Abstracts

Rationale: Depression, the most frequent psychiatric comorbidity in epilepsy (22.9%), is two to three times higher in patients with mesial temporal lobe epilepsy and hippocampal sclerosis (MTLE-HS). In epilepsy, the true role of hippocampus in lifetime depression (LD) is not completely understood. In psychiatry, neuroimaging studies have shown hippocampal volume reduction in major depression disorder (MDD). Additionally, histopathological studies from autopsies of suicidal individuals with MDD have suggested glial cells (GC) proliferation. The aim of this study is to relate hippocampal tissue findings with LD in patients with unilateral MTLE-HS. Methods: Hippocampi specimens were obtained after epilepsy surgery of 58 patients between 2005-2012. Hippocampi neuronal densities and their subfields (H- or S-Nd) were determined by counting cell bodies after NeuN-immunoreaction and compared with autopsies of 13 normal controls individuals (NCI). HS types according to ILAE classification were determined. LD diagnosis, evaluated by a psychiatrist, had to fulfil DSM-IV Axis I criteria for past/current mood disorders (the gold standard at that time). Beck Depression Inventory I (BDI-I) plus ESI-55 quality of life assessment were also applied to quantify pre-surgical symptoms. For statistical histopathological analysis (HA), we established a Z score (Zs) using NCI to estimate Nd regardless age at surgery and epilepsy duration. We used Fisher’s exact text for independent variables, Mann-Whitney for means comparisons and a p-value < 0.05 was significant. Results: Twenty patients (13 women) had LD (depression group, DG) (35.5%), being 10 for each HS side. BDI was higher in DG (mean 16.04; pattern deviation: 10.04) than in non-depression group (NDG) (8.8 and 8.26, p=0.002). Antidepressants (all in therapeutic range) prevailed in DG (60% vs. 2.6%, p < 0.001), especially fluoxetine (45% vs. 0%, p < 0.001). DG was older at surgery (41.2 yrs., p=0.039) and had a longer epilepsy duration (26.9 yrs., p=0.036) than NDG (35 and 20.0 yrs.). DG had lower mental health (43.1 vs. 66.7, p=0.001) and general quality of life (54.1 vs. 64.2, p=0.03). In HA, H-Nd (showed as x103/mm2) was decreased in DG (mean 9.4; pattern deviation: 4.44) as well as in NDG (10.1 and 5.57, p=0.798). Regarding S-Nd (showed as Zs), CA1 had the lowest counting in both (mean -4.59 in DG vs. -4.44 in NDG, p=0.718). The other subfields had an approximately equal reduction (in DG and NDG, respectively: CA2: -2.93vs. -2.78, p=0.508; CA3: -2.97 vs. -2.60, p= 0.456; CA4: -2.79 vs. -2.93, p=0.823). In our series, ILAE type 1 was the most prevalent (75% in DG vs. 86.9% in NDG); however, type 2 was four times higher in DG (20% vs. 5.3%, p=0.311). Conclusions: Although HS is a pathological hallmark in MTLE-HS, its role in comorbid LD is probably not a key feature. In literature, it is speculated that chaperones related to GC might play a major role in this setting. We need more studies to put lights on the physiopathology of LD in MTLE-HS.   Funding: FAPESP, CAPES and CNPq from São Paulo State and Brazilian Federal Goverments supported this work.