Presurgical neuropsychological testing predicts cognitive and seizure outcomes after anterior temporal lobectomy

Abstract

We sought to determine significant predictors of seizure and cognitive outcome following surgery for epilepsy. Participants included 41 patients who had undergone anterior temporal lobectomy (ATL). Higher presurgical verbal/language scores and lower nonverbal memory scores were predictive of seizure-free status following ATL. Overall, the presurgical predictors were 93% accurate in discriminating between seizure-free and non-seizure-free patients postsurgery. Surgery in the nondominant-for-language hemisphere was predictive of higher postsurgical verbal/language and verbal memory scores. Higher presurgical visual/construction, nonverbal memory, and verbal/language scores were predictive of better postsurgical verbal/language functioning. Better presurgical verbal/language functioning was predictive of the same skills postsurgically as well as visual/construction outcomes. Exploratory analyses in a subset of participants (n = 25) revealed that dominant and nondominant intracarotid amobarbital (Wada) memory scores added unique variance only for predicting nonverbal memory following ATL. Presurgical neuropsychological testing provides significant and unique information regarding postsurgical seizure freedom and cognitive outcome in patients who have undergone ATL.

Introduction

Surgical treatment is a widely employed, efficacious treatment for medication-refractory temporal lobe epilepsy (TLE) [1], [2], [3], [4], [5]. Surgical treatment for epilepsy, especially temporal lobectomy, has an estimated success rate of 80%, which is generally defined by seizure cessation or a significant reduction in seizure frequency [6], [7], [8], [9]. Few studies consider cognitive outcome when examining surgical success. In light of the large number of patients experiencing postsurgical cognitive changes [10], it is important to identify predictors of cognitive outcome in addition to seizure outcome.

Commonly studied predictors of postsurgical cognitive and seizure outcomes fall into two broad categories: (1) neurological and disease characteristics, and (2) neuropsychological functioning. With respect to the first category, several studies have highlighted the importance of hemispheric dominance for language and memory, typically assessed using the Wada procedure [11]. Patients with higher Wada memory scores in the nonepileptic (i.e., contralateral) hemisphere tend to have better seizure control postsurgery [12], [13], [14], [15], [16]. Both the integrity of the nonepileptic hemisphere and the reduced functional ability of the removed tissue also predict better memory performance following surgery [17], [18], [19], [20], [21], [22].

Extant literature underscores differential outcomes depending on certain presurgical neurological and seizure characteristics. Positive seizure and cognitive outcomes are associated with surgical removal of a sclerotic hippocampus, typically right temporal resection, older age at seizure onset, and shorter seizure duration. For example, research shows that resection of a sclerotic hippocampus has been associated with seizure reduction and cognitive improvement [23], [24], [25], [26], [27], while removal of an intact hippocampus is not advantageous [28], [29], [30]. Additionally, after resection of the nondominant-for-language (generally right) temporal region, patients often show unchanged or slightly improved verbal memory [31], [32], [33], and right-sided (more than left-sided) surgeries are more frequently associated with improvement in seizure frequency [7]. In contrast, language-dominant (generally left) temporal resection is associated with postsurgical deficits across memory [21], [22], [31], [32], [33], [37] and language [28], [34], [35], [36] domains. Research shows that a longer history of epilepsy is associated with poor seizure outcome after surgery [9], [21], [25], [27], [37], [38], and that duration of epilepsy is directly related to decrements in higher cognitive functions in pharmacoresistant temporal lobe epilepsy [40]. These data are consistent with findings that suggest that patients who are younger at the time of seizure onset have less chance of seizure remission [41] and are at higher risk for cognitive impairment after anterior temporal lobectomy (ATL) [19], [42].

The relevant research literature is smaller for preoperative neuropsychological functioning, but notable for predicting postoperative seizure and cognitive outcomes. Patients with higher preoperative IQ scores tend to have better seizure control [43], [44] and improved cognitive performance postsurgery [10], [44]. Additionally, patients with more impaired preoperative verbal memory scores tend to have better seizure outcomes after ATL [39], [45] and are less likely to show postsurgical verbal memory decline than those with better preoperative verbal memory [30], [46].

Various studies have examined these predictors individually or in limited combinations, but to our knowledge no published study has examined all of these collectively to predict seizure and cognitive outcome. As a result, when clinicians attempt to make predictions about postsurgical outcomes, they are often faced with a host of pre- and postoperative variables to consider but only piecemeal research with which to work when attempting to make evidence-based decisions about patient care. Thus, clinicians are limited when informing patients of the potential risks and benefits of surgical intervention, an important clinical consideration given that neurosurgical risks can be substantial [47]. To address this knowledge gap, the current study examines both seizure outcome and the cognitive effects of surgery, and incorporates measures of both categories of preoperative predictors (neurological/disease characteristics and neuropsychological functioning) to provide a more complete and comprehensive picture for clinicians and patients facing difficult decisions.

The primary aim of the current study was to identify prognostic variables associated with postsurgical seizure remission and preserved or improved cognitive functioning. On the basis of prior research, it was hypothesized that specific seizure-related characteristics (i.e., hippocampal integrity, seizure focus, age at seizure onset, and duration of seizures) and preoperative cognitive abilities would predict postoperative seizure activity and cognitive outcome. Exploratory analyses were also run using Wada memory scores as additional predictors of seizure and cognitive outcome within the subset of patients who were given the Wada procedure prior to surgery.