In a Retrospective Observational Case-Control Study Hodelin Maynard et al. 1) investigated whether quantitative anomalies in cerebral white matter tracts, assessed through automated fiber quantification using diffusion tensor imaging (DTI), are associated with postsurgical clinical outcomes (seizure freedom vs. seizure recurrence) in patients with drug-resistant frontal or temporal lobe epilepsy
The authors set out to explore whether quantitative white matter anomalies, assessed via DTI tractography, correlate with seizure outcomes in drug-resistant epilepsy patients. On paper, the goal is relevant and aligns with modern neuroimaging trends. In practice, however, the study is a textbook example of how not to do translational neuroimaging—a fragile mix of low statistical power, overfitting, and unjustified speculation masquerading as mechanistic insight.
⚠️ Fatal Methodological Flaws
1. Sample Size: Statistically Comatose
The entire study rests on just 20 patients (8 with frontal lobe epilepsy, 12 with temporal lobe epilepsy), and only 19 controls.
Subgrouping by seizure outcome (freedom vs. recurrence) shatters any statistical credibility—splitting a tiny sample into even smaller bits guarantees overfitting and false discovery.
This is not hypothesis testing. It’s hypothesis hallucination.
2. No Control for Multiple Comparisons
The authors explore differences across 4 metrics (FA, MD, Vol, Fib) in multiple tracts, bilaterally, pre- and post-op.
Yet no correction (e.g., Bonferroni, FDR) is applied for multiple comparisons.
With this many comparisons, statistical significance becomes meaningless—they're just cherry-picking from noise.
3. Temporal and Spatial Overinterpretation
Claims about “presurgical MD increases in the contralateral uncinate fasciculus” predicting seizure freedom are statistically hollow and biologically implausible without stronger longitudinal or mechanistic validation.
🔍 Technical Shortcomings and Black Box Analytics
4. Opaque Tractography Pipeline
The study uses “automatic fiber quantification” but fails to describe any quality assurance protocols:
Were outlier tracts removed?
Were results visually confirmed?
What was the inter-rater reliability?
Without transparency, this becomes tractography theatre: complex-sounding techniques with no verifiable reproducibility.
5. Mismatch of Controls and Patients
No demographic or scanner-hardware details are provided to assure that controls are well-matched to patients.
DTI metrics are notoriously sensitive to scanner differences—a few milliseconds of sequence variation can invalidate comparisons.
🧨 Interpretive Overreach
6. From Correlation to Prediction… Without Justification
The authors repeatedly imply predictive or prognostic power of DTI features—but never validate these with ROC analysis, cross-validation, or out-of-sample testing.
Association ≠ prediction ≠ causation. The authors blur all three.
7. Post hoc Just-So Storytelling
The authors reverse-engineer “meaningful” explanations for tract changes (e.g., thalamic radiation anomalies “explaining” seizure outcomes) without any physiological or electrophysiological grounding.
No integration with SEEG, EEG, or histopathology is provided. The result is neuroanatomical astrology.
📉 Conclusion: Overblown, Underpowered, and Ultimately Misleading
This paper is a cautionary tale in modern neuroimaging: dazzling technical language masking a deeply flawed design. With an underpowered sample, statistical negligence, opaque methods, and inflated claims, this work contributes more to neuroimaging confusion than to the clinical management of epilepsy.
If anything, it should be used in journal clubs as an example of how not to infer meaning from DTI data in small clinical cohorts.
🔚 Final Judgment
Scientific merit: 2/10
Methodological rigor: 1/10
Clinical impact: 0/10
Buzzword inflation index: 10/10
Based on the promising results of randomized controlled trials, deep brain stimulation (DBS) and responsive neurostimulation (RNS) are increasingly used in the treatment of patients with drug-resistant epilepsy. Drug-resistant temporal lobe epilepsy (TLE) is an indication of either DBS of the anterior nucleus of the thalamus (ANT) or temporal lobe (TL) RNS, but there are no studies that directly compare seizure benefits and adverse effects associated with these therapies in this patient population.