Type of presurgical deficit
Laterality findings
Change with surgery
Language functions
Naming (auditory/visual)
Both left and right TLE—frequently exhibit deficits (left > right)
Left TLE—decline
Right TLE—no decline
Semantic fluency
Both left and right TLE—frequently exhibit deficits (may vary by type of category)
Left TLE—decline
Right TLE—data is sparse/appears decline possible
Letter fluency
Both left and right TLE—often exhibit deficits
Left and right TLE—no decline/may improve with seizure freedom
Action (verb) fluency
Possibly same pattern as letter fluency/data is sparse
Possibly same/data is sparse
Other language functions
Left TLE—usually normal unless patient has dysphasia secondary to an additional neurological cause (e.g., stroke)
Right TLE—intact core language functions
Left TLE—no decline unless encroaching on classic language areas
Right TLE—no decline
Memory and learning
Material-specific memory deficits may be observed
Left TLE—often exhibits auditory/verbal memory deficits
Right TLE—sometimes (but less consistently) exhibits visual memory deficits
Left TLE—auditory/verbal memory often declines
Right TLE—visual memory often declines (depends more on specific tasks than previously recognized)
Task-specific dissociations in performance
Left TLE—impairment on difficult associational and rote learning paradigms more associated with mesial TL dysfunction. Impairment on semantically related tasks more associated with lateral TL dysfunction (e.g., story recall, easy word pairs)
Right TLE—often exhibits deficits involving object-location recall (particularly when assessed in three dimensions) and route learning
Declines in functioning can occur that are consistent with the specified preoperative patterns
More research is needed to pin down specific structure-function relationships
General memory dysfunction is observed in some TLE patients
Both left and right TLE—sometimes exhibit general or global memory dysfunction at baseline
Decline still seems to reflect the task and material-specific patterns that occur with surgery
Learning and retention patterns may differ by side of seizure onset
Left TLE—often exhibits good initial retention of words, but poor retention over time
Right TLE—often exhibits problems with initial learning of visual material but retain most of what is encoded/learned over time
No data has been published examining the differential learning patterns postoperatively
Visuo-perceptual/visual-spatial processing and object recognition
Visuo-perceptual deficits tend to be infrequent
Right TLE—occasionally has visuo-perceptual deficits
Both left TLE and right TLE—may experience a visual-field cut following surgery
Left TLE—no decline on perceptual tasks
Right TLE—mild decline on some tasks on occasion
Visual-spatial deficits tend to be infrequent
Right TLE patients—may rarely have visual-spatial deficits (e.g., problems judging spatial features)
Left TLE—no decline
Right TLE—very rare decline on some tasks
Category-related object recognition deficits
Left TLE—occasionally sees mild recognition deficits for man-made objects with posterior TL involvement
Right TLE—frequently exhibit recognition deficits for famous persons/faces, landmarks, and animals with anterior TL dysfunction
Left TLE—possible decline on recognition of man-made objects with encroachment on left posterior regions
Right TLE—frequent decline on recognition of famous persons/faces, landmarks, and animals
General intellectual functioning
IQ scores
Typically normal in TLE
Left TLE will sometimes show greater problems with verbal tasks, and right TLE will sometimes show greater problems with perceptual tasks
May see “material-specific” declines in line with presurgical status
May see improvements in processing and attention regardless of hemisphere of seizure focus if patient experiences a reduced seizure burden and a reduction in AEDs
Executive control processes
Complex problem solving, response inhibition, generative fluency, complex attention
Both left and right TLE—often exhibit deficits in one of more of these areas that is assumed to result from the disruption of temporofrontal networks secondary to epileptiform activity
Both left and right TLE—frequently show improved performance in these domains if experiencing seizure freedom
Findings in Frontal Lobe Epilepsy (FLE)
Research examining the neurocognitive functioning of presurgical frontal lobe epilepsy (FLE) patients has been less commonly completed. In part, the lack of research in this area is likely due to greater difficulty obtaining adequate sample sizes for such studies, as FLE patients are believed to reflect only 10–20 % of all surgical referrals (Jokeit & Schacher, 2004). It is usually necessary to collaborate across epilepsy centers or to spend many years obtaining large enough sample sizes to achieve adequate statistical power to answer basic questions. Research is also hampered in this area by limitations in our understanding of FL functions and the adequate development of tests and methods to assess them, as well as the same latent variables that plague epilepsy research in general. In addition, there is a great deal of heterogeneity in terms of pathology and seizure localization within FL regions, which contributes to differing neurocognitive profiles. Complex partial seizures in FLE commonly arise from the frontal poles and from the orbital, medial, and dorsolateral FL regions (Williamson, Spencer, Spencer, Novelly, & Mattson, 1985).
While the conclusions drawn in this section should be viewed as more tentative in nature due to the limited number of completed studies, some clear trends appear to be emerging. For example, there is growing evidence that presurgical patients with FLE often exhibit problems with response inhibition (McDonald et al., 2005; Upton & Thompson, 1996), although this problem has not been universally observed (Helmstaedter, Kemper, et al., 1996). One study that compared a sizeable number of FLE and TLE patients found that FLE patients performed significantly worse than the TLE group on a measure of response inhibition (Stroop Color-Word Interference Test) (Upton & Thompson, 1996). In contrast, Helmstaedter, Kemper, et al. (1996) found that their FLE sample performed worse than TLE patients on all Stroop conditions, which suggests their primary limitation may have involved reading or processing speed. McDonald et al. (2005) added a matched control sample to the previous paradigm of comparing FLE and TLE patients, demonstrating that the FLE group was impaired on the Stroop task from the Delis-Kaplan Executive Functioning System (DKEFS) when compared to the control subjects while the TLE group was not. The left FLE group was more impaired than the right FLE group or either TLE group on this portion of the task. There is at least one study suggesting that response inhibition performance may decline with unilateral FLE resections (Helmstaedter, Kemper, et al., 1996).
Performance on motor tasks is often decreased in FLE patients as compared to controls or other samples of epilepsy patients, and there is some evidence that decline occurs on these tasks with some FLE resections. For example, Helmstaedter, Kemper, et al. (1996) demonstrated that a small sample of FLE patients (n = 23) performed worse than a set of TLE patients on measures of psychomotor speed and motor coordination. Upton and Thompson (1996) compared the performance of presurgical FLE patients to that of TLE patients on a variety of motor tasks, finding that the FLE group performed worse than the TLE group on tasks of motor sequencing and bimanual hand movements (left FLE worse than right FLE). Studies in children with FLE seizure onset have reported greater problems with motor coordination (Hernandez, Sauerwein, Jambaque, et al., 2002) as compared to normal controls or children with other seizure onset.
Performance on complex problem-solving tasks also appears to be impaired in some FLE patients, and this finding has been observed both pre- and postoperatively (Upton & Thompson, 1996). Milner published a classic case study involving patient “K.M.,” who demonstrated severe deficits on the WCST following bilateral resection of the anterior FLs for the control of seizures despite maintaining normal IQ. As noted in the section on TLE, however, complex problem-solving deficits can also be seen in patients with TLE, likely due to the spread of seizure activity to FL regions.
Verbal fluency, as noted in the TLE section, is often impaired in presurgical FLE regardless of laterality of seizure onset and includes both semantic and letter fluency. There is also some data indicating that action (verb) fluency is also decreased in FLE patients. Preliminary evidence exists that performance on these tasks can sometimes yield localizing data when explored in ways that examine component skills required for successful completion (Drane et al., 2006).
Design fluency has also been shown to be decreased in FLE patients relative to other epilepsy patients or healthy controls, with some studies suggesting lateralization to the nondominant hemisphere (Jones-Gotman & Milner, 1977) and others not (McDonald, Delis, Norman, Tecoma, & Iragui, 2005). One study reported worse performance for patients with left FLE (McDonald, Delis, Norman, Tecoma, et al., 2005) as compared to right FLE while another demonstrated that FLE patients performed worse than TLE patients regardless of seizure onset laterality (Suchy, Sands, & Chelune, 2003). A fourth study found that patients with FLE produced a similar number of designs as did patients with TLE but made more design errors (Helmstaedter, Kemper, et al., 1996). Discrepancies across studies may in part reflect differences in the design fluency tasks themselves, as these measures differ in regard to the structure they provide, the presence or absence of concomitant task demands (e.g., shifting attention), and the aspect of performance emphasized (e.g., design generation, self-monitoring, shifting).
Some evidence suggests that FLE patients may perform worse than TLE patients on measures of attention, working memory, and psychomotor speed. For example, Helmstaedter, Kemper, et al. (1996) demonstrated that a small sample of FLE patients performed worse than a set of TLE patients on measures of attention and psychomotor speed. In the FLE group, differences were not related to side of seizure focus or presence of a structural lesion. Upton and Thompson (1996) found that a small group of FLE patients made more errors on a complex visual scanning and tracking measure than did a comparable TLE group.
There are many additional studies with FLE patients demonstrating deficits on a variety of tasks presumed to be sensitive to FL dysfunction, but most of these are isolated findings that have yet to be replicated. Areas of reported dysfunction have included deficient cost estimation (Upton & Thompson, 1996), an elevated rate of questions required to identify objects on the Twenty Questions Test from the DKEFS (Upton & Thompson, 1999), problems with determining temporal order (McAndrews & Milner, 1991), and aspects of social cognition (e.g., humor appreciation, recognition of facial emotion, perception of eye gaze expression) (Farrant, Morris, Russell, et al., 2005). Kemper, Helmstaedter, & Elger (1993) reported that FLE patients performed much worse than TLE patients on a planning task, suggesting that this difference correctly predicted the seizure focus of 80 % of all cases.
Memory performance has not been studied extensively in patients with FL seizure onset, and available results are somewhat mixed. Most studies have either compared performance between FLE and TLE patients with one another or with healthy controls. While some of these studies have failed to demonstrate differences between FLE and controls on memory measures (Delaney, Rosen, Mattson, & Novelly, 1980), others have reported worse functioning for FLE, at least on some types of tasks. FLE patients tend to perform worse on more complex learning paradigms (e.g., list-learning tasks), with their limitations attributed to problems with encoding and/or retrieval.
There have also been some interesting studies suggesting that certain aspects of learning and memory are perhaps more impaired in FLE than in other epilepsy groups. For example, Pigott and Milner (1993) demonstrated that preoperative FLE patients exhibit problems with release from proactive interference (i.e., earlier memories interfere with of learning new information). Milner attributed this deficit to problems with encoding and retrieval mechanisms. McDonald and colleagues (2001) more recently demonstrated this pattern in postoperative FLE patients and provided further evidence that encoding/retrieval deficits may underlie this pattern. In their study, postsurgical TLE patients did not display release from proactive interference, and there was no difference between the TLE and FLE groups in terms of consolidation of stimuli (i.e., they showed similar rates of retention over trials). Milner has also shown that FLE patients have difficulty structuring and segregating events in memory (Milner, 1968b), and her FLE patient samples have also exhibited problems with organization of materials to be learned and have had trouble recalling the temporal order of information (Milner, Petrides, & Smith, 1985). These findings applied to a wide range of stimuli and may be material specific in nature (Milner, Corsi, & Leonard, 1991).
There have been few systematic studies of psychiatric functioning in patients with FLE, although a number of case reports describe wide-ranging interictal behavioral abnormalities. For example, Boone et al. (1988) described a young adolescent girl with FL seizures who experienced reversible behavioral changes including sexual disinhibition, loss of concern for personal hygiene, physical and verbal aggression, and pressured and tangential speech accompanying interictal anterior FL discharges. Patients with anterior cingulate seizure foci have also been reported to develop interictal psychosis, aggression, sociopathic behavior, sexual deviancy, irritability, obsessive-compulsive disorder, and poor impulse control (Devinsky, Morrell, & Vogt, 1995). Additionally, Helmstaedter (2001) has also reported that FLE patients have an elevated rate of behavioral problems compared to other epilepsy patients and controls but noted that these tended to be mild as compared to the findings obtained in other neurological patients with structural FL lesions. Based on the limited studies available, psychiatric conditions such as depression and anxiety appear to be more common in TLE (Gilliam et al., 2004).
In summary, it appears that patients with FLE present with a variety of deficits involving motor functioning, executive control processes, attention, speed of processing, and aspects of memory performance, as well as some possible behavioral abnormalities. These functions have been minimally explored in FLE patients with few studies using a presurgical/postsurgical decline and most seeming to be underpowered. There have been virtually no attempts to explore functions by FL subregion in any epilepsy study, yet this methodology has proven useful in other areas. Similarly, numerous cognitive functions attributed to the FLs have yet to be explored in FLE patients. Some areas of dysfunction observed in patients with FLE have also been observed in patients with TLE, presumably due to seizure spread across large interconnected neural networks. There also appear to be some distinct patterns between patients with FLE and TLE that may yet be useful for confirming the region of seizure onset (Table 4.2).
Table 4.2
Core pre- and postsurgical deficits in frontal lobe epilepsy patients
Type of presurgical deficit | Laterality findings | Change with surgery |
|---|---|---|
General intellectual functioning | Typically normal in FLE | Typically no significant change |
Language | Typically normal apart from verbal fluency deficits (unless neurologic/functional disruption of classic speech regions, e.g., Broca’s area) | Typically no significant change with the exception of verbal fluency performance (see below) |
Motor functioning | Left and right FLE—often exhibit motor deficits contralateral to side of seizure focus (e.g., gross motor speed, fine motor speed, and dexterity) | Both left and right FLE—may show a decline in the motor performance of their contralateral limbs (particularly when surgery encroaches on precentral gyrus region) |
Response inhibition | Both left and right FLE—often exhibit deficits | Both left and right FLE—may decline depending upon location of FL resection |
Complex problem solving | Both left and right FLE—often exhibit deficits | Both left and right FLE—may decline depending upon location of FL resection |
Verbal fluency | Both left and right FLE—often exhibit baseline deficits on all types of verbal fluency tasks | Both left and right FLE—may decline on all verbal fluency tasks, although semantic fluency may improve in some cases |
Design fluency | Both left and right FLE—often exhibit deficits | Both left and right FLE—may decline |
Memory functioning | Both left and right FLE—often exhibit poor learning/encoding and decreased free recall with good recognition memory Both left and right FLE—often exhibit problems with release from proactive interference | Surgery may improve or worsen baseline problems based on location of surgery and postsurgical seizure freedom |
Attention | Both left and right FLE—often exhibit deficits in primary and complex attention | Data is lacking. Any change is likely dependent upon seizure status and AED regimen at follow-up assessment |
Social cognition | Some patients, regardless of laterality, have shown problems with recognizing humor and faux pas errors, recognition of facial emotion, and perception of eye gaze expression | Data is lacking. Theoretically, it appears that some functions could decline depending upon surgical variables, while seizure freedom and decreased AEDs may contribute to mild gains |
Visuo-perceptual, visual-spatial, and constructional praxis | Typically normal on most visuo-perceptual and visual-spatial tasks Often exhibit decreased performance on constructional tasks due to poor organization and planning | Data is lacking. However, no reports of significant declines in these areas exist in the research literature |
Findings in Posterior Cortical Epilepsy (PCE)
Seizures arising from the parietal lobe, the occipital lobe, the occipital border of the temporal lobe, or a combination of these regions are sometimes referred to as posterior cortical epilepsies, as it is difficult to find clear anatomic or pathophysiological differences in these regions (Dalmagro, Bianchin, Velasco, et al., 2005). The occurrence of posterior cortical epilepsies tends to be much rarer, and such conditions have been less well studied (Binder, Lehe, Kral, et al., 2008). Therefore, for the purposes of this chapter, we have decided to consider all of the work related to neurocognitive profiles related to parietal or occipital lobe seizure onset together. Dalmagro and colleagues (2005) reported that just over 5 % of their total referrals for long-term video-EEG monitoring experienced PCE, and of these, approximately half were actually surgical candidates. Overall, this group makes up well under 10 % of the total surgical referrals seen by a standard epilepsy surgical program, making this type of seizure onset even less common than FLE.
Cognitive studies of PCE patients are lacking in general, and there have been no systematic prospective studies of neurocognitive functioning in these patients that include both pre- and postoperative analysis. Studies appearing in the literature tend to involve retrospective analysis of clinical data. For example, one recent study examined retrospective pre- and postsurgical clinical data collected on 28 PCE patients between 1991 and 2000 (Luerding, Boesebeck, & Ebner, 2004). These investigators indicated that mild declines occurred in performance IQ from the WAIS-R regardless of resected hemisphere and also reported declines in some measures of visual-spatial processing. Postsurgical gains were made on some tasks thought to be mediated by the FLs, and there was no decline in WAIS-R verbal IQ. Of note, however, not only was the sample size very small, but this resulted in a pool of subjects with potentially very different lesions (e.g., left temporo-occipital verssu right inferior parietal). Also, only a limited number of subtests from the WAIS-R were available for examination. Overall, while this type of study of neurocognitive function of patients with PCE is sorely needed, such studies cannot definitively answer these questions due to a lack of sufficient power and inadequate coverage of potential domains to be examined. Other retrospective studies of PCE, particularly those involving parietal lobe dysfunction, have reported changes in visual functioning, visual-spatial processing, and visuo-perceptual abilities (Siegel & Williamson, 2000). Sensory changes are sometimes observed when surgical resection extends into the postcentral gyrus, and one study has reported disturbances of body image in a few patients with right inferior parietal corticectomies (Salanova, Andermann, Rasmussen, Olivier, & Quesney, 1995). One very small, retrospective study examining the neurocognitive status of children with occipital lobe (OL) seizure onset suggested that such patients experience an elevated rate of scholastic difficulty, psychiatric disorders (i.e., primarily depression), and cognitive dysfunction involving problems with face processing and making spatial judgments (Chilosi, Brovedani, Moscatelli, Bonanni, & Guerrini, 2006).
A recent study completed in Germany with a small series of OL epilepsy surgical patients prospectively examined visual-field integrity, demonstrating that a significant proportion of these patients experienced visual-field defects postoperatively (i.e., 42 % of OL patients experienced new or increased visual-field defects) (Binder et al., 2008). This study and others have shown that preoperative patients with seizure onset involving the mesial OL are more likely to exhibit baseline visual-field defects (e.g., reports suggest approximately 40–50 %) than those with lateral OL onset (e.g., ranging from 0 to 18 %) (Binder et al., 2008; Blume, Wiebe, & Tapsell, 2005). While focusing on perceptual rather than cognitive testing per se, this type of pre-/postoperative design is exactly what is needed in this area. At present, we lack definitive profiles for preoperative functioning in the posterior cortical epilepsies and have no prospective postsurgical outcome studies available for this patient group. One would assume, based on available lesion studies in other neurological disorders and functional imaging paradigms, that dysfunction in the OL could cause problems with face recognition, object localization, color processing, or object recognition (Kiper, Zesiger, Maeder, Deonna, & Innocenti, 2002) and that lesions in the parietal region could cause deficits involving visual-spatial processing, object recognition, sensory discrimination, arithmetic skills, and aspects of language functioning (Table 4.3).
Table 4.3
Core pre- and postsurgical deficits in posterior cortical epilepsy patients
Type of presurgical deficit | Laterality findings | Change with surgery |
|---|---|---|
General intellectual functioning | Typically normal in PCE at baseline | Limited research suggests possibly mild declines in PIQ for PCE patients regardless of side of surgery/laterality and mild improvements in VIQ |
Language | Depends on seizure focus: Left PLE—may exhibit classic language deficits (e.g., naming, repetition, comprehension) Right PLE and both left and right OLE—unlikely to exhibit language deficits | Typically no significant change with the exception of possible language declines with some left parietal lesions |
Visuo-perception, acuity, and visual fields | Both left and right OLE—often exhibit problems with visuo-perception (including face processing/recognition) Left and right OLE—often exhibit baseline visual-field cuts (much more common for medial OL seizure onset) Left and right OLE—might expect baseline deficits in color processing and object localization, as well as positive visual phenomena (yet epilepsy specific research is absent) | Left and right OLE—often exhibit new or increased visual-field cuts Left and right PCE—in general, may exhibit mild or greater visuo-perceptual decrements depending on aspects of surgery (i.e., location and extent) |
Visual-spatial processing | Both left and right OLE and right PLE—often exhibit problems with visual-spatial judgments | Both left and right PCE—declines in some aspects of visual-spatial processing (limited research) |
Memory | Presumed to be normal | Presumed to remain at baseline apart from possible gains related to improvements in seizure status and reduced AED regimen |
Sensory functioning | Both left and right PLE—may exhibit baseline problems with sensory discrimination | Sensory functioning is often worse in patients if surgery encroaches upon postcentral gyrus Disturbance of body image has been reported in patients with inferior parietal resections |
Motor | Presumed to be normal | Presumed to be normal |
Potential Confounds in the Neuropsychological Assessment of the Epilepsy Surgical Patient
In assessing epilepsy surgical patients, a variety of factors can obscure an individual’s true neurocognitive profile, including disease- and treatment-related variables, as well as limitations associated with current assessment paradigms and our knowledge of brain-behavior relationships. The clinical neuropsychologist must be aware of these issues in order to take them into consideration when interpreting assessment results. Specific factors with the potential to create “noise” in our assessments include the effect of AEDs on brain functioning, problems with the specific tests being employed, comorbid medical and psychiatric conditions, acute ictal/interictal epileptiform activity, and the acute and long-term impact of seizure activity on brain regions distant from the seizure focus. There is no absolute approach to dealing with these issues, and this leads to a number of decisions regarding when and where to conduct the neuropsychological assessment and the selection of appropriate tests.
Effects of AEDs
Many AEDs can have an appreciable impact upon brain functioning, as they function to dampen the neuronal irritability that constitutes a seizure, yet they also more broadly dampen neuronal excitability in general (Drane & Meador, 2002). This can lead to the emergence of cognitive deficits that are unrelated to the epileptic focus. For example, a TLE patient treated with topiramate may present with limitations in primary attention, verbal fluency, and processing speed that have nothing to do with their underlying seizure focus (Kockelmann, Elger, & Helmstaedter, 2003; Ojemann et al., 2001). Presurgical evaluation in such a patient may fail to confirm the lateralization or localization of the seizure focus and will provide a significant underestimation of the patient’s abilities. Knowledge of the effects of AEDs is therefore critical to interpreting neurocognitive results in this patient population, and there may be instances where it is worthwhile to take a patient off of their usual AEDs prior to evaluation.
Problems with Instrument Design
Problems with features of test design and selection can also muddle the interpretation of neurocognitive data. For example, test selection potentially becomes a barrier to discovering accurate brain-behavior relations when we employ measures that require the interaction of multiple cognitive skills controlled by different brain regions without a clear awareness of these relationships. For example, the Family Pictures subtest of the 3rd edition of the Wechsler Memory Scale (Wechsler, 1997) contributes to the Visual Memory Index from this battery, yet there is strong evidence that it loads on a verbal factor (Dulay et al., 2002). This is likely due to the necessity to accurately name the pictured individuals in order to get credit for any aspect of recall on this task. In turn, however, we often see a decline on Family Pictures, as well as the Visual Memory Index to which it contributes, in patients who undergo a dominant (typically left) hemisphere resection (Chapin, Busch, Naugle, & Najm, 2009). If someone were to explore the possibility of material-specific memory deficits in TLE using the Verbal and Visual Memory Indices of the WMS-3, they could easily draw wrong conclusions if they were unaware of this pattern of findings.
Nociferous Cortex Hypothesis
The nociferous cortex or “neural noise” hypothesis suggests that seizure activity can disrupt more expansive neural networks that extend beyond the irritative zone of the seizure (Penfield & Jasper, 1954). As covered earlier, executive control processes thought to be primarily mediated by FL regions (e.g., letter fluency, complex problem solving) can be disrupted in patients with TL seizure onset (Hermann & Seidenberg, 1995). These apparent deficits frequently resolve with the successful control of the TL seizure activity (Martin et al., 2000), and these alterations in the functioning of the FL cortex can be captured with functional neuroimaging (Jokeit et al., 1997; Maccotta et al., 2007). It is also thought that FL seizures will disrupt limbic and TL regions, although less research is available to confirm such patterns. Overall, the potential for distal effects of epileptiform activity can obviously obscure the central seizure focus.
Effect of Comorbid Conditions
Medical and psychiatric conditions that are often comorbid with epilepsy can also introduce greater noise into a patient’s neurocognitive profile. For example, patients with epilepsy experience a higher rate of depression and a slightly elevated rate of psychosis as compared to the general public (Blumer, Montouris, & Hermann, 1995; Manchanda, 2002). Epilepsy patients struggling with mood issues or perhaps even actively psychotic may be less able to actively engage in testing. Similarly, epilepsy sometimes reflects a secondary condition resulting from a more primary medical condition (e.g., brain tumor, stroke, HSV encephalitis) or injury (e.g., traumatic brain injury). The primary disease or injury contributes uniquely to the patient’s pattern of dysfunction and can mask any potential lateralizing/localizing neurocognitive findings. For example, patients with focal TL seizure onset resulting from posttraumatic epilepsy may exhibit significant executive function impairment that is related to potentially widespread cerebral dysfunction resulting from the head trauma.
Effect of Ictal and Interictal Discharges
There is growing awareness that acute ictal or interictal epileptiform discharges can alter neurocognitive profiles. While the impact of such epileptiform activity can sometimes accentuate a profile pattern in the case of focal seizure onset, it can also obscure this pattern when there is secondary generalization or non-focal patterns of interictal discharges (Aarts, Binnie, Smit, & Wilkins, 1984; Aldenkamp & Arends, 2004; Binnie, 2003; Kasteleijn-Nolst Trenite & Vermeiren, 2005).
Practice Effects
Neuropsychological assessment in epilepsy requires repeated testing over time, which necessitates consideration of possible practice effects. Some studies have been completed that examine test-retest changes in either epilepsy patients or in healthy controls and that provide reliable change indices to allow one to determine if a given change is related to the treatment intervention as opposed to a simple practice effect (Martin, Sawrie, Gilliam, et al., 2002). This can be particularly important when one recognizes that a lack of an expected practice effect may reflect a limitation in a postsurgical patient.
Atypical Language Lateralization
Given that epilepsy is often associated with comorbid neurological disorders/injury, it is not surprising that one observes an elevated rate of atypical language lateralization in this population (i.e., right or bilateral language). Many of these cases appear to represent language reorganization that has occurred in individuals experiencing early-life injuries, although a few studies suggest there is likely a subset of patients with rare but naturally occurring atypical language lateralization (Drane, Ojemann, Ojemann, et al., 2009; Knecht, Jansen, Frank, et al., 2003). The possibility of atypical language lateralization must be borne in mind when analyzing neurocognitive data and making outcome predictions.
In summary, it is recommended that the clinical neuropsychologist makes every effort to be aware of potential latent variables and to control for their presence when possible. In this manner, one may be able to better localize or lateralize a seizure event in a patient that otherwise showed no focal findings. One may also gain better insight into the patient’s genuine performance baseline in the absence of seizure activity. Occasionally, it may also be possible to use these variables to one’s advantage such as using ictal or postictal assessment techniques to enhance focal findings.
Decisions About Neuropsychological Assessment in Epilepsy Surgical Patients
Inpatient Versus Outpatient Assessment
There are various pros and cons to either testing presurgical epilepsy patients on the inpatient unit versus in the outpatient clinic, and both practices continued to be employed with regularity by epilepsy surgical programs throughout the world.
Conducting evaluations in the epilepsy monitoring unit (EMU) allows one to be aware of the presence of ictal and interictal epileptiform discharges, as the patient is undergoing continuous video-EEG monitoring. This is probably its largest advantage, as increasing data demonstrates that subclinical and interictal epileptiform activity can have a transient disruptive impact upon neurocognitive functioning (Aldenkamp & Arends, 2004; Kasteleijn-Nolst Trenite & Vermeiren, 2005). It is likely that transient changes in performance sometimes lead to an underestimation of the patient’s baseline functioning (see current clinical vignette), which can obscure change over time (e.g., leading one to miss declines in performance subsequent to surgical intervention or other treatment) and contribute to false predictions regarding outcome. For example, if someone appears to have severely impaired verbal memory due to transient epileptiform activity, we might erroneously predict that their risk of decline is small due to their poor baseline. This sort of error potentially gives the patient, the neurosurgeon, and the rest of the treatment team misinformation for making their decisions regarding the risk associated with surgery. On the other hand, having concurrent electrophysiological data allows the neuropsychologist to explore changes in performance in relation to ictal and interictal discharges. Inpatient testing also gives the examiner a greater span of time to conduct tests.
The downside of inpatient assessment involves potentially dealing with acute changes in medications, as the patient’s standard AED regimen may be altered or discontinued in order to provoke seizures. Similarly, some EMUs will also employ sleep deprivation as a means to induce seizures more rapidly. Depending upon the practice of the inpatient unit, these issues can often be more effectively managed by agreeing upon a standard start time for neurocognitive testing. In our practice, we initiate the baseline testing at the first full day of monitoring, prior to the patient being sleep deprived and often prior to any changes in AED regimen. This also provides us with electrophysiological data from the day of admit to insure that acute seizure activity has not immediately preceded our assessment.
Ideally, the epilepsy neuropsychology service will have dedicated rooms on the inpatient monitoring unit that preserve the uninterrupted, private environment while allowing the patient to continue with their video-EEG monitoring. If such rooms are not available, bedside testing can also work adequately, if proper steps are taken to educate staff about not interrupting. While dealing with additional staff creates additional work for the examiner, having them available to assist with the patient if a seizure occurs is quite advantageous, particularly with patients that experience seizures on a frequent basis.
The advantages and disadvantages of conducting presurgical evaluations in the outpatient setting are essentially the opposite of those just cited for inpatient assessment. The major disadvantage is that one has no objective knowledge of the immediate electrophysiological functioning of the patient that they are assessing. Epileptiform activity may be occurring during the examination, with direct impact upon the assessment, and neither the examiner nor the patient will be aware of its occurrence. Similarly, the patient may have experienced a seizure within the last 24 h yet lack any recollection of this occurrence. In contrast, the patient is less likely to be sleep deprived, and they are most likely continuing with their standard treatment regimen. Our group, as well as others, has suggested that the eventual standard for outpatient neuropsychological assessment of epilepsy should include obtaining simultaneous EEG recordings (Patrikelis, Angelakis, & Gatzonis, 2009).
Financial considerations may also play a role in deciding upon assessment venue, as some payors may incentivize one type of assessment over another. Nevertheless, we have often been successful in getting policies adjusted by providing data to explain our preferences in assessment location depending upon the factors involved. This is an area where standardized policies could conceivably be established by interested practice organizations.
Reliable Change Indices Versus Alternate Test Forms
The serial assessment of epilepsy patients undergoing surgery necessitates correction for practice effects. Since the early 1990s, some neuropsychologists have started using advanced methods of measuring change, including the Reliable Change Indices (RCIs) and standard regression-based (SRB) change score norms (Hermann et al., 1996; Sawrie, Chelune, Naugles, & Luders, 1996). Both are methods that attempt to statistically control for test-retest effects and measurement error, allowing the clinician to better evaluate whether change in performance is actually related to a specific treatment intervention. RCI and SRB scores based on the performance of unoperated patients with epilepsy and on healthy control subjects are available for a number of neurocognitive measures (Dikmen, Heaton, Grant, & Temkin, 1999; Heaton et al., 2001; Temkin, Heaton, Grant, & Dikmen, 1999). Examining the performance of the same type of patient allows for the greatest control of other disease-related variables.
Another approach is to use alternative forms of commonly used tests when available. The major problem involved in using alternative test forms relates to the difficulty we have insuring that each alternative version of a test is actually equivalent to the original.
Choice of Neuropsychological Tests
In general, when putting together a battery to assess epilepsy surgical patients, one wants to cover all of the standard neurocognitive domains. However, the relative emphasis to place on a given domain and the choice of specific tests should be guided by the overarching purposes of the neuropsychological evaluation in the epilepsy surgical setting. More specifically, tests should be chosen that are potentially useful for:
(a)
Confirming the epileptic focus: Knowledge of deficits commonly observed in various epilepsy syndromes preoperatively will aid in the selection of tests that are useful for lateralizing/localizing purposes.
(b)
Demonstrating postoperative change in function: One should include measures that examine functions known to commonly decline with a particular type of surgery (e.g., verbal memory decline following dominant TL resection), allowing one to assess outcome. Of note, postsurgical change can also be positive, as in the case of someone showing improvement in processing speed and attention secondary to becoming seizure-free and discontinuing their AED regimen.
(c)
Insuring valid interpretation: Measures chosen in this area assess the impact of latent variables commonly occurring in the epilepsy surgical setting. For example, by using performance validity measures, one determines whether the evaluation has been invalidated by factors such as poor motivation or task engagement on the part of the patient or the possible effect of other disease-related variables (e.g., subclinical or interictal epileptiform discharges).
(d)
Allowing research to be performed: Measures should be used allowing one to research basic questions about brain-behavior functions and surgical outcome for specific cognitive skills, psychosocial and vocational functioning, emotional/psychiatric processing, seizure freedom, and quality of life.
Table 4.4 lists specific neurocognitive functions proven important to assess for several common types of epilepsy and specific tests that have been used to assess these functions in the context of epilepsy. Table 4.5 presents similar data for motor and sensory functioning, mood, personality, and quality-of-life assessment. Obviously, tests selected for use need to be psychometrically sound (e.g., valid, reliable) and preferably have RCI or SRB scores or alternative forms for repeat assessment. In addition, the following section provides a brief overview of how the findings in each of these neurocognitive domains may be used and integrated in the neuropsychological assessment.
Table 4.4
Core neurocognitive functions to be assessed in epilepsy surgical patients and suggested tests
Neurocognitive domains | Within domain areas to emphasize during assessment | Possible tests to consider |
|---|---|---|
Language | − Naming (e.g., visual, auditory/naming to description, category related) − Verbal fluency (semantic, letter, and action) − Screen reading and other core language tasks | − Boston Naming Test, Columbia Auditory Naming Test, Category- Related Naming Tests − Category fluency tasks (e.g., animals, supermarket items), DKEFS Verbal Fluency, Controlled Oral Word Association Test, action fluency − Recognition Reading Subtest of the Wide Range Achievement Test, American Version of the National Adult Reading Test (AMNART), Wechsler Test of Adult Reading, Token Test, sentence repetition |
Attention | − Primary attention (auditory and visual) − Complex attention (auditory and visual) − Sustained attention (auditory and visual) | − Digit Span Forward (WAIS), Picture Completion (WAIS) − Digit Span Backwards and Letter-Number Sequencing (WAIS), Trail Making Tests, spatial span − Continuous Performance Test (not used as commonly by most epilepsy centers) |
Visual processing | − Visuo-perception − Visual-spatial − Object recognition | − Visual Object and Space Perception (VOSP) Battery, Facial Recognition Test – Judgment of Line Orientation − Famous Faces Test, Category-Related Object Recognition Tests |
Constructional praxis | − Graphomotor copying tasks − Assembly tasks | − Copying simple shapes (e.g., Greek cross, Necker cube), Rey Complex Figure Test (Copy) − Block Design (WAIS) |
Memory and learning | − Auditory/verbal Learning, memory retention, and recognition − List-learning Tasks − Contextual memory − Associative learning − Visual learning, memory retention, and recognition − Simple geometric designs − Face recall − Complex visual designs − Remote recall | − Rey Auditory/Verbal Learning Test, California Verbal Learning Test, Verbal Selective Reminding Test − Logical Memory Subtest (Wechsler Scales), Reitan Story Memory − Verbal Paired Associates (VPA) Subtest (Wechsler Scales; WMS-III VPA appears less helpful than other versions, as it eliminated the easier word pairs) − Visual Reproduction (Wechsler Memory Scales: older versions appear to be more useful for lateralization than the 3rd edition) − Face Recall/Hospital Facial Recognition Task − Rey Complex Figure Test, MCG Complex Figures − Information Subtest (WAIS) |
Executive control processes | − Complex problem solving − Response inhibition − Complex attention/mental flexibility − Abstract reasoning − Generative fluency tasks (verbal and visual) − Metacognition | − Wisconsin Card Sorting Test, Brixton Spatial Anticipation Task, Iowa Gambling Task − Color-Word Interference (Stroop) Test, Hayling Test, Go/No-Go Tasks − Trail Making Test, Mental Control (WMS) − Similarities Subtest/Matrix Reasoning Subtest (WAIS) − DKEFS Verbal Fluency, DKEFS Design Fluency, 5-Point Design Fluency − Cognitive Estimation Tasks |
General intellectual functioning | – Verbal and performance IQ | – Wechsler Adult Intelligence Scale (various editions) – Wechsler ASI |
Academic achievement | – Reading recognition – Reading comprehension – Mathematical skills – Spelling ability | – Wide Range Achievement Test (WRAT)—reading recognition – Gray Oral Reading Test (GORT) – WRAT—arithmetic – WRAT—spelling |
Performance validity testing | – Determine task engagement. This can be disrupted due to issues including poor motivation as well as the impact of acute seizures and epileptiform activity | – Word Memory Test
Related posts: Evaluation and Management of Psychogenic Nonepileptic Attacks
The Wada Test: Current Perspectives and Applications
Psychotherapeutic and Cognitive Remediation Approaches
Developmentally Delayed Children and Adults
 Functional MRI in the Presurgical Epilepsy Evaluation
 Structural and Quantitative MRI in Epilepsy
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|