Objective To identify clinical and cognitive predictors of psychiatric and seizure outcome following temporal lobe epilepsy (TLE) surgery.

Method Medical records of 280 patients who underwent temporal lobe surgery at the National Hospital for Neurology and Neurosurgery, Queen Square, from 1997 to 2007 were reviewed. Pre- and postoperative psychiatric conditions were identified from medical records in addition to information on seizure recurrence and neuropsychological status. Logistic regression analysis was used to identify predictors of having a de novo psychiatric diagnosis in the 4 years following surgery and also to identify factors predicting seizure freedom over this period. Adjustments were made for preoperative cognitive deficits, interictal bilateral EEG abnormalities and chronicity and severity of TLE.

Results The most prevalent postoperative psychiatric diagnoses were mood (14%, of which a third were de novo) and anxiety disorders (5%, of which half were de novo), followed by interictal psychosis (3%, of which a third were de novo). Patients with a preoperative psychiatric history had over five times the odds of a postoperative psychiatric condition (OR 5.80, 95% CI 3.31 to 10.17, p<0.001). Overall, 24% (68/280) patients developed de novo psychopathology in the 4 years following surgery. No clinical or cognitive predictors were independently related to having a de novo diagnosis. From patients with seizure data available, 49% (127/258) remained seizure free for 4 years after surgery. Patients with a history of generalised seizure(s) (OR 0.47, 95% CI 0.25 to 0.90, p=0.02) and those with a preoperative psychiatric diagnosis (OR 0.53, 95% CI 0.28 to 0.98, p=0.04) were significantly less likely to remain seizure free.

Conclusion Postoperative psychopathology is a troublesome complication of TLE surgery that has been neglected in previous surgical series. Our data demonstrate a strong association between a lifetime psychiatric history and the failure to achieve postsurgical seizure freedom. Our findings do not imply that TLE patients with a co-morbid psychiatric diagnosis should be excluded from surgical consideration. Rather, it needs to be considered a pertinent risk factor for seizure outcome. These findings underscore the need for a thorough preoperative neuropsychiatric assessment in order to ensure patients are fully informed and counselled regarding postoperative risks. Prospective studies are warranted to increase our understanding of the mechanism(s) mediating a worse postsurgical seizure outcome.