Dados do Trabalho

Título

Longitudinal effects of pharmacoresponse and presence and side of hippocampal sclerosis on in-vivo neuronal damage in MTLE patients

Resumo

Introduction: Mesial temporal lobe epilepsy (MTLE) is a heterogeneous disease which frequently presents with pharmacoresistance. Reduction in hippocampal N-acetylaspartate (NAA) levels, a marker of neuronal dysfunction measured by proton magnetic resonance spectroscopy (1H-MRS), has been associated to pharmacoresponse and presence of hippocampal sclerosis (HS) in MTLE. Progression of structural damage on MRI, such as HS, has been reported even in patients with good seizure control. However, how neuronal dysfunction behaves in the long run remains an open question.
Objectives: To investigate longitudinal neuronal damage, measured by NAA ratios to creatinine (NAA/Cr), in patients with MTLE with or without adequate seizure control.
Materials and Methods: We measured ipsi- and contralateral hippocampal NAA/Cr using single-voxel 1H-MRS from 161 individuals, namely: 66 pharmacoresponsive (18 MRI-negative, 21 right-HS, and 27 left-HS), 95 pharmacoresistant (20 MRI-negative, 34 right-HS, and 41 left-HS) patients and six healthy controls. We selected up to three 1H-MRS per individual (scan-1 to 3). The mean time interval between scan-1 and scan-2 was 2.1±1.8years, and between scan-2 and scan-3 was 2.1±1.4years. We performed generalized estimated equation models with an identity link function, including pharmacoresponse, HS-side and time as main effects, and a pharmacoresponse or HS-side*time interactions, covarying for age at each scan. P<0.05 was set as significant. Model fit was evaluated using Quasi Likelihood Independence Model Criterion scores.
Results: We found significant effects of pharmacoresponse (p=0.02), side (p=0.01), time (p=0.004), and time*side (p=0.008) interaction on ipsilateral NAA/Cr, as follows: ipsilateral NAA/Cr was reduced in pharmacoresistant patients when compared to pharmacoresponsive patients (p=0.02) and controls (p=0.02). Ipsilateral NAA/Cr was also decreased in MRI-negative and left-HS patients from scan 1 to 3 (p<0.02 for both groups), and between scan-2 and 3 (p=0.027 for both groups). There were no longitudinal changes to ipsilateral NAA/Cr ratios in controls and right-HS patients (all p>0.067). We found no significant longitudinal effects of pharmacoresponse or HS-side on contralateral NAA/Cr (all p>0.19)
Conclusion: Our data suggest unilateral progression of neuronal damage in pharmacoresistant patients and those with left-HS and negative MRI as measured by NAA/Cr reduction.

Keywords: hippocampus, MRI, epilepsy