For the management of temporal lobe epilepsy via the lateral temporal lobe approach, classical anterior temporal lobectomy requires the removal of large chunks of temporal cortex, 6 cm posteriorly from the temporal cusp on the nondominant side and 4.5 cm posteriorly from the temporal cusp on the dominant side, but most temporal lobe epilepsy is caused by medial temporal lobe structures, so classical anterior temporal lobectomy has given way to selective hippocampal amygdala resection. Liu Zhan, Department of Neurosurgery, General Hospital of Ningxia Medical University
The current study shows that hippocampal sclerosis is the main factor causing medial temporal lobe epilepsy. Intraoperative deep electrode monitoring has shown that in patients with medial temporal lobe epilepsy, epileptiform discharges in medial temporal lobe structures are more intensive, and surgical resection of hippocampus and amygdala can result in satisfactory control of medial temporal lobe epilepsy. However, the intraoperative identification of these structures remains a challenge for neurosurgeons.RhotonAL points out that for medial temporal lobe surgery, it is essential for the surgeon to establish the following concepts: 1) the content of the hook gyrus and the extent of surgical resection; 2) the relationship between the hippocampus and the hook gyrus; 3) the relationship between the hook gyrus and the temporal horn of the lateral ventricle; 4) the relationship between the hook gyrus, hippocampus, and amygdala of the temporal lobe 5. the relationship between the hook gyrus and the internal carotid artery, posterior communicating artery, anterior choroidal artery, middle cerebral artery, and Rosenthal’s basilar vein: 6. the role of the choroidal fissure in the temporal horn in temporal lobe surgery; 7. the extent of surgical resection of the amygdala. Due to the deep location and complex anatomical relationships, only by being familiar with the important neurostructural relationships in the medial temporal lobe area and the nerves and vessels encountered in reaching the surgical access to this area can a suitable surgical access be selected to deal with the lesions in this area, improve the surgical success rate and reduce surgical complications.
History of selective hippocampal amygdala resection
The development of selective hippocampal amygdala resection and the evolution of the surgical approach are closely related to the level of understanding of temporal lobe epilepsy, the ability to screen for and localize epileptogenic foci. However, the preoperative determination of the epileptogenic focus was only based on seizure symptomatology and clinical manifestations, and the intraoperative electrical stimulation and gross pathological observation were not sufficient to understand the relationship between seizure symptomatology and brain localization, so epilepsy surgery was basically limited to traumatic epilepsy and seizures caused by lesions in the central region. In the 1930s, EEG began to be used in clinical practice, and preoperative lobe localization and lateralization of the epileptogenic foci could be performed according to the epileptic discharges on EEG, and intraoperative cortical EEG recording was also used to guide the extent of resection. Between 1939 and 1949, Penfield performed temporal lobe resection for temporal lobe epilepsy at the Montreal Neurological Institute in 68 cases, of which only 10 cases had the hook gyrus removed and 2 cases had the parahippocampal gyrus removed. 21.5% of the cases had no significant improvement in seizures and were failures, 25.4% had more than 50% reduction in seizures, and only 27.4% had complete seizure control. During this period, with the accumulation of experience in temporal lobe epilepsy surgery, the expansion of clinicopathologic control analysis, and the intraoperative observation of seizure onset with the help of electrical stimulation of the hippocampus and amygdala, more was learned about temporal lobe epilepsy symptomatology, and the importance of medial temporal lobe structures in temporal lobe epilepsy began to be recognized. In these cases, emphasis was placed on resection of the medial temporal lobe structures, including reoperative removal of the hippocampal structures in cases that had failed in the previous phase, with improved outcomes and control of seizures in some of the reoperated patients. This confirmed the importance of medial temporal lobe structures for temporal lobe epilepsy from surgical practice and introduced the concept of incisural sclerosis, the basic counterpart of current medial temporal lobe sclerosis, as the most common cause of temporal lobe epilepsy. Although the importance of medial temporal lobe structures in temporal lobe epilepsy was demonstrated from several perspectives in the early 1950s, and Niemeyer proposed selective hippocampal amygdala resection via the middle temporal gyrus via the ventricles in 1958 based on clinical and experimental studies at that time, the concept of multiple hippocampal resection was not pursued at that time, and temporal lobectomy with preservation of the hippocampus was more popular. Until the 1980s, temporal lobectomy with limited hippocampal preservation was largely used for temporal lobe epilepsy for various reasons, including: the Montreal Neurological Institute reported the effects of a case of disabling memory impairment after bilateral hippocampal resection. The localization of definitive laterality regarding temporal lobe epilepsy is basically limited to symptomatological analysis and EEG. Cerebral angiography and ventriculography can only respond to larger lesions in the brain and are of limited help in epilepsy surgery, and EEG can only be broadly definitive and definitive of the lobes. Wieser and Yasargil reported the therapeutic value of selective hippocampal amygdala resection via the lateral fissure for temporal lobe epilepsy, systematically analyzed the electrophysiological and clinical manifestations, and proposed the concept of medial temporal lobe epilepsy. After the concept of medial temporal lobe epilepsy syndrome was introduced, the interest in the treatment of temporal lobe epilepsy by selective hippocampal amygdala resection was greatly increased. However, two scholars adopted selective hippocampal amygdala resection mainly based on the results of complex invasive EEG recordings, and the pathological basis was mostly unclear before surgery. The clinical use of MRI has greatly contributed to the development of epilepsy surgery, especially temporal lobe epilepsy surgery, and the clinical use of MRI has enabled the preoperative detection of microscopic lesions associated with temporal lobe epilepsy, such as hippocampal sclerosis, and the preoperative diagnosis of pathologic correlations, such as Medial temporal lobe epilepsy syndrome associated with hippocampal sclerosis, medial temporal lobe epilepsy syndrome associated with focal lesions, etc. Meanwhile, long-range video EEG devices and PET gained great popularity during this period, thus reemerging epilepsy surgery worldwide after the first boom in the 1950s. One can design a surgical plan based on preoperative imaging performance. Because of these many factors, people are no longer reporting cases as vaguely as they once did, but mostly in conjunction with the pathologic basis, allowing for single-patient case reporting. Hippocampal sclerosis-related medial temporal lobe epilepsy syndrome became an established diagnosis preoperatively, and hippocampal sclerosis is gradually becoming a clinical disease entity with medial temporal lobe epilepsy as the main manifestation. The application of micro-neurosurgical techniques and intraoperative navigation has improved the level of surgery, and selective surgery has become more widely used, mainly in hippocampal sclerosis-related medial temporal lobe epilepsy syndromes.
Indications for selective hippocampal amygdala resection
Hippocampal sclerotic medial temporal lobe epilepsy is mostly drug-resistant epilepsy, but the surgical results are good and it is one of the most suitable epilepsy syndromes for surgical treatment, so the screening ability needs to be enhanced. First, recognize the seizureological features of medial temporal lobe epilepsy and do not misdiagnose complex partial seizures as psychogenic seizures. MRI should be routinely performed in patients with epilepsy, including routine whole brain MRI scans, and hippocampal scans, including coronal, sagittal and axial scans, and at least T1, T2 and FAIR sequences. Hippocampal volumetric measurements and mass spectrometry are useful in the diagnosis of hippocampal sclerosis. PET scans are helpful in the definite side. Long-range video EEG monitoring is routinely performed, and preoperative neuropsychological testing is not only valuable for localizing the nociceptive focus, but also as a base value for the assessment of later neurological changes.
Selective hippocampal amygdala resection is indicated for those whose seizures originate from the medial temporal lobe structures. In the early stage, cases were mainly screened based on intracranial electrode recordings, and a significant number of cases were temporal lobe tumors. In recent years, cases of temporal lobe epilepsy due to hippocampal sclerosis have been selected mainly based on MRI scan findings. When carrying out this procedure, the indications should be strictly controlled, mainly selecting temporal lobe epilepsy cases with unilateral or predominantly unilateral hippocampal atrophy: (i) typical temporal lobe originated complex partial seizures; (ii) high-resolution MRI scans showing unilateral hippocampal atrophy; (iii) electrophysiological examination showing interictal and/or ictal epileptic discharges originating from the temporal lobe on the side of hippocampal atrophy shown by MRI.
Selective hippocampal amygdala resection surgical approach
Selective hippocampal amygdala resection not only removes the epileptogenic focus located in the hippocampal amygdala, but also blocks the pathway for seizure discharges to diffuse through the olfactory cortex and hippocampal paracentral gyrus, thus achieving epilepsy control. At present, the main approaches for selective hippocampal amygdala resection are the trans-lateral fissure-insula approach, the lateral temporal gyrus and sulcus approach, and the inferior temporal approach.
The trans-lateral fissure-insula approach, proposed by Yasargil, uses a conventional pterygoid approach with the head turned 35 degrees to the opposite side and the longitudinal axis slightly lower by 15 degrees to keep the lateral fissure at the highest point in the center of the operator’s visual field. A four-hole rhomboid formation craniotomy or free bone flap craniotomy can be used, and the lower edge of the bone window need not be too low to expose the lateral fissure area.
The main branches of the carotid artery are dissected, and the lateral fissure arachnoid is incised in front of the Sylvian vein to expose the internal carotid artery, anterior cerebral artery, and middle cerebral artery from the inside out, and after separating the lateral fissure, the main branches of the communicating artery, anterior choroidal artery, middle cerebral artery, and the anterior 1/3 of the insula are exposed. The lateral temporal pole artery and anterior temporal artery of the M1 segment were identified, and the superior temporal gyrus was incised between them at a distance of 1.5-2 cm near the insula, a few millimeters below which corresponds to the amygdala. The incision was first made deeper inward and downward to reach the tip of the lateral temporal horn, and after finding the temporal horn, the ventricular wall was incised 2 cm backward and the hippocampus and amygdala structures were identified within the temporal horn.
The lateral part of the amygdala was first excised in pieces, and the medial hippocampus was incised along the choroidal fissure, taking care to protect the trunk of the anterior choroidal artery and the optic tract, and the lateral peduncle was incised along the foot of the hippocampus, from the tip of the temporal horn backward to the lateral fossa triangle, and the branch of the inferior temporal artery originating from the posterior cerebral artery and the hippocampal artery originating from the posterior cerebral artery at a right angle were cut by electrocoagulation at the equivalent of the posterior cerebral artery P3. The hippocampus is finally removed at the level of the posterior edge of the cerebral peduncle, that is, the lateral geniculate body.
In this approach, the following points should be noted: ① The position of the entry into the temporal horn. This access requires exposing the inferior insular sulcus, the M2 segment of the middle cerebral artery and its branches after opening the lateral fissure. Sometimes the inferior trunk of the middle cerebral artery passes through the inferior insular sulcus and requires nudging the inferior trunk, or choosing an incision closer to the temporal lobe, with the incision positioned between the temporal pole artery and the anterior temporal artery, just lateral to the inferior insular vein behind the insular threshold. Do not go too close to the insula threshold, otherwise the hook bundle may be damaged and the fiber connection between the temporal lobe and the frontal-orbital gyrus may be damaged; ② the direction of the incision. If the incision is too close to the medial side, it may damage important structures such as the thalamus and basal ganglia, and if the incision is too lateral, it is easy to cut the lateral wall of the temporal horn instead of entering the temporal horn, increasing the chance of damaging the optic radiation. Campero et al. believe that the incision should be made in the anterior part of the inferior insular cricoid sulcus, with the incision direction being inferior-internal and at an angle of 45 degrees to the midpoint of the lateral wall of the cavernous sinus, but the position of the cavernous sinus cannot be judged intraoperatively and the angle is not easy to grasp However, intraoperative puncture can be used to locate the temporal angle, and the application of navigation technique will be more accurate; ③ the choroidal fissure should be opened on the side of the umbilical cord, not on the side of the thalamus, otherwise the drainage veins of the thalamus, the anterior choroidal artery and the posterior lateral choroidal artery will be damaged; ④ the subchondral aspiration hook, do not cross the height of the optic tract, so as not to damage the pallidum.
The trans-lateral fissure hippocampal amygdala resection is technically demanding, requiring proficiency in medial temporal lobe dissection, and requires resection in blocks. Compared with the transcortical approach and anterior temporal lobectomy, the intellectual recovery is superior to the latter, but there is no significant superiority in the recovery of cognitive function. Compared to the transtemporal inferior approach, it reduces the chance of strain on the temporal lobe and venous injury, and can provide a surgical pathway to reach the mid-cerebellar tangential area. Compared to the lateral transtemporal sulcus and gyrus, maximum temporal lobe cortex function is preserved. Because a portion of the temporal trunk will be incised, there may be damage to the optic radiations located at the apex of the temporal horn, resulting in quadrant blindness. Vajkoczy et al. performed 32 hippocampal amygdala resections via this approach and only 1 case of superior quadrant blindness occurred. yasargil et al. reported an even lower incidence of visual field defects, with only 2 cases of superior quadrant visual field defects in 102 procedures, and the analysis was related to the occurrence of postoperative hematoma in the operative area. To reduce damage to the optic radiation, Choi et al. found a safe triangle at the base of the lateral fissure, which is located between Meyer’s collaterals and the optic tract, with the tip of the triangle in the lateral geniculate and the base of the triangle in the amygdala, at the level of the insular threshold or within 5 mm of the inferior insular sulcus, from which there is little risk of damage to the optic radiation by entering the temporal horn. Another location to avoid damage to the optic radiations is to incise the polar plane medial to the insula threshold and remove the lower and lateral parts of the amygdala into the temporal horn. In fact, both of these incisions are more anterior, entering from the anterior wall of the temporal horn.
The lateral temporal lobe approach includes a cortical approach via superior temporal gyrus, middle temporal gyrus or inferior temporal gyrus, and a sulcus approach via superior temporal sulcus and inferior temporal sulcus, which differ in the degree of cortical excision and the passage through the sulcus and sulcus.
The advantages of this approach are that temporal lobectomy provides a better window for the whole hippocampus and its adjacent structures to be removed, clear exposure, easy operation, and lower technical requirements than other surgical approaches, and it can reach more posterior areas while preserving the language function of the lateral temporal cortex. The disadvantages are that the temporal cortex needs to be removed; according to the fiber distribution of the optic radiations, the lateral approach may cause visual field defects, especially the superior temporal gyrus and middle temporal gyrus approaches are more likely to damage the optic radiations; and the distance to reach the temporal horn is far compared with the temporal floor and superior temporal surgical approaches. The following points should be noted during the surgery: (1) the cortices related to memory and language are mostly located in the superior temporal gyrus and middle temporal gyrus, and the use of temporal floor incision through the medial occipitotemporal vein can avoid these functional cortical injuries; (2) to reduce the strain on the temporal cortex during the surgery, because the structure of the middle cranial fossa floor is high on the inside and low on the outside, so the strain on the temporal lobe is heavier when revealing the structures of the medial temporal lobe such as the hook gyrus; (3) to maintain the arachnoid membrane of the brain pool during the surgery The temporal trunk is an important connection between the temporal cortex and other cortices such as the frontal lobe and parietal lobe. If the temporal trunk is not protected during surgery, the neurological function of the temporal lobe will be seriously affected even if the integrity of the temporal cortex is preserved during surgery; ⑤ In this surgical approach, the light of the operating microscope should be projected along the anterior-posterior axis of the head and temporal horn in order to see the structures in the posterior most part of the temporal horn; ⑥ Since the amygdala is not clearly demarcated from the basal ganglia, the sign of superior resection is from the internal jugular artery bifurcation to the inferior chord. (6) Since the amygdala is not clearly demarcated from the basal ganglia, the marker for superior resection is the line from the bifurcation of the internal carotid artery to the inferior choroidal point, i.e., the carotid artery choroidal line or ventrolateral line.
The inferior temporal approach is limited to the anterior hippocampus, amygdala, and parahippocampal gyrus, preserving the coxal gyrus and other parts of the temporal lobe, and is the only procedure that can remove only the medial temporal lobe structures without damaging the rest of the temporal lobe.
The bone window is placed as close as possible to the base of the middle cranial fossa to reduce the strain on the temporal lobe and to enlarge the view through the temporal base. In order to expose the base of the middle cranial fossa more lowly, the zygomatic arch can be removed, the temporalis muscle can be pulled downward, the dura can be cut and the temporal lobe can be lifted from the base of the middle cranial fossa, the hook gyrus can be pulled from underneath the temporal lobe, and the cerebrospinal fluid of the cricoid pool can be aspirated to clearly expose the coxal gyrus, the parahippocampal gyrus, the hook gyrus, the canopy and the cricoid pool. The Labbe vein should be protected at this point. The location of the entry into the temporal horn can be chosen from the lateral parahippocampal and nasal sulci, occipitotemporal sulci, coxal gyrus, and parahippocampal gyrus, but the closer the incision is to the medial side, the more vertical the direction of the track into the ventricle. This approach enters the temporal horn from the temporal floor without the risk of optic radiation injury and reaches the temporal horn at a shorter distance compared to the lateral temporal approach, allowing the management of lesions in the posterior part of the medial temporal lobe. However, the strain on the temporal lobe is heavier, and in the dominant hemisphere, the speech area of the coxal gyrus may be damaged, and there is a risk of damage to the Labbe vein, especially in the dominant hemisphere, which can cause speech impairment.
All of the above surgical approaches emphasize the resection of the medial temporal lobe structures. The author experiences that when removing the medial temporal lobe structures, regardless of the access chosen, it is necessary to go through the temporal horn of the lateral ventricle. It is crucial to be familiar with the anatomical landmarks associated with this. These landmarks include the lateral ventricular lateral sulcus, lateral bulge, hippocampal fissure, choroidal fissure, internal olfactory sulcus, and terminal striae. Theoretically, the trans-lateral fissure approach used by Yasargil et al. has the added advantage of separating the frontal contact fiber tracts with minimal neuronal damage and visual field disturbance, but it undoubtedly requires more surgical skill due to the lack of obvious anatomical landmarks and the interference of the lateral fissure vessels intraoperatively.
Points to note for selective hippocampal amygdala resection
Craniotomy requires precise cranial site. MRI images should be read carefully when designing scalp incisions and bone flaps, paying attention to anatomical landmarks such as the auricle, ear screen, external auditory canal and temporal bone rock in relation to the inferior horn of the lateral ventricle and hippocampus. The application of neuronavigation and open MRI is beneficial for the precise positioning of the craniotomy and cortical incision. Note that the cortical incision should not be too small. Be sure to pay attention to the identification of landmark structures such as the lateral ventricular sulcus, lateral bulge, hippocampal fissure, choroidal fissure, internal olfactory groove, and terminal stripe to prevent disorientation. Strictly use the submural resection technique to avoid damage to the peduncle, optic tract, motoneurotic nerve, and posterior cerebral artery. Patience must be used when using general suction. As long as the surgical indications are selected appropriately, the results of using this procedure for temporal lobe epilepsy are reliable. 01ivier treated 150 cases of temporal lobe epilepsy with this procedure, and the rate of complete control of epilepsy was 81%, and most of them had improved self-perceived memory and quality of life after surgery, and there were no cases of exacerbation of seizures.
Early onset tonic clonic seizures after surgery may be related to cortical damage stimulation during the acute phase. The author’s experience is more likely to be related to the temporary discontinuation of conventional antiepileptic drugs after surgery. Therefore, if there is no obvious vomiting after postoperative anesthesia awakening, the preoperative medication is promptly resumed. In the transitional phase, intramuscular, subcutaneous and intravenous preparations can be applied instead. Early-onset postoperative seizures that are inconsistent with habitual seizure forms, especially tonic-clonic seizures, do not predict poor long-term results of surgery. Regarding postoperative medication, the author prefers to apply carbamazepine at 600-1200 mg daily, mostly 900 mg. An effective follow-up contact pathway must be established with the patient and family after surgery to regularly supervise the patient’s medication and related habits, and only then can good therapeutic results be obtained. In general, if there is no seizure 1 year after surgery, the medication can be considered to be discontinued, but attention should be paid to the gradual and orderly discontinuation of the medication must be guided by the physician.
References
1. Lfiders HO, Comair YG. Epilepsy Surgery. second edition. philadelphia: Lippincott Willianms Wilkins, 2001:606-610.
2.Chen XH. Microanatomy and surgical access to the medial temporal lobe. PhD thesis, Tianjin Medical University. 2009:79-81.
3. Bate H,Eldridge P,Varma T. et a1.The seizure outcome after amygdalohippocampectomy and temporal lobectomy.Eur J Neurol.2007,14(1): 90-94.
4.Briellmann RS,Kalnins RM,Berkovic SF, et al. Hippocampal pathology in refractory temporal lobe epilepsy,T2-weighted signal change reflects dentate gliosis.Neurology,2002,58(2):265-271.
5. Wang F, Sun T, Li GZ, et al. Microanatomical study of the cricoid pool and its surgical access. Chinese Medical Journal,2007,87(3):165-168.
6.Wieser HG,Ortega M,Friedman A,et a1.Long-term seizure outcomes following amygdalohippocampectomy.J Neurosurgery,2003,98(4):75 1-763.
7.Paglioli E,Palmini A,Portuguez M,et a1.Seizure and memory outcome following temporal lobe surgery:selective compared with nonselective approaches for hippocampal sclerosis.J Neurosurg, 2006, 104(1):70-78.
8.Mofino M,Uda T,Naito K,et a1.Comparison of neuropsychological outcomes after selective amygdalohippocampectomy versus anterior temporal Epilepsy Behav,2006,9(1):95-100.
9.Clusmann H,Schramm J,Kral T,et a1.Prognostic factors and outcome after different types ofresection for temporal lobe epilepsy.J Neurosurg,2002, 97(5): 1131-1141.
10.Lutz MT,Chismann H,Elger CE,et a1.NeuropsychologAcal outcome after selective amygdalohippocampectomy with transsylvian versus transcortical approach:a randomized prospective clinical trial of surgery for temporal lobe epilepsy.Epilepsia,2004,45(7):809-816.
11. M. G Yasargil. Microneurosurgery VI B. Ling F. Translation. Beijing: China Science and Technology Publishing, 2005:36-44.
12. Helmstaedter C,Van Roost D,Clusmann H,et a1. Collateral brain damage, a potential source of cognitive impairment after selective surgery for control of mesial temporal lobe epilepsy.J Neurol Neurosurg Psychiatry,2004,75(2):323-326.
13.Moran NF,Lemieux L,Kitchen ND,et a1.Extrahippocampal temporal lobe atrophy in temporal lobe epilepsy and mesial temporal sclerosis.Brain,2001, 124:167-175.
14.Lanerolle NC,Kim JH,Williamson A,et a1.A Retrospective Analysis of Hippocampal Pathology in Human Temporal Lobe Epilepsy:Evidence for A Retrospective Analysis of Hippocampal Pathology in Human Temporal Lobe Epilepsy:Evidence for Distinctive Patient Subcategories.Epilepsia,2003,44(5):677-687.
15.Blumcke I,Pauli E,Clusmann H,et a1.A new clinico-pathological classification system for mesial temporal sclerosis.Acta Neuropathol,2007,512- 514.
16.Wheatley BM. Selective amygdalohippocampectomy: the trans-middle temporal gyrus approach.Neurosurg Focus, 2008,25(3):1-5.
17.Spencer S,Berg AT,Vickrey BG,et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology, 2005,65:912-918.
18.Tanriverdi T,Olivier A,Poulin N,et al. Long-term seizure outcome after mesial temporal lobe epilepsy surgery: corticalamygdalohippocampectomy versus selective amygdalohippocampectomy.J Neurosurg,2008,108:517-524.
19. Tan QF. Experience and lessons learned in the surgical treatment of temporal lobe epilepsy. Chinese Journal of Clinical Neurosurgery, 2006, 8(11):450-451.
20.Helmstaedter C,Richter S,Roske S,et al. Differential effects of temporal pole resection with amygdalohippocampectomy versus selective amygdalohippocampeetomy on material-specific memory in patients with mesial temporal lobe epilepsy.Epilepsia,2008,49:88-97.