Peri-ictal imaging abnormalities in non-convulsive status epilepticus: A systematic review and meta-analysis comparing magnetic resonance imaging and computed tomography perfusion.

OBJECTIVE

To assess and compare the detection rates of peri-ictal abnormalities using magnetic resonance imaging (MRI) and computed tomography perfusion (CTP) in patients with non-convulsive status epilepticus (NCSE).

METHODS

We conducted a systematic literature search in five databases up to February 2025. Studies reporting peri-ictal MRI abnormalities (PMAs) or cerebral perfusion abnormalities (CPAs) in patients with NCSE were included. Meta-analyses of proportions were performed using a random-effects model. Subgroup analyses and meta-regression were used to compare detection rates across imaging modalities.

RESULTS

Nineteen studies were included (15 MRI, 4 CTP), comprising 562 patients for MRI and 72 for CTP. The pooled detection rate of peri-ictal abnormalities was 50.0% (95% confidence interval [CI]: 34.0%-65.0%) for MRI and 79.3% (95% CI: 54.3%-92.5%) for CTP. Among the MRI modalities, arterial spin labeling (ASL) demonstrated the highest detection rate at 88.8% (95% CI: 32.9%-99.2%). CTP showed a significantly higher detection rate than MRI (χ2 = 3.97, p = 0.046); meta-regression indicated increased odds of detection with CTP (odds ratio [OR] = 4.06, 95% CI: 0.97-16.99, p = 0.055). No statistically significant difference was found between ASL and CTP (χ2 = 0.22, p = 0.636).

CONCLUSIONS

CTP demonstrates a higher detection rate than conventional MRI for peri-ictal abnormalities in patients with NCSE, supporting its utility in rapid diagnosis and differential workup. Among MRI sequences, ASL showed the highest detection rates, highlighting its potential role in the diagnostic assessment of NCSE. Although MRI remains essential for clarifying etiology, its effectiveness in detecting PMA is highly dependent on the sequences used.