Hippocampal Depth Electrodes in Epilepsy Surgery: Diagnostic or Damaging?

Commentary

Verbal memory decline from hippocampal depth electrodes in temporal lobe surgery for epilepsy.

Ljung H, Nordlund A, Strandberg M, Bengzon J, Källén K. 2017;58:2143–2152.

OBJECTIVE: To explore whether patients with refractory mesial temporal lobe epilepsy risk aggravated verbal memory loss from intracranial electroencephalography (EEG) recording with longitudinal hippocampal electrodes in the language-dominant hemisphere. METHODS: A long-term neuropsychological follow-up (mean 61.5 months, range 22–111 months) was performed in 40 patients after ictal registration with left hippocampal depth electrodes (study group, n = 16) or no invasive EEG, only extracranial registration (reference group, n = 24). The groups were equal with respect to education, age at seizure onset, epilepsy duration, and prevalence of pharmacoresistant temporal lobe epilepsy (TLE; 75%) versus seizure freedom (25%). Retrospective neuropsychological data from preoperative surgical workup (T1) and prospective followup neuropsychological data (T2) were compared. A ≥1 SD intrapatient decline was considered as clinically relevant deterioration of verbal memory. RESULTS: Significant decline in verbal memory was seen in 56% of the patients in the study group compared to 21% in the reference group. At T1, there were no statistical between-group differences in memory performance. At T2, between-group comparison showed significantly greater verbal memory decline for the study group (Claeson Dahl Learning and Retention Test, Verbal Learning: p = 0.05; Rey Auditory Verbal Learning Test, Total Learning: p = 0.04; Claeson Dahl Learning and Retention Test, Verbal Retention: p = 0.04). An odds ratio (OR) of 7.1 (90% confidence interval [CI] 1.3–37.7) for verbal memory decline was seen if right temporal lobe resection (R TLR) had been performed between T1 and T2. The difference between groups remained unchanged when patients who had undergone R TLR were excluded from the analysis, with a remaining aggravated significant decline in verbal memory performance for the study group compared to the reference group. SIGNIFICANCE: Our results suggest a risk of verbal memory deterioration after the use of depth electrodes along the longitudinal axis of the hippocampus. Until this issue is further investigated, caution regarding depth electrodes in the language-dominant hemisphere hippocampus seems advisable.

Researchers from Sweden contrasted memory and naming functions in temporal lobe epilepsy (TLE) patients who required either unilateral (n = 1) or bilateral hippocampal (n = 15) depth electrode placement for seizure localization (study group: n = 16) with a group of TLE patients who did not undergo any form of invasive monitoring (reference group: n = 24). As Ljung and colleagues were primarily interested in verbal memory outcome, they excluded patients who eventually underwent a left TL resection. They included TLE patients who had never undergone surgery or who had a right TL resection. The authors reported that the study group was lower on several memory and naming measures than the reference group when tested approximately 2 to 10 years after the initial evaluation performed prior to invasive monitoring. Researchers also found that the study group showed a significant decline on some measures (Claeson-Dahl Learning and Retention Test, Rey Complex Figure Test, Boston Naming Test) administered on both occasions while the reference group did not show a decline.

This research study suggests that longitudinally placed hippocampal depth electrodes may significantly harm memory and naming ability, which could have implications for placement of intracranial electrodes for diagnostic and treatment purposes (e.g., neuromodulation). As with much research in the clinical epilepsy setting, however, potentially confounding variables limit the strength of the conclusions that can be drawn. Substantial questions remain, despite extensive efforts by the authors to statistically manage what were inevitable and unavoidable inequalities of retrospective clinical research.

Some may take umbrage with the use of a 1 SD method of assessing significant change, as the use of reliable change index (RCI) methods have dominated the landscape of cognitive change in recent decades in a noble effort to control for practice effects and statistical trends (e.g., regression to the mean; ¹). Nevertheless, questions have been raised about the validity and interpretation of RCI scores, as they could potentially obscure change by creating large hurdles that are sometimes impossible for many study patients to achieve (i.e., baseline scores are often too high or too low for these reliable index score differences to be achieved, making “significant” change impossible to observe in a number of patients; ^{2, 3}). Moreover, as growing evidence is available that interictal epileptiform discharges (IEDs) can affect cognitive performance in a transient yet substantial manner, test–retest indices are potentially confounded in epilepsy patients if simultaneous IEDs are not considered (2, 4, 5). Likewise, establishing statistical thresholds of difference is not the same thing as establishing real-world significance to the patient (6). Therefore, the use of metric is less a concern—and potentially a positive starting point—as it is reasonable to assess even small changes if they occur only in the study group.

More problematic for the current research seemed to be the large number of relevant clinical differences between the study and reference groups, with most of the differences stacked against the study group. These differences included the study group being older at the time of testing with more patients over the age of 50 years (50% vs 17%), and a much greater percentage of bilateral TLE patients in the study group (50% vs 17%). This latter finding suggests that the study group was composed of patients with more severe epilepsy. Researchers also noted that right TLE patients are known to sometimes decline on verbal memory, so including these patients in the current study also reflected a potential confound. They attempted to control for these potential confounding variables statistically; however, the small sample size made this methodology more tenuous when attempting to control for so many critical sample differences, some of which could have been managed using a purely prospective design.

Additionally, the patients for whom the placement of language-dominant hippocampal depths are likely most valuable (i.e., those who went on to have surgery involving that same TL) had to be excluded from consideration, as the researchers could not retrospectively parse the effects of surgery from those having electrode placement. This is an area where future research should prospectively examine different time points in the surgical work-up, which would need to include a timeframe between invasive electrode explantation and the resective surgery, as well as a subsequent postsurgical time-point. This would allow for better control of some of the confounding variables previously cited. One earlier paper followed postsurgical patients who either did or did not undergo invasive monitoring, and the study reported no differences between groups after surgery (7). However, in this study, many patients did not have depth electrodes, and there was no breakdown by type of invasive monitoring and no description of how the depth electrodes were placed.

There were also group differences in the current study regarding changes in the number of antiepileptic drugs from baseline to follow-up testing, which included more patients in the reference group off all medications at follow-up. It does not appear that this variable was ever accounted for in the study design. Nearly all study group TLE patients were evaluated with bitemporal hippocampal depth electrodes (i.e., 15 of 16). The potential effects of unilateral electrode placement may be different than bilateral and should be studied in future work. Both these factors should be given consideration in a future prospective study, as well as a comparison of differing approaches to depth electrode placement.

The follow-up time between electrode placement and postexplantation testing varied widely, with a range of 2 to 10 years. Such findings would be more compelling if they were also observed shortly after electrode placement and at subsequent intervals. With the large number of potentially confounding factors contributing to differences in cognition, these large and varied time intervals of comparison seem problematic. Additionally, studies have demonstrated that the hippocampus can be ablated without leading to naming decline (8), yet one of the measures showing decline in this study was the same visual naming test. This suggests that decline was related to something other than depth electrode placement. Finally, recent studies examining cognitive outcome with bilateral depth electrodes for neuromodulation therapy have not shown such naming and memory declines (9).

Despite the methodological limitations of the current study, its value is enhanced by the importance of the questions raised. Recognition of potential harm caused by treatments is often lacking, particularly in the setting of very complex patients with multiple confounds that could cause cognitive and behavioral variability. Of course, there are methodological issues, so results must be viewed in this context until more complete data can be obtained to fill in the missing gaps and not viewed as setting a new course of practice. If future studies confirm and extend these results, then changes in the use of intracranial electrodes would need to be considered. However, even if proved to be the case, there may yet be circumstances where use of invasive monitoring may offset other risks, if necessary, to achieve a potentially curative surgical approach. Research literature on the long-term behavioral consequences of uncontrolled epilepsy remain mixed (10), yet failing to offer a potential cure to someone with devastating epilepsy may be worse than taking the risk of causing some degree of harm through required evaluation and treatment. More research would also be needed to determine the potential mechanism of harm associated with invasive techniques and if such risks could be mediated by altering routes of electrode placement or other changes to instrumentation and technique.