8.1 Introduction

Temporal lobe epilepsy (TLE) is the most common type of epilepsy in adult and pediatric populations that is responsive to surgery, as has been evidenced by the results of a class I randomized controlled trial (RCT) ¹ and systematic review of class IV non-RCTs. The number needed to treat (NNT) as a measure of impact in TLE surgery is 2, ² suggesting that the surgical procedures utilized in treating TLE are highly effective. The efficacy of surgical resections in TLE is partially due to the focal nature of the epileptogenic zone and the fact that it involves relatively noneloquent areas of the brain that can be safely resected. These are some reasons why resective surgery plays an important role in the management of patients with refractory mesial temporal lobe epilepsy (mTLE). In the long term, medical therapy controls seizures in only 11% of cases with hippocampal sclerosis associated with mTLE. ³ On the other hand, a postoperative cure or significant improvement might be expected in as many as 80% of subjects undergoing resective procedures. ⁴

Anterior temporal lobectomy (ATL) is the standard surgical technique for temporal lobe seizures. However, selective amygdalohippocampectomy (SAH) has emerged as the procedure of choice in many centers. Over time, SAH has proven to cause less disruption of potentially functional temporal neocortex. Patients obtain the same level of seizure control but with a more favorable neuropsychological profile. ⁵ Surgical resection of the temporal lobe often results in verbal, visual, and cognitive dysfunction, particularly when the dominant hemisphere is involved. ⁶

Although all the different operative techniques have refined or intended to refine Niemeyer’s original approach to the amygdala, hippocampus, and parahippocampus, they are associated with their own disadvantages. ⁷ The optic pathways are injured in as many as 53% of patients undergoing SAH by transsylvian or transcortical routes. ⁸ The conventional transventricular approaches often damage the optic radiation lying along the lateral ventricular wall as well as the superior portion of the lateral temporal neocortex.

SAH is an operative procedure originally developed to spare unaffected brain tissue and minimize the memory deficits after temporal lobe surgery. Some studies have indicated significant verbal memory decline after left-sided selective resections, whereas others have shown no such deficits. Some studies have indicated that subtemporal SAH, even on the language-dominant side, does not cause a significant postoperative decline in verbal memory. ⁹ There have been consistent limitations of these reports: sample sizes in these studies are relatively small, standardized preoperative and postoperative neuropsychological measures are lacking, and underlying pathological substrates and/or psychiatric comorbidities have not been accounted for. It is known, for instance, that patients without hippocampal sclerosis have poorer memory outcomes than those with such induration. In these patients, resection of the hippocampus is related to significant impairment in long-term aspects of verbal memory. Significant deterioration in short-term aspects of verbal memory was reported in patients after temporolateral resection. ¹⁰

Patients undergoing ATL can show deficits in both short- and long-term verbal memory functions. To ameliorate these complications, several methods of SAH have been developed to achieve comparable seizure control and better cognitive outcomes than those following ATL.

8.2 Historical Evolution of Some Surgical Techniques for SAH

In 1958, Paulo Niemeyer in Brazil, ¹¹ without the use of the surgical microscope, described in detail and performed a middle temporal gyrus (T2) corticotomy to provide access to the ventricular temporal horn. This procedure has been called the “transventricular approach.” Once the amygdala and hippocampus are reached and identified, the structures are selectively removed. In this procedure, however, the stem of the superior temporal gyrus is severed to provide access.

Many years later, in 1982, Wieser and Yaşargil ¹² proposed a transsylvian surgical route through the temporal stem in the depth of the sylvian fissure to reach the temporal horn for selective removal of the amygdala and hippocampus, trying to avoid injury to the lateral temporal cortex. However, the transsylvian approach has not been widely used, given the high degree of skill required to provide wide exposure of the sylvian fissure.

In 1984, at the Montreal Neurological Institute (MNI), André Olivier described a surgical technique to reach the temporal horn by resecting the anterior portion of the superior temporal gyrus. ¹³ A few years later, he modified his own surgical technique to a less invasive transulcal-T1 approach.

After these original and historical procedures, a number of selective or minimalistic surgical techniques have been described to remove the mesial structures of the temporal lobe. ¹⁴ Different technologies and surgical techniques have been developed to provide access with maximal security and achieve complete removal (en bloc resections or removal by subpial aspiration) or disconnection of the amygdala and hippocampus via the preferential pathway. ¹⁵ With most of these newer techniques, the hippocampal formation and adjacent structures (amygdala, parahippocampal gyrus [PHG]) are removed by subpial suction.

The effect that the choice of a particular approach or the extent of the resection has on outcome in terms of seizure reduction and neuropsychological profiles remains an unanswered question. The question should be considered on two levels. The first level is the decision of the surgeon to remove certain structures to a defined degree using one of several standardized approaches. The second and far less addressed question is that of how much of the intended resection volume was really removed. ¹⁶ Ultimately, a meaningful comparison of the superiority of one versus another approach proves difficult not only by the multidimensionality of outcome, which has not been modeled into a quantitative measure, but by case-specific (expected) inconsistencies in each surgical procedure.

In patients undergoing selective resection of the mesial temporal structures for left temporal lobe seizures, postoperative verbal memory loss may still be a problem. Cognitive function after SAH using the transsylvian or transcortical approaches was compared. Both approaches resulted in a significant decline in verbal memory when performed on the left-side surgery. Phonemic fluency was significantly improved after surgery with the transcortical but not with the transsylvian approach. ¹⁷ When the epileptogenic zone is on the left side, irrespective of preserving the temporal lobe neocortex, there might be a decline in memory. For these reasons, new surgical approaches and techniques utilizing diverse technology (Visualase, cooling) are under development and awaiting evidence from controlled trials.