Comparison of minimally invasive and traditional surgical approaches for refractory mesial temporal lobe epilepsy: A systematic review and meta‐analysis of outcomes

Abstract

Magnetic resonance‐guided laser interstitial laser therapy (MRgLITT) and radiofrequency ablation (RFA) represent two minimally invasive methods for the treatment of drug‐refractory mesial temporal lobe epilepsy (mTLE). We performed a systematic review and a meta‐analysis to compare outcomes and complications between MRgLITT, RFA, and conventional surgical approaches to the temporal lobe (i.e., anterior temporal lobe resection [ATL] or selective amygdalohippocampectomy [sAHE]). Forty‐three studies (13 MRgLITT, 6 RFA, and 24 surgery studies) involved 554, 123, 1504, and 1326 patients treated by MRgLITT, RFA, ATL, or sAHE, respectively. Engel Class I (Engel‐I) outcomes were achieved after MRgLITT in 57% (315/554, range = 33.3%–67.4%), RFA in 44% (54/123, range = 0%–67.2%), ATL in 69% (1032/1504, range = 40%–92.9%), and sAHE in 66% (887/1326, range = 21.4%–93.3%). Meta‐analysis revealed no significant difference in seizure outcome between MRgLITT and RFA (Q = 2.74, p = .098), whereas ATL and sAHE were both superior to MRgLITT (ATL: Q = 8.92, p = .002; sAHE: Q = 4.33, p = .037) and RFA (ATL: Q = 6.42, p = .0113; sAHE: Q = 5.04, p = .0247), with better outcome in patients at follow‐up of 60 months or more. Mesial hippocampal sclerosis (mTLE + hippocampal sclerosis) was associated with significantly better outcome after MRgLITT (Engel‐I outcome in 64%; Q = 8.55, p = .0035). The rate of major complications was 3.8% for MRgLITT, 3.7% for RFA, 10.9% for ATL, and 7.4% for sAHE; the differences did not show statistical significance. Neuropsychological deficits occurred after all procedures, with left‐sided surgeries having a higher rate of verbal memory impairment. Lateral functions such as naming or object recognition may be more preserved in MRgLITT. Thermal therapies are effective techniques but show a significantly lower rate of Engel‐I outcome in comparison to ATL and sAHE. Between MRgLITT and RFA there were no significant differences in Engel‐I outcome, whereby the success of treatment seems to depend on the approach used (e.g., occipital approach). MRgLITT shows a similar rate of complications compared to RFA, whereas patients undergoing MRgLITT may experience fewer major complications compared to ATL or sAHE and might have a more beneficial neuropsychological outcome.

Resective, Ablative and Radiosurgical Interventions for Drug Resistant Mesial Temporal Lobe Epilepsy: A Systematic Review and Meta-Analysis of Outcomes

“…This means that with the small sample sizes present in most publications, studies may have been underpowered to detect statistically significant differences in seizure freedom between interventions. A meta-analysis comparing seizure free outcomes between ATLR, SAH, RF-TC and LITT was recently published which concluded that LITT was significantly less effective than open surgery ( 67 ). There are, however, significant methodological differences between our study.…”

Epileptic discharges initiate from brain areas with elevated accumulation of α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptors

“… 31 Furthermore, the cumulative evidence indicated that the selective treatment to the HIP or to the mesial temporal structures were inferior to the anterior temporal lobectomy in terms of the chance of seizure freedom. 32 , 33 The finding of the elevated uptake of [ 11 C]K-2 in the lateral temporal cortex might be useful for the prediction of treatment outcomes.…”

Ordinal regression increases statistical power to predict epilepsy surgical outcomes

“…We test this hypothesis using the problem of predicting surgical outcomes after stereotactic laser amygdalohippocampectomy (SLAH). A recent meta‐analysis of 10 studies found that the presence of mesial temporal sclerosis (MTS) predicted a better chance (63% vs 42%) of seizure freedom after SLAH when MTS was present than when absent 4 . However, most individual studies did not detect this effect, as their low sample size limited statistical power.…”

“…A recent meta-analysis of 10 studies found that the presence of mesial temporal sclerosis (MTS) predicted a better chance (63% vs 42%) of seizure freedom after SLAH when MTS was present than when absent. 4 However, most individual studies did not detect this effect, as their low sample size limited statistical power. We specifically hypothesize that using ordinal logistic regression we will meaningfully increase the statistical power to detect the known relation between MTS and seizure outcome in each study.…”