Frontal lobe epileptic seizures are accompanied by elevated pitch during verbal communication

The objective of our study was to assess alterations in speech as a possible localizing sign in frontal lobe epilepsy. Ictal speech was analyzed in 18 patients with frontal lobe epilepsy (FLE) during seizures and in the interictal period. Matched identical words were analyzed regarding alterations in fundamental frequency (?o) as an approximation of pitch. In patients with FLE, ?o of ictal utterances was significantly higher than ?o in interictal recordings (p = 0.016). Ictal ?o increases occurred in both FLE of right and left seizure origin. In contrast, a matched temporal lobe epilepsy (TLE) group showed less pronounced increases in ?o, and only in patients with right‐sided seizure foci. This study for the first time shows significant voice alterations in ictal speech in a cohort of patients with FLE. This may contribute to the localization of the epileptic focus. Increases in ?o were interestingly found in frontal lobe seizures with origin in either hemisphere, suggesting a bilateral involvement to the planning of speech production, in contrast to a more right‐sided lateralization of pitch perception in prosodic processing.     

Anatomic Correlates of Interhippocampal Seizure Propagation Time

The relation between interhippocampal seizure propagation time (IHSPT) and anatomic alterations in the human epileptic hippocampus may provide insight into the pathophysiology of temporal lobe epilepsy (TLE). Using depth electrode recordings, we measured the time required for spontaneous seizures with onset in one hippocampus to become manifest in the contralateral hippocampus in 50 patients who underwent resection of the temporal lobe of seizure origin. Cell densities in individual hippocampal subfields were determined and correlated with mean IHSPT for each patient. Mean IHSPT was significantly and inversely correlated with cell counts in CA4 only (r = -0.38, p less than 0.01, Pearson's product correlation; r = -0.52, p less than 0.001, Spearman's rank order correlation). In 5 patients with bilateral independent hippocampal seizure onset who had temporal lobectomy and a diagnosis of mesial temporal sclerosis, mean IHSPT was consistently longer from the sclerotic temporal lobe than to it. These observations suggest that anatomic changes associated with chronic epilepsy alter propagation patterns. Because CA4 is believed to modulate the output of dentate granule cells and also has commissural connections to the contralateral homotopic area, the association of decreased CA4 cells with prolongation of IHSPT suggests that the observed anatomic alterations may actively (through increased inhibition) or passively (through decreased recruitment) interfere with various routes of seizure propagation.     

Interictal Aggression in Epilepsy: The Buss-Durkee Hostility Inventory

Summary: Adult patients with left, right, or bilateral temporal lobe epilepsy or absence epilepsy, and normal controls completed the Buss-Durkee Hostility Inventory (BDHI), a standardized questionnaire of aggressive tendencies. Patients with left temporal lobe seizure foci scores higher on the Suspición scale than did other patients or controls (p < 0.05). Factor analysis scale scores identified three factors: hostile feelings, covert aggression, and overt aggression. The groups differed on their pattern of factor scores (p < 0.01): patients with left temporal lobe epilepsy scored higher than other groups on hostile feelings, normal controls scored higher on Covert aggression, and bitemporal patients scores higher on Overt aggression. Patients with absence seizures did not differ from controls. Lateralization of the seizure focus in patients with temporal lobe epilepsy may alter expression of aggressive behavior.     

In vivo 31P nuclear magnetic resonance spectroscopy of human temporal lobe epilepsy.

We performed localized 31P nuclear magnetic resonance (NMR) 1H-image-guided in vivo spectroscopy to study regional high-energy phosphate levels in the brains of normal controls and in patients with intractable unilateral temporal lobe epilepsy. We did not observe differences in intracellular pH between controls and patients. The phosphocreatine/inorganic phosphate ratio was reduced by 50% in the epileptogenic temporal lobe compared with controls (p less than 0.005) and by 35% when compared with the unaffected contralateral temporal lobe (p less than 0.05). We did not observe differences in the ratio of phosphomonoesters to phosphodiesters between controls and patients. These findings suggest that in vivo 31P NMR spectroscopy yields a distinctive interictal metabolic profile in patients with intractable unilateral temporal lobe epilepsy and may permit noninvasive lateralizing evidence of the seizure focus.     

Hyperreligiosity in temporal lobe epilepsy: redefining the relationship

Seventy-six subjects with the primary diagnosis of complex partial seizures with a unilateral temporal lobe focus were examined for the presence of hyperreligiosity. Fifty-one subjects had a left temporal lobe seizure focus and 25 had a right temporal lobe seizure focus. In addition to the temporal lobe epilepsy (TLE) groups, two reference groups were also examined. The first consisted of 31 subjects with primary generalized seizures. The second control group consisted of 27 subjects with documented pseudoseizures and no objective evidence of recurrent epileptic seizures. In all cases, the documentation of seizures or pseudoseizures as well as the localization of the seizure focus was accomplished through simultaneous videotaped recording of the 16-channel scalp EEG tracing and the concurrent overt behavior during the ictal phase. The results of this study fail to support the hypothesis that individuals with TLE are characteristically hyperreligious during the interictal state. There were no significant group differences between the left vs. the right TLE groups, nor were there any significant group differences between the two TLE groups vs. the two comparison groups.     

Nuclear Magnetic Resonance Imaging, a New Approach to the Investigation of Refractory Temporal Lobe Epilepsy

Nuclear magnetic resonance is a noninvasive technique of imaging the brain using signals produced by magnetic fields and radiowaves. Sixteen patients with refractory temporal lobe epilepsy (TLE) had magnetic resonance imaging (MRI) during investigation for possible surgical management. These images plus those of 10 normal controls, were interpreted blind and the results from the TLE patients compared to the EEG and computerized tomographic (CT) scan findings. Fourteen patients compared with one normal control had one or more of the following: small temporal lobe, small mesial temporal structures, small hemisphere, and focal mesial temporal prolonged spin-spin relaxation time. These findings correlated with the electrographically determined seizure focus in 11 patients. Enlarged or asymmetrical temporal horns were seen with equal frequency in patients and controls and did not correlate with the seizure focus. CT scans were less likely to show these changes than MRI. Correlation of the MRI changes with histopathology of resected temporal lobes was poor. MRI reveals structural and possibly functional alterations which may aid in the localization of the seizure focus in the temporal lobes of patients with refractory complex partial seizures.     

Interictal temporal lobe epileptiform discharges and their relationship to secondarily generalized epilepsy

In a previous study of epileptic phenomena in 19 patients with partial complex seizures, it was noted that seizures of left temporal lobe origin had a higher incidence of secondary generalization. To evaluate this observation further, we retrospectively reviewed reports of EEGs for evidence of focal interictal epileptiform discharges (FIED) of temporal lobe origin and correlated this finding with seizure type. Of 3,276 EEG reports reviewed, 195 showed FIED. The medical records of these patients were reviewed and 79 had sufficient information therein to enable seizure classification. Of 79 patients, 61 had secondary generalized seizures, 45 with left temporal FIED, 16 with right FIED. Of 79 patients, 18 had partial seizures, 13 with right temporal FIED, 5 with left FIED (p less than 0.001). These findings suggest that seizures of left temporal lobe origin may have a higher incidence of secondary generalization. The significance of this observation and its relevance to medical and surgical treatment of complex partial seizures is discussed.     

Patterns of cerebral glucose metabolism in patients with partial seizures

We analyzed local cerebral metabolic rates of glucose (LCMRglu) in 20 regions from 22 patients with complex partial seizures, who were taking neither phenytoin nor phenobarbital and who had normal CTs. Results were compared with data from 19 normal controls. Ten patients had left temporal, eight right temporal, and four bitemporal or generalized EEG discharges. There were no significant differences between patient and control values in any of 20 regions of interest. LCMRglu was depressed at the site of the epileptic focus: L/R ratio was 0.85 +/- 0.12 (p less than 0.003 compared with control), 0.92 +/- 0.08 (p less than 0.05), and 0.84 +/- 0.1 (p less than 0.001), respectively, in mesial, superior, and inferior temporal regions for patients with left temporal foci; 1.7 +/- 0.96 (p less than 0.04), 1.1 +/- 0.1 (NS), and 1.15 +/- 0.04 (p less than 0.001) for patients with right temporal foci. Patients with left temporal EEG foci had significantly lower values than patients with right temporal foci in left superior frontal and thalamic as well as left temporal regions, while patients with right-sided EEG foci had depressed LCMRglu (compared with patients with left temporal EEG foci) restricted to right mesial temporal lobe. The patients with left temporal foci tended to have longer seizure histories (22.7 +/- 5.4 versus 11 +/- 5.6 years; p less than 0.001). There was an inverse correlation between length of seizure history and mean LCMRglu (r = 0.38; 0.1 greater than p greater than 0.05). Our study suggests that LCMRglu is not depressed in regions beyond the epileptic focus when patients are not taking drugs known to decrease cerebral glucose metabolism.     

Low incidence of abnormal (18)FDG-PET in children with new-onset partial epilepsy: a prospective study

OBJECTIVE: Patients with refractory partial epilepsy often exhibit regional hypometabolism. It is unknown whether the metabolic abnormalities are present at seizure onset or develop over time. METHODS: The authors studied 40 children within 1 year of their third unprovoked partial seizure with EEG, MRI, and [(18)F]-fluorodeoxyglucose ((18)FDG)-PET (mean age at seizure onset = 5.8 years, range 0.9 to 11.9 years; mean epilepsy duration = 1.1 years, range 0.3 to 2.3 years; mean number of seizures = 30, range 3 to 200). The authors excluded children with abnormal structural MRI, except four with mesial temporal sclerosis and two with subtle hippocampal dysgenesis. (18)FDG-PET was analyzed with a region of interest template. An absolute asymmetry index, [AI], greater than 0.15 was considered abnormal. RESULTS: Thirty-three children had a presumptive temporal lobe focus, five frontotemporal, and two frontal. Mean AI for all regions was not different from 10 normal young adults, even when children less likely to have a temporal focus were excluded. Eight of 40 children (20%) had focal hypometabolism, all restricted to the temporal lobe, especially inferior mesial and inferior lateral regions. Abnormalities were ipsilateral to the presumed temporal lobe ictal focus. CONCLUSIONS: Abnormalities of glucose utilization may be less common and profound in children with new-onset partial seizures than in adults with chronic partial epilepsy. Although these patients' prognosis is uncertain, resolution of epilepsy after three documented seizures is uncommon. If the subjects develop a higher incidence of hypometabolism in the future with planned follow-up studies, metabolic dysfunction may be related to persistent epilepsy rather than present at seizure onset.     

Time from ictal subdural EEG seizure onset to clinical seizure onset: an electrocorticographic time factor associated with temporal lobe epileptogenicity

Long-term subdural video/electroencephalographic (EEG) monitoring was performed in a series of patients with medically intractable complex partial seizures, in a study of diagnostic accuracy, to test the hypothesis that the time from ictal subdural EEG seizure onset to clinical seizure onset (ECOT) is correlated with temporal lobe epileptogenicity and confirm measures of validity of ECOT for predicting seizure-free outcome following anterior temporal lobectomy and amygdalohippocampectomy (ATL/AH). In 34 patients with refractory temporal lobe epilepsy, subdural EEG monitoring localized the ictal epileptogenic focus to a single temporal lobe. In each patient, ECOT was analysed for correlation with temporal lobe epileptogenicity as measured by seizure interval in hours. Patients in whom ECOT was equal to or less than the mean (i.e. subdural EEG seizure onset preceding clinical seizure onset by at least 11.7 seconds) had a significantly greater likelihood of becoming seizure-free following ATL/AH compared to patients in whom ECOT was greater than the mean (i.e. subdural EEG seizure onset preceding clinical seizure onset by less than 11.7 seconds) (x(2) = 5.78, p<0.05). The validity of ECOT for predicting seizure-free outcome following ATL/AH is confirmed to have sensitivity of 55.0%, specificity of 85.7%, false positive rate of 15.4%, false negative rate of 42.9%, diagnostic value of 84.6% and diagnostic accuracy of 67.6%. In addition, a significant correlation, described by a second order polynomial relationship, was found between the natural exponential function of ECOT and seizure interval [f(x=0.415x(2) -25.554x + 267.036, r= 0.731, df= 32, t =6.05, p<0.001, where f(x)=e(ECOT) and x= seizure interval). This result provides the epileptologist with a quantitative tool capable of predicting seizure interval based on ECOT. The capability of ECOT to predict seizure interval may allow the patient and epileptologist to anticipate future seizure onset based on ECOT, potentially facilitating accurate timing of ictal seizure focus localization techniques and clinical intervention to abort seizure onset using various available central and peripheral nervous system stimulation therapeutic strategies. The results suggest a relationship between ECOT and seizure interval. Fundamental pathophysiologic processes involved in the transition from ictal EEG to clinical seizure onset may be responsible for temporal lobe epileptogenicity.     

Combined depth and subdural electrode investigation in uncontrolled epilepsy

We used both depth and subdural electrodes to obtain localization of the seizure focus in 47 medically refractory epileptic patients. Seizures were localized in 33 patients. Onset was consistently localized by the depth electrodes in 23 patients, was variable or simultaneous in depth and subdural electrodes in 6 (in the same lobe), and was consistently localized to subdural electrodes in 4. All patients localized with subdural electrodes were extratemporal and 3 of the 4 had lesions on imaging studies which helped guide location of electrode placement. Eighty-seven percent of temporal lobe seizures began in hippocampus (recorded by the depth electrode), and 80% were eventually propagated to the ipsilateral temporal neocortex (recorded by the subdural electrode). In 8 patients with bilateral temporal depth and subdural recording, seizures never spread to the contralateral neocortex before the ipsilateral neocortex. Subdural electrodes were 20% less sensitive than depth electrodes in detection of seizures beginning in hippocampus but were accurate when lateralized. Variable or simultaneous unilateral neocortical versus hippocampal temporal lobe seizure onset, determined by the combined study, was significantly correlated with less favorable seizure control after anteromedial temporal lobectomy and hippocampectomy.     

Acute postoperative seizures after frontal lobe cortical resection for intractable partial epilepsy

PURPOSE: To evaluate the incidence and prognostic importance of acute postoperative seizures (APOSs) occurring in the first week after a focal corticectomy in patients with partial epilepsy of frontal lobe origin. METHODS: We retrospectively evaluated 65 patients who underwent a frontal lobe cortical resection for intractable partial epilepsy between April 1987 and December 2000. All patients were followed up for a minimum of 1 year after surgery. RESULTS: APOSs occurred in 17 (26%) patients. None of the following factors was shown to be significantly associated with the occurrence of APOSs: gender, duration of epilepsy, etiology for seizure disorder, use of subdural or depth electrodes, surgical pathology, or postoperative risk factor for seizures. Patients with APOSs were older at seizure onset and at the time of surgery (p = 0.003 and p = 0.05, respectively). At last follow-up, patients who had APOSs had a seizure-free outcome similar to that of individuals without APOSs (47.1% vs. 50.0%; p = 0.77). Patients with APOSs appeared less likely to have a favorable outcome [i.e., fewer than three seizures per year and >95% decrease in seizure activity (58.8 vs. 70.8%; p = 0.35)]. This result may not have reached statistical significance because of the sample size. No evidence suggested that precipitating factors or the timing of APOSs was an important prognostic factor. CONCLUSIONS: The presence of APOSs after frontal lobe surgery for intractable epilepsy does not preclude a significant reduction in seizure tendency. These findings may be useful in counseling patients who undergo surgical treatment for frontal lobe epilepsy.     

Significance of chronic epilepsy in glial tumors and correlation with surgical strategies.

PURPOSE: This study was designed to compare the frequency of postoperative epilepsy in patients with acute and chronic pre-operative epilepsy and with frontal or temporal lobe glial tumors based on the hypothesis that patients with chronic epilepsy do worse. METHODS: We compared the clinical and diagnostic characteristics of the patients (n = 73) who had seizures preoperatively with those of the patients (n = 153) who did not. Among those who have had seizures preoperatively, we compared those (n = 32, chronic seizure group) who had seizures a year or more prior to surgery with those (n = 41, acute seizure group) who had seizures less than a year prior to surgery. RESULTS: Among the various factors, the frequency of benign pathology and favorable neurological state were higher in the seizure group than in the non-seizure group (p < 0.05). Complex partial seizures and low-grade tumors were frequent in the chronic seizure group, whereas simple partial seizures and high-grade tumors were frequent in the acute seizure group. Seizure-free rate was significantly higher in the acute seizure group than in the chronic group (p < 0.05). Also, the difference of seizure control rate between surgical strategies was statistically significant (p < 0.05). CONCLUSION: This study indicates that preoperative seizure duration and frequency have a close relationship with the frequency of postoperative epilepsy in patients with glial tumors. A longer duration may allow the formation of epileptogenic foci, leading to chronic epilepsy, and eventually have a negative effect on the prognosis of the patients. Factors including histopathological characteristics of the tumor, its location, seizure duration/frequency, and symptomatology should be taken into account when deciding on surgical strategies.     

Localizing Value of Seizure Manifestations of Temporal Lobe Epilepsies and the Consequence of Analyzing Their Sequential Appearance

We investigated the localizing and lateralizing value of principal seizure manifestations in temporal lobe epilepsies (signal symptoms, oroalimentary automatisms, somatomotor manifestations, unilateral dystonic posturing, ictal speech, motionless stare) of 223 complex partial seizures in 50 patients. All the patients had invasive long-term monitoring with the combined implantation of intracerebral electrodes in and subdural electrodes on the bilateral temporal lobes. Postoperative freedom from seizures was ascertained for longer than one year. We found that 35 patients had amygdalohippocampal seizures and 15 had lateral temporal seizures. The value of the manifestations was established in relation to the site and side of seizure origin and to the progression of seizure discharges within the unilateral temporal lobe or to the contralateral cerebral hemisphere. Several signs among the manifestations were found to be reliable in predicting the site or side of the temporal lobe seizure focus. We emphasized the importance of investigating sequential changes of seizure manifestations in relation to ictal EEG findings by means of simultaneous recording.     

Day/night patterns of focal seizures

PURPOSE: In many patients with epilepsy seizures occur with a day/night pattern. Our aims were to compare day/night patterns in seizure frequency among patients with different epileptogenic regions. METHODS: We analyzed video-EEG recordings in 15 patients with temporal lobe (TLE) and 11 with extratemporal lobe epilepsy (XTLE). Each seizure was classified according to subject group (TLE vs XTLE), sleep/wake state, and time of day of seizure occurrence (grouped into 6 x 4-hour "bins"). RESULTS: Of 90 seizures, 41 occurred in TLE and 49 in XTLE patients. There were day/night patterns of seizure occurrence in each group, with differences in the patterns between groups. In TLE, 50% of seizures occurred between the hours of 15:00 and 19:00 (17% would be expected by chance in each 4-hour "bin": F=3.59, P<0.006). In XTLE, there was a peak between 19:00 and 23:00 (47%: F=4.72, P<0.0018). The effect of time on seizures was least pronounced in the XTLE patients who had more than one epileptogenic region. The proportion of seizures occurring from sleep was significantly less in TLE (19%) than in XTLE patients (41%) [P<0.04, Fisher's exact test]. CONCLUSIONS: There are clear day/night patterns of seizure occurrence in epilepsy, with differences in the patterns between TLE and XTLE. There is an additional interaction with sleep/wake state, with relatively few seizures occurring from sleep in TLE compared with XTLE. Thus, it appears that both sleep/wake state and day/night or circadian rhythms may affect seizure proclivity, with different effects depending on the location of the epileptogenic region.     

Temporal Neocorticectomy in Management of Intractable Epilepsy: Long-Term Outcome and Predictive Factors

We report the results of a long-term follow-up study of 50 patients who underwent removal of temporal neocortex with preservation of deeper limbic structures as surgical therapy for intractable temporal lobe epilepsy. The follow-up period ranged from 3 to 15 years. Preoperative EEG investigations were based on interictal discharges alone. Three factors were predictive of a good outcome: (a) A clear unilateral anterior-midtemporal focus (p less than 0.01), (b) stereotypical onset of temporal lobe seizure (p less than 0.005), and (c) greater volume of tissue removed at operation (p less than 0.05). Overall results showed that 62% of patients experienced an outcome of "cure" or "almost cure," as classified according to a modified version of Crandall's criteria (Crandall's I and II). Those who experienced a significant reduction in seizures but who continued to have intractable epilepsy (Crandall's III) were not considered to have had a good result. Overall outcome compares favorably with other that of centers using different surgical approaches and indicates that neocorticectomy is a suitable procedure in a highly selected population even when limited resources are available.     

Lateralizing value of behavioral arrest in patients with temporal lobe epilepsy

Analysis of ictal semiology is essential to presurgical evaluation of patients with epilepsy. To assess the localizing value of behavioral arrest in temporal lobe epilepsy (TLE), we analyzed 107 video/EEG monitoring-documented seizures of 107 adult patients with TLE for a set of defined seizure phenomena with respect to frequency and sequence of occurrence in relation to epileptogenic (mesial vs extramesial, left vs right) origin. Behavioral arrest was observed more frequently in left-sided temporal seizures: 25.7% of left-sided mesial seizures and 25.0% of left-sided extramesial seizures exhibited behavioral arrest, whereas only 3.4% of right-sided mesial seizures and 10.5% of right-sided extramesial seizures were associated with behavioral arrest. In addition, occurence of behavioral arrest within the sequence of seizure phenomena was remarkably consistent, being observed mainly as the first apparent feature of seizure onset. Thus, behavioral arrest is a valuable early indicator of a left-sided temporal epileptogenic focus in adult patients with TLE.     

Absence of simple partial seizure in temporal lobe epilepsy: its diagnostic and prognostic significance

The diagnostic and prognostic significance of the absence of simple partial seizures (SPS) immediately preceding complex partial seizures (CPS) was examined in patients with temporal lobe epilepsy. The status of self-reported SPS in 193 patients with temporal lobe epilepsy who had surgical therapy more than 2 years ago was reviewed. Before surgery, 37 patients never experienced SPS before CPS (Group A), 156 patients either always or occasionally had SPS before CPS (Group B). The frequency of mesial temporal sclerosis (MTS) was lower and the age at onset of epilepsy was higher in Group A. The seizure focus was in the language-dominant temporal lobe in 73% of the cases in Group A, compared with 40% in Group B. The surgical outcome did not differ between the two groups. The findings suggest that temporal lobe seizures without preceding SPS tend to originate in the language-dominant temporal lobe that contains a pathologic etiology other than MTS, especially in the lateral temporal lobe. The surgical outcome in patients without SPS is similar to that in patients with SPS.     

Role of the frontal lobes in the propagation of mesial temporal lobe seizures.

The depth ictal electroencephalographic (EEG) propagation sequence accompanying 78 complex partial seizures of mesial temporal origin was reviewed in 24 patients (15 from the University of Pittsburgh Epilepsy Center and 9 from UCLA). All patients were monitored with bilateral mesial frontal and mesial temporal depth electrodes and later received anterior temporal lobectomy. Ictal EEG records were categorized according to sequence of spread from the temporal focus to the other regions. Although propagation patterns varied both within and between patients, certain features were notable: (a) It was very common for seizure activity to spread initially to the ipsilateral frontal lobe (observed in 22 of 24 patients). (b) The most common mode of spread (15 of 24 patients) was initiating temporal lobe----ipsilateral frontal lobe----contralateral frontal lobe----contralateral temporal lobe. (c) Occasionally, seizure discharges invaded the frontal lobes but failed to invade the contralateral temporal lobe (2 of 24 patients). (d) Seizure activity occasionally invaded the contralateral temporal lobe prior to invading the frontal lobes (2 of 24 patients). Other notable features included (i) a clear tendency for mesial temporal seizure discharges initially to invade orbitofrontal (as opposed to anterior cingulate) cortex and (ii) the emergence of a period of clear asymmetry in the frontal lobes during which high-amplitude, rapid discharges were present on the side ipsilateral to the initiating temporal lobe. These results suggest that the prefrontal region, especially the orbitofrontal cortex, is strongly influenced by mesial temporal ictal activity. This region appears to be frequently involved in the propagation of seizures initiated in the mesial temporal lobe and may play a role in the interhemispheric propagation of mesial temporal seizures.     

Long term outcome of temporal lobe epilepsy surgery: analyses of 140 consecutive patients

OBJECTIVE: To analyse the long term results of temporal lobe epilepsy surgery in a national epilepsy surgery centre for adults, and to evaluate preoperative factors predicting a good postoperative outcome on long term follow up. METHODS: Longitudinal follow up of 140 consecutive adult patients operated on for drug resistant temporal lobe epilepsy. RESULTS: 46% of patients with unilateral temporal lobe epilepsy became seizure-free, 10% had only postoperative auras, and 15% had rare seizures on follow up for (mean (SD)) 5.4 (2.6) years, range 0.25 to 10.5 years. The best outcome was after introduction of a standardised magnetic resonance (MR) imaging protocol (1993-99): in unilateral temporal lobe epilepsy, 52% of patients became seizure-free, 7% had only postoperative auras, and 17% had rare seizures (median follow up 3.8 years, range 0.25 to 6.5 years); in palliative cases (incomplete removal of focus), a reduction in seizures of at least 80% was achieved in 71% of cases (median follow up 3.1 years, range 1.1 to 6.8 years). Most seizure relapses (86%) occurred within one year of the operation, and outcome at one year did not differ from the long term outcome. Unilateral hippocampal atrophy with or without temporal cortical atrophy on qualitative MR imaging (p < 0.001, odds ratio (OR) 5.2, 95% confidence interval (CI) 2.0 to 13.7), other unitemporal structural lesions on qualitative MR imaging (p < or = 0.001, OR 6.9, 95% CI 2.2 to 21.5), onset of epilepsy before the age of five years (p < 0.05, OR 2.9, 95% CI 1.2 to 7.2), and focal seizures with ictal impairment of consciousness and focal ictal EEG as a predominant seizure type (p < 0.05, OR 3.4, 95% CI 1.2 to 9.1) predicted Engel I-II outcome. Hippocampal volume reduction of at least 1 SD from the mean of controls on the side of the seizure onset (p < 0.05, OR 3.1, 95% CI 1.1 to 9.2) also predicted Engel I-II outcome. CONCLUSIONS: Outcome at one year postoperatively is highly predictive of long term outcome after temporal lobe epilepsy surgery. Unitemporal MR imaging abnormalities, early onset of epilepsy, and seizure type predominance are factors associated with good postoperative outcome.