CHAPTER

Neuropsychological Evaluation

Phillip D. Ruppert and Sarah E. Cook

In the United States, there are approximately 200,000 new epilepsy diagnoses made each year, and approximately 3 million Americans carry a diagnosis of epilepsy at any given time. Many of these individuals experience epilepsy-related cognitive symptoms that have significant psychological, social, and economic consequences. Thus, identification and management of cognitive symptoms are important considerations in the care of individuals with epilepsy.

Neuropsychologists make distinct contributions to the interdisciplinary epilepsy care team. Neuropsychologists are clinical psychologists who have received specialized training in brain/behavior relationships, psychometric theory, psychopathology, and psychological interventions. Thus, neuropsychologists are uniquely capable of providing an objective, evidence-based assessment of an intervention for cognitive and psychological disorders in individuals with epilepsy. For these reasons, neuropsychologists are commonly consulted in the care of epilepsy patients, and the National Association of Epilepsy Centers has stated that neuropsychological services are an essential component of collaborative interdisciplinary care teams at specialized epilepsy centers.

This chapter discusses the role of neuropsychology in the care of epilepsy patients. First, we provide a brief overview of the common causes of cognitive dysfunction in epilepsy. Next, we list some of the most common referral questions posed to neuropsychologist working with epilepsy populations and describe the methods and tools used by the neuropsychologist to address these questions. Then we cover topics relevant to one of the most common reasons for referrals to neuropsychologists in epilepsy centers, evaluation of epilepsy patients for epilepsy surgery. We then turn our attention to the topic of interventions for cognitive symptoms in epilepsy patients. Finally, we briefly discuss the contributions of the neuropsychological evaluation to the identification of patients with psychogenic nonepileptic seizures.

A number of factors can impact cognitive functioning in epilepsy patients. These include relatively fixed factors, such as the nature and location of the underlying pathology causing the seizures, the age of onset for this pathology and for the seizures, and the age at which seizure treatment was initiated. Factors pertaining to the disease course are also relevant. Specifically, higher lifetime number of generalized seizures, higher number of episodes of status epilepticus, and head injuries secondary to falling during seizures are all associated with increased cognitive morbidity. Finally, more remediable or transitory factors affecting cognition include side effects of some antiepileptic drugs (AEDs), subclinical epileptiform activity, quality of sleep, recency of last seizure, and mood dysfunction.

COMMON REFERRAL QUESTIONS

Specific reasons for referral to neuropsychology will vary for each individual case, but referral is generally made in situations where an objective assessment of cognitive and/or psychological factors will aid in the diagnosis or treatment planning for the patient. When a person with epilepsy is being considered as a candidate for palliative resection surgery, neuropsychological evaluation is often requested to obtain a cognitive baseline for comparison to postsurgical performance, to assist with the lateralization/localization of a seizure focus, and to assist in the prediction of risk for postoperative cognitive decline. In the postsurgical context, neuropsychological evaluation is commonly requested to quantify the degree of cognitive change, if any, after surgery. If postsurgical deficits are identified, results of the neuropsychological evaluation can also guide the selection of specific interventions and compensatory techniques that may help the patient to better cope with these deficits. Outside of the surgical context, the neuropsychological evaluation may be helpful for documenting areas of cognitive strengths and weakness for individuals seeking academic or occupational accommodations. In addition to questions about cognitive function, neuropsychologists are also frequently asked to provide assessment of a patient’s mood, as psychiatric distress is a common comorbidity in epilepsy. Finally, neuropsychologists are occasionally asked to assist with the identification of patients who are at risk for psychogenic nonepileptic seizures.

GENERAL ASSESSMENT STRATEGIES

An outpatient neuropsychological assessment is a relatively lengthy procedure, lasting anywhere from 3 to 8 (or more) hours. It typically begins with a detailed clinical interview where the neuropsychologist will ask the patient and family about the existing cognitive complaints. Medical, family, social, and psychiatric histories are also obtained. Based upon history obtained from records and the interview, a battery of tests is selected and administered. Many neuropsychologists have a flexible approach to assessment, meaning that they select tests based upon specific considerations for each patient rather than give the same battery to every patient. This flexible approach can be an advantage in a managed care environment where testing hours may be limited. However, there can be benefits of having a “fixed” battery for specific populations, such as in presurgical epilepsy evaluations, to promote the collection of the same variables for clinical research purposes. Depending on the setting, the neuropsychologist or a trained psychometrist will administer the tests in a standardized fashion on a one-on-one basis to the patient in a nondistracting room. Breaks are encouraged to minimize the effect of fatigue or providers may schedule testing on more than one day. Following administration, the neuropsychologist or psychometrist will score the tests using standardized criteria and each performance will be compared to established normative values. The neuropsychologist then interprets the findings and produces a report for the referral source.

Common Tests of Specific Cognitive Domains

The Ruff Figural Fluency Test (RFFT) (9) requires the patient to draw as many unique designs as possible on a page within a limited amount of time. The test measures novel generation of visual designs and resistance to perseveration.

Memory

The Wechsler Memory Scale (WMS), currently in its fourth edition (11), is a battery of multiple memory tests that primarily assesses a patient’s learning and memory ability for verbal and visual information.

The California Verbal Learning Test (CVLT) (12), Rey Auditory Verbal Learning Test (RAVLT) (13), and Hopkins Verbal Learning Test (HVLT) (14) are all similar tests of verbal learning and memory. These tests require the patient to learn a word list over successive trials of exposure to the list. Delayed memory for the list is tested using delayed recall and recognition tasks.

The Rey Complex Figure Test (RCFT) (15) is a test of visual memory. In this test, the patient is shown a complex figure made up of multiple, detailed visual components. The patient initially draws a copy of the figure to facilitate memory encoding. Delayed memory is tested by having the patient draw the figure from memory after a delay.

The Brief Visuospatial Memory Test (BVMT) (16) is a test of visual learning and memory. The patient learns visual figures by drawing them repeatedly after successive trials of presentation of the figures. Delayed memory is tested by having the patient draw the figures from memory after a delay.

Language

The Boston Naming Test (BNT) (17) is a test of confrontation naming. In this test, the patient is presented with line drawings of everyday objects and is required to give the name for each object. The test progresses in difficulty from very commonly encountered objects to objects that are less frequently encountered in daily life.

Verbal fluency tests are tasks that require the patient to generate as many words as possible that meet specified criteria under a given time limit. For example, phonemic or literal fluency requires that the patient give as many words that he/she can think of that begin with a specific letter in a minute (18). Semantic or categorical fluency requires the patient to give as many words that he/she can think of that belong to a specific semantic category (eg, tools) in a minute (19). Verbal fluency tasks measure a number of cognitive abilities, including expressive language, information retrieval from memory, executive control, selective attention/inhibition, and response generation.

Visual-Perceptual

The Judgment of Line Orientation Test (JOLO) (20) is a measure of visual-spatial perception and orientation. The task presents the patient with a target pair of angled lines as well as a separate semicircular array of multiple lines. The patient is asked to tell which lines in the array match the same angle/orientation as the target pair of lines.

The Benton Facial Recognition Test (FRT) (21) is a test of visual perception and specifically assesses perception of human facial features. The test presents the patient with a picture of a target individual. The patient is then asked to identify the same individual from an array of pictures that contain both the target and the multiple foils. Foils often look similar to the target individual. Test difficulty is manipulated by varying the lighting and visual angle of target and foil faces.

Motor

The Finger Tapping Test (FTT) (22) measures motor speed in the left and the right hand. The task requires the patient to rapidly and repeatedly tap a lever with his/her left and right index finger during a specified time limit. The test was designed to detect lateralized motor weakness.

The Grooved Pegboard Test (GPT) (23) primarily measures motor speed and dexterity in the left and the right hand but also can measure visual perception and fine-grained somatosensory perception. The task requires the patient to sequentially place pegs into a pegboard as quickly as possible. The pegs have a rounded side and a grooved side, so the pegs can only be placed into the board in a specific orientation. The test was designed to detect lateralized motor weakness.

Mood and Personality

The Minnesota Multiphasic Personality Inventory (MMPI) (24) and Personality Assessment Inventory (PAI) (25) are self-report questionnaires measuring mood symptoms and personality characteristics. The MMPI and PAI are useful for identifying and characterizing certain Axis I and Axis II psychiatric disorders. Both tests also contain scales that measure aspects of response validity (eg, random responding, symptom over-endorsement, and symptom-minimization). Specific MMPI and PAI scales have also demonstrated mild-to-moderate usefulness in identifying individuals with psychogenic nonepileptic seizures.

The Beck Depression Inventory (BDI) (26) is a brief self-report questionnaire assessing common symptoms of depression.

The Beck Anxiety Inventory (BAI) (27) is a brief self-report questionnaire assessing common symptoms of anxiety.

POTENTIAL CONFOUNDS TO NEUROPSYCHOLOGICAL TEST PERFORMANCE IN EPILEPSY

In addition to these general factors, there are factors specific to epilepsy populations that may confound test performance and interpretation. In some clinical contexts, such as a presurgical epilepsy evaluation, these factors could complicate interpretation of lateralization of the cognitive profile. One potential confound is the cognitive side effects of some AEDs. Such effects are often dose-dependent and have been shown to occur more frequently with AED polytherapy as opposed to monotherapy (28). The most common cognitive side effects of these medications are a reduction in processing speed and/or problems with attention. Therefore, any task that is timed may be negatively affected in individuals taking AEDs. For individuals with epilepsy, there is also the risk for their performance to be affected by the occurrence of a seizure during the evaluation or in the hours preceding the assessment. This is not a common occurrence but can more likely happen in individuals with a high frequency (eg, daily) of seizures. Thus, for practical purposes, parts of the evaluation may have to be completed in the context of a recent seizure. In addition to concerns about seizures and postictal confusion, test performances may be affected by subclinical epileptiform discharges. Such events last only momentarily and are often not detected by the individual with epilepsy or others around them. Such brief events could potentially interfere with test performance, particularly on tasks of attention, speed of processing or reaction time, and learning (29). To mitigate the possible effects of these confounds, efforts are made to conduct the neuropsychological testing in a comfortable, nondistracting environment (ie, a quiet room without substantial noise adjacent to it, little to no distracting room décor, the examiner conscious of their attire and hygiene, and comfortable seating). Subjective pain ratings may be taken throughout the evaluation to determine potential association with test performance. Examinees are typically given the option of taking breaks at regular intervals to reduce the possibility of fatigue, including taking breaks for meals or snacks during lengthy evaluations. Patient effort and motivation levels are often evaluated with formal or embedded tests specifically designed to detect performance validity.

If a seizure occurs during an assessment, it is customary to cease testing until the patient demonstrates resolution of any postictal confusion. While such judgments are inherently subjective and there is no consensus in the literature on how long the neuropsychologist should wait after a seizure has occurred before resuming the evaluation, common practice is to resume testing after a short period of time (eg, a few minutes) after a less severe seizure, such as a partial seizure, and to wait longer periods (eg, several hours) after more severe seizures, such as complex partial or generalized seizures. Regarding the potential for subclinical seizures to affect performance, the neuropsychological evaluation is designed to measure the same ability (eg, attention) at multiple time points throughout the evaluation, so that the effects of a subclinical event at any given time during testing is reduced in the overall summary of scores.

ASSESSMENT OF COGNITIVE FUNCTION IN THE PRESURGICAL CONTEXT

One of the most common reasons for referral to neuropsychology in an epilepsy center is for presurgical characterization of cognitive function. Testing in this context has historically served multiple purposes, including establishment of a presurgical cognitive baseline for comparison to postsurgical function, assistance with lateralization/localization of seizure focus, and prediction of postsurgical cognitive outcomes. Literature and rationale relevant to the use of neuropsychological testing for each of these purposes is discussed in the following section. As the majority of the patients presenting for epilepsy surgery have intractable epilepsy of medial temporal lobe onset (30), the preponderance of available neuropsychological literature on presurgical evaluation is weighted toward such samples. Thus, the discussion here is focused on neuropsychological contributions to the presurgical evaluation of patients with temporal lobe epilepsy (TLE).

Baseline Testing

Cognitive changes after surgery are common, with many patients showing measurable declines in memory or other abilities and a smaller set of patients actually showing cognitive improvements. Baseline neuropsychological testing provides unique contributions to the clinical management of surgical epilepsy patients in this context. Most importantly, neuropsychological testing allows for objective assessment of cognitive ability. As is discussed later in this chapter, patients’ subjective ratings of their cognitive abilities do not always accurately reflect their actual cognitive performances. Thus, the value of neuropsychological testing is that it utilizes tests with established reliability and validity that are administered and scored using standardized procedures. Test scores can then be compared to normative samples of healthy peers that account for the effects of age and demographic factors (eg, level of education and ethnicity) on cognitive test performances. Scores can also be compared to distinct clinical populations. When baseline testing is available, a patient can serve as his/her own baseline for comparison to postsurgical performances. This can be useful for characterizing the nature of cognitive changes after surgery to inform treatment planning (eg, neuropsychological intervention) or for identifying alternative explanations for a patient’s perception of cognitive change (eg, depression/anxiety). Finally, baseline and postsurgical cognitive data can be very useful for patients in making informed decisions about their academic or occupational goals.

Localization of Seizure Focus

It should be noted that much of the literature on neuropsychological lateralization of TLE assumes typical cerebral lateralization of language functioning to the left hemisphere. In fact, many of these studies attempt to control for this by excluding patients who have shown atypical language representation on Wada testing or by limiting samples to only include right-handed patients. Thus, the ability for neuropsychological testing to lateralize seizure focus is likely reduced in the context of patients who have atypical language representation or in other situations in which atypical functional organization is likely to have occurred (eg, focal left hemisphere brain injury at a very young age).

Prediction of Risk of Memory Decline

Risk factors for postoperative memory decline have generally been conceptualized according to one of two models, termed the functional reserve model and the functional adequacy model (44). The functional reserve model holds that the degree of postsurgical memory decline for a given patient depends upon the capacity (ie, functional reserve) of the remaining contralateral temporal lobe to support memory function after surgery. Early support for this model came from the observations of relatively preserved memory function in most unilateral patients and profound amnesia syndromes in patients having bilateral resections or in isolated cases of unilateral resection who were later determined to have significant structural abnormality in the remaining contralateral temporal lobe (45). Additional support for the functional reserve model has come from studies using the Wada test (discussed later). Alternatively, the functional adequacy model holds that the functional ability of the tissue to be resected in the ipsilateral temporal lobe determines the nature and extent of memory decline after surgery. The rationale for this model is that resection of functionally intact left temporal structures will result in decline in verbal memory abilities and resection of functionally impaired structures should have minimal effect on memory. This model has received support from presurgical functional (ie, neuropsychology scores and fMRI) and structural markers (ie, hippocampal volumes on MRI). Ultimately, the risk for postoperative memory decline is likely best determined through consideration of both ipsilateral and contralateral variables. We now provide a brief overview of some of the most commonly studied predictors of memory decline.

Neuropsychological Testing

Multiple studies have supported the utility of presurgical neuropsychological data for estimating relative risk for postsurgical memory declines. Overall, these studies have indicated that patients with higher preoperative memory scores are at a higher risk for postoperative memory decline. This effect has been most robustly demonstrated for verbal memory in patients having left temporal resections, with higher presurgical verbal memory scores indicating higher risk and greater magnitude of verbal memory decline following surgery (46–50).

MRI

Multiple studies have shown that presurgical hippocampal volumes ipsilateral to the seizure focus are predictive of postoperative memory outcomes. Specifically, memory decline becomes less likely with increased ipsilateral medial temporal sclerosis (MTS) (50–53).

Notably, presurgical MRI and neuropsychology data are not always in perfect agreement. Specifically, there may be times when MRI shows left MTS but the patient still demonstrates intact verbal memory on testing. Predicting memory outcome in these cases is not easy. It has been suggested that verbal memory in such patients may have reorganized to be supported by the right temporal lobe, essentially reducing risk for postsurgical declines. This view has received support from the “age of onset’ effect, which refers to the finding that patients with left MTS, intact verbal memory, and younger age of epilepsy onset generally have better memory outcomes after left temporal lobectomy than similar patients with more recent age of epilepsy onset. A possible explanation for the age of onset effect is that reorganization may be more likely to occur when left medial temporal damage occurs at a young age. However, this age of onset effect is not universal, and significant verbal memory decline still occasionally occurs in young onset, left MTS patients with intact verbal memory (54). In these cases, Wada testing may be informative for predicting the likelihood of memory declines.

Wada Test