MRI Brain for New-Onset Seizures: Without Contrast Initially
For new-onset seizures, order MRI brain WITHOUT contrast as the initial study. Contrast is not routinely necessary and should be reserved for specific clinical scenarios where neoplasm or inflammatory conditions are suspected 1.
Primary Recommendation
The American College of Radiology (ACR) guidelines explicitly state that intravenous contrast is not routinely necessary for initial evaluation of new-onset seizures 1. The standard epilepsy protocol MRI sequences (T1-weighted, T2-weighted, FLAIR, and gradient echo sequences) are sufficient to detect the vast majority of epileptogenic lesions without gadolinium 1.
Evidence Supporting Non-Contrast Approach
High Diagnostic Yield Without Contrast
- 100% of epileptogenic lesions were detected on non-contrast sequences in a study of 103 adult patients with new-onset seizures, with zero lesions detected exclusively on post-contrast imaging 2
- In pediatric patients, only 5% of cases receiving gadolinium actually required it, with the remaining 95% showing no added diagnostic benefit 3
- Non-contrast MRI has a negative predictive value of 98.4% for excluding brain lesions, meaning if the non-contrast study is negative, adding contrast changes management in less than 2% of cases 4
Cost and Safety Considerations
- The unnecessary use of contrast in patients without specific indications results in excess costs of approximately $103,680 per 1,000 patients 2
- Avoiding unnecessary gadolinium exposure reduces potential risks including nephrogenic systemic fibrosis in patients with renal impairment and gadolinium deposition in brain tissue 2
When to Add Contrast
Contrast IS indicated in the following specific scenarios 1, 5:
- Suspected neoplasm: New-onset seizures in adults, particularly with focal neurologic deficits or concerning imaging findings on non-contrast sequences
- Suspected infection: Clinical signs of CNS infection (fever, meningismus, immunocompromised state) 3, 6
- Inflammatory conditions: Suspected autoimmune encephalitis or other inflammatory processes 1
- Inadequate non-contrast images: When initial sequences are technically limited or show equivocal findings requiring further characterization 1
In pediatric patients younger than 2 years, contrast was deemed essential in only 1.8% of cases, all involving documented or suspected infection 6.
Optimal MRI Protocol (Without Contrast)
The epilepsy-specific protocol should include 1:
- Coronal T1-weighted imaging (3 mm slices) perpendicular to the hippocampal long axis for detecting hippocampal sclerosis
- High-resolution 3D T1-weighted gradient echo (1 mm isotropic voxels) for cortical malformations
- Coronal and axial FLAIR sequences (or 3D FLAIR) for signal abnormalities and cortical dysplasia
- Coronal T2-weighted sequences for hippocampal internal architecture
- Gradient echo or susceptibility-weighted imaging for detecting calcifications, hemorrhage, and vascular malformations
Clinical Algorithm
Step 1: Order MRI brain without contrast for all new-onset seizures as initial study 1
Step 2: Review clinical context for high-risk features:
- Focal neurologic deficits 1
- Immunocompromised state
- Fever or signs of infection 3, 6
- Known or suspected malignancy
- Progressive symptoms
Step 3: If high-risk features present OR non-contrast MRI shows concerning findings requiring further characterization, then add contrast 1, 5
Step 4: If non-contrast MRI is negative and no high-risk features, contrast adds no diagnostic value and should be omitted 2, 4
Common Pitfalls to Avoid
- Don't reflexively order contrast for all seizure patients—this represents overutilization in 72-92% of cases 3, 2
- Don't skip the specialized epilepsy protocol sequences—standard brain MRI may miss subtle cortical dysplasia or hippocampal sclerosis 1
- Don't use contrast as a substitute for proper non-contrast sequences—the diagnostic information comes primarily from T2/FLAIR imaging, not enhancement 2, 4
Emergency vs. Non-Emergency Settings
In acute/emergency settings, non-contrast CT may be performed first to rapidly exclude hemorrhage, mass effect, or acute stroke requiring immediate intervention 1. However, MRI without contrast remains the definitive study and should follow once the patient is stabilized, as CT detects only 30% of epileptogenic lesions compared to MRI 1.