Gender, Memory, and Hippocampal Volumes: Relationships in Temporal Lobe Epilepsy

Gender, Memory, and Hippocampal Volumes: Relationships in Temporal Lobe Epilepsy

Epilepsy & Behavior 1, 112–119 (2000) doi:10.1006/ebeh.2000.0051, available online at on Gender, Memory, and Hippocampal V...

71KB Sizes 0 Downloads 40 Views

Epilepsy & Behavior 1, 112–119 (2000) doi:10.1006/ebeh.2000.0051, available online at on

Gender, Memory, and Hippocampal Volumes: Relationships in Temporal Lobe Epilepsy Michelle Bengtson, Ph.D., Roy Martin, Ph.D., 1 Stephen Sawrie, Ph.D., Frank Gilliam, M.D., Edward Faught, M.D.S., Richard Morawetz, M.D.S., and Ruben Kuzniecky, M.D.S. Epilepsy Center, Department of Neurology, University of Alabama at Birmingham, 312 CIRC, 1719 6th Avenue South, Birmingham, Alabama 35294-0021 Received January 18, 2000; revised March 20, 2000; accepted for publication March 21, 2000

Previous research has suggested bilateral hippocampal support for verbal memory in women with early left-hemisphere injury and that women experience better verbal memory outcome following anterior temporal lobectomy (ATL). The present study investigated two issues: (1) Do women have better verbal memory outcome following ATL compared with men? (2) Are verbal memory abilities differentially supported by the right and left hippocampus in males and females? Verbal memory performance [Wechsler Memory Scale: Logical Memory (LM) savings score] was assessed in 70 patients who underwent ATL. MRI volumetric measurements of the left and right hippocampus were performed. No LM savings score difference was found between groups preoperatively although a statistically significant gender effect (P < 0.04) was found for postoperative LM savings scores. Females displayed better postoperative memory performance, regardless of side of surgery. Preoperative verbal memory performance was not associated with right or left hippocampal volumes in either left or right ATL females, although the right hippocampus was positively associated with memory performance for left ATL males. Hippocampal volumes were not associated with postoperative LM savings scores for any group. Results suggest that prose recall was only modestly influenced by gender and that bilateral hippocampal support for prose recall was not present in our female patients. © 2000 Academic Press Key Words: temporal lobe epilepsy; memory; neuropsychology; gender differences; magnetic resonance imaging volumetry; hippocampus.

There is an emerging line of investigation considering potential gender differences in relation to functional brain asymmetry. Several studies have suggested that the neuroanatomical representation of cognitive functions such as language, visuospatial ability, memory, and attention may vary depending on gender (1– 4). The ability with which verbal and nonverbal skills are executed may be a function of gender, in that females tend to demonstrate a greater proficiency on verbal tasks, while males demonstrate relatively superior nonverbal/spatial performance (2, 5–7). Hart and

O’Shanick (8) reported that neurologically intact females performed better on verbal memory measures as compared with visual memory measures whereas the opposite was found for males. Differential performances reported between genders across cognitive domains have led to hypotheses regarding functional brain asymmetry being more pronounced in adult males than in females (3, 9 –11). Dichotic listening paradigms and tachistoscopic research with neurologically intact adults have found greater verbal and visual process functional asymmetry (12, 13), as well as greater structural asymmetry in adult males than females (14). Other studies have demonstrated a larger degree of right-hemisphere specialization for spatial functions in men than in women (13, 15).


To whom all correspondence should be addressed. Fax: (205) 975-6255.


1525-5050/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.


Gender–Memory Relationships after ATL

It was demonstrated more than three decades ago that cognitive deficits following unilateral anterior temporal lobectomy (ATL) were associated with not only side of surgery, but also the gender of the patient (16). Geckler et al. (17) reported that following left ATL, women demonstrated superior performance on verbal memory measures as compared with men. Powell et al. (18) demonstrated a propensity for men to experience declines in verbal memory following left ATL, while women demonstrated significant visual memory decline following right temporal lobectomy. Trenerry et al. (19) reported that following left ATL, females experienced an improvement in their ability to recall story content from a verbal memory test while men suffered a decline. These studies have supported the idea that males and females differ in cognitive abilities and that they are differentially at risk for memory decline following ATL. Previous studies have shown that preoperative verbal memory abilities are related to the integrity of the left hippocampus. Studies examining MRI-based left hippocampal volumes in left-hemisphere speechdominant temporal lobe epilepsy (TLE) patients with temporal seizure foci have demonstrated that atrophic left hippocampal tissue was associated with impaired preoperative verbal memory (20, 21). Also, changes in verbal memory performance following ATL have been associated with the preoperative status of the left hippocampus. In general, removal of normal or mildly sclerotic left hippocampus is associated with verbal memory decline following resection of the left temporal lobe in both males and females. Trenerry et al. (19) reported that women were at less risk for postsurgical decline and may in fact experience postsurgical verbal memory improvements. That report also found that preoperative verbal memory performance was significantly correlated with both left and right hippocampal volumes in left temporal lobe women but not men. These authors suggested that verbal memory superiority in women undergoing left ATL is due to greater bilateral hippocampal representation of verbal memory abilities compared with their male counterparts. The question remains as to whether females’ verbal memory abilities are less lateralized in general or whether such verbal memory is necessarily more plastic secondary to early left-hemisphere injury. The purpose of the present study was to investigate gender-related associations between preoperative MRI hippocampal volumetry and verbal memory outcome following temporal lobectomy. Specifically, we sought to reexamine previous research assessing relationships between male and female presurgical verbal

memory performance with MRI hippocampal volumes. Based on previous research we expected that females would demonstrate superior verbal memory performance both pre- and post-ATL compared with males and that both right and left hippocampal volumes would be associated with verbal memory performance in left ATL females.

METHODS Patients Data were obtained from 22 women and 18 men who underwent left ATL and 16 women and 14 men who underwent right ATL for medical refractory epilepsy. Long-term interictal and ictal scalp EEG recording was obtained in all patients during inpatient video monitoring using digital equipment. The guidelines of the American EEG Society for long-term monitoring were followed (22). Patients with intracarotid sodium amytal testing (Wada testing) evidence of mixed or right-hemisphere speech dominance were excluded. No patient in this group had evidence of a mass lesion (e.g., tumor, hamartoma, arteriovenous malformation) on MRI. For the purposes of this study, mesial temporal sclerosis was not considered a structural lesion. All seizures were found to originate from a single temporal lobe.

MRI Protocol MRI studies were performed on a 1.5 T Philips ACS scanner (Best, The Netherlands) (23). A fast scout scan (axial and coronal images, 90 seconds) was obtained to position the subject. Angulation correction was performed by adjusting the angle of acquisition to be parallel with the interhemispheric fissure. A localizing series of 5-mm sagittal T1-weighted spin echo images were obtained. A three-dimensional set of 110 to 130 coronal 1.5 mm images through the entire brain were obtained perpendicular to the long axis of the hippocampus (TR ⫽ 20 milliseconds, TE ⫽ 6 milliseconds, matrix size ⫽ 218 ⫻ 256, FOV ⫽ 23 cm). If any image had motion artifact, it was repeated. The three-dimensional images were transferred to a Philips Gyroview workstation and analyzed using commercial software (ISG Technologies Inc., Malton Canada). Volumetric measurements were performed with an interactive hand contouring device. Each image was enlarged times 2. The hippocampus was measured. The volume Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

114 for each slice was calculated by multiplying the area outlined by slice thickness. Total hippocampal volume was calculated by adding each slice volume. Individual variation in the volume of the hippocampus was normalized by correcting for total intracranial volume based on previously published techniques (24). The area of supratentorial brain parenchyma on the midsagittal image was used to normalize volumes. In addition, the difference (DHV) between the two volumes was derived by subtracting the left hippocampal volume from the right hippocampal volume. Our normalization sample consisted of 17 healthy volunteers (9 women, 8 men; mean age 35 years, age range 24 – 41). Anatomic guidelines for the hippocampus were established using Duvernoy’s anatomic sections (25). Presence of digitations established the head of the hippocampus. Separation of the hippocampal head and amygdala was determined by use of the lateral ventricle. The entire head and body of the hippocampus was measured, but parahippocampal gyrus was excluded. To minimize partial volume effect, the three-dimensional MRI data were resampled as 1-mmthick images (no gap), using multiple planes of reconstruction to follow the anatomic curvature.

Epilepsy Surgery All patients underwent en bloc resections of either the left or right anterior temporal lobe. All surgical procedures were performed by a single neurosurgeon. Resections were begun below a line through the midportion of the middle temporal gyrus. Anterior resection extended between 4.0 and 4.5 cm of the left temporal lobe and between 4.0 and 5.5 cm of the right temporal lobe, and was carried mesially into the temporal horn of the adjacent lateral ventricle. Resection of the amygdala, orbicua, adjacent uncus, and anterior two-thirds of the hippocampus was performed and sent for pathological analysis. This procedure has been reported elsewhere (26).

Neuropsychological Variables All patients were referred for neuropsychological evaluations prior to surgery and again approximately 6 months following surgery. The neuropsychological evaluations included a full assessment of general intellectual functioning utilizing the Wechsler Adult Intelligence Scale—Revised (WAIS-R) (27). The Logical Memory (LM) subtest of the Wechsler Memory Scale Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

Bengston et al.

(WMS) (28) was used as our measure of verbal memory. The LM savings score has been shown to be sensitive to hippocampal volume loss (29, 30) and was presented as the primary variable in the Trenerry et al. (19) paper. The Logical Memory test consists of an oral recitation of two complex prose paragraph stories. After each respective story the patient is asked to recite the story aloud to the examiner, in as close to the same words as the patient can recall. Approximately 30 minutes later, without a repeat presentation, the patient is again asked to recite the stories from memory. On each occasion the patient’s recitations are recorded verbatim and scored according to standard criteria. A LM savings score was calculated by dividing the delayed recall score by the immediate recall score and multiplying that total by 100 [(LM delay/LM immediate) ⫻ 100 ⫽ savings score]. The LM savings scores from the preoperative testing session and the postoperative testing were our dependent variables.

Statistical Analyses A series of ANOVAs were employed to examine between-group differences across the demographic variables and MRI hippocampal volumetric values. A 2 (gender) X 2 (side of surgery) X 2 (pre-, post-ATL testing) repeated-measures ANOVA was performed to examine changes in WMS LM saving scores. Univariate and post hoc t tests were also performed as needed. Correlational analyses were performed in each group to explore potential differential relationships with the MRI-based hippocampal volume measurements. If significant between-groups differences were found then those effects would be controlled by use of partial correlation coefficient statistics. Power analysis was performed due to the large number of correlations and also due to the varying sample sizes among the dependent variables. We set power level to detect a medium effect size (power ⫽ 70%). When a correlation involved the left ATL males, the effect size of r ⫽ 0.43 was required for significance (N ⫽ 18, P ⬍ 0.05). When a correlation involved left ATL females, the effect size of r ⫽ 0.40 was required for significance (N ⫽ 22, P ⬍ 0.05). For right ATL males, the effect size of r ⫽ 0.51 was required (N ⫽ 14, P ⬍ 0.05). The right ATL females required an effect size of r ⫽ 0.42 (N ⫽ 16, P ⬍ 0.05). Only correlations that exceeded their sample specific effect sizes were presented or discussed as statistically significant.


Gender–Memory Relationships after ATL TABLE 1 Mean (SD) Demographic, Neuropsychological, and Hippocampal Volume Values for Left Temporal Lobectomy Patients a

Age Education Seizure onset Seizure duration Verbal IQ Performance IQ LM ss—pre LM ss—post HV left HV right HV asym

Males (n ⫽ 18)

Females (n ⫽ 22)

32.0 (9.6) 12.1 (3.4) 14.8 (11.0) 16.5 (12.3) 86.2 (9.2) 90.3 (14.0) 49 (28.5) 47 (23.6) 2501 (440) 3398 (510) 896 (433)

35.0 (9.0) 12.2 (1.8) 13.5 (8.7) 20.7 (10.8) 86.4 (9.3) 89.1 (10.9) 59 (24.4) 64 (23.3) 2448 (372) 3403 (336) 952 (388)


WAIS-R Verbal IQ and Performance IQ: LM ss, logical memory savings score; HV, hippocampal volume (mm 2); asym, asymmetry score (right HV ⫺ left HV).

RESULTS Demographic, Neuropsychological, and MRI Volume Data Tables 1 and 2 provide demographic, preoperative IQ score, Logical Memory savings score, and hippocampal volumetric data for both men and women who underwent left or right ATL, respectively. No significant between-group differences were found across any of the demographic variables or the IQ scores. A 2 (gender) X 2 (side of surgery) X 2 (pre-, postATL) repeated-measures ANOVA with LM savings score (pre- and postoperative) as the dependent variable revealed no significant interaction findings. Additional univariate analyses for the preoperative LM savings score revealed no significant effects across the gender or side of surgery factors. However, for the LM postoperative savings score separate main effects were found for side of surgery [F(1, 68) ⫽ 5.1, P ⬍ 0.03] and gender [F (1, 68) ⫽ 4.4, P ⬍ 0.04] factors. Males, regardless of surgery side, did not recall as much information on the LM savings score compared with females, and the left ATL group had lower LM savings scores compared with the right ATL group. No interaction effects were found. When examining the postoperative LM savings scores, males (combined left and right ATL patients) displayed a modest 2-point gain in the LM savings score from pre- to postoperative testing while females showed a 5-point gain. Right ATL patients improved by 5 points on their LM savings score compared with a smaller gain of 3 points for the

left ATL group. None of these LM savings scores represented more than a 7% change from pre- to postoperative testing, reflecting relatively small magnitude of change (see Tables 1 and 2). Two additional univariate analyses were performed on a larger ATL series from our center (n ⫽ 92, left ⫽ 49, right ⫽ 43) who were over the age of 16 and who had completed the WMS LM at both pre- and postoperative evaluations. This larger group did not have MRI hippocampal volumetric data and did not differ from our smaller study group in terms of age (mean age ⫽ 29.5, SD ⫽ 7.7) [F (1,160) ⫽ 1.2, P ⬎ 0.28], years of education (12.7, SD ⫽ 2.3) [F (1,160) ⫽ 1.2, P ⬎ 0.72], age at seizure onset (11.9, SD ⫽ 8.9) [F(1, 160) ⫽ 0.37, P ⬎ 0.55], WAIS-R Verbal IQ (86.5, SD ⫽ 10.3) [F (1,160) ⫽ 0.07, P ⬎ 0.80], WAIS-R Performance IQ (86.8, SD ⫽ 11.5) [F (1,160) ⫽ 0.004, P ⬎ 0.95], or seizure free status post-ATL [Mann–Whitney U ⫽ 3052, z ⫽ ⫺ 1.3, P ⬎ 0.18]. However, the groups differed with regard to seizure duration (17.6, SD ⫽ 8.3) [F(1,160) ⫽ 5.4, P ⬍ 0.02] with the group who had MRI volumetric analysis having epilepsy for an average of 3 more years than the no-MRI group (20.3 years vs 17.6 years). Examination of LM savings scores for the larger patient series revealed a significant time (LM savings scores at pre- and post-ATL) X side of surgery effect [F (1,88) ⫽ 6.4, P ⬍ 0.01]. Follow-up analysis revealed that LM savings score declined from pre- to post-ATL for the left ATL group (pre-LM savings score: 52, post-LM savings score: 47) while a performance increase was seen for the right ATL group (pre-LM

TABLE 2 Mean (SD) Demographic, Neuropsychological, and Hippocampal Volume Values for Right Temporal Lobectomy Patients

Age Education Seizure onset Seizure duration Verbal IQ Performance IQ LM ss—pre LM ss—post HV left HV right HV asym

Males (n ⫽ 14)

Females (n ⫽ 16)

35.2 (7.2) 12.6 (3.6) 12.9 (10.9) 22.2 (11.5) 93.4 (12.0) 95.1 (12.0) 59 (27.1) 66 (28.5) 3263 (346) 2621 (679) ⫺609 (540)

32.7 (10.1) 12.0 (1.8) 10.9 (8.6) 21.7 (9.3) 88.1 (8.9) 86.0 (7.5) 71 (28.0) 75 (20.0) 3372 (478) 2547 (529) ⫺867 (483)

a WAIS-R Verbal IQ and Performance IQ: LM ss, Logical memory savings score; HV, hippocampal volume (mm 2); asym, asymmetry score (right HV ⫺ left HV).

Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.


Bengston et al.

TABLE 3 Correlations between Hippocampal Volumes, Gender, and LM Savings Scores for the Left and Right ATL Groups Right ATL group

Left ATL group

Males (n ⫽ 14)

Females (n ⫽ 16)

Males (n ⫽ 18)

Females (n ⫽ 22)

⫺0.04 ⫺0.22 ⫺0.15

⫺0.11 0.16 0.13

0.48 0.13 0.52*

⫺0.30 0.37 0.06

⫺0.27 ⫺0.31 ⫺0.35

⫺0.14 0.17 0.10

0.26 0.30 0.48

⫺0.14 ⫺0.14 ⫺0.31

LM savings score—pre HA a LH RH LM savings score—post HA LH RH

* P ⬍ 0.04. a HA, hippocampal asymmetry; LH, left hippocampal volume; RH, right hippocampal volume.

savings score: 60, post-LM savings score: 70). However, no significant interaction or gender effects were found. Left ATL males and females performed statistically similar at both the preoperative time [left ATL males: 49 (SD ⫽ 28), left ATL females: 54 (SD ⫽ 19)] and postoperative time [males: 43 (SD ⫽ 36), females: 51 (SD ⫽ 28)], although left ATL females had higher LM savings scores at both testings. Right ATL males improved from a preoperative LM savings score of 59 (SD ⫽ 26) to a postoperative score of 71 (SD ⫽ 19) while right ATL females improved from a preoperative score of 60 (SD ⫽ 20) to a postoperative LM savings score of 68 (SD ⫽ 22). MRI Hippocampal Volumes Left ATL patients had significantly smaller left hippocampal volumes compared with right ATL patients [F(1, 67) ⫽ 69.9, P ⫽ 0.0001], while right ATL patients had significantly smaller right hippocampal volumes [F(1,67) ⫽ 43.0, P ⬍ 0.0001]. No significant interaction effects or gender differences were found for left or right hippocampal volumes. LM Savings Scores and Hippocampal Volumes Pearson correlation coefficients were examined for each of four groups (left ATL males, left ATL females, right ATL males, right ATL females) across the LM savings scores (pre- and postoperative). Table 3 lists the correlation coefficient values. When examining the hippocampal volume and LM savings score correlation at the preoperative time only one significant correlation was found. A single statistically significant relationship was found between the right hippocamCopyright © 2000 by Academic Press All rights of reproduction in any form reserved.

pal volume and the preoperative LM savings score for left ATL males. A statistical trend (P ⬍ 0.06) was observed for the hippocampal asymmetry value and the savings score for left ATL males. No correlations were statistically significant for the left ATL females or the right ATL male or female patients. Postoperative LM savings scores were not significantly associated with any of the three hippocampal volume values across the left or right ATL groups. These null results held regardless of gender. Additional correlational analysis collapsing across the entire patient sample found that the left hippocampal volume was significantly associated with LM savings score (r ⫽ 0.26, P ⬍ 0.03) and the hippocampal asymmetry value was significantly related to the postoperative LM savings score (r ⫽ 0.25, P ⬍ 0.04).

DISCUSSION The present study had two central goals: first, to examine the potential moderating influence of gender on verbal memory outcome following ATL, and second, to examine whether associations between hippocampal volumes and LM savings scores are influenced by gender. The idea of gender-based differences in neuroanatomic representation of verbal memory abilities has recently been reported as an explanation of why females experience a more promising outcome with respect to verbal memory capacity than do men following ATL (19). Females may have a greater propensity for bilateral hippocampal contribution to certain forms of verbal memory while males may exhibit a more symmetric memory representation within the hippocampal memory system. The results of the

Gender–Memory Relationships after ATL

present study were not in complete agreement with these previous conclusions. We found that females, regardless of side of surgery, showed statistically better memory performance than males following ATL. However, when examining a larger patient series, who had not undergone MRI volumetric analysis, we found no gender differences in LM savings score performance at pre- or postoperative testings. Thus our results are not consistently supportive of previous reports showing better memory outcome for females (19, 31, 32). Trenerry et al. (19) found gender differences on the WMS Logical Memory task while Berenbaum et al. (31) found female preference on a task of word list learning. Although females in the present study demonstrated larger postoperative gains on the LM savings score than their male counterparts, the difference was modest in magnitude, with both groups demonstrating small performance improvements. Gender-based differences in verbal memory performance appear partially moderated by task type, with word list learning and paired associate memory paradigms more consistently demonstrating gender differences compared with prose recall or verbal recognition tasks (31, 32). Second, contrary to an earlier report we failed to find a relationship between the right and left hippocampal MRI-based volumes and the LM savings scores in left ATL females. Unexpectedly, we found that in left ATL males the right hippocampal volume was statistically related to preoperative LM performance. Neither male nor female left ATL patient performance on the verbal memory measure was related to left hippocampal volume or to the hippocampal asymmetry value. Trenerry et al. previously found that MRI-based hippocampal volumes were unrelated to verbal memory performance in right ATL patients and in left ATL males. We found similar null results for the right ATL groups. An explanation of the right hippocampal volume–left ATL male correlation is not readily apparent from review of the literature on sex differences. Interestingly, only when we combined across gender and ATL factors for a single statistical analysis (i.e., entire patient series) did we find the expected left hippocampal volume and LM savings score relationship consistent with previous MRI volumetric reports (29, 30). Issues of statistical restriction of range of our groups or statistically chance findings may have accounted for the present finding. However, relevant to comparisons with the Trenerry et al. (19) findings was that we found virtually no correlation between right hippocampal volume and left ATL female performance on the verbal memory task even

117 though this was our largest patient group. A statistical trend (P ⬎ 0.10) was noted between the left hippocampal volume and LM savings score for the left ATL females. Trenerry and colleagues’ (19) assertion that “verbal memory abilities are more plastic following early left hemisphere insult” (p. 74) in left ATL females is not supported by the present data. Additionally, contrary evidence for this gender plasticity hypothesis was suggested by Berenbaum et al. (31) who found that histopathologically-determined left hippocampal cell loss was related to word list learning performance for both males and females. There are potential contributing factors to the differences between our current conclusions and those reported by Trenerry et al. (19). First, the sample size in the present study was approximately one-half that of the ATL patients in the Trenerry et al. (19) report. Although our sample size was smaller than that of Trenerry et al. (19) we did set statistical power a priori and reported only significant results meeting those criteria. When examining the magnitude of statistical effects from the present study we did not approach the correlation size for the left ATL females compared with Trenerry et al. (19). Even with a larger sample size this correlation would not likely be significant given its near-zero relationship between right hippocampal volume and female gender. It may also be important to consider that demographically our sample was similar to that of Trenerry et al. (19) in most respects including WAIS-R IQ, education, and age at seizure onset. The present patient sample appears reasonably consistent demographically and clinically with other reported ATL patient series (34). All patients from both series had pathology-confirmed evidence of mesial temporal lobe sclerosis. Other possible explanations for the studies’ discrepant findings could include variation in patient surgery selection criteria and extent of surgery resection. Surgical resection parameters from our center appear larger compared with those described by Trenerry et al. (19) and may account for some variability of memory outcome. However, more extensive hippocampal resection does not necessarily result in more pronounced verbal memory loss following ATL (35). Another factor that is often highlighted as a potential confound for this type of research includes different interinstitutional hippocampal volumetric techniques, as well as different interinstitutional definitions of normal hippocampal measurements. Several researchers have noted that while MRI volumetric studies will be useful presurgically for patients with intractable epilepsy as well as in research attempting Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

118 to study the neuroanatomic correlates of cognitive functioning, great caution must be taken when comparing interinstitutional results. A wide range of “normal” volumes exists in studies of young adults. Hence, the volumetric measurement procedure is highly technique dependent and determination of “normal” volumes will be institution specific (36). In trying to empirically use MRI volumetry for research and clinical purposes, several conditions should be considered: (1) There is a linear relationship between hippocampal volumes and total intracranial volumes in normal young adults. (2) Male hippocampal volumes are generally larger than those of women largely as a function of increased intracranial volumes. (3) Hippocampal volumes remain relatively stable throughout young adulthood but in later life demonstrate a negative relationship to age-related cerebral atrophy. (4) Hippocampal volumes are largely variable depending on the hippocampal boundaries included in volume measurement; it is also possible that variability will likely be a function of differing image acquisition patient positioning. (5) Finally, some variation may be found to exist as a function of sample racial and ethnic makeup (36 –38). These are potential confounds that could continue to be major contributing factors in looking at conflicting interinstitutional conclusions. Another consideration is the possibility that either Trenerry and colleagues’ conclusions or those of the current paper were the result of spurious results. We acknowledge that only one of 12 correlations with pre-LM savings scores was significant. However, the strength of the association was similar in magnitude to that of Trenerry et al. (19) and we reported correlations as significant only if they met a priori power analysis set points. Neuropsychological evaluation of patient deficits continues to be a necessary addition to the medical treatment regimen in temporal lobectomy patients, as research has found that patients tend to make the mistaken assumption that their memory has improved when they have experienced positive surgical outcomes, when in reality objective assessment does not reveal such positive outcomes (32, 39). Obviously the current study was only an initial step in corroborating earlier findings suggesting neuroanatomically based gender differences in memory functioning. Future research should strive to make sense of the conflicting results between studies for a clearer understanding of the relationship between volumetric measurement and neuropsychological performance and the promise it holds for counseling epilepsy surgery candidates. Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

Bengston et al.

ACKNOWLEDGMENTS The authors thank Erhan Bilir and Susan Ho for their efforts with the MRI volumetric measurements.

REFERENCES 1. 2. 3. 4. 5. 6.


8. 9.


11. 12. 13.

14. 15. 16. 17.




Lansdell H. The effect of neurosurgery on a test of proverbs. Am Psychol 1961;16:448. McClone J. Sex differences in functional brain asymmetry. Cortex 1978;14:122– 8. McClone J. Sex differences in human brain asymmetry: a critical review. Behav Brain Sc 1980;3:215– 63. Kimura D. Right temporal lobe damage. Arch Neurol 1963;8: 264 –71. Este WK. Learning theory and intelligence. Am Psychol 1974; 29:740 –9. Mann VA, Sasanuma S, Sakuma N, Masaki S. Sex differences in cognitive abilities: a cross-cultural perspective. Neuropsychologia 1990;28:1063–77. Weinderholt WC, Cahn D, Butters NM, Salmon DP, KritzSilverstein D, Barrett-Connor E. Effects of age, gender and education on selected neuropsychological tests in an elderly community cohort. J Am Geriatr Soc 1993;41:639 – 47. Hart RP, O’Shanick GJ. Forgetting rates for verbal, pictorial, and figural stimuli. J Clin Exp Neuropsychol 1993;15:245– 65. Desmond DW, Glenwick DS, Stern Y, Tatemichi TK. Sex differences in the representation of visuospatial functions in the human brain. Rehab Psychol 1994;39:3–14. Howard R, Fenwick P, Brown D, Norton R. Relationship between CNV asymmetries and individual differences in cognitive performance, personality and gender. Int J Psychopathol 1992;13:191–7. Kramer JH, Delis DC, Daniel M. Sex differences in verbal learning. J Clin Psychol 1988;44:907–15. Hannay J, Malone D. Visual field effects and short-term memory for verbal material. Neuropsychologia 1976;14:203–9. McGlone S, Davidson D. Relation between cerebral speech laterality and spatial ability with special reference to sex and hand preference. Neurpsychologia 1973;11:105–13. Wada J, Clarke R, Hamm A. Cerebral hemispheric asymmetry in humans. Arch Neurol 1975;32:239 – 46. McGlone S, Kertesz A. Sex differences in cerebral processing of visuospatial tasks. Cortex 1973;9:313–20. Lansdell H. A sex difference in effect of temporal-lobe neurosurgery on design preference. Nature 1962;194:852– 4. Geckler C, Chelune G, Trenerry M, Ivnik R. Gender related differences in cognitive status following temporal lobectomy. Arch Clin Neuropsychol 1993;8:226 –7. Powell GE, Polkey CE, McMillan T. Relationships between neuropathology and cognitive functioning in temporal lobectomy patients. J Neurol Neurosurg Psychiatry 1987;50:167–76. Trenerry MR, Jack CR Jr, Cascino GD, Sharbrough FW, Ivnik RJ. Gender differences in post-temporal lobectomy verbal memory and relationships between MRI hippocampal volumes and preoperative verbal memory. Epilepsy Res 1995;20: 69 –76. Lencz T, McCarthy G, Bronen RA, Scott T, Inserni J, Sass K, Novelly R, Kim J, Spencer D. Quantitiative magnetic resonance


Gender–Memory Relationships after ATL


22. 23.

24. 25. 26.

27. 28. 29.


imaging in temporal lobe epilepsy: relationship to neuropathology and neuropsychological function. Ann Neurol 1992;31: 629 –37. Loring DW, Murro AM, Meador KJ, Lee GP, Gratton CA, Nichols ME, Gallagher BB, King DW, Smith JR. Wada memory testing and hippocampal volume measurements in the evaluation for temporal lobectomy. Neurol 1993;43:1789 –93. American Epilepsy Society. Guidelines for long-term monitoring for epilepsy. J Clin Neurophysiol 1994;11:2–5. Bilir E, Craven W, Hugg J, Gilliam F, Martin R, Faught E, Kuzniecky R. MRI-volumetric measurement of the limbic system: anatomic determinants. Neuroradiology 1998;40:138 – 44. Jack JC. MRI-based hippocampal volume measurements in epilepsy. Epilepsia 1994;35:S21–9. Duvernoy HM. The human hippocampus. Munich: Bergmann Verlag, 1988. Kuzniecky R, Burgard S, Faught, Morawetz R, Bartolucci A. Predictive value of magnetic resonance imaging in temporal lobe epilepsy. Arch Neurol 1993;31:138 – 46. Wechsler DA. Wechsler Adult Intelligence Scale—Revised. New York: Psychological Corp., 1981. Wechsler DA. Standardized memory scale for clinical use. J Psychol 1945;19:87–95. Martin RC, Hugg JW, Roth DL, Bilir E, Gilliam FG, Faught E, Kuzniecky RI. MRI extrahippocampal volumes and visual memory: correlations independent of MRI hippocampal volumes in temporal lobe epilepsy patients. J Int Neuropsychol Soc 1999;5:534 – 42. Sass KJ, Sass A, Westerveld M, Lencz T, Rosewater KM, Novelly RA, Kim JH, Spencer DD. Russell’s adaptation of the Wechsler Memory Scale as an index of hippocampal pathology. J Epilepsy 1992;5:24 –30.










Berenbaum SA, Baxter L, Seidenberg M, Hermann B. Role of the hippocampus in sex differences in verbal memory: memory outcome following left anterior temporal lobectomy. Neuropsychology 1997;11:585–91. McGlone J. Memory complaints before and after temporal lobectomy: do they predict memory performance or lesion laterality? Epilepsia 1994;35:529 –39. Tranhan EE, Quintana JW. Analysis of gender effects upon verbal and visual memory performance in adults. Arch Clin Neuropsychol 1990;5:325–34. Strauss E, Loring D, Chelune G, Hunter M, Hermann B, Perrine K, Westerveld M, Trenerry M, Barr W. Predicting cognitive impairment in epilepsy: findings from the Bozeman epilepsy consortium. J Clin Exp Neuropsychol 1995;17:909 –17. Loring DW, Lee GP, Meador KJ, Smith JR, Martin RC, Ackell AB, Flanigan HF. Hippocampal contribution to verbal recent memory following dominant-hemisphere temporal lobectomy. J Clin Exp Neuropsychol 1991;13:575– 86. Jack CR Jr, Theodore WH, Cook M, McCarthy G. MRI-based hippocampal volumetrics: data acquisition, normal ranges, and optimal protocol. Magn Res Imaging 1995;13:1057– 64. Jack CR Jr, Sharbrough FW, Twomey CK, et al. Preoperative MR imaging-based volume measurements of the hippocampal formation and anterior temporal lobe in epileptic patients. Radiology 1989;173:287. Jack CR Jr, Sharbrough FW, Cascino GD, et al. Magnetic resonance image-based hippocampal volumetry: correlation with outcome after temporal lobectomy. Ann Neurol 1992;31:138 – 46. Sawrie S, Martin R, Kuzniecky R, Faught E, Morawetz R, Jamil F, Viikinsalo M, Gilliam F. Subjective versus objective memory change after temporal lobe epilepsy surgery. Neurology 1999; 53:1511–17.

Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.