Objective In temporal lobe epilepsy surgery, there is a trend towards becoming more selective in order to achieve seizure control with an optimal neuropsychological outcome. The present study evaluated whether mesial resection length matters for memory outcome after selective amygdalo-hippocampectomy (SAH). Therefore, a sub-analysis of the larger SFB/TR3/A1 multicentre randomised trial on seizure outcome and mesial resection length in temporal lobe surgery was performed.

Methods A homogeneous group of 67 selectively operated patients with hippocampal sclerosis as the sole pathology were allocated to a short (2.5 cm, n=34) or a long (3.5 cm, n=33) mesial resection. Repeated memory assessment and three-dimensional MRI data sets served as dependent within group variables, and intraoperatively determined resection lengths (short/long), resected hippocampal volumes (small/large) and side of surgery were independent between group variables.

Results Resection length did not have a significant effect on seizure or on memory outcome. The resected hippocampal volume also did not affect seizure outcome but it did make a difference with regard to memory outcome. Outcome in verbal learning and memory was poorer after resection of larger left hippocampal volumes. Figural memory outcome was poorer with larger resected volumes on either side.

Interpretation The data indicate that in SAH, mesial resection length and resected volumes have no differential effect on seizure outcome. The findings on memory outcome are best explained by suggesting that hippocampal volumes take the degree of preoperative pathology into account whereas resection length does not. This suggests resection of non-pathological functional tissues as the basis for memory outcome after SAH.