Epilepsy surgery for focal cortical dysplasia: Seizure and quality of life (QOLIE-89) outcomes
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.246263
Source of Support: None, Conflict of Interest: None
Keywords: Drug resistant epilepsy, epilepsy surgery, focal cortical dysplasia, quality of life, seizure outcome
Focal cortical dysplasia (FCD) is a leading cause of drug resistant epilepsy (DRE) and a common substrate, especially in children. FCD is included in the spectrum constituting malformations of cortical development. FCD results from disturbances in the neuronal migration at an early in-utero developmental stage. Since the initial report by Taylor et al., in 1971, on the disorganized cortex with dysmorphic neurons, it became an area of intense research. FCDs need not be focal and need not be cortical. FCDs may or may not have balloon cells.
With better understanding of the natural history of DRE and the advent of high-resolution MRI, these lesions are increasingly being recognized. It has also been proven beyond doubt that tailored surgical resection has the potential to cure or control the DRE in the majority of the patients. If surgical intervention is done at an early stage of the disease, there is a significant chance of potentially curing epilepsy; it also helps in protecting the developing brain from repetitive epileptic attacks which hamper normal brain development. With improved seizure control and the expected normal brain development, surgery has the potential to improve quality of life (QOL) among these patients.
Studies analyzing the QOL objectively in patients affected by DRE are limited, especially those conducted in a uniform cohort of patients with a single pathological substrate. In this study, we have analyzed the clinical, EEG, MRI, neuropsychological, surgical and histo-pathological features in patients with DRE and FCD to identify the outcome of surgery in terms of seizure control and QOL, using the “Quality Of Life In Epilepsy-89 (QOLIE-89) questionnaire”. We have attempted to identify the predictive factors responsible for seizure cure/control after surgical intervention in these patients. To the best of our knowledge, this is the first study from India reporting the pre-operative and post-operative QOL indices along with seizure outcome among people with DRE and FCD.
This study was conducted at the National institute of Mental health and Neurosciences (NIMHANS), Bengaluru. This study was approved by the NIMHANS Institution Ethics Committee. The duration was extending from January 2008 to December 2016. In this study, the patients were recruited in both prospective and retrospective manner. From August of 2012 onwards, the patients were recruited in a prospective manner. All patients with DRE and FCD, diagnosed with electro-clinical and neuroimaging concordance and confirmed on histo-pathological examination, were included.
A detailed history of the patients, including the age of onset of seizures, the duration of epilepsy, and the frequency, semiology and change in character of seizures, if any, were noted. The peri-natal history of birth asphyxia or delayed cry, a family history of epilepsy and a history of neonatal or febrile seizures, were recorded. Scores based on the “Seizure Severity Outcome Scale” were recorded based on the frequency of seizures. The treatment history, including the intake of anti-epileptic drugs, their dosage and the duration of administration were recorded. Neuropsychological testing was performed on all the patients. These tests were designed to identify deficits of specific lobar function and other associated evidence of discrete cerebral dysfunction. The tests were administered after ascertaining the languages known, the level of education and the handedness of the patient.
In persons suffering from epilepsy, the quality of life is influenced by various factors like the AED usage, psychological or cultural factors, as well as the intrinsic nature of the disease itself. These patients have awareness of epilepsy as well as the side effects of AEDs. Most of the patients have a negative impact, either physically or psychologically, related to the ingestion of AEDs. They have a natural tendency to worry about seizure recurrence, seizure frequency, sustaining bodily injuries during seizures, embarrassment arising out of the seizure episode, and many more factors. To understand these issues, patients in this study underwent a routine neuropsychological assessment, and in addition, a questionnaire for assessment of the impact of epilepsy on the quality of their lives was also filled by them. It was done using a widely accepted questionnaire method, QOLIE-89 (Quality of Life In Epilepsy-89). This questionnaire consists of a total of 17 multi-item scales that tapped the following health related parameters: health perception (HP), overall QOL (OQOL), pain (P), physical functions (PF), role limitations due to emotional problems (RL_E), role limitation due to physical problems (RL_P), lack of energy/fatigue (EF), emotional well-being (EWB), seizure worry (SW), medication effects (ME), health discouragement (HD), attention/concentration (AC), memory (M), language (L), social support (SS) and social isolation (SI). In field testing, raw scores for each perspective were recorded. The overall score was obtained using a weighted average from the multi-item scale. For all prospective patients, the scores were recorded prior to surgery. Post-operative scores were recorded for all patients at the last follow-up visit along with the recording of Engel's outcome score. Pre-operative scores for patients enrolled in the retrospective cohort were recorded based on their subjective experience before surgery in comparison to their post-operative response for the same item concerned. The histological features were reported in accordance with the International League Against Epilepsy (ILAE) 2011 recommendations. Following surgery, details regarding the immediate post-operative deficits/seizure recurrence and post-operative morbidity were noted and entered into a pre-designed proforma. Patients were also contacted via telephonic interviews and their current status with respect to seizure control and antiepileptic drug dosages were ascertained.
Statistical analysis was performed using the Satistical Package for the Social Sciences, version 22 (SPSS v22.0; IBM). Descriptive statistics such as percentages were calculated for the categorical variables such as gender, while mean, median, range and standard deviation were calculated for continuous data such as age, seizure severity score, QOLIE-89 score and the Engel outcome scale. Comparison of the two groups (good versus bad outcome) was done using Mann Whitney U-Test. The statistical tests used were the Pearson's chi Square test, and the Mann-Whitney U test (for assessment of the non-parametric data) for uni-variate analysis. After applying the Pearson chi-square and Fisher's exact test for outcome with each single variable of interest, a multivariate binary logistic regression analysis with backward elimination method was carried out for all interested variables. Significance was established at the 95% level. A comparison between the pre-operative and post-operative measurements was done using Wilcoxan signed rank test.
A total of 52 patients fulfilled the inclusion criteria of the study, of which 14 patients were recruited in a retrospective, and the rest of the 38 patients in a prospective manner. The demographic parameters are depicted in [Table 1]. Prior to surgery, around 70% of the patients were having one to ten episodes of habitual focal seizures daily. The frequencies of seizure severity scores are shown in [Table 2]. All of the patients had localization related focal seizures at presentation. Of these, 61.5% had secondary generalization of these seizures at least once in their course of illness. The number of anti-epileptic drugs (AEDs) tried in their lifetime to control these seizure episodes ranged between five and eight drugs in 86.5% of the patients. A history of febrile seizures was present in 17 (32.7%) patients. A family history of seizures was noted in only 4 (7.7%) patients. One (1.9%) patient had encephalitis early in life; while nine (17.3%) patients had a history of birth hypoxia with delayed cry at the time of birth, exposing them to the risk of hypoxic brain damage early in life.
The scalp electroencephalogram (EEG) identified the epileptic discharges (random epileptic discharges [REDs]) in 39 (75%) patients, of which 18 (34.6%) had bilateral abnormalities and 21 patients (40.6%) had unilateral abnormalities. In rest of the 13 (25%) patients, the scalp EEG did not detect any abnormality. Clinical and EEG concordance was noted in 35 (67.3%) patients. The sensitivity of long term VEEG was superior in detecting and localizing epileptogenic focal discharges among these patients. The ictal run of spikes was the most common abnormality found in these patients (n = 28; 53.8%), followed by the ictal built-up of theta/delta waves, which was seen in 19 (36.5%) patients. Clinical and VEEG concordance was found in 45 (86.4%) patients. Inter-ictal VEEG had bilateral discharges in 6 (11.5%) patients only. Most of the unilateral focal discharges (n = 23; 44.2%) were from the temporal lobes. In 4 patients, no clinical attacks could be witnessed during the long term VEEG recording. Among these four patients, the inter-ictal single photon emission computed tomography (SPECT) helped in localization of the epileptogenic zone (EZ) in one patient; the positron emission tomography-computed tomography (PET-CT) helped in localization of the EZ in another, and the other two patients were operated based on the risks and benefits directed by the imaging and EEG findings. The most common area that showed the onset of clinical episodes was the temporal lobe [21 (40.4%)], followed by the frontal 12 (23.1%) lobe. A multi-lobar clinical onset was noted in 15 (28.8%) patients.
MRI revealed the lesions in all the 52 patients [Figure 1]. Not all of these MRI scans were concordant with the dysplastic epileptogenic cortex localized on VEEG/EEG. In 5 (9.6%) patients, lesions were present which were non-concordant with the dysplastic part of the brain having the EZ within. All of these five MRI scans had bilateral temporal abnormalities with or without evidence of dysplasia. Lateralization and localization was decided upon the electrophysiological and other ancillary imaging studies. The lesion was left hemispherical in 25 (48.1%) patients, right hemispherical in 22 (42.3%) patients, and bilateral in 5 (9.6%) patients. The most common lobar involvement was temporal (n = 23; 44.2%) followed by frontal (n = 15; 28.8%), parietal (n = 3; 5.8%) and occipital (n = 2; 3.8%). Multi-lobar dysplasia was found in rest of the nine (17.3%) patients. Focal thickening of the cortex (n = 39; 75%) and blurring of grey-white matter (n = 44; 84.6%) were the most common abnormalities identified. The trans-mantle sign was noted in 22 (42.3%) patients; while T1-hyperintensity was seen in only 5 (9.6%) patients. T2 hyperintensity was noted in 36 (69.2%) patients. Imaging based dual or double pathology, including hippocampal sclerosis, was identified in 15 (28.8%) patients. Cerebellar atrophy was noted in four patients, among whom two patients also had cerebral atrophy. 75% patients had their dysplasia in eloquent cortices. 33 (63.46%) patients required ancillary tests like PET-CT, SPECT, magneto-encephalography (MEG) and/or functional-MRI for a precise localization of their EZ and/or its relation with the eloquent cortices. Ictal- SPECT was done in 8 patients, of whom 5 patients showed findings concordant with VEEG and imaging; 3 had non-concordant findings, and one patient had an inter-ictal SPECT which was normal. PET-CT was available for 20 patients, out of which 2 had a normal scan; 14 patients had concordant findings, and 2 each respectively, had either non-concordant or concordant findings for the whole lobe (not just the lesion). 18 patients underwent functional-MRI to delineate the eloquent cortex based upon the individual case. 16 of all the 38 prospective cases underwent a MEG scan, of which only 4 patients showed lesional concordance, 10 patients had a lobar abnormality which was concordant, and one patient had bilateral inter-ictal discharges (IEDs) and dipole-cluster (DC) formation.
Neuro-psychological assessment and QOLIE-89
Thirteen (25%) patients had neuropsychological deficits limited to one cerebral hemisphere. 32 (61.5%) patients had neuropsychological deficits distributed in bilateral cerebral hemispheres. Neuropsychological assessment could not unveil any deficits in 7 patients, i.e., in 13.5% of the patients.
Higher scores on QOLIE-89 indicate a better health [Table 3]. Among the 17 health domains measured by this questionnaire, and after comparing the mean values of different QOLIE-89 domains using WICOXON Signed-Rank Test for paired samples to establish significance as shown in [Table 4], maximum changes in favour of improvement after surgery were seen in four individual domains. These domains included: Seizure worry (18.73 vs. 86.91; P < 0.01) followed by role limitation due to emotional factors (23.46 vs. 94.90; P < 0.05), overall quality of life (24.92 vs. 84.28; P < 0.05), and lastly, role limitation due to physical factors (25 vs. 80; P < 0.05). The total score or the final QOLIE-89 score significantly improved from a mean of 38.33 in the pre-operative period to a mean of 75.21 after surgery, with an absolute improvement of 49.03% (P < 0.001). Two individual parameters showed the most significant improvement after surgery namely; emotional well-being (34.07 vs. 75.12; P < 0.001) and medication effects (42.51 vs. 78.73; P < 0.001); and ultimately, final overall scores of QOLIE-89 ( 38.33 vs. 75.21; P < 0.001). Minimum and insignificant improvement in scores were seen in four different domains, which included energy/fatigue (42.78 vs. 55.19; P = 0.09), followed by pain (67.83 vs. 88.03; P = 0.08), attention/concentration (50.10 vs. 68.27; P = 0.08) and language (57.30 vs. 79.15; P = 0.09).
24 patients harboring an extra-temporal (n = 28) dysplastic EZ underwent either a lesionectomy or extended lesionectomy. Resection extended to include even an entire lobe (lobectomy; n = 1) or multiple lobes (n = 2), with or without multiple sub-pial transections (MST; n = 3). One MST was done in the sensory cortex of a patient who underwent a functional posterior quadrant disconnection; rest of the two MSTs were done in the motor cortices of patients who underwent lesionectomy in the frontal lobe. Two patients in the extra-temporal group underwent a posterior quadrant disconnection. One patient underwent a functional hemispherotomy to prevent multiple falls, and biopsy from the lesion at the disconnection margin revealed FCD. Details of surgical procedures have been mentioned in [Table 5].
FCD type II constituted the most common type (n = 26, 50%) of cortical dysplasia in this series, especially the type IIB, which was noted in 18 (34.6%) patients. Seven (13.5%) patients had a multi-lobar FCD. One of these patients with multi-lobar dysplasia had type IIIA FCD in the temporal lobe and type IC FCD extending posteriorly into the adjacent parietal lobe. As seen in [Table 6], Type II FCD was mostly distributed in the frontal lobe (n = 12; 46.15% of all type II FCDs). The next most common was type III FCD, which was present in 16 (30.7%) patients. A dual pathology was seen in seven patients on histopathological examination. Four of these patients had FCD IIA with hippocampal sclerosis (HS) and one patient had FCD IIB with HS. One patient each had a gliotic scar or encephalitis along with HS, respectively. The double pathology was proven only in two patients. One patient had FCD IA with gangliocytoma in the left parietal lobe; the other one had a multi-lobar (temporal and occipital) FCD IA with a gliotic scar. When two different and independent lesions (but not including hippocampal sclerosis) existed within one or multiple lobes, this combination was referred to as “double pathology”. An example was a cavernoma on one side and a ganglioglioma in the other lobe. Only electrophysiological studies like the VEEG can reveal the most likely source of epileptogenesis. On the other hand, “dual pathology” means that hippocampal sclerosis is combined with another independent principal lesion, which may or may not reside in the same temporal lobe. This second principal lesion may be a tumour, encephalitis, a vascular malformation, or even a malformation of cortical development (MCD) like FCD IIA or IIB. For our study purpose, this second principal lesion had to be FCD, for it to be labeled as dual pathology. However, FCD I combined with HS is not a dual pathology, as this will be defined as the Type IIIA group. [Figure 2], [Figure 3], [Figure 4] represent the histopathological appearance of cortical dyslamination, thickening of cortical ribbon and the dysmorphic and balloon cells present in these patients, respectively.
Follow-up was available for a total of 50 patients with the follow-up range being from 3 months to 9 years with a mean duration of 3.70 ± 2.22 years. Two patients were lost to follow-up. If they could have been reviewed, they would have completed three years after their surgery, as mentioned in [Table 7]. Of the 50 patients available for follow-up, most of them (n = 35; 70%) had a seizure severity score of 2 after their surgery at the last available follow-up, indicating seizure control with or without AEDs. Only 4 (8%) patients had a score of either 8 or 9; none of the patients had a score of 10 or more. In 25 (50%) patients, at the last follow-up, AEDs were either reduced in dosage or numbers, or in both. 3 patients were completely off their AEDs at their last follow-up visit. 7 patients needed to increase the dose or the number of AEDs or had to be switched to another drug. In 15 patients (30%), the numbers or dosages of AEDs were unchanged at their last follow-up. A total of 14 patients were eligible for a 6-year follow-up, 37 patients were eligible for a 3-year follow up, and 50 patients were eligible for a 2-year follow-up. All patients eligible for follow-ups were available excepting for the two patients who were at a three-year follow-up. They were lost to follow up, as mentioned previously; and, 2 patients have not yet completed a two-year follow-up visit beyond surgery. A favorable outcome of Engel's IA was attained in 70% (n = 30 out of 50) patients at a 2-year follow-up visit; and, in 72.22% (n = 26 out of 35) patients at a 3-year follow-up visit. 12 out of 14 patients in whom a 6-year follow-up was available had an Engel's grade IA outcome.
Acute postoperative seizures (APOS) within 24 hours were noted in 8 (15.4%) patients. Operative site hematoma was seen in 3 patients, but none of these patients required a re-exploration and evacuation of the hematoma. Two patients suffered from meningitis that required administration of antibiotics, thus prolonging their hospital stay. One patient developed bone flap osteomyelitis after wound infection and cerebrospinal fluid (CSF) leak, which required removal of the infected bone flap. She underwent a titanium mesh cranioplasty eight months after her primary surgery. 11 patients developed motor weakness of grade 1-2/5 immediately after surgery that gradually improved in the follow-up period. All of them reported in the outpatient department with the muscle power in the affected limb being at least 4+ to 5/5. In 3 of these patients with motor deficits, an upper motor neuron facial palsy was also noted. All these patients were expected to have an aggravation of their neurological deficits following surgery as the lesions were either in the motor strip or adjacent to the motor cortex. Non-disabling visual field cuts were seen in 3 patients, which correlated with the location of the lesion. Similarly, three other patients developed dysphasia after surgery from which they gradually improved.
To study the characteristics and variables of patients that may influence the seizure outcome, patients were divided into two groups depending upon their final Engel's outcome at the last follow-up after surgery. Group A was assigned to the patients with a favorable outcome which was defined as Engel's IA grade; while, rest of the patients whose outcome was other than Engel's IA grade were categorized in Group B. [Table 8] shows the comparison of means of usual variables between the two groups using Mann Whitney U-test. Only demographic or statistically significant variables are shown in this table. It was noted that prior to surgery, these two groups did not significantly differ in their gender (48.5% vs. 73.33% males; P = 0.548), mean age at presentation (7.60 vs. 9.79 years P = 0.695), duration of epilepsy (12.26 vs. 15.50 years; P = 0.420) and number (6.60 vs. 6.53; P = 0.965) of AEDs tried in the life-time before surgery. However pre-operatively, these two groups did differ in their seizure severity scores (SS score; 8.6 vs 9.6; P = 0.020), age at surgery (20.26 vs. 24.93 years; P = 0.285) and three of the sub-scales in quality of life assessment viz. emotional well being (31.54 vs. 39.20; P = 0.039), health discouragement (36.00 vs. 26.66; P = 0.013) and language function (56.00 vs. 61.33; P = 0.043) at a significant level. A higher score indicates a superior health parameter, which means patients in group A had lesser health discouragement than those in group B but their emotional well being and language function related parameters were worse-off when compared to group B. Interestingly, despite of higher frequency of seizures in group B, both groups did not differ in their seizure worry and medication effects prior to surgery. Rest of the parameters were comparable between the two groups prior to surgery. After surgery, the most significant improvement was seen among two variables which ultimately determined the final outcomes after surgery. These two were seizure severity score (1.80 vs. 5.80; P < 0.001) and final QOLIE-89 score (79.15 vs. 66.0; P < 0.0001). In addition, improvement in health perception (84.04 vs. 63.33; P = 0.024), health discouragement (76.00 vs. 57.33; P = 0.006) and overall QOL scores (90.42 vs. 68.90; P = 0.002) between the two groups after surgery was noted. None of the remaining parameters, neither in QOLIE-89 assessment nor in general variables, differed in between the two groups. It should be noted that the mean age at surgery did not differ between these two groups but when the whole pre-operative cohort was dichotomized based on the cut off age of 20 years (which was based on the fact that this value indicated the median age at surgery in the whole cohort), younger patients showed a significantly better outcome (78.26% vs. 62.965; P = 0.013), as shown in [Table 9]. Likewise, as shown in [Table 8] and [Table 9], the other factors predicting a favorable outcome (Engel's IA) were: a seizure severity score of 9 or more, lack of an early post-operative seizure, FCD type II, and focal discharges on VEEG without propagation. These factors were found to be the significant factors that determined a favorable outcome on both uni-variate as well as multi-variate analysis. Dyscognitive seizures of temporal lobe origin when presenting as a single semiology, temporal lobe FCDs, absence of mental retardation, lack of the need to perform ancillary tests to localize the EZ were other important factors predicting a good outcome when tested on univariate analysis only. Factors that predicted a significant improvement in the QOL among these patients were lower seizure severity scores (both before and after surgery), sense of better emotional well being as well as better language function prior to surgery, higher pre-operative as well as post-operative health perception scores, and lack of health discouragement after surgery. Not all patients with Engel IA outcome always had a better improvement in their QOL. Factors such as surgical complications, e.g., persistent weakness of limbs, language dysfunction and social factors played an integrated role in achieving a significantly better QOL on both uni- as well as on multivariate analysis.
Age at surgery and duration of epilepsy
In the current series of 52 patients with FCD, the mean age at onset of seizures was 7.94 ± 6.23 years (range: 1 month–40 years, median age: 5 years), while the mean age at surgery in this group was 20.88 ± 12.51 years (range: 1–60 years, median age: 20 years). In one series, the authors studied 22 patients with FCD with an equal number of male and female patients (11 patients each of both the genders) whose mean age at surgery was found to be 26.6 years ± 17.6 years. However, a meta-analysis of a large number of patients (n = 2014) that included 37 studies, which reported on seizure outcome after resection of FCD, found the mean age of the patients at surgery to be 15.4 ± 8.3 years. This mean age at surgery was almost a decade less than the value quoted in the series with 22 patients (26.6 years vs. 15.4 years) mentioned earlier, and almost half a decade lesser than our series (15.4 years vs. 20.88 years). There is variability among the studies as far as age at surgery and duration of epilepsy are considered. Other studies, found the mean duration of epilepsy before surgery to be 21 years, with the mean age at the time of surgery being 28 years. In a large series of children suffering from DRE, the mean time from the seizure onset to surgery was only 5.7 years. We found the mean duration of epilepsy to be 12.95 ± 9.56 years (range: 11months–39 years, median duration: 12 years) prior to surgery in these patients. Whether or not this duration of epilepsy before surgery, and thus, the age at surgery affects outcome, is one of the important questions to be answered in order to predict the surgical outcome as well to identify the changing trends in the surgical management. This answer would also give a concrete evidence regarding subjecting these patients to surgery as early as possible in their life. An analysis of 16 studies with 469 patients found that adults suffering from DRE due to FCD with age more than 18 years (at the time of surgery) were equally benefited by resection when compared to the pediatric population. In another study, the authors lowered their cut-off age to 6 years in order to dichotomize the group; the difference between the groups was still found to be insignificant, indicating that there was no difference in the outcome based on the patient's age at the time of surgery. However, in present series, both an early age of the patients at the time of onset of epilepsy (younger than 8 years), and therefore, a longer duration of epilepsy (more than 12 years), were both not found to be associated with a significant impact on the outcome, when compared to the group with a later onset (>8 years) and a shorter duration (less than 12 years) of epilepsy (72.41% vs. 66.66%; P = 0.662; and, 72.00% vs. 68.00, P = 0.758, respectively).
Operating at a younger age with a lesser seizure frequency showed a trend towards a favorable outcome in the current cohort. Patients who were suffering from fewer seizures (seizure severity score of less than 9) had a better outcome compared with those who were suffering from more frequent seizures (90.00% vs. 56.66%; P = 0.012). This may reflect the re-organization of the epileptogenic network when epilepsy is of a long-standing duration in a particular patient. Patients who were operated at a younger age (less than 20 years at the time of surgery) also had a favorable outcome when compared to patients who were operated in the third decade or beyond, in their life (78.26% vs. 62.96% P = 0.013). Female patients showed a trend towards a better outcome (81.88%) compared to the male ones (60.71%), but this difference did not achieve a statistical significance (P = 0.106). None of the other factors like a family history of seizures (71.73% vs. 50%; P = 0.574), the presence of febrile seizures (76.47% vs. 56.25%; P = 0.191) in childhood, the past history of encephalitis/meningitis (71.42% vs. 0%; P = 0.300), or even the number of AEDs tried in one's life time (P = 0.574), were found to be significantly affect the outcome in terms of freedom from seizures.
Seizure semiology and electrophysiology
Patients presenting with a single seizure and a stereotypical semiology of dyscognitive seizures (DCS) without secondary generalization had significantly better outcome when compared with those who patients who had polymorphic seizures (100.00% vs. 61.53%; P = 0.006), such as generalized seizures, absence seizures or myoclonic jerks in addition to DCS, prior to surgery. Our result was in accordance with the results of the meta-analysis mentioned earlier, in which the authors noted the presence of partial seizures favoring a good outcome after surgery; this meta-analysis did not find any correlation between the patient's age and the EEG or VEEG localization of the ictal onset zone with seizure freedom. However, on multivariate analysis, the present study did not correlate with the seizure semiology and surgical outcome (P = 0.09).
Antonio studied the importance of finding rhythmic epileptic discharges (REDs) on scalp EEGs in 74 (34 had FCD and 40 had non-dysplastic lesions) patients suffering from DRE. It was found that 44% of 34 patients suffering from DRE secondary to FCD had REDs, which correlated with continuous epileptiform discharges (CEDs) recorded on the electrocorticogram (ECoG). None of the patients with a non-dysplastic lesion (40 patients) had REDs. They concluded that the presence of REDs and CEDs were highly specific and sensitive indicators for demonstrating the presence of dysplastic lesions; and, that FCD is likely to be noted in the region where REDs are found. Neither detection of these abnormalities, that is, REDs (61.53% in normal vs. 72.97% in abnormal EEGs; P = 0.493) on surface/scalp EEG nor its laterality (81% in unilateral spiking, 62.5% in bilateral spiking, 61.53% in normal EEGs; P = 0.337) or concordance (72.72% vs. 75%; P = 0.548) with the lesion, as seen on MRI, correlated with a significantly better outcome in terms of seizure freedom in our series.
The seizure episodes originating from the temporal lobe (85.71% vs. 58.62%; P = 0.05), and spikes localized to one hemisphere on VEEG recording (88.6% vs. 0%; P < 0.001) were the two most significant factors that predicted a favorable outcome. On multivariate analysis, focal discharges without propagation on ictal VEEG also consistently helped in predicting a favorable outcome. In a large number of patients, Nathan et al., concluded that EEG or VEEG based localization of the ictal onset zone was not helpful in predicting seizure freedom. Concordance of the clinical attacks (with 75% good outcome when a concordance of attacks was noted, 25% good outcome when non-concordant attacks were noted, and 66.7% good outcome in each category when no attacks could be recorded or when findings were difficult to interpret; P = 0.148) with VEEG findings did not correlate with a favorable outcome. Similarly, concordant spikes on VEEG (P = 0.475) and the types of discharges (P = 0.681) were not found to be associated with a favorable outcome. Absence of contralateral discharges was associated with a better outcome (85% vs. 63.5%) but with no statistical significance (P = 0.061).
In the current series, the MRI could identify the presence and location of FCD in all (100%) of the patients. Multiple studies have shown the identification of FCDs by the presence of a radial sub-cortical hyper-intensity known as the ‘trans-mantle sign’ on the T2-weighted images of MR studies. The sign helps in the diagnosis of FCDs and is, therefore, considered as a proxy marker of a favorable outcome. In our study, the presence of the ‘trans-mantle sign’ did not predict a better seizure freedom (72.4% vs. 66.7%; P = 0.662). Absence of FCDs in the eloquent cortex (100% vs. 59.45%; P = 0.013) and lack of the need to perform ancillary tests (83.33% vs. 62.50%; P = 0.006) like the ictal-SPECT, PET-CT or magnetoencephalogram (MEG) were the only two imaging related variables which were in favor of predicting an Engel IA outcome on the univariate analysis. Rest of the imaging findings like the presence of an MRI lesion in concordance with VEEG spikes, the presence of T1-or T2- lesional hyper-intensities (68.88% vs. 80%; P = 1.000 and 0% vs. 72.91%; P = 0.099, respectively), identification of lesions suggesting a dual/double pathology (68.75% vs. 72.22%; P = 0.747) on MRI, did not influence the outcome significantly. To our knowledge, MEG was utilized for the first time in India for evaluation of patients with DRE and FCD. MEG helped in the localization of source (which serves as the proxy of EZ) in 62.5% cases, with maximum spatial resolution being obtained in the area of the dysplastic cortex (10 out of 16 cases). Only in 4 patients (25%), MEG localized the source that led to the spatial resolution of the lesion.
Six patients had evidence of a residual lesion on the post-operative MR scan. All of them belonged to Group B and had a poor outcome. The reasons for incomplete resection were: Missing the peri-ventricular extension of the lesion (an 8-year old boy with right parietal lobe FCD [Engel IIA]); the residual lesion was in the posterior resection margin, which was extending into the eloquent cortex. Thus, a more aggressive resection could not be done (a 12-year old boy harboring left frontal FCD with residual lesion in the left motor cortex [Engel IIA], and another 17-year old male patient with a left frontal FCD and the residual lesion being in the left motor cortex [Engel IV]; multi-lobar dysplasia without (a 23-year old male [Engel IVA]) or with (a 4-year old girl [Engel IIA]) a double pathology; and, the dominant temporal lobe involvement (a 60-year old female patient with left temporal lobe FCD [Engel IIIA]). Five patients with a non-favorable outcome had undergone complete resection, as evidenced by their post-operative scans, but persistent discharges were recorded either during the intra-operative ECoG recording (n = 2) or during their post-operative EEG (n = 3) at a 3-month follow-up visit.
SPECT and PET-CT
In three patients, the ictal VEEG was difficult to interpret. Therefore, their surgery was based on the ictal SPECT (a 60-year old female patient with left temporal FCD); the PET-CT findings (an 11-year old girl with right multi-lobar dysplasia); or, on the clinical and scalp EEG findings (a 6-year old boy with left parietal FCD with double pathology). This resulted in Engel IA outcome in both pediatric patients. However, the 60-year old elderly female had an Engel IIIA outcome. This 60-year old lady initially had Engel IA outcome for 6 months and then started having her habitual seizures. Post-operatively, she was evaluated with MEG, which showed a dipole cluster at the posterior margin of the resection and, therefore, she was advised a re-do surgery; but she preferred to be on add-on medical therapy rather than undergo re-surgery. Experience with these patients suggests that an early age at surgery before the development of extensive epileptic networks is responsible for a good prognosis. Among the 4 patients in whom no clinical seizures were noted on a long term VEEG evaluation, surgery was guided by concordance with either an ictal-SPECT (a 28-year old male patient with left frontal FCD), with PET-CT (a 36-year old male patient with left temporal FCD, and another one-year old boy with left multi-lobar dysplasia), or with repeated scalp EEG concordant findings. All these three patients had an Engel IA outcome initially. However, the one-year old boy was one of those two patients who were lost to follow-up.
Out of the 4 patients with non-concordant clinical episodes that were assessed with a VEEG, only one patient (a 31-year old female patient with right parietal FCD) had a good outcome. In the latter patient, surgery was guided by concordant hypo-metabolism identified on the PET-CT imaging. Rest of the three patients had either an Engel IIIA outcome (a 16-year old female patient with left frontal FCD, where surgery was decided on the basis of clinical and scalp EEG findings), an Engel IVA outcome (a 29-year old male patient with a dual pathology in the right parietal lobe, where surgery was guided by the concordance between ictal-SPECTand imaging), and an Engel IIA outcome (a 8-year old boy with right parietal FCD, where surgery was again decided upon the clinical and scalp EEG findings).
Neuro-psychology and quality of life (QOL) assessment
Absence of mental retardation (71.87% vs. 52.94%; P = 0.05) on the pre-operative intelligence quotient (IQ) assessment was the only neuropsychological factor associated significantly with a favorable outcome. Whether the patient had no deficits (85.71% good outcome) on neuropsychological assessment or had unilateral (61.53% good outcome), bilateral (70% good outcome), lobar (63.15% good outcome) or diffuse (70.83% good outcome) deficits did not appear to make any significant difference (P = 0.559 and P = 0.478 for laterality and extent of deficits, respectively) in the outcome after surgery in terms of seizure control. To the best of authors’ knowledge, there is no published data showing the objective measurements of quality of life in Indian epilepsy patients (as a uniform group, as far as pathology is concerned) per se. Thus, the study on QOL for patients suffering from DRE secondary to a single uniform pathology like FCD is also difficult to perform in the current scenario. To what extent control of seizures, either with the use of AEDs or surgery, affects their QOL is another pertinent question that needs to be answered. At the time of tabulating the results of this project, we did not find any literature on the objective comparison of improvement in the QOL among patients suffering from FCD before and after surgery. Using QOLIE-89 method of assessment, we noted a highly significant improvement in the QOL after surgery (final QOLIE-89 score 38.33 vs. 75.21 with an absolute improvement of 49.03%; P < 0.001) in this cohort. Seizure worry (18.73 vs. 86.91; P < 0.01), role limitation due to emotional factors (23.46 vs. 94.90; P < 0.05), overall quality of life (24.92 vs. 84.28; P < 0.05) and role limitation due to physical factors (25 vs. 80; P < 0.05) were, significantly as well as quantitatively, maximally improved factors among the patients. Emotional well-being (34.07 vs. 75.12; P < 0.001), final overall scores of QOLIE-89 ( 38.33 vs. 75.21; P < 0.001) and medication effects (42.51 vs. 78.73; P < 0.001) were the three individual parameters with most significant improvement (but not necessarily showing maximum improvement) after surgery. Energy/fatigue, pain, attention/concentration and language function were minimally and insignificantly improved parameters.
Surgical techniques and the use of intra-operative adjuncts
There has been a well-established favorable correlation between the extent of complete surgical resection and seizure-free outcome. According to this series in which 469 patients with FCD were studied, 60% of patients were seizure-free after 1 year of surgery. Complete resection has been defined,,, as lack of abnormalities in the postoperative MR imaging; resection of EZ confirmed on ECoG; or, obtaining resection margins with a normal histology. Utilization of the invasive monitoring and the age of the patient had no influence on the chances of seizure freedom after surgery. Localization to the temporal lobe and complete resection have been noted in various studies to be positive predictors of seizure free outcome. The extent of resection has been identified by means of studying the sequential brain samples at the resection margin till they show normal tissue. With this method, the authors in some studies have found that 71% patients achieved a seizure-free outcome., Others have found that the seizure semiology, temporal vs. extra-temporal location, intracranial VEEG, post-resection ECoG, and histological findings did not affect the seizure free outcome. Complete resection of FCD, however, was an indicator of a good seizure outcome. When cortical resection was incomplete, multiple sub-pial transection (MST) did not improve outcome in this series. Seizure freedom rates of even 92% have been reported in small and specific subgroups.
A study from the Cleveland Clinic questioned the usefulness of sub-dural electrodes in identifying the ictal onset zone; thus, it also questioned the extent of help the procedure offered in performing a complete resection in FCD. Of the 48 patients studied, complete resection led to a 65% seizure free outcome, and incomplete resection led to freedom from seizures in only 8% of the patients. Surgical resection of brain tissue with the intention to cure epilepsy, like any other surgical procedure, is associated with a learning curve for the surgeon. The initial series of FCD patients in one study had Engel's class I outcome of 21% and Engel's class II outcome of 22% only, which improved to nearly 50% in a later series.,
Surgical outcome of right hemispherical lesions (82.6%) tends to be better than the outcome in left hemispherical lesions (59.3%). This difference between the hemispheres may be considered as a marker for the surgical strategy to be conservative on the left side leading to an incomplete resection in the dominant hemisphere, combined with a more liberal approach and extensive and complete resection on the side of the non-dominant hemisphere.,
Patients with Type II FCD (76.92% vs. 62.5%; P = 0.045) were associated with a significantly better outcome, as has been noted in various studies. Presence of dual (57.14% vs. 72.09%; P = 0.415) or double pathology (50% vs. 70.83%; P = 0.514) does appear to affect the outcome in a less favorable manner but this trend lacked statistical significance, possibly due to the presence of an insignificant number of patients with a dual (n = 7) or double pathology (n = 2). As far as location of the dysplastic lesion is concerned, temporal lobe FCDs (79.16% vs. 61.53%; P = 0.005) had a significantly better outcome compared to FCDs affecting other lobes, but not as an independent factor.
Fauser et al., studied 120 patients with FCDs retrospectively and noted that the age at onset of epilepsy was not significantly different for varied lobar distribution of FCD, like the temporal, extra-temporal or multi-lobar involvement. Secondly, they also found that the presence of cyto-architectural abnormalities (FCD Type IB, IIA and IIB) leads to an earlier onset of seizures compared (P = 0.001) to its absence (FCD type IA). They also noted that only in patients with dual pathology (additional presence of hippocampus sclerosis), febrile seizures were more frequent (P = 0.02) compared with those without it. Also, the degree of hippocampal sclerosis was more severe (Wyler grade 3-4) when patients with a dual pathology had febrile seizures in their historical or clinical presentation.
In one of the Indian series of 153 epilepsy patients who underwent surgery with the intention of achieving cure of epilepsy, patients with isolated focal cortical dysplasia constituted 15.11% of histological specimens. This figure was next to medial temporal sclerosis (MTS) [24.8%] and tumours (19.6%) when counted for isolated pathologies. However, majority of the patients with a dual pathology (which constitutes 20.9%, n = 32 cases) had FCDs with either a ganglioglioma (n = 15) or a dysembryoplastic neuroectodermal tumor (DNET; n = 12). It gives an impression that either in isolation or when manifesting a dual pathology, patients with FCD constitute approximately 36% of patients with DRE who are being operated upon for their epilepsy. Another study involving 57 cases of FCDs reported isolated (49.12%) and dual pathologies (50.67%) in their series. At a follow up of 3 years (with three patients lost to follow up), 51% patients had Engel class I and 26% had class II outcome.
Patients who suffered from an acute post-operative seizure (APOS) within 24 hours of surgery appeared to have a poor long-term outcome. In the present series, an APOS was the single most important and significant independent factor in the identification of a surgical failure or a poor outcome (87.5% vs. 19%; P < 0.001) on a long-term basis. According to one study from Mayo clinic, the incidence of APOS within 7 days of anterior temporal lobectomy, that was performed as a treatment for intractable seizures, was 20% among 160 patients. The patients in that series who suffered from APOS had a lower rate of a favorable outcome compared to those patients who suffered from no such event (62.5% vs 83.6%, P < 0.05). Our findings matched with the published data although the incidence of APOS was assessed within 24 hours (and not for 7 days) in our study; moreover, FCD was the uniform etiological substrate in our study. We noted APOS in approximately 15% of patients in this cohort. In that sense, hyper-acute post-operative seizures (HPOS) might be the appropriate expression for the events noted in our series.
In the present series, among patients with DRE and FCD who underwent surgical intervention at an early age, those patients with a lower pre-operative score and focal discharges without propagation on ictal VEEG had a significantly better outcome. Absence of early post-operative seizures within 24 hours after surgery and the presence of Type II FCD were significant independent variables that predicted the attainment of Engel IA outcome. Surgery improved the quality of life in these patients significantly with almost 50% improvement in the overall QOLIE-89 scores after surgery. The parameters of quality of life that improved significantly were seizure worry, emotional well being, medication effects, the role limitation due to emotional factors and the overall quality of life. Based upon our study, we conclude that an early age referral to the epilepsy surgery centers has the potential for achieving a better cure of epilepsy in these patients. This fact gives these patients a chance to lead a life with better quality and acceptance in the society. The current series illustrates that epilepsy surgery centers in India have the capability and potential to evaluate and treat patients suffering from DRE and FCD and may achieve success rates that are comparable to that of established epilepsy centers world-wide.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8], [Table 9]