Bilateral stereotactic radiofrequency amygdalohippocampectomy for a patient with bilateral temporal lobe epilepsy



To search for a method for treatment of bilateral temporal lobe epilepsy (BTLE), we report one patient with BTLE experienced bilateral stereotactic radiofrequency amygdalohippocampectomy (SAHE). Neuropsychological examinations were performed before and 5 days, and 6, 18, and 48 months after operation. No seizure occurred in the follow-up time, and no long-term memory and intelligence deficits were found except for a transient decline of the scores immediately after operation. Because severe damage of memory could be caused by bilateral resection surgery, bilateral SAHE should be considered as a possible approach for the treatment of BTLE. However, further studies with accumulation of cases are needed, especially in the detailed assessment of neuropsychological function.

About 20–35% of patients with temporal lobe epilepsy (TLE) have bilateral discharges (Cukiert et al., 2000), and in at least 10–20% of them, bilateral temporal lobe epilepsy (BTLE) has been confirmed (Spencer et al., 1990). If patients were proved as BTLE, bilateral resection would results in the complication of severe dysmnesia (Scoville & Milner, 2000). Vagus nerve stimulation (VNS) could reduce seizure frequency, but few cases became seizure-free (Alsaadi et al., 2001). Deep brain stimulation (DBS) in the bilateral anterior thalamic nucleus has been performed without neuropsychological damage, but the effect of seizure control was unstable (Oh et al., 2012). Thermocoagulation of epileptic foci in the bilateral medial temporal lobe is not a new technique. As reported in the 1950–1970s, bilateral coagulation confined to the amygdala was safe according to simple neuropsychological studies (Parrent & Lozano, 2000). However, because of lack of advanced techniques and equipment, some severe complications occurred, especially in the coagulation of bilateral hippocampus (Cheung & Chan, 2003).

With the development of magnetic resonance imaging (MRI) and stereo-electroencephalography (SEEG)–guided thermocoagulation, stereotactic radiofrequency amygdalohippocampectomy (SAHE) underwent an resurgence in Canada (Blume et al., 1997; Parrent & Blume, 1999), France (Catenoix et al., 2008), and the Czech Republic (Malikova et al., 2011). A 34-year-old woman with bilateral epileptic foci was reported without change of seizure frequency after 18 coagulations in the left entorhinal cortex, inferior temporal, and parahippocampal gyrus (Catenoix et al., 2008). There was no bilateral coagulation being reported in recent years. A 43-year-old man with BTLE is described in this paper to show how about the effect of seizure control and protection of function with bilateral SAHE under limited coagulation.

Case Report

On April 3, 2008, a 43-year-old man was admitted to the neurosurgery department of Navy General Hospital, Beijing, China, because of paroxysmal attacks of convulsion and loss of consciousness for 41 years intermittently. At the age of 2 years, the patient had one seizure for several seconds without any reason. After 4 years without seizures, the seizures recurred and increased in frequency to 1–2 times per month. The patient's seizures were initially well controlled with use of antiepileptic drugs (AEDs), but later this treatment failed to control the seizures. The seizure frequency increased to 10 times per month in the 3 years before admission. AEDs were increased to 1,200 mg/day of carbamazepine and 3,000 mg/day of valproate. Before the attacks, the patient had an aura first, such as singultus, acid regurgitation, and disturbance, then doing something unconsciousness. During convulsion and loss of consciousness, he was found foaming at his mouth and sometimes with urinary incontinence. No one who witnessed the seizure noted to which side the patient turned his eyes and head during convulsions. The duration of attacks was about 1 min. MRI (T1, T2, and fluid-attenuated inversion recovery [FLAIR]) examination was negative. Long-term 41-channel video scalp EEG showed independent discharges coming from both sides during the interictal period, as no seizure was captured in all four monitoring hours. Dipole source localization showed that the epileptic discharges were accumulated in the bilateral medial temporal lobe. All scores of indexes in the Wechsler Adult Intelligence Scale-Revised (WAIS-R) and Wechsler Memory Scale-Revised (WMS-R) were under the lower limit for the normal population, especially in “Performance IQ” and “Attention.” During 72 h monitoring of invasive SEEG, interictal discharges were coming from the bilateral medial temporal lobe independently, as shown in Fig. 1A. Two attacks were captured. During the first attack, the patient turned his eyes and head to the right side and the rhythmic discharge activities originated from the first two contacts of electrode situated in the medial part of left temporal lobe as shown in Fig. 1B. During another attack, the patient turned his eyes and head to the left side, and the rhythmic discharge activities originated from the first contact of the electrode situated in the medial part of right temporal lobe as shown in Fig. 1C.

Figure 1.

SEEG recordings of the interictal and ictal discharges from both medial temporal lobes independently. The first six lines (LX1–LX6) represent the six contacts of deep electrode implanted into the medial part of left temporal lobe, and the 7th–12th lines (RX1–RX6) represent the six contacts of deep electrode implanted into the medial part of right temporal lobe. (A) Interictal epileptic discharges were from LX1, LX2, and RX1 independently. (B) During the first attack, the patient turned his eyes and head to the right side, and the rhythmic discharge activities originated from LX1 and LX2. (C) During another attack, the patient turned his eyes and head to the left side and the rhythmic discharge activities originated from RX1.

After confirming the diagnosis of BTLE, the patient was advised that his condition was unsuitable for resection surgery. As he was eager to be treated, he was advised to try bilateral SAHE, which is effective for seizure control with less damage of memory, but still has the risk for bad result or memory deficit. Under the agreement of the ethics committee of our hospital, the patient signed the informed consent form for the operation. On April 21, 2008, bilateral SAHE was performed with a transfrontal approach under guidance of intracranial depth electrodes. Lesions (75°C, 120 s) were made with an interval of 3 mm within epileptic zone according to the results of intraoperative recordings. There were six lesions on the left and three on the right. MRI examination was performed 4 days after operation (Fig. 2). Neuropsychological examinations were performed 5 days after operation. All indexes were influenced by coagulations except “Performance IQ” and “Attention.” The memory damage in “Delayed Recall” and “Verbal MQ” was more severe than IQ. No other complications were found and the patient was discharged on April 26, 2008.

Figure 2.

Postoperative MRI examination.

After 18 months of seizure freedom, the patient began to reduce the dose of AEDs gradually. Two and half a years after surgery, AEDs were reduced to 600 mg/day of carbamazepine, and 1,500 mg/day of valproate. Long term 41-channel scalp EEG showed nearly normal, and MRI showed bilateral strip lesions with limited volume passing through the amygdala, hippocampal head, and part of the parahippocampal gyrus. Three years postoperatively, some auras recurred when the dose of valproate was decreased to 1,000 mg/day. Scalp EEG showed sharp and slow waves in the bilateral temporal lobe. After returning the valproate dose to 1,500 mg/day, the auras disappeared. In a further consultation at 4 years, abnormal discharges in the right temporal lobe disappeared, but sharp and slow waves were still evident on the left. Neuropsychological analyses were made at 6, 18, and 48 months postoperatively. The scores were increased compared with those before and immediately after operation.

The index scores of WAIS-R and WMS-R at different times are showed in Table 1.

Table 1. WAIS-R and WMS-R scores before and after operation
 Before operation5 days after operation6 months after operation18 months after operation48 months after operation
Full-scale IQ7772828482
Verbal IQ7968848281
Performance IQ7580828582
Global MQ7367847572
Verbal MQ7861898174
Visual MQ7963827875
Delayed recall7659857972


For BTLE, what we were deeply worried about was the damage of memory after surgery. For safety, we set the target point at the center of amygdala. Although the transoccipital approach is more popular for SAHE, as reported recently, because coagulation should be produced bilaterally, and the volume of lesions must be controlled to spare the hippocampal body for reserving memory and related functions, extension of lesions to the hippocampal body through a transoccipital trajectory was not suitable. In this case, the ictal discharges were coming primarily from the anterior part of hippocampus and its junction with amygdala, as well as parahippocampus gyrus under the target point during SEEG monitoring, so we chose a transfrontal approach passing through the target point to the skull base during operation so as to coagulate the complexes of amygdala, hippocampal head, and part of parahippocampal gyrus together. Except for the coagulation of epileptic focus, the disconnecting of hippocampus might be one of the reasons for the postoperative seizure control. Although AEDs should be taken with the effective dosage, the outcome of seizure control and neuropsychology safety is optimistic in this case.

The Wechsler Memory Scale (WMS) scores were recorded by a specific neuropsychologist. The patient never knew the correct answers after each test. Even so, there was still a little bit of a practice effect for this patient. However, compared with the man named HM suffered from severe dysmnesia after bilateral resection of temporal lobe reported by Scoville and Milner (2000), this patient was lucky. Although some of memory functions were affected transiently 5 days after operation, intelligence and memory in our this patient was finally improved within 6–18 months. The impact of intelligence and memory is related to the acute reaction of bilateral hippocampus at the early postoperative time (Fig. 2), After 6 months, the acute phase faded away and the gradually recovered medial structures of temporal lobes worked well for memory and intelligence.

On the other hand, studies have shown that stimulation of the hippocampus above the threshold for eliciting an afterdischarge on the EEG results in memory impairments (Suthana et al., 2012). Before the operation, the frequently interictal and ictal epileptic discharges in both hippocampi played a role as stimulation that corresponded to the lower neuropsychological scores. Along with the disappearance of discharges postoperatively, the scores increased to a level greater than that before operation.

In general, bilateral SAHE in this patient with BTLE was effective and safe. SAHE should be considered as a possible approach for the treatment of BTLE. The transfrontal approach could avoid the damage to hippocampal body and thereby protect memory function. Intraoperative recording makes the coagulation precisely in epileptic zone. The limited volume of the lesion in both amygdalohippocampi is effective in seizure control without neuropsychological damage. However, because this is only one case, we cannot conclude that bilateral SAHE is the best approach for BTLE. Any conclusions should be made according to further accumulation of cases.


Thanks to my adviser Pro. Zhao Quanjun and all colleagues at the Navy General Hospital of PLA for giving me the help during the clinical work. Thanks to Dr Wu Zhaohui for offering the EEG results of the patient. Thanks to Dr Wang Haiying in the neuropsychological department of our hospital for providing neuropsychological assessment of the patient. Thanks to the staff in the radiology department for offering the images of the patient. We were supported by the Grant Agency of the Science and technical committee of Beijing No. Z111107058811018.


Author Huimin Luo has received support from Author Quanjun Zhao. The remaining authors have no potential conflicts of interest to report. 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.


  • Image of creator

    Huimin Luo is a Neurosurgeon at the Department of Neurosurgery, Navy General Hospital of PLA, Beijing, China.