECTOMY IN HEMIPLEGIC ADULT PATIENTS WITH REFRACTORY EPILEPSY ASSOCIATED WITH EARLY MIDDLE CEREBRAL ARTERY INFARCTS
Cristine Mella Cukiert
Cássio Roberto Forster
Valeria Antakli Mello
José Augusto Burattini
Pedro Paulo Mariani
Epilepsy Surgery Program, Hospital Brigadeiro, São Paulo SP.
Purpose: To study the outcome after hemispherectomy (HP) in a homogeneous adult patient population with refractory hemispheric epilepsy.
Methods: Fourteen adult patients submitted to HP were studied. Patients had to be at least 18 years old, had refractory epilepsy, clearly focal lateralized seizures and unilateral porencephalus consistent with early middle cerebral artery infarct on MRI. All patients were submitted to functional hemispherectomy. We analyzed age of seizure onset, age by the time of surgery, gender, seizure type and frequency, interictal and ictal EEG findings, MRI and IQ scores pre-operatively; seizure frequency, drug regimen and IQ outcome were studied post-operatively.
Results: All patients had frequent daily seizures pre-operatively. All patients had unilateral simple partial motor seizures (SPS); 11 patients had secondarily generalized tonic-clonic seizures (GTC) and 5 patients had complex partial seizures (CPS), pre-operatively. All patients had hemiplegia and hemianopsia. Twelve patients had unilateral EEG findings and in 2 epileptic discharges were seen exclusively over the apparently normal hemisphere. Twelve patients were seizure-free after surgery and two patients had at least 90% improvement in seizure frequency. Pre- and post-operative mean general IQ was 84 and 88, respectively. Five of the twelve Engel I patients were receiving no drugs at last follow-up. There was no mortality or major morbidity.
Conclusions: Our results suggest that well selected adult patients might also get good results after HP. Although good results were obtained in our adult series, the same procedure yielded a much more striking result if performed earlier in life.
OUTCOME AFTER HEMISPHERECTOMY IN HEMIPLEGIC ADULT PATIENTS WITH REFRACTORY EPILEPSY ASSOCIATED TO EARLY MIDDLE CEREBRAL ARTERY INFARCTS
Hemispherectomy (HP) has been used to treat refractory hemispheric epileptic syndromes. Patients with early vascular insults, Rasmussen syndrome, hemispheric cortical disease, hemimegaloencephaly and Sturge-Weber syndrome are the most suitable candidates for this procedure. Hemispherectomy is the largest neurosurgical procedure. In well selected patients, a 70-95% seizure remission rate should be expected (Tinuper et al 1988, Carreno et al 2002, Guzzetta et al 2006).
Hemispherectomy is usually performed in children. This is due to the fact that refractory epilepsy is usually present early in these patients; in some of them catastrophic epilepsy may be seen (Gonzalez-Martinez et al 2005). Some technical variants do exist, but the main common objective is to perform complete intra- and inter-hemispheric cortical disconnection (Carson et al 1996, Villemure and Daniel 2006). In most series, patients submitted to HP were usually hemiplegic and hemianoptic preoperatively and patients with dominant hemisphere lesions usually showed language deficits that did not worsen after early surgery. Results obtained from pediatric series have clearly shown that seizure-free kids also disclosed a marked post-operative neuropsychological and motor improvement.
The outcome after HP in adults has been poorly described in the literature, and was reported in heterogeneous patient populations or as part of pediatric series. We describe the outcome after HP in a homogeneous hemiplegic adult patient population with refractory epilepsy associated to middle cerebral artery infarcts.
Fourteen adult patients submitted to HP at the Hospital Brigadeiro Epilepsy Surgery Program were studied.
Patients had to be at least 18 years old, had refractory epilepsy treated with at least 3 monotherapy regimens and polytherapy, clearly focal lateralized (usually motor) seizures and an unilateral extensive intracranial lesion (porencephalus) consistent with middle cerebral artery infarct on MRI.
Patients younger then 18 years old, with bilateral lesions on MRI, with Rasmussen or Sturge-Weber syndromes or hemispheric cortical dysplasia were excluded from this series.
Pre-operative work-up included: clinical history, neurological examination, inter-ictal and ictal video-EEG recording, IQ testing and high resolution 1.5T MRI. Post-operatively, MRI and IQ testing were obtained in all patients.
All patients were submitted to functional hemispherectomy. The procedure included temporal lobectomy and fronto-parietal cortical resection. The frontal pole and occipital lobe were left in place and fully disconnected; the insular cortex was disconnected and left in place. Due to the presence of the vascular lesion, a small amount of cortex was actually removed. No drain was left in place. All procedures were carried out by the same surgeon (AC).
We analyzed age of seizure onset, age by the time of surgery, gender, seizure type and frequency, interictal and ictal EEG findings, MRI and IQ scores pre-operatively; seizure frequency and IQ outcome were studied post-operatively. The anti-epileptic drug therapy was analyzed pre- and post-operatively.
Age by the time of surgery ranged from 19-39 years (mean=27 years; 8 men) and age of seizure onset ranged from 3-18 months (mean= 7 months). All patients in this series had at least 2 years of follow-up; mean follow-up time was 64 months (range 24-122 months). All patients had frequent daily seizures. All patients had unilateral simple partial motor seizures (SPS); 11 patients had secondarily generalized tonic-clonic seizures (GTC) and 5 patients had complex partial seizures (CPS), pre-operatively.
All patients had hemiplegia and hemianopsia. All patients were able to walk by the time of surgery. Twelve patients had unilateral interictal and ictal EEG findings, contralateral to the motor deficit; in two of them secondary bilateral synchrony could be seen. In two patients, epileptic discharges were seen exclusively over the apparently normal hemisphere. Eight patients had left middle cerebral artery infarcts; all these patients had variable degrees of language impairment, but all of them were able to speak.
Twelve patients were seizure-free after surgery (Engel I) and two patients had at least 90% improvement in seizure frequency (Engel II). These two Engel II patients presented with face motor simple partial seizures. These seizures were very likely coming from poorly disconnected insular cortex. Pre-operatively, mean general IQ was 84 (60-110), mean verbal IQ was 82 (55-105) and mean executive IQW was 85 (69-95). After surgery, mean general IQ was 88 (62-112), mean verbal IQ was 86 (57-107) and mean executive IQ was 86 (69-95) (Table I). The two patients in whom epileptic discharges were seen exclusively over the apparently normal hemisphere were rendered seizure-free after surgery.
All patients were receiving high dose anti-epileptic drug polytherapy before surgery. Five of the twelve Engel I patients were receiving no drugs at last follow-up; one Engel I patient was under anti-epileptic drug monotherapy and six Engel I and the two Engel II patients remained under polytherapy after surgery.
There was no mortality or major morbidity.
All patients in this series benefit from surgery and 85% of them were rendered seizure free. This high post-operative seizure-free rate has also been seen in some pediatric series (Basheer et al 2007, Terra-Bustamante et al 2007). Although good post-operative results might also be expected in an adult population, these findings were very poorly discussed in the literature so far. McClelland and Maxwell (2007) was probably the single paper to address it directly, although reporting on a smaller and heterogeneous population. Most importantly, the present series showed that it is possible in this patient population to adequately select surgical candidates and avoid operating on patients with bilateral disease. The patients in this series in whom epileptic discharges were seen exclusively over the apparently normal hemisphere were rendered seizure-free after surgery and this neurophysiological finding might not represent a contra-indication for the procedure. This is the larger series of adult patients with a homogeneous etiology submitted to HP published so far. Mc Clelland and Maxwell (2007) studied a smaller heterogeneous adult population submitted to anatomical hemispherectomy with a long follow-up; their data suggested that birth trauma was related to a better outcome when compared to other etiologies (tumor, AVM, encephalitis). The outcome regarding seizure frequency would very likely be the same after either anatomical or functional hemispherectomy (as in our series). The rationale to perform functional hemispherectomy and not anatomical hemispherectomy would be to reduce the complication (acute and delayed) rate and not to improve seizure control.
The major reason for the presence of residual seizures post-operatively is incomplete disconnection. In our series, this was especially true regarding the insular cortex which, as noted by Cats et al (2007), when incompletely disconnected, was able to generate motor simple partial seizures involving the face area. We did not re-operate any patient in this series although we did consider doing so in the two patients with residual simple partial seizures in the face, which we attributed to incomplete insular disconnection. Fortunately, these seizures disappeared spontaneously over the first 6 months of post-operative follow-up.
Our adult patients had improvement of motor and cognitive function post-operatively, but this improvement was not as evident as that seen in our younger population (Fonseca et al 2004). Hemispherectomy diminished spasticity, especially in the contralateral lower limb and this was associated to gait improvement. The absence of seizures and epileptic discharges led to a 5% overall cognitive improvement post-operatively. All patients were able to speak by the time of surgery, although patients with left hemisphere pathology had functional but lower output and less sophisticated language skills. Speech improvement occurred after right or left operations and reflected on the post-operative verbal IQ scores. Actually, from a functional point of view, all these patients were operated on the non-dominant hemisphere: either on a primarily non-dominant (usually right) or on a secondarily non-dominant (usually left with an early insult and complete function transfer) hemisphere. Speech improvement was probably related to better functioning of the remaining hemisphere after disappearance of the abnormal epileptic activity or secondary bilateral synchrony. Although our patients had refractory epilepsy and daily seizures, none of them was severely mentally retarded. This was probably related to the fact that they have a long lasting exclusively unilateral disease; severe mental retardation is usually seen when there are bilateral hemispheric lesions. Thirty-five percent of the patients were receiving no drugs at last follow-up visit and although the remaining patients were still under anti-epileptic drug treatment, all of them were under o more modest regimen; this might also has led to cognitive and quality of life improvement (Griffiths et al 2007).
The presence of secondary bilateral synchrony or contralateral discharges on interictal and ictal EEG was not related to a worse outcome in our patients. This has already been noted by others in patients with gross hemispheric cortical lesions (Smith et al 1991, Wyllie et al 2007).
A number of operative complications were reported after HP (Strowitzki et al 1994). In the present series we had no major morbidity or mortality. Although a mean follow-up time of around 5 years would not be useful to rule out the future occurrence of delayed cortical siderosis, we believe that this follow-up duration is reliable regarding stability of seizure frequency outcome in this patient population. Our patients had frequent daily seizures pre-operatively and the absence of seizures over these initial years would very likely represent their final outcome (Engel I). Additionally, none of our patients presented with post-operative hydrocephalus and this is possibly a reliable finding while defining their final outcome. Ten percent (10%) of our pediatric patients submitted to hemispherectomy developed post-operative hydrocephalus that needed shunting. None of our adult patients in this series developed hydrocephalus. Although it would be possible that they would develop delayed hydrocephalus, this was not seen in our pediatric patients, in whom post-operative hydrocephalus initially developed within the first post-operative year. On the other hand, half of the patients had post-operative aseptic meningitis (de Almeida et al 2006), characterized by high fever, neck stiffness and preserved consciousness and cognitive status; no patient received antibiotic therapy. We saw no cognitive decline in any patient.
Our results suggest that well selected adult patients might also get good results after HP. Our patients might also have been good surgical candidates in childhood. Hemispherectomy is a highly underused surgical procedure, especially in patients with hemiplegic epileptic syndromes. Although good results were obtained in our adult series, the same procedure yields a much more striking result if performed earlier in life.
- Basheer SN, Connolly MB, Lautzenhiser A, Sherman EM, Hendson G, Steinbok P (2007). Hemispheric surgery in children with refractory epilepsy: seizure outcome, complications, and adaptive function. Epilepsia 48:133-40.
- Carreño M, Kotagal P, Perez Jiménez A, Mesa T, Bingaman W, Wyllie E (2002). Intractable epilepsy in vascular congenital hemiparesis: clinical features and surgical options. Neurology 59:129-31.
- Carson BS, Javedan SP, Freeman JM, Vining EP, Zuckerberg AL, Lauer JA, Guarnieri M (1996). Hemispherectomy: a hemidecortication approach and review of 52 cases. J Neurosurg. 84:903-11.
- Cats EA, Kho KH, Van Nieuwenhuizen O, Van Veelen CW, Gosselaar PH, Van Rijen PC. Seizure freedom after functional hemispherectomy and a possible role for the insular cortex: the Dutch experience (2007). J Neurosurg. 107(4 Suppl):275-80.
- de Almeida AN, Marino R Jr, Aguiar PH, Teixeira MJ (2006). Postoperative fever after hemispherectomy: the role of non-infectious factors. Seizure 15:340-3.
- Fonseca LF, Melo RP, Cukiert A, Burattini JA, Mariani PP, Brandão R, Ceda L, Baldauf CM, Argentoni M, Forster C, Baise C (2004). Early functional hemispherectomy in hemimegalencephaly associated to refractory epilepsy. Arq Neuropsiquiatr. 62:1063-7.
- González-Martínez JA, Gupta A, Kotagal P, Lachhwani D, Wyllie E, Lüders HO, Bingaman WE (2005). Hemispherectomy for catastrophic epilepsy in infants. Epilepsia 46:1518-25.
- Griffiths SY, Sherman EM, Slick DJ, Eyrl K, Connolly MB, Steinbok P (2007). Postsurgical health-related quality of life (HRQOL) in children following hemispherectomy for intractable epilepsy. Epilepsia 48:564-70.
- Guzzetta F, Battaglia D, Di Rocco C, Caldarelli M (2006). Symptomatic epilepsy in children with poroencephalic cysts secondary to perinatal middle cerebral artery occlusion. Childs Nerv Syst. 22:922-30.
- McClelland S 3rd, Maxwell RE (2007). Hemispherectomy for intractable epilepsy in adults: the first reported series. Ann Neurol. 61:372-6.
- Smith SJ, Andermann F, Villemure JG, Rasmussen TB, Quesney LF (1991). Functional hemispherectomy: EEG findings, spiking from isolated brain postoperatively, and prediction of outcome. Neurology 41:1790-4.
- Strowitzki M, Kiefer M, Steudel WI (1994). Acute hydrocephalus as a late complication of hemispherectomy. Acta Neurochir (Wien). 131:253-9.
- Terra-Bustamante VC, Inuzuka LM, Fernandes RM, Escorsi-Rosset S, Wichert-Ana L, Alexandre V Jr, Bianchin MM, Araújo D, Santos AC, Oliveira dos Santos R, Machado HR, Sakamoto AC. Outcome of hemispheric surgeries for refractory epilepsy in pediatric patients (2007). Childs Nerv Syst. 23:321-6.
- Tinuper P, Andermann F, Villemure JG, Rasmussen TB, Quesney LF (1988). Functional hemispherectomy for treatment of epilepsy associated with hemiplegia: rationale, indications, results, and comparison with callosotomy. Ann Neurol. 24:27-34.
- Villemure JG, Daniel RT (2006). Peri-insular hemispherotomy in paediatric epilepsy. Childs Nerv Syst. 22:967-81.
- Wyllie E, Lachhwani DK, Gupta A, Chirla A, Cosmo G, Worley S, Kotagal P, Ruggieri P, Bingaman WE (2007). Successful surgery for epilepsy due to early brain lesions despite generalized EEG findings. Neurology 69:38-97.
Patient Age (Years) Age Sz Onset (Months) Sz Pattern Sz Frequency EEG MRI Neuropsychology – IQ
I 19 5 M-SPS / GTC daily IPSO / SBS left G: 82 , V: 75 , P: 90
II 21 8 M-SPS / GTC / CPS daily IPSO / SBS left G: 60 , V: 55, P: 70
III 21 12 M-SPS / GTC daily IPSO right G: 90 , V: 90 , P: 90
IV 25 18 M-SPS / GTC daily IPSO left G: 62 , V: 58 , P: 71
V 26 4 M-SPS daily IPSO right G: 95 , V: 92 , P: 97
VI 27 3 M-SPS / CPS daily IPSO left G: 85 , V: 84 , P: 90
VII 27 2 M-SPS / CPS daily IPSO left G: 65 , V: 59 , P: 78
VIII 27 3 M-SPS / GTC / CPS daily contra only left G: 65 , V: 61 , P: 69
IX 28 6 M-SPS / GTC daily IPSO / SBS right G: 94 , V: 97 , P: 90
X 29 7 M-SPS / GTC daily contra only right G: 92 , V: 94 , P: 78
XI 30 12 M-SPS / GTC daily IPSO / SBS left G: 90 , V: 89 , P: 95
XII 30 4 M-SPS / GTC daily IPSO right G: 100 , V: 98 , P: 89
XIII 31 5 M-SPS / GTC daily IPSO right G: 110 , V: 105 , P: 90
XIV 39 13 M-SPS / GTC / CPS daily IPSO left G: 95 , V: 95 , P: 95
Table I: Summary of the preoperative clinical data. Sz: seizure; M-SPS: motor simple partial seizures; GTC: generalized tonic-clonic seizures; CPS: complex partial seizures; IPSO: ipsolateral; SBS: secondary bilateral synchrony; contra: contralateral; G: general IQ; V: verbal IQ; P: performance IQ.
Patient Engel Neuropsychology – IQ Follow-up (months)
I I G: 83 , V: 77 , P: 89 122
II II G: 62 , V: 57, P: 70 113
III I G: 95 , V: 95 , P: 89 99
IV I G: 67 , V: 65 , P: 73 91
V I G: 95 , V: 92 , P: 96 90
VI I G: 89 , V: 87 , P: 90 73
VII I G: 72 , V: 67 , P: 82 50
VIII I G: 66 , V: 62 , P: 69 49
IX I G: 98 , V: 100 , P: 91 47
X I G: 93 , V: 97 , P: 80 39
XI I G: 95 , V: 94 , P: 95 38
XII I G: 110 , V: 107 , P: 92 36
XIII II G: 112 , V: 107 , P: 92 28
XIV I G: 95 , V: 96 , P: 95 24
Table II: Post-operative outcome regarding seizures and cognitive function. G: general IQ; V: verbal IQ; P: performance IQ.
We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Disclosure of Conflicts of Interest
None of the authors has any conflict of interest to disclose.