Statistics from Altmetric.com
A new way to look at epilepsy surgery outcomes
Three fields of neurosurgery have truly lived up to the aspirations of evidence-based medicine with multiple randomised controlled trials published in top impact journals: epilepsy surgery and movement disorder surgery and to a lesser degree stereotactic radiosurgery.1–3
The JNNP paper of Vogt et al 4adds another layer of granularity to the epilepsy surgery literature and shows quite clearly that randomised prospective trials in neurosurgery—so badly needed in many areas—are feasible and very useful.
Comparing two different approaches and hence techniques of performing selective amygdala-hippocampectomies (SAH) in a randomised prospective fashion is not an esoteric enterprise. First, the field naturally is moving towards more tailored minimally invasive surgical procedures in epilepsy surgery, relegating gross volumetric techniques like anterior temporal lobectomy hopefully to neurosurgical history. Second, epilepsy surgery is badly underused partly due to the fear of patients when facing major craniotomies plus neuropsychological sequelae and we owe it to our patients to provide solid data on the procedures we offer and recommend. As the major side effect of SAH besides visual field defects is neuropsychological—as so lucidly worked out by this very same group— the study focuses on neuropsychological outcomes after subtemporal versus transsylvian approach in patients with mesial temporal lobe epilepsy.
The authors had previously published the same comparison but used a matched-pair design. In that study they found significant differences between surgical approaches as to neuropsychological scores pertaining to memory and learning.
These differences became insignificant on the group level in this study. Only using post-hoc analysis (questionable in randomised trials to begin with) did the authors find worse performance in memory tasks (free delayed recall) and learning in the subtemporal group. They then discuss the pros and cons of the two different study designs, that is, matched pairs versus randomised prospective.
The paper sets the tone for future studies in neurosurgery comparing techniques on the basis of validated rigorous outcome measures in order to truly advance the field.5
The interpretation of the results calls for a bit of caution: the study is quite small (N=47) not allowing a powerful multivariate analysis. The seizure outcome is slightly on the lower side with ILAE 1a being 62% coming from a very experienced group.
It is interesting that only the post-hoc analysis revealed differences between the two surgical approaches. While intuitively plausible that the subtemporal access fares worse for memory and learning as the temporal lobe needs more retraction, these data should be taken with a grain of salt. Post-hoc analysis is a way of ‘corriger la fortune’ for statistical purists.6
The differences between the previous and the current study (matched pairs vs prospective randomised) are well explained by the authors. Though randomised prospective studies need large cohorts to detect minor differences which matched-pair design can get by without, the level of evidence is clearly higher for the first eliminating selection bias.
The biggest criticism comes from a different realm—technological advance: open selective amygdala-hippocampectomy might soon be replaced by interstitial stereotactic laser amygdala-hippocampectomy with equal efficiency but significantly less morbidity avoiding craniotomies and sparing the lateral temporal lobe completely.7 The Stereotactic Laser Ablation in Temporal Epilepsy (SLATE) trial is currently conducted to provide prospective multi-institutional data.
This is not to diminish the current study which was started before the advent of interstitial laser amygdala-hippocampectomy. What is going to stay for the field is that randomised prospective trials of surgical techniques are feasible, render superior evidence and are the standard we should strive for.
Funding The author has not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.