While there are several techniques for temporal lobectomy, the main variant between procedures is the extent of the medial temporal and lateral temporal resections. We prefer the methodology initially described by Spencer and modified for children, which involves an anterolateral cortical resection, followed by a medial resection, (e.g.), a two-step procedure. Patients are placed supine on the table with a roll under the ipsilateral shoulder. The head is placed in 3-point fixation, and positioned parallel to the floor with the vertex downward 10-15 degrees. A question mark or reverse-question mark shaped incision is made, beginning inferiorly at the zygoma just anterior to the tragus, and carried superiorly just over the superior temporal ridge. The musculature is split in a "T" shape leaving a cuff of tissue at the superior temporal ridge for later reapproximation. The craniotomy should permit the maximal amount of exposure anteriorly and inferiorly, so that access to the anterior temporal lobe and the floor of the middle fossa is facilitated. Either a U-shaped dural incision based inferiorly or a C-shaped dural incision based anteriorly is performed and an anterior temporal resection is performed that extends up to 5 cm on the non-dominant side, and to 4 cm on the dominant hemisphere, measured along the middle temporal gyrus from the temporal pole. The resection is performed along the superior margin of the middle temporal gyrus and extended anteriorly across the superior temporal gyrus anterior to the sphenoid wing. The resection is deepened posteriorly until entry into the temporal horn of the lateral ventricle, which is then protected using a cottonoid pledget to block access of blood into the ventricular system. The anterolateral cortex is then resected using an ultrasonic aspirator. Once the lateral neocortex is removed, further aspiration is performed on the lateral part of the parahippocampal gyrus to simplify the second step.

The second portion of the procedure consists of the resection of the medial structures and is often performed using microscopic magnification. By unroofing the temporal horn anteriorly, the hippocampal complex is identified. The choroidal fissure and the attachments between the hippocampus and the amygdala are incised, maintaining the integrity of the arachnoidal border between the superior and inferior compartments. The hippocampus is then elevated medial-to-lateral and anterior-to-posterior, coagulating and taking sharply the feeding vessels from the anterior choroidal and posterior cerebral arteries as they pass through the pia arachnoid into the hippocampal complex. Care must be taken because these hippocampal or Ammon’s horn vessels can be stretched. Damage to the parent artery will result in an ischemic complication to the corticospinal tracts or internal capsule. Once mobilized, the hippocampus can be resected posteriorly through the tail at the level of the tectum of the brain stem.

Two other methods for temporal lobectomy have been described. The en bloc, or one-stage removal maintains the anatomic relationships between the structures, and was advocated to study the pathophysiologic and anatomic abnormalities of the medial temporal structures. The procedure consists of a subpial dissection along the Sylvian fissure above the superior temporal gyrus and sectioning medially through the isthmus of the temporal lobe. The dissection is then carried inferiorly to identify the medial structures, which are reflected laterally along the entire specimen. The anterior temporal lobe and hippocampus can then be removed as one unit.

The tailored temporal lobectomy, as advocated by Silbergeld and Ojemann, tailors the resection of the lateral neocortex to avoid language areas. With this approach, the resection is based on the patient’s interictal epileptiform activity and the location of eloquent cortical tissue that is identified intraoperatively in awake patients using cortical stimulations. Once the mapping is completed, the anterior temporal lobectomy is performed to avoid functional tissue and to resect abnormal areas. The medial portion, including the hippocampal complex, is removed as necessary depending on the preoperative and intraoperative studies. This method is difficult to perform in children under the age of 15 years, but may be useful in older children in the dominant temporal lobe due to their inability to cooperate in the operative setting. The authors have performed the awake craniotomy in children as young as 11 years old but in a child who was particularly mature. While he tolerated it well, overall, it is not recommended except in individual cases.

Major morbidity is rare with temporal lobectomies, and the mortality rate is less than 0.5%. Classic complications include:

1. homonymous superior quadrantanopsia (secondary to interruption of the optic tract and radiations during the posterior resection of the temporal lobe);
2. language or verbal deficits (occur when language areas of the dominant hemisphere are resected--this is often not a problem in children less than 9 years of age because language is not developmentally complete until 10 years of age);
3. manipulation hemiplegia (likely due to spasm of the Sylvian vessels in en bloc resections -injury of the perforating vessels of the internal capsule with anterior choroidal artery manipulation may induce stroke or ischemia to the posterior limb of the internal capsule, which also can lead to motor deficits); and
4. infection and postoperative meningitis, both infectious and noninfectious, have also been known to occur following temporal lobectomy.

The optimal outcome for epilepsy surgery occurs when there is specific localization of a seizure focus. Seizure-free or marked reduction in seizure outcome rates in children who underwent temporal lobectomy range from 78-87%. While there is little evidence of different outcomes with the different approaches in adults, preoperative selection factors likely contribute most to good outcomes. With temporal lobectomies in children, there are often improved neuropsychological outcomes, in conjunction with improved seizure control and resultant decreased medication requirements. Improvements in intellectual function and psychosocial adjustment are observed in most patients with a reduction or elimination of seizures following surgery. This is especially true in children in whom in one study, there was on average a 10-point gain of full-score IQ scoring and rehabilitation to an age-appropriate psychosocial development and where the best outcome was observed when surgery was performed in adolescence or even earlier.