Predicting Naming Loss After Epilepsy Surgery
A study of patients who underwent epilepsy surgery in the language-dominant temporal lobe identified key risk factors for early naming decline and found that the degree of early decline predicts long-term recovery.
July 14, 2025 — For many patients with epilepsy, surgery on the temporal lobe offers a path toward seizure relief but at a cognitive cost.
A long-term study of 153 patients who underwent temporal lobe epilepsy surgery found that naming function decline was common after dominant left-sided resections, but recovery was possible and often predicted by the extent of early postoperative deterioration.
Among 66 patients with left temporal lobe epilepsy (LTLE), 42% (28 patients) showed a significant naming decline 6 to 12 months after surgery, as measured by the Boston Naming Test (BNT). In contrast, only 3% (3 of 87) of patients with right temporal lobe epilepsy exhibited decline. Of the LTLE patients who declined early, 39% (11 of 28) recovered to their preoperative naming ability at 2 or more years post-surgery, with a mean follow-up of 8.7 years (standard deviation, 4.7; range, 2.0–23.1).
Risk factors for early naming decline included later age at seizure onset, lower verbal memory performance before surgery, and posterior temporal resections of 40 mm or more from the temporal pole. Each additional year in age at seizure onset increased the risk of naming decline by 6.9% (odds ratio [OR], 1.069; 95% confidence interval [CI], 1.004–1.139; P = .037). A presurgical verbal memory z-score below –0.95 also significantly predicted decline (OR, 0.351; 95% CI, 0.135–0.913; P = .032).
Among patients with seizure onset after age 16 and posterior resections, 94% experienced naming decline. By comparison, patients with early seizure onset and posterior resections had an 18% risk. Patients with anterior resections and late seizure onset had a 44% risk.
The strongest predictor of long-term naming outcome was the degree of early decline. Patients who lost fewer than 10 BNT items at early follow-up were more likely to recover. Those with larger declines were unlikely to return to baseline. At late follow-up, 26% of LTLE patients had persistent naming deficits averaging 18.8% below preoperative performance.
Katrin Walther, MD, of the Department of Neurology, Universitätsklinikum Erlangen in Germany, and co–lead author of the study, stated, “Patients with a decline of more than 10 items at the BNT at early FU were at high risk for a persisting naming decline.”
Neither seizure outcome nor antiseizure medication status influenced recovery. Demographic variables such as sex, age at surgery, education level, and pathology (e.g., hippocampal sclerosis, tumor) were not significantly associated with persistent deficits or recovery.
Although naming recovery was observed in some patients over time, early decline provided a reliable predictor of long-term outcome. The authors noted, “For 25% of patients in our cohort with resection in the language-dominant temporal lobe, naming ability remains on average 19% below the preoperative level several years after surgery.”
The findings support incorporating cognitive risk factors into preoperative counseling and highlight the importance of early postoperative cognitive assessment for identifying patients who may benefit from long-term rehabilitation strategies.
The authors reported no conflicts of interest.
Source: https://conexiant.com/neurology/articles/predicting-naming-loss-after-epilepsy-surgery/
Journal of Neurology, Neurosurgery, & Psychiatry