MRI-Negative Epilepsy Linked to Lower Seizure Freedom After Surgery

Patients with drug-resistant epilepsy who show no abnormalities on magnetic resonance imaging (MRI) scans are less likely to achieve seizure freedom following surgery compared with those who have visible lesions, according to results published in Epilepsia.

While MRI has been a key predictor of surgical success, the criteria used to classify patients as MRI-negative are inconsistent across the literature. To address this, investigators reviewed studies spanning 3 decades to assess how MRI-negative status is defined, whether advanced imaging analysis can improve detection, and how MRI findings relate to long-term surgical outcomes.

The researchers analyzed 246 studies encompassing 17,899 patients, of whom 10,463 were designated MRI-negative and 7436 MRI-positive (both cohorts median age, 32 years; IQR, 27-36). Compared with MRI-positive patients, those who were MRI-negative underwent invasive monitoring more often (75% vs 54%; P <.05), were offered surgery less frequently (73% vs 84%; odds ratio [OR], 1.14; P <.001), and had less worse post-operative seizure outcomes (61% vs 72%; P <.05).

Advanced post-processing of MRI scans, such as voxel-based morphometry and 3D reconstructions, significantly improved diagnostic yield. Across 34 studies involving 1272 patients, the proportion of scans reclassified as positive increased from 42% to 81% after post-processing, representing a 39% increase in lesion detection (OR, 11.10; 95% CI, 7.45-16.53).

Analysis of surgical outcomes across 77 studies (n=5395) confirmed a persistent gap in seizure freedom rates. Among MRI-positive patients, 75% achieved seizure freedom compared with 58% of those with MRI-negative findings (P <.01). Histopathologic patterns also revealed notable differences. While focal cortical dysplasia type II was similarly common in both groups, MRI-negative patients had slightly lower rates of seizure control than MRI-positive patients (66% vs 71%). Hippocampal sclerosis, however, was nearly twice as prevalent in MRI-positive cohorts and was associated with substantially better outcomes (87% vs 75%).

The review further exposed a wide variation in how studies defined MRI-negative status. Fewer than one-third of studies provided detailed MRI acquisition parameters, reader expertise, or use of post-processing, while others relied on non-MRI modalities such as positron emission tomography or single-photon emission computed tomography. The authors cautioned that this lack of standardization contributes to inconsistent classification and may limit surgical success in patients incorrectly labeled as MRI-negative.

Limitations of the analysis included methodological variability across studies, incomplete reporting of MRI protocols, and evidence of publication bias.

“[O]ur findings suggest that MRI post-processing should be systematically performed to evaluate for prevalent epileptogenic lesions… before ascribing MRI-negative status to patients with medically intractable seizures,” the study authors concluded.

Disclosures: This research was supported by the Natural Sciences and Engineering Research Council of Canada, Epilepsy Canada, Brain Canada, and Fonds de Recherche.