Short-term neurocognitive outcomes following anterior temporal lobectomy

Epilepsy & Behavior 62 (2016) 140–146

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Short-term neurocognitive outcomes following anterior temporal lobectomy Philip S. Lee a,⁎, Jamie Pardini a, Rick Hendrickson b, Vincent Destefino a, Alexandra Popescu b, Gena Ghearing b, Arun Antony b, Jullie Pan b, Anto Bagic b, Danielle Wagner a, R. Mark Richardson a a b

Department of Neurological Surgery, University of Pittsburgh Medical Center, UPMC Presbyterian, Suite B400, 200 Lothrop Street, Pittsburgh, PA 15213, USA Department of Neurology, University of Pittsburgh Medical Center, 8111 Kaufmann Medical Building, 3471 Fifth Avenue, Pittsburgh, PA 15213, USA

a r t i c l e

i n f o

Article history: Received 25 February 2016 Revised 28 April 2016 Accepted 18 June 2016 Available online xxxx Keywords: Epilepsy Surgery Neuropsychology Verbal memory

a b s t r a c t Changes in cognitive function are a well established risk of anterior temporal lobectomy (ATL). Deficits in verbal memory are a common postoperative finding, though a small proportion of patients may improve. Postoperative evaluation typically occurs after six to 12 months. Patients may benefit from earlier evaluation to identify potential needs; however, the results of a formal neuropsychological assessment at an early postoperative stage are not described in the literature. We compared pre- and postoperative cognitive function for 28 right ATL and 23 left ATL patients using repeated measures ANOVA. Changes in cognitive function were compared to ILAE seizure outcome. The mean time to postoperative neuropsychological testing was 11.1 weeks (SD = 6.7 weeks). There was a side × surgery interaction for the verbal tasks: immediate memory recall (F(1,33) = 20.68, p b 0.001), short delay recall (F(1,29) = 4.99, p = 0.03), long delay recall (F(1,33) = 10.36, p = 0.003), recognition (F(1,33) = 5.69, p = 0.02), and naming (F(1,37) = 15.86, p b 0.001). This indicated that the left ATL group had a significant decrement in verbal memory following surgery, while the right ATL group experienced a small but significant improvement. For the right ATL group, there was a positive correlation between ILAE outcome and improvement in immediate recall (r = −0.62, p = 0.02) and long delay recall (r = −0.57, p = 0.03). There was no similar finding for the left ATL group. This study demonstrates that short-interval follow-up is effective in elucidating postoperative cognitive changes. Right ATL was associated with improvement in verbal memory, while left ATL resulted in a decrement in performance. Improvement in the right ATL group was related to improved seizure outcome. Short-interval follow-up may lend itself to the identification of patients who could benefit from early intervention. © 2016 Elsevier Inc. All rights reserved.

1. Introduction Surgery is often the treatment of choice for patients with medically refractory focal epilepsy. Anterior temporal lobectomy (ATL) with amygdalohippocampectomy is the most common epilepsy surgery. Although generally considered safe, ATL can result in a variety of cognitive changes in up to 44% of patients [1]. Cognitive deficits may even be severe enough to warrant interventions or accommodations to perform activities of daily life. Postoperative neuropsychological testing typically is performed and compared to preoperative results in order to quantify the cognitive effects of surgical intervention. However, optimal testing intervals at which to detect significant cognitive changes and to direct patients toward appropriate cognitive rehabilitation are unknown.

Abbreviations: ATL, anterior temporal lobectomy; ILAE, International League Against Epilepsy. ⁎ Corresponding author. Tel.: +1 412 647 6777; fax: +1 412 647 6483. E-mail address: [email protected] (P.S. Lee).

http://dx.doi.org/10.1016/j.yebeh.2016.06.019 1525-5050/© 2016 Elsevier Inc. All rights reserved.

Several studies have examined the cognitive effects of anterior temporal lobectomy or so-called selective amygdalohippocampectomy. As expected, these surgeries may result in memory and other deficits. Deficits appear to be dependent on the side of surgery, matching with functional lateralization. For instance, it is well established that left temporal lobe surgery can result in verbal memory deficits [2]. Indeed, a recent meta-analysis found that 44% of patients undergoing left temporal lobe surgery had postoperative verbal memory deficits [1]. Though findings have not been as consistent, there is some evidence that suggests that right temporal lobe surgery can result in visual memory deficits [3]. Most often, the extent of cognitive change is not bothersome to the patient, especially if a patient's seizures are stopped or the seizure burden is significantly reduced. Interestingly, data also suggest that cognitive function can improve following epilepsy surgery. Specifically, a small percentage of patients show improvement in verbal and/or visual memory, regardless of the side of the temporal lobe surgery [4]. Improvements in attention and psychomotor speed are most common and may be related to the reduction or elimination of seizure activity and reduction in use of antiepileptic medications that are known to

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146

have a multitude of cognitive side effects [5–7]. Though the mechanisms by which both positive and negative effects of ATL occur are still being elucidated, the above findings are consistent, and this experience is used to frame discussions of operative risks with prospective patients. A primary motivator for patients to pursue epilepsy surgery appears to be the stigma associated with (often public) seizures, as the fear of potential postoperative cognitive deficits does not differ between patients who chose medical versus surgical management [8]. Thus, many patients choose to undergo epilepsy surgery despite the risks of changes in cognitive function. In most studies of outcome, patient satisfaction and neuropsychological outcome is measured at six months to one year following surgery. Following patients to longer intervals often shows that memory for some will improve over time to better-thanpreoperative levels. For instance, Grammaldo et al. [9] found that patients undergoing left ATL showed impairment in verbal memory at one year. However, many of those impaired did regain some function or return to their postoperative baseline at two years, resulting in no significant difference pre- and postoperatively for the total group. Similarly, Helmstaedter et al. [10] found that patients undergoing ATL showed memory deficits following surgery but that recovery of memory function at long-term follow-up after 2–10 years was related to successful treatment of seizures. In contrast, those patients who continued to have seizures were more likely to continue to show memory deficits. For patients who experience lasting cognitive impairments following surgery, there is evidence that memory training is helpful in improving function [11]. In fact, intervention as early as within three to 15 days of surgery has been shown to mitigate the effects of memory decline following temporal lobe resection [12]. Taken together, it has been well established that epilepsy surgery can lead to cognitive deficits and sometimes to improvements. Though in some instances deficits can be transient, there are certainly patients who experience lasting impairment. However, there is no clear consensus on the best interval for evaluating cognitive function postoperatively to determine whether these impairments exist, in order to expeditiously direct patients toward the appropriate cognitive rehabilitation therapy. Understanding the timing of these deficits is important, as early intervention can potentially minimize the effect on a patient's quality of life. Therefore, the purpose of this study was to compare preoperative neuropsychological performance in patients undergoing epilepsy surgery with early postoperative performance. We hypothesized that short-term follow-up evaluation beginning at six weeks after resective epilepsy surgery would reveal changes in cognitive function that inform patient care.

2. Methods The Institutional Review Board of the University of Pittsburgh Medical Center approved this study, and all patients signed informed consent for inclusion. All data were deidentified for use in analyses. All patients undergoing ATL for medically intractable epilepsy had pre- and postoperative neuropsychological evaluation by a licensed neuropsychologist. Patients were administered standardized neuropsychological measures evaluating language (Boston Naming Test), verbal (California Verbal Learning Test or Rey Auditory Verbal Learning Test) and visual memory (Rey Complex Figure Test), processing speed and motor coordination (Grooved Pegboard), attention, and executive function (Wechsler Adult Intelligence Scale Digit Span and Digit-Symbol Coding Subtests, Trail-making Test). In some cases, the measure of verbal memory administered varied across subjects according to the neuropsychologist's preference or patient's level of preoperative cognitive function. However, the same measure was always used for the pre- and postoperative evaluation of a given patient. Performance on these measures was converted to z-scores, to allow for direct comparisons across subjects. Otherwise, repeat performance was compared based on scaled scores or t-scores.

141

Preoperative seizure frequency was estimated for all patients from their outpatient neurology clinic visit documentation. Seizure frequency was categorized as occurring either daily (i.e., ≥ one seizure per day), weekly (i.e., ≥one seizure per week but bdaily), monthly (i.e., ≥one seizure per month but bweekly), or N monthly (i.e., ≥one seizure every two months but b monthly). No patients in this study had seizures characterized less frequently than one every two months. Postoperative outcome was characterized according to the International League Against Epilepsy (ILAE) outcome classification scheme [13]. During the interval between pre- and postoperative neuropsychological evaluation, there was no change in antiepileptic medications for any of the patients. Spearman correlation was performed to compare preoperative seizure frequency and postoperative ILAE outcome. Mann–Whitney test was used to compare the groups for ordinal data. Chi-squared test was used to compare the groups for nominal data. Analysis of variance (ANOVA) was performed to evaluate differences in neuropsychological function across the groups. Repeated measures ANOVA was used to compare pre- and postoperative neuropsychological functioning within the groups. If a measure demonstrated a significant change on repeated measures ANOVA, a reliable change index (RCI) score was calculated using the formula, RCI = (posttest score − pretest score) / standard error of measurement [14]. The RCI was calculated to address possible confounds such as practice effects or the result of statistical anomalies such as regression to the mean. Since there was no typical control group in this study, the standard error of measurement was taken from the results of both clinical groups. A p-value of ≤0.05 was considered to be statistically significant. Standard statistical processing software was used to perform the analyses (SPSS version 22; IBM, Armonk, NY). 3. Results A total of 53 patients (33 male, 20 female) underwent surgery at the University of Pittsburgh Medical Center from September 2011 until June 2015. No patients from this time interval were excluded from analysis. Patients were divided into groups according to side of surgery. Twenty-eight patients (53%) underwent right ATL, and 25 (47%) underwent left ATL. The majority of patients (24) underwent surgery because of mesial temporal sclerosis (MTS). The pathology and recommendation for surgery for these patients were based on the decision of a multidisciplinary surgical epilepsy team; decisions were based on clinical seizure semiology, neuroimaging demonstrating MTS with an associated lateralized seizure focus on MEG or SPECT, scalp EEG, and neuropsychological profile. For the remaining patients, eligibility for surgery was based on the recommendation of the surgical epilepsy team following invasive intracranial seizure monitoring demonstrating a unilateral mesial temporal seizure focus or the presence of a clear

Table 1 Diagnostic patient characteristics and surgery by group.

Neuroimaging Normal MTS Cortical dysplasia/heterotopia Tumor Astrocytoma Oligodendroglioma Ganglioglioma Cavernous malformation Preresective intracranial monitoring None Subdural grid and depth electrodes Stereo-EEG electrodes Surgery type Asleep Asleep with ECoG Awake with mapping

Right ATL

Left ATL

10 10 1

6 14 4

1 1 4 1

1

18 7 3

16 7 2

19 8 1

7 1 17

142

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146

brain abnormality (e.g., tumor, cavernous malformation). Preresection characteristics for the patients are displayed in Table 1. There was a significant difference between the groups in neuroimaging findings, with the left ATL group being more likely to have MTS (χ2 (2, N = 53) = 9.29, p = 0.01). In contrast, there was no difference in preresection intracranial monitoring (χ2 (2, N = 53) = 0.15, p = 0.93). All patients undergoing ATL had resection of both mesial and lateral structures. The extent of resection was consistent with established clinical practice, generally a maximum of the anterior 6 cm of nondominant and 4.5 cm of dominant temporal lobe. Language-dominant resections were further restricted by positive language sites when identified with awake mapping or by sparing of the majority of the superior temporal gyrus during asleep resection. Mesial resection included the amygdala and at least the anterior 2/3 of the hippocampus. Intracranial electrodes were placed one week prior to resection for patients who underwent this diagnostic study. The one exception to this was a patient from the left ATL group who developed a transient neurologic deficit several days after implantation because of mass effect; this patient had his electrode array removed emergently and two months passed before resection was performed. The number of patients undergoing asleep resection with no intraoperative electrocorticography (ECoG), asleep resection with intraoperative ECoG, or awake resection with language mapping is noted in Table 1. There was a significant difference in likelihood of awake functional mapping, as this was more common in patients undergoing left ATL (χ2 (2, N = 53) = 24.89, p b 0.001). The mean age of the patients was 42.8 years (SD = 12.9 years), and the mean level of education was 13.4 years (SD = 2.3 years). There was no difference in age between the left ATL (M = 45.9, SD = 13.9) and right ATL (M = 40.2, SD = 11.6) groups (t(51) = − 1.63, p = 0.11). Nor was there a difference in level of education between the left ATL (M = 13.5, SD = 1.9) and right ATL (M = 13.1, SD = 2.6) groups (t(51) = − 0.71, p = 0.48). Twenty-six (93%) of the right ATL patients were right-handed, while 22 (88%, 1 ambidextrous) of the left ATL patients were right-handed. Formal language dominance testing consisting of either intracarotid sodium amobarbital testing, magnetoencephalography, or functional magnetic resonance imaging was obtained for 37 (70%) of the patients. Sixteen right ATL patients showed left hemisphere, one showed right hemisphere, and one showed bilateral dominance. For the left ATL patients, 16 showed left hemisphere, one showed right, and two had bilateral dominance. Preoperative seizure frequency for the right ATL patients was as follows: nine (28%) with daily seizures, nine (28%) with weekly, six (21%) with monthly, three (11%) with Nmonthly, and one (4%) without clear documentation; for the left ATL patients: eight (32%) with daily, 10 (40%) with weekly, two (8%) with monthly, three (12%) with N monthly, and two (8%) without clear documentation. The ILAE outcome following surgery is listed in Table 2. The mean time to follow-up for ILAE outcome was 10.6 weeks (SD = 6.5 weeks). There was no difference between the groups in preoperative seizure frequency (U = 289.5, p = 0.67) or ILAE outcome (U = 298.5, p = 0.21). In addition, there

Table 2 Number of patients per ILAE outcome classification. ILAE classification

Right ATL

Left ATL

1 2 3 4 5 6

20 2 1 2 3 0

20 2 2 1 0 0

1: seizure-free, no auras; 2: only auras, no other seizures; 3: 1–3 seizure days per year, ±auras; 4: 4 seizure days per year to 50% reduction of baseline seizures, ±auras; 5: b50% reduction of baseline seizure days to 100% increase of baseline seizure days, ±auras; 6: N100% increase of baseline seizure days, ±auras.

was no correlation between preoperative seizure frequency and ILAE outcome (r = −0.01, p = 0.93). Preoperative neuropsychological data were analyzed using ANOVA, with group (left ATL vs. right ATL) as the primary factor. Preoperative testing indicated that there were no group differences in preoperative cognitive performance, though there was a trend toward slightly worse list memory learning performance in the left ATL group (F(1,42) = 3.41, p = 0.07). Preoperative neuropsychological performance and ANOVAs comparing the effect of group are displayed in Table 3. As noted above, there was a significant difference in the presence of MTS in the left and right ATL groups. For this reason, we performed a subgroup analysis using ANOVAs with group (left ATL vs. right ATL) as the primary factor. There were no significant group differences in the MTS subgroup, and there was no longer a trend toward a difference in memory learning performance. This indicated that, although the left ATL group was more likely to have MTS, the presence of MTS did not appear to differentially affect their preoperative functioning. The mean time to obtaining postoperative neuropsychological testing was 11.1 weeks (SD = 6.7 weeks). Pre- and postoperative

Table 3 Preoperative performance on neuropsychological measures. Measure Verbal Memory Immediate Recalla Right ATL Left ATL Verbal Memory Short Delay Recalla Right ATL Left ATL Verbal Memory Long Delay Recalla Right ATL Left ATL Verbal Memory Recognitiona Right ATL Left ATL Boston Naming Testa Right ATL Left ATL Grooved Pegboard — dominant handb Right ATL Left ATL Grooved Pegboard — nondominant handb Right ATL Left ATL Rey Complex Figure Copyc Right ATL Left ATL Rey Complex Figure Delayed Recallc Right ATL Left ATL Trail-making Test A Timeb Right ATL Left ATL Trail-making Test A Errorsc Right ATL Left ATL Trail-making Test B Timeb Right ATL Left ATL Trail-making Test B Errorsc Right ATL Left ATL WAIS Digit Spand Right ATL Left ATL WAIS Digit-Symbol Codingd Right ATL Left ATL a b c d

z-Score. t-Score. Raw score. Scaled score.

Mean (SD)

F

Sig

3.41

0.07

2.56

0.12

2.69

0.11

1.10

0.30

0.11

0.74

0.81

0.37

2.18

0.15

1.05

0.31

2.26

0.14

0.26

0.61

1.48

0.23

0.05

0.82

0.09

0.76

0.55

0.46

0.09

0.76

−1.24 (1.28) −1.94 (1.15) −1.39 (0.95) −1.86 (1.08) −1.34 (1.18) −1.97 (1.30) −1.39 (2.11) −0.84 (1.24) −1.56 (1.79) −1.73 (1.85) 36.92 (14.24) 40.55 (13.26) 36.20 (13.83) 42.18 (13.90) 30.98 (4.43) 32.18 (3.55) 12.15 (5.89) 14.66 (5.60) 42.58 (10.28) 44.18 (11.51) 0.15 (0.37) 0.05 (0.21) 45.23 (9.02) 44.55 (11.75) 0.69 (1.44) 0.82 (1.33) 9.46 (3.27) 8.82 (2.61) 7.31 (3.31) 7.59 (3.03)

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146

neuropsychological testing data were compared using repeated measures ANOVA, with group (left ATL vs. right ATL) and surgery (preoperative performance vs. postoperative performance) as factors. The main effects of surgery for each task are listed in Table 4. There was a significant main effect of surgery on time to perform simple motor sequencing on the Trail-making Test (F(1,37) = 7.77, p = 0.01). This indicated a significant improvement in cognitive processing speed following surgery, regardless of the side of resection. The pre- and postoperative differences in performance by side of surgery and interaction effects are shown in Table 5. Interaction effects between the factors (group × surgery) were evaluated. There were several interaction effects, which indicated that the side of surgery influenced postoperative cognitive outcome. Specifically, there was a side × surgery interaction for all of the verbal tasks: immediate verbal memory recall (F(1,33) = 20.68, p b 0.001), recall after a short delay (F(1,29) = 4.99, p = 0.03), recall after a long delay (F(1,33) = 10.36, p = 0.003), verbal memory recognition (F(1,33) = 5.69, p = 0.02), and verbal naming (F(1,37) = 15.86, p b 0.001). These interactions indicated that the leftsided group had a significant decrement in performance following surgery. The right-sided group, however, experienced a small but significant improvement in verbal performance. The changes in performance for these tasks are demonstrated in Fig. 1. In order to characterize the number of patients who showed significant postsurgical change on these specific measures, the RCI was calculated for each patient. The percentage of patients in each group showing significant improvement (i.e., RCI N 1.96) following surgery is listed in Table 6. To explore the possible confound of language lateralization on changes in performance, a subgroup analysis was performed examining changes in patients with documented left-lateralized language only. This demonstrated similar findings to the total group analysis. The pre- and postoperative differences in performance by side of surgery and interaction effects for this subgroup are shown in Table 7. Analysis of this subgroup was consistent with the total group results. The only exception was that the change in verbal memory recognition was no longer significant when examining only the documented left language-dominant patients. There were not enough patients with right lateralized or bilateral language function to perform statistically valid comparisons with these groups. Given the interaction effects, correlations were performed to further examine the relationship between change in neurocognitive performance on verbal tasks and seizure outcome. This revealed that the degree of improvement in verbal performance for the right-sided group was correlated to their ILAE outcome. Specifically, there was a negative correlation between degree of improvement and ILAE outcome for

Table 4 Main effects of surgery. Measure

F

Sig

Verbal Memory Immediate Recalla Verbal Memory Short Delay Recalla Verbal Memory Long Delay Recalla Verbal Memory Recognitiona Boston Naming Testa Grooved Pegboard — dominant handb Grooved Pegboard — nondominant handb Rey Complex Figure Copyc Rey Complex Figure Delayed Recallc Trail-making Test A Timec Trail-making Test A Errorsc Trail-making Test B Timec Trail-making Test B Errorsc WAIS Digit Spand WAIS Digit-Symbol Codingd

0.58 1.91 1.57 3.20 2.67 3.00 0.51 2.70 0.47 7.77 0.64 3.01 0.41 0.39 0.44

0.45 0.18 0.22 0.08 0.11 0.09 0.48 0.11 0.50 0.01⁎ 0.43 0.09 0.47 0.53 0.51

a

z-Score. t-Score. Raw score. d Scaled score. ⁎ Statistically significant. b c

143

Table 5 Change in pre- and postoperative performance on neuropsychological measures and interaction effects. Measure Verbal Memory Immediate Recalla Right ATL Left ATL Verbal Memory Short Delay Recalla Right ATL Left ATL Verbal Memory Long Delay Recalla Right ATL Left ATL Verbal Memory Recognitiona Right ATL Left ATL Boston Naming Testa Right ATL Left ATL Grooved Pegboard — dominant handb Right ATL Left ATL Grooved Pegboard — nondominant handb Right ATL Left ATL Rey Complex Figure Immediate Recallc Right ATL Left ATL Rey Complex Figure Copyc Right ATL Left ATL Trail-making Test A Timeb Right ATL Left ATL Trail-making Test A Errorsc Right ATL Left ATL Trail-making Test B Timeb Right ATL Left ATL Trail-making Test B Errorsc Right ATL Left ATL WAIS Digit Spand Right ATL Left ATL WAIS Digit-Symbol Codingd Right ATL Left ATL

Mean change (SD)

F

Sig

20.68

b0.001⁎

4.99

0.03⁎

10.36

0.003⁎

5.69

0.02⁎

15.86

b0.001⁎

0.01

0.92

0.29

0.60

0.65

0.43

0.62

0.44

0.27

0.61

0.00

0.98

0.00

1.00

3.24

0.08

1.28

0.27

0.09

0.77

+0.84 (1.07) −0.60 (0.80) +0.20 (1.33) −0.86 (1.29) +0.86 (0.82) −0.38 (1.31) +0.31 (2.24) −2.15 (3.48) +0.88 (1.49) −2.15 (2.93) +2.88 (11.40) +3.25 (10.15) +1.78 (9.02) +0.25 (8.55) −1.47 (3.87) −0.50 (2.99) −1.28 (5.52) +0.08 (4.43) +5.89 (9.12) +4.05 (12.76) −0.05 (0.52) −0.05 (0.22) +3.00 (9.73) +3.00 (11.72) +0.21 (1.08) −0.50 (1.36) +0.16 (1.61) −0.56 (2.20) +0.39 (2.40) +0.15 (2.58)

a

z-Score. t-Score. Raw score. d Scaled score. ⁎ Statistically significant. b c

immediate verbal memory recall (r = − 0.62, p = 0.02) and recall after a long delay (r = −0.57, p = 0.03). This indicated that greater improvement in verbal memory performance was associated with better seizure control. There was no similar relationship found for the leftsided group, and there was no significant correlation between preoperative seizure frequency and change in verbal memory performance for either group. The relationship between verbal memory improvement and ILAE outcome is demonstrated in Fig. 2. 4. Discussion In this study, we demonstrated that short-term follow-up was sensitive in detecting changes in motor sequencing and verbal memory following ATL. There were no significant changes in most domains of neurocognitive function following epilepsy surgery, reinforcing the fact that epilepsy surgery is safe, as the likelihood of significant postoperative cognitive impairment is low. In fact, there was a significant main effect of surgery on improved cognitive processing speed. Consistent with prior work, we found decreased measures of verbal memory

144

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146

b)

-1.00

-1.00

Left ATL

-3.00

-3.00

-3.00 Preop Immediate Postop Immediate Recall Recall Error Bars: +/- 1 SE

-1.00

-2.00

-2.00

Preop Long Delay Postop Long Delay Recall Recall

Preop Short Delay Postop Short Delay Recall Recall

Error Bars: +/- 1 SE

Error Bars: +/- 1 SE

d)

Right ATL

0.00

Left ATL

-2.00

e) 0.00

0.00

Right ATL Left ATL

Mean (Z-Score)

Mean (Z-Score)

Right ATL

0.00

Mean (Z-Score)

Mean (Z-Score)

c)

Right ATL Left ATL

0.00

Mean (Z-Score)

a)

-1.00

-2.00

-3.00

Right ATL Left ATL

-1.00

-2.00

-3.00

-4.00

-5.00

-4.00

Preop Boston Naming

Preop Recognition Postop Recognition

Postop Boston Naming

Error Bars: +/- 1 SE

Error Bars: +/- 1 SE

Fig. 1. Change in performance for a) immediate verbal memory recall, b) recall after a short delay, c) recall after a long delay, d) verbal memory recognition, and e) verbal naming.

performance in patients undergoing left ATL. In contrast, patients undergoing right ATL typically exhibited improved verbal performance. Moreover, this improvement in verbal memory performance was related to better postoperative seizure outcome. Overall, the results of this study are consistent with the largest metaanalysis of neuropsychological outcome following epilepsy surgery by Sherman et al. [1], which demonstrated a risk to verbal memory particularly in patients undergoing left-sided temporal lobe surgery. Furthermore, this study demonstrates that neuropsychological change is measurable on short-term follow-up from epilepsy surgery. Early follow-up is important, as the early identification of cognitive deficits can aid in the early initiation of interventions. Research has demonstrated the benefits of early intervention in the neurocognitive treatment of preterm infants [15], cognitive decline [16], and following traumatic Table 6 Percentage of patients with significant verbal task performance improvement following surgery. Measure List Memory Learning Right ATL Left ATL List Memory Short Delay Right ATL Left ATL List Memory Long Delay Right ATL Left ATL List Memory Recognition Right ATL Left ATL Boston Naming Test Right ATL Left ATL

Percent showing improvement 36 8 14 20 39 16 25 12 32 8

brain injury [17,18]. Similarly, early intervention with individualand group-based interventions is effective in diminishing the negative cognitive effects of epilepsy surgery [11,12]. This study differs from prior reports in that we found a significant improvement in verbal memory for a large portion of our patient population. There was no change in verbal memory performance for all comers following surgery. However, there were changes in verbal memory performance that were dependent on the side of surgery, as a greater proportion of right ATL patients had improvement in verbal memory than left ATL patients. We found postoperative rates of memory improvement in up to 39% of right ATL patients, a rate significantly higher than typically reported in the literature. For instance, Baxendale et al. [4] noted that, in a large study of almost 250 patients, only 22% of right ATL and 9% of left ATL patients experienced these benefits. One possible explanation for the larger improvements noted in this specific population could be due to practice effects, given the short-interval follow-up. Indeed, there is evidence for global improvement in cognitive performance of patients with epilepsy at repeat measurement after seven months [19]. However, improvement was specific to verbal memory performance, which suggests that a global practice effect does not explain all of the variance. In addition, surgery led to an improvement in simple motor sequencing, regardless of side of surgery. Similar to other studies, this suggests that surgical resection of epileptic foci may have led to improvement in this task as a function of improved executive function [19,20]. There are several explanations for why patients improve following ATL. First, it is generally accepted that epileptogenic regions are nonfunctional or, at the least, dysfunctional. For example, functional reorganization of mesial temporal structures after the onset of seizures appears to be essential in the preservation of function. Specifically, patients undergoing left temporal lobectomy with activation of the ipsilateral anterior hippocampus during memory encoding tasks are more likely to have postoperative deficits, while those with greater bilateral

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146 Table 7 Change in pre- and postoperative performance on neuropsychological measures and interaction effects for patients with left lateralized language. Measure

Mean change (SD)

Verbal Memory Immediate Recalla Right ATL Left ATL Verbal Memory Short Delay Recalla Right ATL Left ATL Verbal Memory Long Delay Recalla Right ATL Left ATL Verbal Memory Recognitiona Right ATL Left ATL Boston Naming Testa Right ATL Left ATL Grooved Pegboard — dominant handb Right ATL Left ATL Grooved Pegboard — nondominant handb Right ATL Left ATL Rey Complex Figure Immediate Recallc Right ATL Left ATL Rey Complex Figure Copyc Right ATL Left ATL Trail-making Test A Timeb Right ATL Left ATL Trail-making Test A Errorsc Right ATL Left ATL Trail-making Test B Timeb Right ATL Left ATL Trail-making Test B Errorsc Right ATL Left ATL WAIS Digit Spand Right ATL Left ATL WAIS Digit-Symbol Codingd Right ATL Left ATL

F

Sig

18.83

b0.001⁎

6.19

0.03⁎

12.16

0.003⁎

+1.36 (1.18) −0.50 (0.69) +0.62 (1.70) −1.14 (1.22) +1.25 (0.97) −0.59 (1.18) 1.55

0.23

5.81

0.03⁎

1.29

0.27

0.07

0.79

1.17

0.30

0.16

0.70

0.08

0.79

3.69

0.07

0.54

0.47

1.47

0.24

0.43

0.52

2.94

0.10

+0.32 (1.31) −1.15 (2.02) +0.66 (0.42) −2.02 (3.28) +6.44 (9.36) +1.83 (9.05) +1.00 (9.08) −0.08 (9.44) −0.39 (6.56) 0.65 (4.78) −1.44 (4.07) +0.40 (3.37) +6.56 (10.67) +5.00 (14.21) +0.22 (0.44) −0.08 (0.29) +4.78 (11.03) +1.08 (11.64) +0.56 (1.51) −0.08 (0.90) +0.44 (1.88) −0.18 (2.32) +1.22 (1.64) −0.58 (2.81)

a

z-Score. t-Score. Raw score. d Scaled score. ⁎ Statistically significant. b c

or ipsilateral posterior hippocampal activation are more likely to show postoperative improvement [21]. Moreover, patients who show reorganization of memory function to regions outside of the epileptic focus on functional MRI show fewer memory deficits following surgery [22]. These findings suggest the possibility that removal of the dysfunctional neural elements that lead to epilepsy also removes impedances to memory function [23]. This hypothesis is strengthened by our finding that improved memory function was positively correlated with ILAE outcome. Indeed, prior research has demonstrated the relationship between seizure freedom and improved working memory [24] and verbal memory [10,25]. This study examined cognitive function at one point in time, three months postoperatively on average, leaving the question of durability of these findings unanswered. All subjects at our center are asked to return for a second postoperative neuropsychological evaluation at one year, but we have found it difficult to achieve consistent follow-up at this time point, possibly because patients who are doing well do not see the need to return for repeat testing. Our results, however, indicate the importance of acquiring both short- and long-interval data in order to establish new cognitive baselines and examine postoperative stability of these functions. In addition, earlier neuropsychological testing at one month postoperatively may be useful, as effective interventions have been implemented within that timeframe. Patients in this study were administered neurocognitive measures based on clinician judgment and task familiarity. Thus, some patients were administered measures that were similar in design. For instance, verbal memory was assessed by either the RAVLT or CVLT. Although the CVLT has been shown to be sensitive to changes in memory following temporal lobe surgery, some research suggests that it may be less sensitive to subtle verbal memory changes than the RAVLT [26]. This is based on the fact that the CVLT has some items that share semantic categories, thereby possibly acting as a memory cue for some patients. Patients in the present study were administered the same measures pre- and postoperatively; therefore, the effect that this had on a given individual's performance should be the same for each of their measurements. However, patients who were evaluated with the CVLT may demonstrate even greater memory deficits, or lack of improvement, than that seen in this study if only the RAVLT was utilized. Taken together, the results of this study demonstrate that shortinterval follow-up is effective in elucidating cognitive changes following epilepsy surgery. In some instances, this lends itself to the identification of patients who could benefit from early intervention. In other cases, being able to identify the lack of change, or even cognitive benefit, may influence decisions such as return to work or school. Furthermore, the results of this study show that epilepsy surgery is associated with few cognitive deficits, and in some cases, improved seizure control is related to improved cognitive function.

Immediate Verbal Recall Long Delay Verbal Recall

2.00

1.00

0.00

-1.00 1

2

3

4

Mean Change in Performance (Z-scores)

b)

a) Mean Change in Performance (Z-scores)

145

Immediate Verbal Recall Long Delay Verbal Recall

1.00

0.00

-1.00

-2.00 1

2

3

4

ILAE Outcome

ILAE Outcome

Error Bars: +/- 1 SE

Error Bars: +/- 1 SE

Fig. 2. Change in verbal memory and ILAE outcome for a) right ATL and b) left ATL.

5

146

P.S. Lee et al. / Epilepsy & Behavior 62 (2016) 140–146

Acknowledgments This research was funded in part through a grant (UN2014-73622) from the Walter L Copland Fund of the Pittsburgh Foundation. Ethical publication statement We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Disclosure of conflicts of interest None of the authors has any conflict of interest to disclose. References [1] Sherman EMS, Wiebe S, Fay-McClymont TB, Tellez-Zenteno J, Metcalfe A, Hernandez-Ronquillo L, et al. Neuropsychological outcomes after epilepsy surgery: systematic review and pooled estimates. Epilepsia 2011;52:857–69. [2] Lee TMC, Yip JTH, Jones-Gotman M. Memory deficits after resection from left or right anterior temporal lobe in humans: a meta-analytic review. Epilepsia 2002;43:283–91. [3] Vaz SAM. Nonverbal memory functioning following right anterior temporal lobectomy: a meta-analytic review. Seizure 2004;13:446–52. [4] Baxendale S, Thompson PJ, Duncan JS. Improvements in memory function following anterior temporal lobe resection for epilepsy. Neurology 2008;71:1319–25. [5] Martin R, Kuzniecky R, Ho S, Hetherington H, Pan J, Sinclair K, et al. Cognitive effects of topiramate, gabapentin, and lamotrigine in healthy young adults. Neurology 1999;52:321–7. [6] Meador KJ, Loring DW, Ray PG, Murro AM, King DW, Nichols ME, et al. Differential cognitive effects of carbamazepine and gabapentin. Epilepsia 1999;40:1279–85. [7] Meador KJ, Loring DW, Vahle VJ, Ray PG, Werz MA, Fessler AJ, et al. Cognitive and behavioral effects of lamotrigine and topiramate in healthy volunteers. Neurology 2005;64:2108–14. [8] Anderson CT, Noble E, Mani R, Lawler K, Pollard JR. Epilepsy surgery: factors that affect patient decision-making in choosing or deferring a procedure. Epilepsy Res Treat 2013;2013:309284. [9] Grammaldo LG, Di Gennaro G, Giampà T, De Risi M, Meldolesi GN, Mascia A, et al. Memory outcome 2 years after anterior temporal lobectomy in patients with drug-resistant epilepsy. Seizure 2009;18:139–44. [10] Helmstaedter C, Kurthen M, Lux S, Reuber M, Elger CE. Chronic epilepsy and cognition: a longitudinal study in temporal lobe epilepsy. Ann Neurol Oct 2003;54:425–32.

[11] Radford K, Lah S, Thayer Z, Miller LA. Effective group-based memory training for patients with epilepsy. Epilepsy Behav 2011;22:272–8. [12] Helmstaedter C, Loer B, Wohlfahrt R, Hammen A, Saar J, Steinhoff BJ, et al. The effects of cognitive rehabilitation on memory outcome after temporal lobe epilepsy surgery. Epilepsy Behav 2008;12:402–9. [13] Wieser HG, Blume WT, Fish D, Goldensohn E, Hufnagel A, King D, et al. ILAE commission report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001;42:282–6. [14] Jacobson NS, Truax P. Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. J Consult Clin Psychol 1991;59:12. [15] Spittle A, Orton J, Anderson P, Boyd R, Doyle LW. Early developmental intervention programmes post-hospital discharge to prevent motor and cognitive impairments in preterm infants. Cochrane Database Syst Rev 2012;12, CD005495. [16] Naismith SL, Glozier N, Burke D, Carter PE, Scott E, Hickie IB. Early intervention for cognitive decline: is there a role for multiple medical or behavioural interventions? Early Interv Psychiatry 2009;3:19–27. [17] Ponsford J, Willmott C, Rothwell A, Cameron P, Ayton G, Nelms R, et al. Impact of early intervention on outcome after mild traumatic brain injury in children. Pediatrics 2001;108:1297–303. [18] Ponsford J, Willmott C, Rothwell A, Cameron P, Kelly A-M, Nelms R, et al. Impact of early intervention on outcome following mild head injury in adults. J Neurol Neurosurg Psychiatry 2002;73:330–2. [19] Hessen E, Lossius MI, Gjerstad L. Repeated neuropsychological assessment in wellcontrolled epilepsy. Acta Neurol Scand 2013;127:53–60. [20] Tang Y, Yu X, Zhou B, Lei D, Huang XQ, Tang H, et al. Short-term cognitive changes after surgery in patients with unilateral mesial temporal lobe epilepsy associated with hippocampal sclerosis. J Clin Neurosci 2014;21:1413–8. [21] Bonelli SB, Powell RHW, Yogarajah M, Samson RS, Symms MR, Thompson PJ, et al. Imaging memory in temporal lobe epilepsy: predicting the effects of temporal lobe resection. Brain 2010;133:1186–99. [22] Bonelli SB, Thompson PJ, Yogarajah M, Powell RHW, Samson RS, McEvoy AW, et al. Memory reorganization following anterior temporal lobe resection: a longitudinal functional MRI study. Brain 2013;136:1889–900. [23] Sanyal SK, Chandra PS, Gupta S, Tripathi M, Singh VP, Jain S, et al. Memory and intelligence outcome following surgery for intractable temporal lobe epilepsy: relationship to seizure outcome and evaluation using a customized neuropsychological battery. Epilepsy Behav 2005;6:147–55. [24] Wisniewski I, Wendling AS, Steinhoff BJ. Impact of side of lesion, seizure outcome and interictal epileptiform discharges on attention and memory after surgery in temporal lobe epilepsy. Epileptic Disord 2011;13:27–35. [25] Helmstaedter C, Elger CE. Cognitive consequences of two-thirds anterior temporal lobectomy on verbal memory in 144 patients: a three-month follow-up study. Epilepsia 1996;37:171–80. [26] Loring DW, Strauss E, Hermann BP, Barr WB, Perrine K, Trenerry MR, et al. Differential neuropsychological test sensitivity to left temporal lobe epilepsy. J Int Neuropsychol Soc 2008;14:394–400.