PDF: Wiebe 2001
(2001)Objective
To compare the efficacy and safety of anterior temporal lobe resection versus optimized medical therapy in patients with drug-resistant temporal-lobe epilepsy
Study Summary
• Freedom from all seizures (including auras): 38% surgical vs 3% medical (P<0.001)
• NNT = 2 for freedom from seizures impairing awareness; NNT = 3 for freedom from all seizures
• Quality of life (QOLIE-89) was significantly better in the surgical group (P<0.001)
• Surgical complications occurred in 4 patients (10%): thalamic infarct, wound infection, and verbal memory decline in 2 patients
• One death occurred in the medical group (sudden unexplained death in epilepsy)
Intervention
Anterior temporal lobe resection (6.0–6.5 cm nondominant or 4.0–4.5 cm dominant lateral resection + mesial resection including amygdala and 1.0–3.0 cm of hippocampus) followed by 1 year of optimized medical therapy vs optimized medical therapy alone for 1 year
Inclusion Criteria
• Age ≥16 years
• Temporal-lobe seizure semiology for >1 year
• Seizures ≥1/month despite ≥2 antiepileptic drugs (including phenytoin, carbamazepine, or valproic acid)
• No brain lesions requiring urgent surgery
• No progressive CNS disorders, pseudoseizures, or IQ <70
• No prior epilepsy surgery or bilateral temporal focus
Study Design
Arms: Array
Patients per Arm: Surgery: 40; Medical: 40
Outcome
• Freedom from all seizures including auras: 38% surgery vs 3% medical (P<0.001)
• NNT for seizure freedom (impairing awareness): 2 (95% CI 1.5–3)
• NNT for freedom from all seizures: 3 (95% CI 2–5)
• QOLIE-89 score: significantly higher in surgical group (P<0.001)
• Median seizure improvement: 100% (surgical) vs 34% (medical) (P<0.001)
• Surgical complications: 4/40 (10%) — thalamic infarct, wound infection, verbal memory decline
• Deaths: 0 surgical vs 1 medical (sudden unexplained death in epilepsy)
Bottom Line
Surgery is dramatically superior to prolonged medical therapy for temporal-lobe epilepsy. At one year, 58% of surgical patients were free of seizures impairing awareness compared to only 8% of medically treated patients (NNT = 2). Surgery also significantly improved quality of life. This was the first randomized trial demonstrating the superiority of epilepsy surgery, ending decades of reliance on observational data and establishing Level 1 evidence for surgical referral in drug-resistant temporal-lobe epilepsy.
Major Points
- First-ever randomized controlled trial comparing epilepsy surgery to medical therapy — prior evidence was entirely observational
- Dramatic efficacy difference: 58% of surgical patients were seizure-free (impairing awareness) at 1 year vs only 8% of medical patients (P<0.001)
- Freedom from ALL seizures including auras: 38% surgical vs 3% medical (P<0.001)
- NNT of 2 for seizure freedom — one of the most effective interventions in all of neurology
- Quality of life (QOLIE-89) was significantly better in the surgical group (P<0.001)
- Surgical complications were relatively infrequent: 4/40 (10%) had adverse effects (thalamic infarct, wound infection, verbal memory decline in 2)
- One death in the medical group (sudden unexplained death in epilepsy — SUDEP), zero in the surgical group
- Of 36 patients who actually underwent surgery, 64% were seizure-free, suggesting even stronger per-protocol efficacy
Study Design
- Study Type
- Single-center, randomized, controlled, parallel-group trial
- Randomization
- Yes
- Blinding
- Assessor-blinded; participants and treating physicians unblinded
- Sample Size
- 80
- Follow-up
- 1 year
- Centers
- 1
- Countries
- Canada
Primary Outcome
Definition: Freedom from seizures impairing awareness at 1 year
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 8% | 58% | - | <0.001 |
Limitations & Criticisms
- Single-center study limits generalizability
- Unblinded for participants and treating physicians; potential bias in self-reported outcomes
- Relatively short follow-up (1 year); long-term outcomes unknown
- Small sample size (n=80)
- 71% had mesial temporal sclerosis — results may not generalize to non-lesional TLE
- No detailed cognitive outcome assessment reported in primary publication
- 10% of the surgical group had 4 of 40 patients with adverse effects but no formal safety comparison
Citation
N Engl J Med 2001;345:311-318