The cognitive profile of occipital lobe epilepsy and the selective association of left temporal lobe hypometabolism with verbal memory impairment *†Alex A. Knopman, ‡§Chong H. Wong, *Richard J. Stevenson, *Judi Homewood, ‡¶Armin Mohamed, #**Ernest Somerville, ‡¶Stefan Eberl, ‡¶Lingfeng Wen, ‡¶Michael Fulham, and ‡§Andrew F. Bleasel Epilepsia, 55(8):e80–e84, 2014 doi: 10.1111/epi.12623
Alex Knopman is a Clinical Neuropsychologist completing a Ph.D. at Macquarie University, Sydney, NSW, Australia.
We investigated the cognitive profile of structural occipital lobe epilepsy (OLE) and whether verbal memory impairment is selectively associated with left temporal lobe hypometabolism on [18F]-fluorodeoxyglucose positron emission tomography (FDGPET). Nine patients with OLE, ages 8–29 years, completed presurgical neuropsychological assessment. Composite measures were calculated for intelligence quotient (IQ), speed, attention, verbal memory, nonverbal memory, and executive functioning. In addition, the Wisconsin Card Sorting Test (WCST) was used as a specific measure of frontal lobe functioning. Presurgical FDG-PET was analyzed with statistical parametric mapping in 8 patients relative to 16 healthy volunteers. Mild impairments were evident for IQ, speed, attention, and executive functioning. Four patients demonstrated moderate or severe verbal memory impairment. Temporal lobe hypometabolism was found in seven of eight patients. Poorer verbal memory was associated with left temporal lobe hypometabolism (p = 0.002), which was stronger (p = 0.03 and p = 0.005, respectively) than the association of left temporal lobe hypometabolism with executive functioning or with performance on the WCST. OLE is associated with widespread cognitive comorbidity, suggesting cortical dysfunction beyond the occipital lobe. Verbal memory impairment is selectively associated with left temporal lobe hypometabolism in OLE, supporting a link between neuropsychological dysfunction and remote hypometabolism in focal epilepsy. KEY WORDS: Occipital lobe epilepsy, Neuropsychological functioning, Surgical series, Focal epilepsy, Positron emission tomography.
Cognitive comorbidity is increasingly recognized as a feature of occipital lobe epilepsy (OLE), a rare condition accounting for 1–5% of patients investigated for epilepsy surgery.1 Subtle impairments in visuospatial and perceptual abilities have been reported in small samples of both genetic OLE and structural OLE.2,3 In addition, wider deficits in intellectual functioning, attention, episodic memory, and higher-level executive skills have been reported in genetic OLE,2 but the broader cognitive profile of structural OLE has received limited attention3–5 and remains largely unknown. The first aim of our study was to provide a comprehensive cognitive profile of structural OLE in a well-defined homogeneous surgical series. The second aim of our study was to
e81 Left Temporal Lobe Hypometabolism in OLE explore the etiology of neuropsychological impairment in OLE. We investigated the link between verbal memory deficit and left temporal lobe dysfunction—an area beyond the seizure focus—with quantitative [18F]-fluorodeoxyglucose positron emission tomography (FDG-PET). We hypothesized that deficits would be evident in a number of cognitive domains in line with the known epileptic network extending beyond the occipital cortex, and the more widespread cognitive deficits identified in genetic OLE. We also hypothesized verbal memory would be negatively correlated with hypometabolism of the left temporal lobe, with the strength of this relationship significantly stronger than that between left temporal lobe hypometabolism and executive skills, such as abstract reasoning, problem solving, set-shifting, and phonemic fluency.
executive functioning composite, the number of perseverative errors on the Wisconsin Card Sorting Test7 (WCST) was used as a standalone measure of cognitive flexibility.
Statistical analysis Psychometric scores were converted to age-adjusted zscores (mean 0, standard deviation [SD] 1) from published normative tables and a mean z-score for each cognitive domain calculated. The cutoff for mild cognitive impairment was set at z = 1.0, moderate impairment at z = 1.64 and severe impairment at z = 1.96. Nonparametric Wilcoxon signed-rank tests were used to determine whether cognitive domains were statistically different from the population mean (z = 0). Alpha was set at 0.05 (onetailed), as cognitive domain measures were presumed independent and expected to be impaired based on similar research on idiopathic OLE. Spearman correlation was used to investigate the relationship between temporal lobe hypometabolism and neuropsychological measures, with alpha set at 0.05. The strength of the correlation between verbal memory and left temporal lobe hypometabolism relative to WCST or executive functioning and left temporal lobe hypometabolism was examined using the Williams test.9 Here alpha was set at 0.05 (one-tailed), given the expected relationships between the cognitive variables and functioning of the brain’s regions of interest, based on similar research.10,11
Patient selection The study was approved by the ethics committees of Western and Central Sydney Area Health Services. Patients were drawn from the epilepsy surgery database at both Westmead and Royal Prince Alfred Hospitals, Sydney, Australia. All patients between 1994 and 2004 with a diagnosis of OLE established by clinical assessment, video–electroencephalography (EEG), ictal single proton emission computed tomography (SPECT), magnetic resonance imaging (MRI), interictal FDG-PET, and intracranial EEG (iEEG) were included in the study. One patient with missing memory data was excluded from the second part of the study (patient 9), leaving a final sample of five adults and three children. All patients were receiving antiepileptic polypharmacotherapy, of English-speaking background, right-handed, and with left hemisphere language dominance assumed. No surgical OLE patients over this period had IQs below 70, psychiatric comorbidity, brain injury, or dual pathology. All patients underwent epilepsy surgery restricted to the occipital cortex and were followed for a minimum of 24 months to ascertain surgical outcome. Seven patients became seizure-free (Engel class I),6 one patient was almost seizure-free (Engel class II), and one patient had worthwhile improvement (Engel class III). See Table S1 for demographic and clinical data for each patient. Neuropsychological measures Standardized neuropsychological tests were administered as part of presurgical investigations. Tests were chosen on the basis of age-appropriateness and ability to measure six cognitive domains routinely assessed in comprehensive neuropsychological workups: IQ, Speed, Attention, Verbal Memory, Nonverbal Memory, and Executive Functioning. See Data S1 for description of the cognitive measures and structure of cognitive domain composites. In addition to an
FDG-PET Presurgical FDG-PET was analyzed voxel-by-voxel using statistical parametric mapping (SPM) software (SPM version 2; Wellcome Department of Cognitive Neurology, London, United Kingdom) to identify significant regions of hypometabolism relative to age-matched healthy controls. The extent (volume) of significant hypometabolism in the left temporal lobe was determined by counting the number of contiguous voxels. Detailed procedures for FDG-PET acquisition and image processing have been reported elsewhere.8 See supporting information for further details (Data S1).
Results As a group, each cognitive domain for OLE patients was significantly poorer than the population mean for agematched controls. The mean value for each cognitive domain was: IQ: 83.3 (SD 10.5) corresponding to a z-score of 1.11 (SD = 0.7; z = 2.43, p = 0.008); Speed: 1.32 (SD = 1.1; z = 2.24, p = 0.01); Attention: 1.00 (SD = 0.7; z = 2.40, p = 0.009); Verbal Memory: 0.88 (SD = 1.1; z = 1.83, p = 0.03); Nonverbal Memory: 0.61 (SD = 1.1; z = 1.72, p = 0.04); and Executive Functioning: 1.2 (SD = 0.8; z = 2.55, p = 0.006). Mild impairments were evident for IQ, speed, attention, and executive functioning (Fig. 1). See Table S2 for Epilepsia, 55(8):e80–e84, 2014 doi: 10.1111/epi.12623
e82 A. A. Knopman et al.
Figure 1. Mean (and SD) for each cognitive domain. Mild impairments (z ≤ 1.0) were evident for IQ, speed, attention, and executive functioning. Epilepsia ILAE
results for individual patients. Table S3 details for each cognitive domain the percentage of patients falling into each classification of impairment or otherwise. Verbal memory performance was highly variable: three subjects demonstrated moderate impairment (patients 3, 6, and 7), each with left-sided OLE, and one subject demon-
strated severe impairment, with right-sided OLE (patient 1). The remaining patients demonstrated intact verbal memory. WCST was mildly impaired relative to age-matched peers (z = 1.59, SD = 1.6). All patients demonstrated hypometabolism, both in and beyond the epileptogenic occipital lobe (Fig. 2). Temporal lobe hypometabolism was found in all patients except patient 5, with two patients demonstrating bitemporal, three patients right temporal, and two patients left temporal hypometabolism. Typically, the pattern involved hypometabolism of the ipsilateral occipital lobe and ipsilateral or bilateral temporal lobe (patients 1–4 and 7–8). Patient 4 mainly showed occipital and ipsilateral parietal hypometabolism and patient 5 chiefly demonstrated an ipsilateral frontocentral pattern. See Table S2 for temporal lobe hypometabolism voxel counts. Verbal memory was significantly correlated with the volume of left temporal lobe hypometabolism (r = 0.90, p = 0.002). Executive functioning was correlated with verbal memory but this relationship just missed statistical significance (r = 0.70, p = 0.052). WCST was only weakly correlated with verbal memory (r = 0.22, p = 0.64). Neither executive functioning (r = 0.61, p = 0.11) nor WCST (r = 0.08, p = 0.87) was significantly correlated with left temporal lobe hypometabolism. The correlation
Figure 2. FDG-PET hypometabolism for each patient relative to age-matched healthy controls. All patients demonstrated hypometabolism extending beyond the occipital cortex. Temporal lobe hypometabolism was identified for all patients except patient 5. See Table S2 for temporal lobe voxel counts. Epilepsia ILAE Epilepsia, 55(8):e80–e84, 2014 doi: 10.1111/epi.12623
e83 Left Temporal Lobe Hypometabolism in OLE between verbal memory and left temporal lobe hypometabolism was significantly stronger than that between executive functioning (z = 1.85, p = 0.03) or WCST (z = 2.58, p = 0.005) and left temporal lobe hypometabolism. Of note, nonverbal memory was not correlated with the volume of right temporal hypometabolism (r = 0.02, p = 0.95).
Discussion We found that structural OLE was associated with significantly reduced ability relative to the age-matched population in every cognitive domain investigated. As a group, mild impairment was evident in IQ, processing speed, attention, and executive functioning. Verbal memory was variable and four subjects demonstrated moderate to severe impairment. These results support the hypothesis of more widespread cognitive impairment in structural OLE. In line with the hypothesis, poorer verbal memory was associated with left temporal lobe hypometabolism. This relationship was detectable in a very small sample and was significantly stronger than that between left temporal lobe hypometabolism and multiple measures of executive functioning, known to be related to the integrity of the frontal cortex.10 Combined, these results suggest left temporal lobe hypometabolism is specific to verbal memory impairment rather than a marker of broader cognitive dysfunction. Right temporal hypometabolism was not related to nonverbal memory, consistent with reports in the temporal lobe epilepsy literature of this anatomical-clinical relationship being questionable.12 Verbal memory is known to be correlated with left temporal lobe hypometabolism in patients with temporal lobe epilepsy.11 However, the present study is the first to report on the metabolic correlates of verbal memory in patients with extratemporal lobe epilepsy. It is also important to highlight that the cortical region of interest in this study was beyond the epileptic focus, with the surgical outcome providing strong evidence of the critical epileptogenic abnormality being restricted to the occipital cortex. Only two studies have examined the cognitive comorbidity associated with metabolic abnormality beyond the epileptic focus, both in patients with temporal lobe epilepsy, and both studies correlating prefrontal hypometabolism with nonmemory abilities.13,14 Seizures in OLE commonly spread forward into posterior temporal, parietal, and frontal brain cortices.1 In addition, SPECT and FDG-PET abnormalities extend beyond the occipital cortex, most commonly into regions of the temporal lobe.4 The current results support the assertion that hypometabolism extending into the left temporal lobe is associated with impaired verbal memory. The underlying pathophysiology of hypometabolism distant from the seizure focus remains incompletely understood. However, diaschisis appears to be a major contributor, with the occipital lobe dysfunction related to the epilepsy interrupting normal neuronal activity in the
occipitotemporal and occipitoparietal pathways.14 Additional remote metabolic depression may occur as a result of repeated seizure propagation along the same pathways. Our results and those of others15 support the hypothesis that regional hypometabolism is associated with regional dysfunction. Moreover, improvement in remote hypometabolism has been reported following successful surgery in patients with temporal lobe epilepsy with accompanying improvement in cognitive functioning.15 It would be valuable to address whether cognitive functions improve referable to extraoccipital regions following successful OLE surgery. There are some limitations to our study. The sample size is relatively small but intractable OLE is a rare condition and the largest neuropsychologically focused study to date involving refractory OLE included 20 patients but with no neuropsychological measures beyond visuoperceptual measures.3 Missing data precluded inclusion of a visuoperceptual cognitive domain included in similar studies,2,3 but verbal memory is not affected by potential visual field or other visuoperceptual deficits. We were unable to control for antiepileptic medication and thus we cannot rule out the contribution of medication, at least in part, to reductions in cognition. In summary, structural OLE is associated with mild impairments in speed, attention, and executive functioning in patients presenting for surgery. Some patients demonstrate moderate to severe impairment in verbal memory, a finding not expected with focal occipital lesions. Verbal memory impairment is correlated specifically with the volume of left temporal lobe hypometabolism. This observation adds to emerging findings of the link between cognitive dysfunction and cerebral glucose hypometabolism. Cognitive deficits in OLE related in part to metabolic disturbances beyond the occipital cortex, have implications for the functional etiology of cognitive dysfunction in other forms of focal epilepsy.
Acknowledgments This work was supported in part by University of Sydney Postgraduate Award, Millennium Institute Stipend, and Pfizer Neuroscience Research Grant to Dr Chong H. Wong.
Disclosure None of the authors has any conflicts of interest to disclose. 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.
References 1. Salanova V, Andermann F, Olivier A, et al. Occipital lobe epilepsy: electroclinical manifestations, electrocorticography, cortical stimulation and outcome in 42 patients treated between 1930 and 1991. Brain 1992;115:1655–1680. Epilepsia, 55(8):e80–e84, 2014 doi: 10.1111/epi.12623
e84 A. A. Knopman et al. 2. Bilo L, Santangelo G, Improta I, et al. Neuropsychological profile of adult patients with nonsymptomatic occipital lobe epilepsies. J Neurol 2013;260:445–453. 3. Piazzini A, Saetti C, Turner K, et al. Visuoperceptive impairment in adult patients with occipital lobe epilepsies. Epilepsy Behav 2009;15:256–259. 4. Aykut-Bingol C, Bronen RA, Kim JH, et al. Surgical outcome in occipital lobe epilepsy: implications for pathophysiology. Ann Neurol 1998;44:60–69. 5. Battaglia D, Chieffo D, Tamburrini G, et al. Posterior resection for childhood lesional epilepsy: neuropsychological evolution. Epilepsy Behav 2012;23:131–137. 6. Engel J Jr, Van Ness PC, Rasmussen TB, et al. Outcome with respect to epileptic seizures. In Engel JJ (Ed) Surgical treatment of the epilepsies. New York: Raven Press, 1993:609–621. 7. Heaton R, Chelune G, Talley J, et al. Wisconsin Card Sorting Test (WCST) manual, revised and expanded. Odessa, FL: Psychological Assessment Resources; 1993. 8. Wong CH, Bleasel A, Wen L, et al. The topography and significance of extratemporal hypometabolism in refractory mesial temporal lobe epilepsy examined by FDG-PET. Epilepsia 2010;51:1365–1373. 9. Williams EJ. The comparison of regression variables. J R Stat Soc Series B Stat Methodol 1959;21:396–399. 10. Alvarez JA, Emory E. Executive function and the frontal lobes: a metaanalytic review. Neuropsychol Rev 2006;16:17–42. 11. Weintrob DL, Saling MM, Berkovic SF, et al. Verbal memory in left temporal lobe epilepsy: evidence for task-related localization. Ann Neurol 2002;51:442–447.
12. Kneebone AC, Lee GP, Wade LT, et al. Rey Complex Figure: figural and spatial memory before and after temporal lobectomy for intractable epilepsy. J Int Neuropsychol Soc 2007;13:664–671. 13. Jokeit H, Seitz RJ, Markowitsch HJ, et al. Prefrontal asymmetric interictal glucose hypometabolism and cognitive impairment in patients with temporal lobe epilepsy. Brain 1997;120:2283–2294. 14. Takaya S, Hanakawa T, Hashikawa K, et al. Prefrontal hypofunction in patients with intractable mesial temporal lobe epilepsy. Neurology 2006;67:1674–1676. 15. Takaya S, Mikuni N, Mitsueda T, et al. Improved cerebral function in mesial temporal lobe epilepsy after subtemporal amygdalohippocampectomy. Brain 2009;132:185–194.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Demographic and clinical data for each patient. Table S2. Results for individual patients. Table S3. Percent of patients who show cognitive impairment by domain. Data S1. Supplementary methods.
The cognitive profile of occipital lobe epilepsy and the selective association of left temporal lobe hypometabolism with verbal memory impairment.
We investigated the cognitive profile of structural occipital lobe epilepsy (OLE) and whether verbal memory impairment is selectively associated with ...