Abstract
Commentary
The observation that therapeutic control of psychiatric and epileptic disorders is correlated initially was addressed in a post-surgical study in 1963. Examining a large group of patients who underwent temporal lobectomy, Falconer et al. demonstrated that postoperative improvements in seizure control corresponded to improvements in the psychiatric domain (9). It is interesting to note that these surgical patients were largely identified from a psychiatric hospital and most had a preoperative psychiatric disorder. Subsequent studies of postoperative cohorts from epilepsy practices have demonstrated increases in psychiatric symptoms postoperatively (particularly shortly after surgery) that are independent of seizure control (10,11).
More recently, the presence or absence of a psychiatric disorder has been identified as a predictor of a worse outcome in epilepsy (12), especially as a predictor of postsurgical seizure outcome, as illustrated herein by Kanner et al. The study center was uniquely qualified to perform this trial, as all presurgical patients undergo a semistructured interview, based on the Structured Clinical Interview for DSM IV disorders. In this retrospective study of 100 consecutive patients who underwent anterior temporal lobectomy (ATL), the authors examined the lifetime psychiatric history, age of onset and duration of epilepsy, extent of surgical resection, and Engel class seizure outcomes. Patients generally remained on at least one AED indefinitely following surgery. Fifty-six percent of the subjects had a presurgical lifetime history of a psychiatric disorder, consistent with the high prevalence reported in other studies (1,–4). Mood and anxiety disorders accounted for most of the psychiatric history. Two years after the surgery, 44% of the subjects in the cohort were taking a psychotropic drug—a sobering statistic. The use of psychotropic medications postsurgically was not associated with seizure outcome.
The study's findings reflect the theme that psychiatric disorders and epilepsy outcome are concordant. The absence of a lifetime history of psychiatric illness strongly predicted a class 1A (i.e., no seizures, no auras) postsurgical outcome. In fact, those without a psychiatric history were 13 times more likely to have a class 1A outcome than those with a lifetime history of psychiatric illness. When assessing class 1A and class 1B (i.e., only auras but no disabling seizures) together, patients without a history of psychiatric disorders were seven times more likely to experience this outcome. The effect of an absence of psychiatric illness was the most predictive when considering classes 1A, 1B, and IC (i.e., disabling seizures caused by medication discontinuation or having had up to 3 disabling seizures during the first postsurgical year only) together, as patients without a history of psychiatric disorder were 16 times more likely to have a favorable outcome than those with the presence of a psychiatric history. Other nonpsychiatric factors also were associated with favorable seizure outcomes, including a larger resection of mesial structures, presence of mesial temporal sclerosis (MTS), and absence of a presurgical history of generalized tonic–clonic seizures.
There is other evidence that the presence of psychiatric disorders is associated with worse seizure outcomes, both medical and surgical. In a large cohort of subjects with newly diagnosed epilepsy, a history of psychiatric comorbidity was associated with progression to refractory disease (12). In another study, which examined postsurgical temporal lobectomy patients, a past psychiatric history plus new postsurgical psychiatric symptomatology predicted poor seizure outcome (10). Interestingly, although epilepsy surgery is associated with a high prevalence of postsurgical psychiatric symptomatology, a past history of psychiatric disorders does not seem to predict postsurgical psychiatric symptoms (10).
Why are psychiatric history and postsurgical seizure outcome related? As the authors discuss, the mechanisms remain largely unknown, despite ample evidence for common pathologic mechanisms in epilepsy and depression. Perhaps, mood disorders are a marker for a more extensively damaged cortex (i.e., a larger or more diffuse epileptogenic zone). It is also important to consider the possibility of nonepileptic seizures, particularly as postoperative anxiety disorders may mimic seizures. This effect was accounted for in the current study, as all patients who experienced more than three seizures postsurgically underwent video-EEG monitoring to rule out the possibility of postsurgical nonepileptic seizures.
Since the publication of the paper by Kanner and colleagues, similar findings have been reported, highlighting the increasing interest in and recognition of the importance of co-morbid psychiatric disorders and epilepsy surgery (13). What do these findings mean to the clinician? Clearly, it is critical to recognize the need to systematically and thoroughly explore the psychiatric history of all patients with epilepsy, particularly those individuals considering epilepsy surgery. While such a history should not prevent a temporal lobectomy, the possibility that psychiatric disorders may be a marker for a more extensive epileptogenic zone should be considered during the presurgical evaluation, and there may as well be potential implications for presurgical counseling and postsurgical medication decisions.
A number of additional issues that remain to be explored, including outcomes of epilepsy surgeries other than temporal lobectomy and the potential contribution of laterality, were not addressed by Kanner et al. Overall, however, the results of this and similar studies demonstrate the need for greater rigor in the psychiatric assessment of presurgical patients with epilepsy