Summary:Purpose: Decreased memory function represents the area of greatest neuropsychological morbidity after anterior temporal lobectomy (ATL), particularly for left ATL candidates. We wished to identify easily derived demographic and neuropsychological predictors of risk of pre-to postoperative memory decline using only information available preoperatively.Methods: We assessed decline in memory as measured by the California Verbal Learning Test (CVLT) by deriving multiple regression equations using the following measures as independent variables: age at onset, chronological age at time of surgery, sex, Full Scale IQ (FSIQ), level of education, and preoperative memory scores. In all, 203 patients (93 males, 110 females), undergoing ATL (107 left, 96 right) with preoperative and 6-month postoperative testing, were examined.Results: The combination of age, FSIQ, sex, side of surgery and preoperative score was highly predictive (p-values < 0.0001) of postoperative memory scores. Higher postoperative scores were associated with higher preoperative score, younger chronological age, higher FSIQ, female sex, and right side of resection. Reliable change index (RCI) values were used to estimate meaningful decline on the total score across five trials. Logistic regression analysis showed preoperative score and age to be predictors of RCI decline for left-sided resections. Sensitivity of decline (290th centile RCI) prediction was 56%, and specificity was 95%. Validation in 30 patients from a separate population of patients undergoing left ATL produced similar figures.
Conclusions:The derived regression equations can accurately predict verbal memory decline on a list-learning task in -50% of individual patients undergoing ATL, and falsepositive prediction errors are very rare. Key Words: Epilepsy-Memory-Surgery-Temporal lobectomy.Decreased memory function represents the greatest potential neuropsychological morbidity after performance of anterior temporal lobectomy (ATL), particularly for verbal memory measures in patients undergoing left-sided resection (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20). Right-sided resections pose less consistent risk of memory impairment, possibly reflecting the absence of currently available adequate measures of nonverbal memory. Verbal memory deficit in association with right temporal lobe epilepsy (RTLE) is relatively rare and may be due to the presence of concomitant left mesial temporal abnormality, as demonstrated by magnetic resonance imaging (MRI) spectroscopy (21). Identified risk factors for postoperative memory decline after left ATL are older age at time of surgery, later age at onset of unprovoked seizures, and better preoperative memory performance (1)(2)(3)11,22 Males have been shown to be at greater risk than females in some series (23,24). Better preoperative memory performance is also a risk factor for post-ATL decline for right-sided resections ( I 1). For left-sided resections, the principal risk factor is the absence of hippocampal sclerosis (HS...