Epilepsy surgery: how accurate are multidisciplinary teams in predicting outcome?

Area of Science:

• Neurosurgery
• Epileptology
• Medical Decision Making

Background:

• Epilepsy surgery is elective, requiring patients to balance risks and potential benefits for informed consent.
• Accurate prediction of postoperative outcomes is crucial for patients considering epilepsy surgery.

Purpose of the Study:

• To evaluate the precision of multidisciplinary team predictions regarding postoperative seizure freedom in epilepsy surgery candidates.
• To compare team predictions with logistic regression models for outcome prediction.

Main Methods:

• A multidisciplinary team provided preoperative outcome predictions for 94 temporal lobe epilepsy patients.
• Patients were followed up one year post-surgery to assess actual outcomes.

Main Results:

• Team predictions showed accuracy for group outcomes with 30-60% seizure-free likelihood.
• Logistic regression models outperformed team estimates for predicting very high (>70%) or very low (<20%) seizure-free chances.
• Non-localizing scalp EEG predicting the need for invasive EEG was a significant indicator of poor outcomes.

Conclusions:

• Logistic regression models can enhance clinical predictions for extreme outcome probabilities (very good or very poor seizure freedom).
• These models may counteract the 'regression to the mean' tendency observed in team decision-making, improving patient counseling.
• Improved prediction accuracy supports better-informed decisions regarding epilepsy surgery.